Combined CLP 4

CLP-4 Vector Calculus
CLP-4 Vector Calculus

Joel Feldman
University of British Columbia
Andrew Rechnitzer
Elyse Yeager
August 23, 2022

Cover Design: Nick Loewen — licensed under the CC-BY-NC-SA 4.0 Li-
cense.
Source files: A link to the source files for this document can be found at the
CLP textbook website. The sources are licensed under the CC-BY-NC-SA 4.0
License.
Edition: CLP4 Vector Calculus: May 2021

Website: CLP-4
©2016 – 2021 Joel Feldman, Andrew Rechnitzer, Elyse Yeager
This work is licensed under the Creative Commons Attribution-NonCommercial-
ShareAlike 4.0 International License. You can view a copy of the license here.
Preface
This text is a merger of the CLP Vector Calculus textbook and problembook.
It is, at the time that we write this, still a work in progress; some bits and
pieces around the edges still need polish. Consequently we recommend to the
student that they still consult text webpage for links to the errata — especially
if they think there might be a typo or error. We also request that you send us
an email at clp@ugrad.math.ubc.ca
Additionally, if you are not a student at UBC and using these texts please
send us an email (again using the feedback button) — we’d love to hear from
you.
Joel Feldman, Andrew Rechnitzer and Elyse Yeager
v
Feedback about the text
The CLP-4 Vector Calculus text is still undergoing testing and changes. Be-
cause of this we request that if you find a problem or error in the text then:
1. Please check the errata list that can be found at the text webpage.
2. Is the problem in the online version or the PDF version or both?

3. Note the URL of the online version and the page number in the PDF
4. Send an email to clp@ugrad.math.ubc.ca. Please be sure to include
• a description of the error
• the URL of the page, if found in the online edition
• and if the problem also exists in the PDF, then the page number in
the PDF and the compile date on the front page of PDF.
vi
Contents
Preface v
Feedback about the text vi
1 Curves 1
1.1 Derivatives, Velocity, Etc. . . . . . . . . . . . . . . . 5
1.2 Reparametrization . . . . . . . . . . . . . . . . . 20
1.3 Curvature . . . . . . . . . . . . . . . . . . . . 23
1.4 Curves in Three Dimensions . . . . . . . . . . . . . . 32
1.5 A Compendium of Curve Formula . . . . . . . . . . . . 45
1.6 Integrating Along a Curve . . . . . . . . . . . . . . 46
1.7 Sliding on a Curve . . . . . . . . . . . . . . . . . 49
1.8 Optional — Polar Coordinates . . . . . . . . . . . . . 55
1.9 Optional — Central Forces . . . . . . . . . . . . . . 62
1.10 Optional — Planetary Motion . . . . . . . . . . . . . 65
1.11 Optional — The Astroid . . . . . . . . . . . . . . . 67
1.12 Optional — Parametrizing Circles . . . . . . . . . . . . 69
2 Vector Fields 73
2.1 Definitions and First Examples . . . . . . . . . . . . . 73
2.2 Optional — Field Lines . . . . . . . . . . . . . . . 85
2.3 Conservative Vector Fields . . . . . . . . . . . . . . 92
2.4 Line Integrals . . . . . . . . . . . . . . . . . . . 103
2.5 Optional — The Pendulum . . . . . . . . . . . . . . 120
3 Surface Integrals 123

3.1 Parametrized Surfaces . . . . . . . . . . . . . . . . 123
3.2 Tangent Planes . . . . . . . . . . . . . . . . . . 135
3.3 Surface Integrals . . . . . . . . . . . . . . . . . . 143
3.4 Interpretation of Flux Integrals . . . . . . . . . . . . . 166
3.5 Orientation of Surfaces . . . . . . . . . . . . . . . . 174
4 Integral Theorems 178

4.1 Gradient, Divergence and Curl . . . . . . . . . . . . . 178
4.2 The Divergence Theorem . . . . . . . . . . . . . . . 202
vii
CONTENTS viii
4.3 Green’s Theorem . . . . . . . . . . . . . . . . . . 235

4.4 Stokes’ Theorem . . . . . . . . . . . . . . . . . . 246
4.5 Optional — Which Vector Fields Obey ∇ × F = 0 . . . . . . 274
4.6 Really Optional — More Interpretation of Div and Curl . . . . 280
4.7 Optional — A Generalized Stokes’ Theorem . . . . . . . . 289
5 True/False and Other Short Questions 299

5.1 True/False and Other Short Questions . . . . . . . . . . 299
5.2 Exercises . . . . . . . . . . . . . . . . . . . . . 299
A Appendices 312
A.1 Trigonometry . . . . . . . . . . . . . . . . . . . 312
A.2 Powers and Logarithms. . . . . . . . . . . . . . . . 315
A.3 Table of Derivatives . . . . . . . . . . . . . . . . . 317
A.4 Table of Integrals . . . . . . . . . . . . . . . . . . 318
A.5 Table of Taylor Expansions . . . . . . . . . . . . . . 319
A.6 3d Coordinate Systems . . . . . . . . . . . . . . . . 320
A.7 ISO Coordinate System Notation . . . . . . . . . . . . 324
A.8 Conic Sections and Quadric Surfaces . . . . . . . . . . . 329
A.9 Review of Linear Ordinary Differential Equations . . . . . . 332
B Hints for Exercises 339
C Answers to Exercises 359
D Solutions to Exercises 392

Chapter 1
Curves
We are now going to study vector-valued functions of one real variable. That
is, we are going to study functions that assign to each real number t (typically
in some interval) a vector1 r(t). For example

r(t) = x(t), y(t), z(t)
might be the position of a particle at time t. As t varies, r(t) sweeps out a

curve.
r(0)
r(1)
r(2)
While in some applications t will indeed be “time”, it does not have to be.
It can be simply a parameter that is used to label the different points on the
curve that r(t) sweeps out. We then say that r(t) provides a parameterization
of the curve.
Example 1.0.1 Parametrization of x2 + y 2 = a2 . While we will often
use t as the parameter in a parametrized curve r(t), there is no need to call
it t. Sometimes it is natural to use a different name for the parameter. For
example, consider the circle x2 + y 2 = a2 . It is natural
to use the angle θ in
the sketch below to label the point a cos θ , a sin θ on the circle.
y
` ˘
a cos θ , a sin θ
x2 ` y 2 “ a2
θ
x
That is,
r(θ) = a cos θ , a sin θ 0 ≤ θ < 2π
2 2 2
is a parametrization of the circle x +y = a . Just looking at the figure above,
it is clear that, as θ runs from 0 to 2π, r(θ) traces out the full circle.
1 We are going to use boldface letters, like r, to designate vectors. When writing by hand,
it is clearer to use arrows, like ~

r, instead.
1
CHAPTER 1. CURVES 2
However beware that just knowing that r(t) lies on a specified curve does
not guarantee that, as t varies, r(t) covers the entire curve. For example, as t
runs over the whole real line, π2 arctan(t) runs over the interval (−1, 1). For all
t, r !
2 4
arctan(t) , 1 − 2 arctan2 (t)

r(t) = x(t), y(t) = a
π π
is well-defined and obeys x(t)2 + y(t)2 = a2 . But this r(t) does not cover the
entire circle because y(t) is always positive.
Example 1.0.2 Parametrization of (x − h)2 + (y − k)2 = a2 . We can tweak
the parametrization of Example 1.0.1 to get a parametrization of the circle of
radius a that is centred on (h, k). One way to do so is to redraw the sketch of
Example 1.0.1 with the circle translated so that its centre is at (h, k).
y
` ˘
h ` a cos θ , k ` a sin θ
a a sin θ
θ
ph, kq
px ´ hq2 ` py ´ kq2 “ a2
x
We see from the sketch that

r(θ) = h + a cos θ , k + a sin θ 0 ≤ θ < 2π
is a parametrization of the circle (x − h)2 + (y − k)2 = a2 .

A second way to come up with this parametrization is to observe that we can
turn the trig identity cos2 t+sin2 t = 1 into the equation (x−h)2 +(y −k)2 = a2
of the circle by
• multiplying the trig identity by a2 to get (a cos t)2 + (a sin t)2 = a2 and
then
• setting a cos t = x − h and a sin t = y − k , which turns (a cos t)2 +
(a sin t)2 = a2 into (x − h)2 + (y − k)2 = a2 .

2 2
Example 1.0.3 Parametrization of x
a2 + yb2 = 1 and of x2/3 + y 2/3 = a2/3 .
2 2
We can build parametrizations of the curves xa2 + yb2 = 1 and x2/3 +y 2/3 = a2/3
from the trig identity cos2 t + sin2 t = 1, like we did in the second part of the
last example.
2
y x2
• Setting cos t = x
a and sin t = b turns cos2 t+sin2 t = 1 into a2 + yb2 = 1.
x
13 y
13
• Setting cos t = a and sin t = a turns cos2 t + sin2 t = 1 into
2/3 2/3
x y
a2/3
+ a2/3
= 1.
So

r(t) = a cos t , b sin t 0 ≤ t < 2π
3 3

r(t) = a cos t , a sin t 0 ≤ t < 2π
CHAPTER 1. CURVES 3
2 2
give parametrizations of xa2 + yb2 = 1 and x2/3 + y 2/3 = a2/3 , respectively. To
see that running t from 0 to 2π runs r(t) once around the curve, look at the
figures below.
y y
t = π/2
y2
t “ π{2
x2
a2
` b2
“1 x2/3 + y 2/3 = a2/3
t“0 t=0
t“π x t=π x
t “ 3π{2
t = 3π/2
The curve x2/3 + y 2/3 = a2/3 is called an astroid. From its equation, we
would expect its sketch to look like a deformed circle. But it is probably not so
obvious that it would have the pointy bits of the right hand figure. We will not
explain here why they arise. The astroid is studied in some detail in Example
1.1.9. In particular, the above sketch is carefully developed there.
Example 1.0.4 Parametrization of ey = 1 + x2 . A very easy method that
can often create parametrizations for a curve is to use x or y as a parameter.
Because
we can solve ey = 1 + x2 for y as a function of x, namely y = ln 1 +
2
x , we can use x as the parameter simply by setting t = x. This gives the
parametrization
r(t) = t , ln(1 + t2 )

−∞<t<∞

2 2 2
Example 1.0.5 Parametrization of x + y = a , again. It is also quite
common that one can use either x or y to parametrize part of, but not all of,
a curve. A simple example is the circle x2 + y 2 = a2 . For each −a < x < a,
there are two points on the circle with that value of x. So one cannot use x
to parametrize the whole circle. Similarly, for each −a < y < a, there are two
points on the circle with that value of y. So one cannot use y to parametrize
the whole circle. On the other hand
p
r(t) = t , a2 − t2 −a < t < a
p
r(t) = t , − a2 − t2 −a < t < a
provide parametrizations of the top half and bottom half, respectively, of the
circle using x as the parameter, and
p
r(t) = a 2 − t2 , t −a < t < a
p
2
r(t) = − a − t , t 2 −a < t < a
provide parametrizations of the right half and left half, respectively, of the
circle using y as the parameter.
Example 1.0.6 Unparametrization of r(t) = (cos t, 7−t). In this example,
we will undo the parametrization r(t) = (cos t, 7 − t) and find the Cartesian
equation of the curve in question. We may rewrite the parametrization as
x = cos t
CHAPTER 1. CURVES 4
y =7−t
Note that we can eliminate the parameter t simply by using the second equation
to solve for t as a function of y. Namely t = 7 − y. Substituting this into the
first equation gives us the Cartesian equation
x = cos(7 − y)

Curves often arise as the intersection of two surfaces. For example, the
2 2
intersection of the ellipsoid x2 + y2 + z3 = 1 with the paraboloid z = x2 + 2y 2
is the blue curve in the figure below.
z “ x2 ` 2y 2
y2 z2
x2 ` 2
` 3
“1
One way to parametrize such curves is to choose one of the three coordinates
x, y, z as the parameter, and solve the two given equations for the remaining
two coordinates, as functions of the parameter. Here are two examples.
Example 1.0.7 The set of all (x, y, z) obeying
x3 − e3y =0
2 y
x −e +z =0
is a curve. We can choose to use y as the parameter and think of
x3 = e3y
x2 + z = ey
as a system of two equations for the two unknowns x and z, with y being
treated as a given constant, rather than as an unknown. We can now solve the
first equation for x, substitute the result into the second equation, and finally
solve for z.
x3 = e3y =⇒ x = ey
x2 + z = ey =⇒ e2y + z = ey =⇒ z = ey − e2y
So
r(y) = ey , y , ey − e2y

is a parametrization for the given curve.

Example 1.0.8 The previous example was rigged so that it was easy to solve
for x and z as functions of y. In practice it is not always easy, or even possible,
CHAPTER 1. CURVES 5
to do so. A more realistic example is the set of all (x, y, z) obeying
y2 z2
x2 + + =1
2 3
x2 + 2y 2 = z
which is the blue curve in the figure above. Substituting x2 = z − 2y 2 (from

the second equation) into the first equation gives
3 z2
− y2 + z + =1
2 3
or, completing the square,
3 1 3 2 7
− y2 + z+ =
2 3 2 4
If, for example, we are interested in points (x, y, z) on the curve with y ≥ 0,
this can be solved to give y as a function of z.
r
2 3 2 14
y= z+ −
9 2 12
Then x2 = z − 2y 2 also gives x as a function of z. If x ≥ 0,

r
4 3 2 14
x= z− z+ +
9 2 6
r
4 4 2 1
= − z − z
3 9 3
The other signs of x and y can be gotten by using the appropriate square
roots. In this example, (x, y, z) is on the curve, i.e. satisfies the two original
equations, if and only if all of (±x, ±y, z) are also on the curve.
1.1 Derivatives, Velocity, Etc.

This being a Calculus text, one of our main operations is differentiation. We
are now interested in parametrizations r(t). It is very easy and natural to
extend our definition of derivative to r(t) as follows.
Definition 1.1.1 The derivative of the vector valued function r(t) is defined
to be
rptq
rpt ` hq ´ rptq
dr r(t + h) − r(t)
r0 (t) = (t) = lim
dt h→0 h
rpt ` hq

when the limit exists. In particular, if r(t) = x(t) , y(t) , z(t) , then
r0 (t) = x0 (t) , y 0 (t) , z 0 (t)

That is, to differentiate a vector valued function of t, just differentiate each of

its components. ♦
And of course differentiation interacts with arithmetic operations, like ad-
dition, in the obvious way. Only a little more thought is required to see that
CHAPTER 1. CURVES 6
differentiation interacts quite nicely with dot and cross products too. Here are
some examples.
Example 1.1.2 Let
a(t) = t2 ı̂ı + t4 ̂ + t6 k̂

b(t) = e−t ı̂ı + e−3t ̂ + e−5t k̂
γ(t) = t2
s(t) = sin t
We are about to compute some derivatives. To make it easier to follow what

is going on, we’ll use some colour. When we apply the product rule
d
f (t) g(t) = f 0 (t) g(t) + f (t) g 0 (t)

dt
we’ll use blue to highlight the factors f 0 (t) and g 0 (t). Here we go.
γ(t) b(t) = t2 e−t ı̂ı + t2 e−3t ̂ + t2 e−5t k̂
gives
d
γ(t)b(t) = 2te−t −t2 e−t ı̂ı + 2te−3t −3t2 e−3t ̂ + 2te−5t −5t2 e−5t k̂

dt
= 2t e−t ı̂ı + e−3t ̂ + e−5t k̂ + t2 − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂

= γ 0 (t)b(t) + γ(t)b0 (t)
and
a(t) · b(t) = t2 e−t + t4 e−3t + t6 e−5t
gives
d
a(t) · b(t) = 2te−t −t2 e−t + 4t3 e−3t −3t4 e−3t + 6t5 e−5t −5t6 e−5t

dt
= 2te−t + 4t3 e−3t + 6t5 e−5t + −t2 e−t −3t4 e−3t −5t6 e−5t

= 2tı̂ı + 4t3 ̂ + 6t5 k̂ · e−t ı̂ı + e−3t ̂ + e−5t k̂

+ t2 ı̂ı + t4 ̂ + t6 k̂ · − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂

= a0 (t) · b(t) + a(t) · b0 (t)
and
 
ı̂ı ̂ k̂
a(t) × b(t) = det  t2 t4 t6 
e−t e−3t e−5t
= ı̂ı t4 e−5t − t6 e−3t ) − ̂(t2 e−5t − t6 e−t ) + k̂(t2 e−3t − t4 e−t )
gives
d
a(t) × b(t)
dt
= ı̂ı 4t3 e−5t − 6t5 e−3t ) − ̂( 2te−5t − 6t5 e−t ) + k̂( 2te−3t − 4t3 e−t )
+ ı̂ı −5t4 e−5t +3t6 e−3t ) − ̂(−5t2 e−5t +t6 e−t ) + k̂(−3t2 e−3t +t4 e−t )
= 2tı̂ı + 4t3 ̂ + 6t5 k̂ × e−t ı̂ı + e−3t ̂ + e−5t k̂

CHAPTER 1. CURVES 7
+ t2 ı̂ı + t4 ̂ + t6 k̂ × − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂

= a0 (t) × b(t) + a(t) × b0 (t)
and
a s(t) = (sin t)2 ı̂ı + (sin t)4 ̂ + (sin t)6 k̂

d
a s(t) = 2(sin t) cos tı̂ı + 4(sin t)3 cos t ̂ + 6(sin t)5 cos t k̂

=⇒
dt
= 2(sin t)ı̂ı + 4(sin t)3̂ + 6(sin t)5 k̂ cos t

= a0 s(t) s0 (t)

Of course these examples extend to general (differentiable) a(t), b(t), γ(t)
and s(t) and give us (most of) the following theorem.
Theorem 1.1.3 Arithmetic of differentiation. Let
• a(t), b(t) be vector valued differentiable functions of t ∈ R that take values
in Rn and
• α, β ∈ R be constants and
• γ(t) and s(t) be real valued differentiable functions of t ∈ R

Then
d
(a) α a(t) + β b(t) = α a0 (t) + β b0 (t) (linear combination)

dt
d
(b) γ(t)b(t) = γ 0 (t)b(t) + γ(t)b0 (t) (multiplication by scalar function)

dt
d
(c) a(t) · b(t) = a0 (t) · b(t) + a(t) · b0 (t) (dot product)

dt
d
(d) a(t) × b(t) = a0 (t) × b(t) + a(t) × b0 (t) (cross product)

dt
d
(e) a s(t) = a0 s(t) s0 (t) (composition)

dt
Let’s think about the geometric significance of r0 (t). In particular, let’s
think about the relationship between r0 (t) and distances along the curve. The
derivative r0 (t) is the limit of r(t+h)−r(t)
h as h → 0. The numerator, r(t + h) −
r(t), is the vector with head at r(t + h) and tail at r(t).
r(t)
r(t + h) − r(t) ≈ r′ (t) h
r(t + h)
When h is very small this vector
• has the essentially the same direction as the tangent vector to the curve
at r(t) and
• has length being essentially the length of the part of the curve between
r(t) and r(t + h).
Taking the limit as h → 0 yields that
CHAPTER 1. CURVES 8
• r0 (t) is a tangent vector to the curve at r(t) that points in the direction
of increasing t and
• if s(t) is the length
of the part of the curve between r(0) and r(t), then
ds dr (t).
dt (t) = dt
This is worth stating formally.
Lemma 1.1.4 Let r(t) be a parametrized curve.
a Denote by T̂(t) the unit tangent vector to the curve at r(t) pointing in
the direction of increasing t. If r0 (t) 6= 0 then
r0 (t)
T̂(t) =
|r0 (t)|
b Denote by s(t) the length of the part of the curve between r(0) and
r(t).Then
ds sptq
dr
(t) = (t)
rp0q
dt dt
Z T
dr
s(T ) − s(T0 ) = (t) dt
rptq
T0 dt T̂ptq
c In particular, if the parameter happens to be arc length, i.e. if t = s, so
that ds
ds
= 1, then
dr
(s) = 1
ds T̂(s) = r0 (s)
As an application, we have the

Lemma 1.1.5 If r(t) = x(t) , y(t) , z(t) is the position of a particle at time

t, then
position at time t = r(t) = x(t) , y(t) , z(t)

ds
velocity at time t = v(t) = r0 (t) = x0 (t) , y 0 (t) , z 0 (t) = (t) T̂(t)
dt
ds
speed at time t =
p
(t) = |v(t)| = |r0 (t)| = x0 (t)2 + y 0 (t)2 + z 0 (t)2
dt
acceleration at time t = a(t) = r00 (t) = v0 (t) = x00 (t) , y 00 (t) , z 00 (t)

and the distance travelled between times T0 and T is

Z T Z Tp
dr

s(T ) − s(T0 ) = (t) dt = x0 (t)2 + y 0 (t)2 + z 0 (t)2 dt
T0 dt T0
Note that the velocity v(t) = r0 (t) is a vector quantity while the speed
ds 0
dt (t) = |r (t)| is a scalar quantity.
Example 1.1.6 Circumference of a circle. In general it can be quite

difficult to compute arc lengths. So, as an easy warmup example, we will
compute the circumference of the circle x2 + y 2 = a2 . We’ll also find a unit
tangent to the circle at any point on the circle. We’ll use the parametrization

r(θ) = a cos θ , a sin θ 0 ≤ θ ≤ 2π
of Example 1.0.1. Using Lemma 1.1.4, but with the parameter t renamed to θ
r0 (θ) = a − sin θ , cos θ

CHAPTER 1. CURVES 9
r0 (θ)
T̂(θ) = 0
= − sin θ , cos θ
|r (θ)|
ds 0
(θ) = r (θ) = a
dθ
Z Θ
0
s(Θ) − s(0) = r (θ) dθ = aΘ
0
As1 s(Θ) is the arc length of the part of the circle with 0 ≤ θ ≤ Θ, the
circumference of the whole circle is
s(2π) = 2πa
which is reassuring, since this formula has been known2 for thousands of years.
y T̂pθq
` ˘
a cos θ , a sin θ
x2 ` y 2 “ a2
θ
x
The formula s(Θ) − s(0) = aΘ also makes sense — the part of the circle
Θ
with 0 ≤ θ ≤ Θ is the fraction 2π of the whole circle, and so should have length
Θ
2π × 2πa. Also note that

r(θ) · T̂(θ) = a cos θ , a sin θ · − sin θ , cos θ = 0
so that the tangent to the circle at any point is perpendicular to the radius
vector of the circle at that point. This is another geometric fact that has been
known3 for thousands of years.
Example 1.1.7 Arc length of a helix. Consider the curve
r(t) = 6 sin(2t)ı̂ı + 6 cos(2t)̂ + 5tk̂
where the standard basis vectors ı̂ı = (1, 0, 0), ̂ = (0, 1, 0) and k̂ = (0, 0, 1).
We’ll first sketch it, by observing that
• x(t) = 6 sin(2t) and y(t) = 6 cos(2t) obey x(t)2 + y(t)2 = 36 sin2 (2t) +
36 cos2 (2t) = 36. So all points of the curve lie on the cylinder x2 +y 2 = 36
and

• as t increases, x(t), y(t) runs clockwise around the circle x2 + y 2 = 36
and at the same time z(t) = 5t just increases linearly.
Our curve is the helix

1 You might guess that Θ is a capital Greek theta. You’d be right.
2 The earliest known written approximations of π, in Egypt and Babylon, date from
1900--1600BC. The first recorded algorithm for rigorously evaluating π was developed by
Archimedes around 250 BC. The first use of the symbol π, for the ratio between the circum-
ference of a circle and its diameter, in print was in 1706 by William Jones.
3 It is Proposition 18 in Book 3 of Euclid’s Elements. It was published around 300BC.
CHAPTER 1. CURVES 10
t“π
π
t“ 2
y
t“0
x
We have marked three points of the curve on the above sketch. The first
has t = 0 and is 0ı̂ı + 6̂ + 0k̂. The second has t = π2 and is 0ı̂ı − 6̂ + 5π
2 k̂, and
the third has t = π and is 0ı̂ı + 6̂ + 5π k̂. We’ll now use Lemma 1.1.4 to find
a unit tangent T̂(t) to the curve at r(t) and also the arclength of the part of
curve between t = 0 and t = π.
r(t) = 6 sin(2t)ı̂ı + 6 cos(2t)̂ + 5tk̂

r0 (t) = 12 cos(2t)ı̂ı − 12 sin(2t)̂ + 5k̂
ds q p
(t) = r0 (t) = 122 cos2 (2t) + 122 sin2 (2t) + 52 = 122 + 52

dt
= 13
r0 (t) 12 12 5
T̂(t) = = cos(2t)ı̂ı − sin(2t)̂ + k̂
|r0 (t))| 13 13 13
Z π
0
s(π) − s(0) = r (t) dt = 13π
0

Example 1.1.8 Velocity and acceleration. Imagine that, at time t, a
particle is at

t t
r(t) = h + a cos 2π ı̂ı + k + a sin 2π ̂
T T
As |r(t) − hı̂ı − k ̂| = a, the particle is running around the circle of radius a
t t
centred on (h, k). When t increases by T , the argument, 2π T , of cos 2π T and
t

sin 2π T increases by exactly 2π and the particle runs exactly once around
the circle. In particular, it travels a distance 2πa. So it is moving at speed
2πa
T . According to Lemma 1.1.5, it has

0 2πa t 2πa t
velocity = r (t) = − sin 2π ı
ı̂ + cos 2π ̂
T T T T
ds 2πa
speed = (t) = |r0 (t)| =
dt T
2
4π 2 a

00 4π a t t
acceleration = r (t) = − 2 cos 2π ı̂ı − sin 2π ̂
T T T2 T
4π 2
= − 2 r(t) − hı̂ı − k ̂
T
Here are some observations.
• The velocity r0 (t) has dot product zero with r(t) − hı̂ı − k ̂, which is the
radius vector from the centre of the circle to the particle. So the velocity
is perpendicular to the radius vector, and hence parallel to the tangent

vector of the circle at r(t).
• The speed given by Lemma 1.1.5 is exactly the speed we found above,
just before we started applying Lemma 1.1.5.
• The acceleration r00 (t) points in the direction opposite to the radius vec-
tor.
y r1 ptq
rptq
r2 ptq
ph, kq
x

Example 1.1.9 Perimeter of the astroid. In this example, we find the
perimeter of the astroid4
x2/3 + y 2/3 = a2/3
A geometric construction of this curve, as well as a derivation of its equation
is given in the optional section 1.11 later. We’ll start by finding a convenient
parametrization.
• To do so, notice that x2/3 + y 2/3 = a2/3 looks somewhat like the equation
of the circle x2 + y 2 = a2 .
• The standard parametrization of the circle, namely x = a cos t, y = a sin t
works because of the elementary trig identity cos2 t + sin2 t = 1.
• If we can arrange that x(t)2/3 = a2/3 cos2 t and y(t)2/3 = a2/3 sin2 t, then
the same elementary trig identity will give x(t)2/3 + y(t)2/3 = a2/3 , as
desired.
• But of course its easy to arrange that: just solve x(t)2/3 = a2/3 cos2 t
for x(t), namely x(t) = a cos3 t, and solve y(t)2/3 = a2/3 sin2 t for y(t),
namely y(t) = a sin3 t.
Our parametrization is
r(t) = a cos3 tı̂ı + a sin3 t ̂
By Lemma 1.1.4
r(t) = a cos3 tı̂ı + a sin3 t ̂
r0 (t) = −3a sin t cos2 tı̂ı + 3a sin2 t cos t ̂
ds p
(t) = r0 (t) = 9a2 sin2 t cos4 t + 9a2 sin4 t cos2 t

dt q
= 3a sin2 t cos2 t(cos2 t + sin2 t)

= 3a sin t cos t
r0 (t) sin t cos t
T̂(t) = = − cos tı̂ı + sin t ̂
|r0 (t))| | sin t cos t|

= sgn sin t cos t − cos tı̂ı + sin t ̂

Here sgn sin t cos t means “the sign of sin t cos t”, i.e +1 when sin t cos t > 0
and −1 when sin t cos t < 0. So
T̂(t)
( )
1 if sin t > 0, cos t > 0 or sin t < 0, cos t < 0
= − cos tı̂ı + sin t ̂
−1 if sin t > 0, cos t < 0 or sin t < 0, cos t > 0
( )
1 if 0 < t < π2 or π < t < 3π2

= π 3π
− cos tı̂ı + sin t ̂
−1 if 2 < t < π or 2 < t < 2π
Before we go on to sketch the astroid and compute its perimeter, we can

make a few observations that will simplify our lives.
• The signs of both components of r(t) are the same as the signs of the
components of cos tı̂ı +sin t ̂; and the signs of both components of r0 (t) are
the same as the signs of the components of − sin tı̂ı +cos t ̂. Consequently
the astroid looks somewhat like a circle in that
◦ when 0 ≤ t ≤ π2 , r(t) lies in the first quadrant and moves upward
and to the left as t increases and
◦ when π2 ≤ t ≤ π, r(t) lies in the second quadrant and moves down-
ward and to the left as t increases and
◦ when π ≤ t ≤ 3π2 , r(t) lies in the third quadrant and moves down-
ward and to the right as t increases and
◦ when 3π
2 ≤ t ≤ 2π, r(t) lies in the fourth quadrant and moves
upward and to the right as t increases and
◦ r(2π) = r(0) so that the astroid is a closed curve that circumnavi-
gates the origin exactly once as t runs from 0 to 2π.
• Something weird happens at those values of t where sin t cos t changes
sign5 , i.e. at t = 0, π2 , π, 3π
2 , etc. Namely T̂ (t) flips. To be precise

lim T̂ (t) = lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = ı̂ı
t→0− t→0− t→0−

lim T̂ (t) = lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = −ı̂ı
t→0+ t→0+ t→0+
and

lim T̂ (t) =lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = ̂
t→π/2− t→π/2− t→π/2−

lim T̂ (t) = lim sgn sin t cos t lim − cos tı̂ı + sin t ̂ = −̂
t→π/2+ t→π/2+ t→π/2+
and so on. This signals cusps in the curve at t = 0, i.e. at r(0) = aı̂ı,
and at t = π2 , i.e. at r( π2 ) = a̂, and so on. So while the astroid looks
somewhat like a circle, it has cusps at ±aı̂ı and ±a̂. Here is the sketch.
y
t = π/2
x2/3 + y 2/3 = a2/3
t=0
t=π x
t = 3π/2
• The astroid is invariant under reflections in the x-axis and in the y-axis.
That is, x2/3 + y 2/3 = a2/3 is invariant under x → −x and also under
y → −y. So to find the whole perimeter, it suffices to find the arc length
of the part of the astroid in the first quadrant, and then multiply by 4.
Z π/2 Z π/2
ds
perimeter = 4 dt = 4 3a sin t cos t dt
0 dt 0
Z π/2 h cos(2t) iπ/2
= 6a sin(2t) dt = 6a −
0 2 0
= 6a

0
Example 1.1.10 r (t) = 0. In the last example, we found that the astroid had
cusps at those points r(t) where the velocity r0 (t) vanished. In this example,
we will explore a little further what can happen when r0 (t) = 0.
Suppose that you are out for a walk and that your position at time t is r(t).
If at some time you have nonzero velocity, it is very hard for you to change your
direction of motion discontinuously6 . On the other hand, when r0 (t) = 0, you
are not moving at all and it is easy for you to turn and leave in any direction
you choose. You could reverse direction completely, or make a sharp left turn,
or not change direction at all. Here are examples of all of these. They all have
r0 (t) = 0. They are sketched below.
r1 (t) = (t5 , t2 ) r01 (t) = (5t4 , 2t)

( ) ( )
(t2 , 0) if t ≥ 0 0 (2t, 0) if t ≥ 0
r2 (t) = r2 (t) =
(0, t2 ) if t ≤ 0 (0, 2t) if t ≤ 0
3
r3 (t) = (t , 0) r03 (t) = (3t2 , 0)
r1 ptq r2 ptq
r3 ptq
t“0 t“0 t“0

4 Astroid should not be confused with asteroid, though both words derive from the Greek
word for star.
5 Like a cross-walk sign.
6 For your velocity to jump discontinuously, your acceleration has to be infinite, which
requires an infinite force. You might not look so healthy afterwards

Example 1.1.11 Corkscrew. We’ll find the arc length of

√
r(t) = t cos tı̂ı + t sin t ̂ + t k̂ 0≤t≤ 2
We’ll first sketch it, by observing that

• x(t) = t cos t, y(t) = t sin t and z(t) = t obey x(t)2 + y(t)2 = t2 = z(t)2 .
So all points of the curve lie on the cone x2 + y 2 = z 2 and

• as t increases, x(t), y(t) runs counterclockwise around a “circle” whose
radius increases linearly with t and at the same time z(t) also increases
linearly.
Our curve is the “corkscrew”
z
y “ z, x “ 0
x
By Lemma 1.1.4
r(t) = t cos tı̂ı + t sin t ̂ + t k̂

r0 (t) = [cos t − t sin t]ı̂ı + [sin t + t cos t]̂ + k̂
so that
ds
(t) = r0 (t)

dt q
cos2 t−2t sin t cos t+t2 sin2 t + sin2 t+2t sin t cos t+t2 cos2 t +1

=
p
= 2 + t2
Our goal, stated at the beginning of this example, was to compute

√ √
√ Z 2 0
Z 2 p
s( 2) − s(0) = r (t) dt = 2 + t2 dt
0 0
To evaluate the integral, we’ll use three techniques that√ you learned in your
first integral calculus course. First, motivated by the 2 + t2 , we’ll use the
trigonometric substitution
√ √
dt = 2 sec2 u du 2 + t2 = 2 1 + tan2 u = 2 sec2 u

t = 2 tan u
√ π
When t = 0, u = 0 and when t = 2, tan u = 1 so that u = 4 and
√ Z π/4 √ √ Z π/4
2
s( 2) − s(0) = 2 sec2 u 2 sec u du = 2 sec3 u du
0 0
You may have evaluated this integral in first year. There are several ways of
doing so. Perhaps the most straight forward, but also most tedious, method is
to rewrite the integral as
√ Z π/4
cos u
s( 2) − s(0) = 2 du
0 cos4 u
We recognize that this is a trigonometric integral that contains an odd power

of cos u, so we substitute w = sin u, dw = cos u du, cos2 u = 1 − w2 . When
u = 0, w = 0 and when u = π4 , w = √12 so that
√
√ Z 1/ 2
dw
s( 2) − s(0) = 2 2
0 (1 − w2 )
The integrand is now a rational function, i.e. a ratio of polynomials. So we

apply partial fractions.
√
√ Z 1/ 2
dw
s( 2) − s(0) = 2 2
0 [(1 − w)(1 + w)]
Z 1/ 2 h√
1 1 1 i2
= + dw
2 0 1−w 1+w
√
1 1/ 2 h
Z
1 2 1 i
= + + dw
2 0 (1 − w)2 (1 − w)(1 + w) (1 + w)2
√
1 1/ 2 h
Z
1 1 1 1 i
= 2
+ + + 2
dw
2 0 (1 − w) 1 − w 1 + w (1 + w)
√
1h 1 1 i1/ 2
= − ln |1 − w| + ln |1 + w| −
2 1−w 1+w 0
√ √
1 h 2w 1+w i 1/ 2 1 h √ 2 + 1i
= 2
+ ln = 2 2 + ln √
2 1−w 1−w 0 2 2−1
≈ 2.2956
Ooof!
1.1.1 Exercises
Exercises — Stage 1
Questions 1.1.1.1 through 1.1.1.5 provide practice with curve parametriza-
tion. Being comfortable with the algebra and interpretation of these descrip-
tions are essential ingredients in working effectively with parametrizations.
1. Find the specified parametrization of the first quadrant part of the
circle x2 + y 2 = a2 .
a In terms of the y coordinate.
b In terms of the angle between the tangent line and the positive
x-axis.
c In terms of the arc length from (0, a).
2. Consider the following time-parametrized curve:
π
r(t) = cos t , (t − 5)2
4
√
List the three points (−1/ 2, 0), (1, 25), and (0, 25) in chronolog-
ical order.
3. At what points in the xy-plane does the curve (sin t, t2 ) cross itself?
What is the difference in t between the first time the curve crosses
through a point, and the last?
4.
y
x
a
A circle of radius a rolls along the x-axis in the positive direction,

starting with its centre at (a, a). In that position, we mark the top-
most point on the circle P . As the circle moves, P moves with it. Let
θ be the angle the circle has rolled — see the diagram below.
a Give the position of the centre of the circle as a function of θ.
b Give the position of P as a function of θ.
P
θ
5. The curve C is defined to be the intersection of the hyperboloid

1
x2 − y 2 + 3z 2 = 1
4
and the plane
x + y + z = 0.
When y is very close to 0, and z is negative, find an expression
giving z in terms of y.
6. A particle traces out a curve in space, so that its position at time t is
1
r(t) = e−t ı̂ı + ̂ + (t − 1)2 (t − 3)2 k̂
t
for t > 0.
Let the positive z axis point vertically upwards, as usual. When is
the particle moving upwards, and when is it moving downwards? Is
it moving faster at time t = 1 or at time t = 3?
7. Below is the graph of the parametrized function r(t). Let s(t) be the
arclength along the curve from r(0) to r(t).
r(t + h)
r(t)
r(0)
Indicate on the graph s(t + h) − s(t) and r(t + h) − r(t). Are the
quantities scalars or vectors?
8. What is the relationship between velocity and speed in a vector-valued
function of time?
9. ∗. Let r(t) be a vector valued function. Let r0 , r00 , and r000 denote
dr d2 r d3 r
dt , dt2 and dt3 , respectively. Express
d
(r × r0 ) · r00

dt
in terms of r, r0 , r00 , and r000 . Select the correct answer.
a (r0 × r00 ) · r000
b (r0 × r00 ) · r + (r × r0 ) · r000
c (r × r0 ) · r000
d 0
e None of the above.
10. Show that, if the position and velocity vectors of a moving particle are
always perpendicular, then the path of the particle lies on a sphere.
11. ∗. Find the speed of a particle with the given position function
√
r(t) = 5 2 tı̂ı + e5t ̂ − e−5t k̂
Select the correct answer:

a |v(t)| = e5t + e−5t

√
b |v(t)| = 10 + 5et + 5e−t
p
c |v(t)| = 10 + e10t + e−10t
d |v(t)| = 5 e5t + e−5t

e |v(t)| = 5 et + e−t

12. Find the velocity, speed and acceleration at time t of the particle
whose position is
r(t) = a cos tı̂ı + a sin t ̂ + ct k̂
Describe the path of the particle.

13. ∗.
a Let
r(t) = t2 , 3, 13 t3

Find the unit tangent vector to this parametrized curve at t = 1,

pointing in the direction of increasing t.
b Find the arc length of the curve from (a) between the points
(0, 3, 0) and (1, 3, − 31 ).
q
14. Using Lemma 1.1.4, find the arclength of r(t) = t, 32 t2 , t3 from
t = 0 to t = 1.
15. Find the length of the parametric curve
x = a cos t sin t y = a sin2 t z = bt
between t = 0 and t = T > 0.

16. A particle’s position at time t is given by r(t) = (t + sin t, cos t) 7 .
What is the magnitude of the acceleration of the particle at time t?
3

17. ∗. A curve in R3 is given by the vector equation r(t) = 2t cos t, 2t sin t, t3
a Find the length of the curve between t = 0 and t = 2.
b Find the parametric equations of the tangent line to the curve
at t = π.

18. ∗. Let r(t) = 3 cos t, 3 sin t, 4t be the position vector of a particle as
a function of time t ≥ 0.
a Find the velocity of the particle as a function of time t.
b Find the arclength of its path between t = 1 and t = 2.
19. The plane z = 2x + 3y intersects the cylinder x2 + y 2 = 9 in an
ellipse.
a Find a parametrization of the ellipse.
b Express the circumference of this ellipse as an integral. You
need not evaluate the integral8 .
20. ∗. Consider the curve
1 1
r(t) = cos3 tı̂ı + sin3 t ̂ + sin3 t k̂
3 3
π
a Compute the arc length of the curve from t = 0 to t = 2.
b Compute the arc length of the curve from t = 0 to t = π.

21. ∗. Let r(t) = 31 t3 , 12 t2 , 12 t , t ≥ 0. Compute s(t), the arclength of the

curve at time t.

22. ∗. Find the arc length of the curve r(t) = tm , tm , t3m/2 for 0 ≤
a ≤ t ≤ b, and where m > 0. Express your result in terms of m, a,
and b.
7 The particle traces out a cycloid — see Question 1.1.1.4
8 The indefinite integral involved is one of a class of integrals called elliptic integrals
because of their connections to arc lengths of ellipses. In general, elliptic integrals cannot
be expressed in terms of elementary functions. You can easily find discussions of elliptic
integrals using your favourite search engine.
23. Let C be the part of the curve of intersection of the parabolic cylinder
x = y 2 and the hyperbolic paraboloid 3z = 2xy with y ≥ 0.
a Write a vector parametric equation for C using x as the param-
eter.
b Find the length of the part of C between the origin and the
point (9, 3, 18).
c A particle moves along C in the direction for which x is increas-
ing. If the particle moves with constant speed 9, find its velocity
vector when it is at the point (1, 1, 23 ).
d Find the acceleration vector of the particle of part (c) when it
is at the point (1, 1, 32 ).
24. If a particle has constant mass m, position r, and is moving with
velocity v, then its angular momentum is L = m(r × v).
aparticle with mass m = 1 and position function r = (sin t, cos t, t),
For
find dL
dt .

25. ∗. A particle moves along the curve C of intersection of the surfaces
z2 = 12y and 18x = yz in the upward direction. When the particle is
at (1, 3, 6) its velocity v and acceleration a are given by
v = 6ı̂ı + 12 ̂ + 12 k̂ a = 27ı̂ı + 30 ̂ + 6 k̂
z
a Write a vector parametric equation for C using u = 6 as a pa-
rameter.
b Find the length of C from (0, 0, 0) to (1, 3, 6).
c If u = u(t) is the parameter value for the particle’s position at
time t, find du
dt when the particle is at (1, 3, 6).
d2 u
d Find dt2 when the particle is at (1, 3, 6).
1
26. ∗. A particle of mass m = 1 has position r0 = 2 k̂ and velocity
2
v0 = π2 ı̂ı at time 0. It moves under a force
F(t) = −3tı̂ı + sin t ̂ + 2e2t k̂.
a Determine the position r(t) of the particle depending on t.

b At what time after time t = 0 does the particle cross the plane
x = 0 for the first time?
c What is the velocity of the particle when it crosses the plane
x = 0 in part (b)?
27. ∗. Let C be the curve of intersection of the surfaces y = x2 and
z = 23 x3 . A particle moves along C with constant speed such that
dx
dt > 0. The particle is at (0, 0, 0) at time t = 0 and is at (3, 9, 18) at
time t = 27 .
a Find the length of the part of C between (0, 0, 0) and (3, 9, 18).
b Find the constant speed of the particle.
c Find the velocity of the particle when it is at 1, 1, 23 .

d Find the acceleration of the particle when it is at 1, 1, 23 .

28. A camera mounted to a pole can swivel around in a full circle. It is

tracking an object whose position at time t seconds is x(t) metres east
of the pole, and y(t) metres north of the pole.
In order to always be pointing directly at the object, how fast
should the camera be programmed to rotate at time t? (Give your
answer in terms of x(t) and y(t) and their derivatives, in the units
rad/sec.)
√
29. A pipe of radius 3 follows the path of the curve r(t) = ( 2 3 2 t3/2 , 12 t2 , t+
2), for 0 ≤ t ≤ 10.
What is the volume inside the pipe? What is the surface area of
the pipe?
30. A wire of total length 1000cm is formed into a flexible coil that is a
circular helix. If there are 10 turns to each centimetre of height and
the radius of the helix is 3 cm, how tall is the coil?
31. A projectile falling under the influence of gravity and slowed by air
resistance proportional to its speed has position satisfying
d2 r dr
2
= −g k̂ − α
dt dt
dr
where α is a positive constant. If r = r0 and dt = v0 at time t = 0,
find r(t).
1.2 Reparametrization
There are invariably many ways to parametrize a given curve. Kind of trivially,
one can always replace t by, for example, 3u. But there are also more substan-
tial ways to reparametrize curves. It often pays to tailor the parametrization
used to the application of interest. For example, we shall see in the next couple
of sections that many curve formulae simplify a lot when arc length is used as
the parameter.
Example 1.2.1 Here are three different parametrizations of the semi-circle
x2 + y 2 = r2 , y ≥ 0.
• The first uses the polar angle θ as the parameter. We have already seen,
in Example 1.0.1, the parametrization
y
` ˘
r cos θ , r sin θ
x2 ` y 2 “ r 2

r1 (θ) = r cos θ , r sin θ
θ 0≤θ≤π
x
• The second uses x as the parameter.
√ Just solving x2 +y 2 = r2 , y ≥ 0 for y
as a function of x, gives y(x) = r2 − x2 and so gives the parametrization
y
` ? ˘
x, r 2 ´ x2
x2 ` y 2 “ r 2
p
x

r2 (x) = x , r 2 − x2
−r ≤x≤r
x
• The third uses arc length from (r, 0) as the parameter. We have seen, in
Example 1.1.6, that the arc length from (r, 0) to r1 (θ) is just s = rθ. So
the point on the semicircle that is arc length s away from (r, 0) is
y
` ˘ s
r cos sr , r sin sr r3 (s) = r1
x2 ` y 2 “ r 2 r
s s
s = r cos , r sin
r r
x 0 ≤ s ≤ πr

We shall see that, for some purposes, it is convenient to use parametrization
by arc length. Here is a messier example in which we reparametrize a curve so
as to use the arc length as the parameter.
Example 1.2.2 We saw in Example 1.1.9, that, as t runs from 0 to π2 , r(t) =
a cos3 tı̂ı + a sin3 t ̂ runs from (a, 0) to (0, a) along the astroid x2/3 + y 2/3 =
a2/3 . Suppose that we want a new parametrization R(s) chosen so that, as s
runs from 0 to some appropriate value, R(s) runs from (a, 0) to (0, a) along
x2/3 + y 2/3 = a2/3 , with s being the arc length from (a, 0) to R(s) along
x2/3 + y 2/3 = a2/3 .
y
Rpsq
rptq
s
x
π ds 3a
We saw, in Example 1.1.9, that, for 0 ≤ t ≤ 2 , dt = 2 sin(2t) so that the
arclength from (a, 0) = r(0) to r(t) is
Z t
3a 3a
sin(2t0 ) dt0 =

s(t) = 1 − cos(2t)
0 2 4
π
which runs from 0, at t = 0, to 3a2 , at t = 2 . This is relatively clean and we can
invert s(t) to find t as a function of s. The value, T (s), of t that corresponds
to any given 0 ≤ s ≤ 3a 2 is determined by
3a 1 4s
s= 1 − cos 2T (s) ⇐⇒ T (s) = arccos 1 −
4 2 3a
and
R(s) = r T (s) = a cos3 T (s) ı̂ı + a sin3 T (s) ̂

We can simplify cos3 T (s) and sin3 T (s) by just using trig identities to con-

3a

vert the cos 2T (s) in s = 4 1−cos 2T (s) into cos T (s) ’s and sin T (s) ’s.
3a 3a
1 − 2 cos2 T (s) − 1

s= 1 − cos 2T (s) =
4 4
2s
⇐⇒ cos2 T (s) = 1 −

3a
3a 3a
1 − 1 − 2 sin2 T (s)

s= 1 − cos 2T (s) =
4 4
2s
⇐⇒ sin2 T (s) =
3a
Consequently the desired parametrization is
3/2 3/2
2s 2s 3a
R(s) = a 1 − ı̂ı + a ̂ 0≤s≤
3a 3a 2
which is remarkably simple.
1.2.1 Exercises
Z t
1. A curve r(s) is parametrized in terms of arclength. What is |r0 (s)| ds
1
when t ≥ 1?
2. The function
√
s+1 s+1 3
r(s) = sin ı̂ı + cos ̂ + (s + 1)k̂
2 2 2
is parametrized in terms of arclength, starting from the point P .

What is P ?
3. A curve R = a(t) is reparametrized in terms of arclength as R =
b(s) = a(t(s)). Of the following options, which best describes the
relationship between the vectors a0 (t0 ) and b0 (s0 ), where t(s0 ) = t0 ?
You may assume a0 (t) and b0 (s) exist and are nonzero for all t, s ≥
0.
A they are parallel and point in the same direction
B they are parallel and point in opposite directions

C they are perpendicular
D they have the same magnitude
E they are equal
4. ∗.
a Let
r(t) = (2 sin3 t, 2 cos3 t, 3 sin t cos t)
Find the unit tangent vector to this parametrized curve at t =
π/3, pointing in the direction of increasing t.
1 Since we specified the derivatives are nonzero, there’s no messiness about vectors being
parallel to a zero vector.

b Reparametrize the vector function r(t) from (a) with respect to

arc length measured from the point t = 0 in the direction of
increasing t.
5. ∗. This problem is about the logarithmic spiral in the plane
r(t) = et (cos t, sin t), t∈R
a Find the arc length of the piece of this spiral which is contained
in the unit circle.
b Reparametrize the logarithmic spiral with respect to arc length,
measured from t = −∞.
6. Define
1 arctan t
r(t) = √ ,√ , arctan t
1 + t2 1 + t−2
for 0 ≤ t. Reparametrize the function using z = arctan t, and
describe the curve it defines. What is the geometric interpretation of
the new parameter z?
7. Reparametrize the function r(t) = ( 12 t2 , 13 t3 ) in terms of arclength
from t = −1.
1.3 Curvature
So far, when we have wanted to approximate a complicated curve by a simple
curve near some point, we drew the tangent line to the curve at the point.
That’s pretty crude. In particular tangent lines are straight — they don’t
curve. We will get a much better idea of what the complicated curve looks like
if we approximate it, locally, by a very simple “curvy curve” rather than by a
straight line. Probably the simplest “curvy curve” is a circle1 and that’s what
we’ll use.
Definition 1.3.1
a The circle which best approximates a given curve near a given point is
called the circle of curvature or the osculating circle2 at the point.
b The radius of the circle of curvature is called the radius of curvature
at the point and is normally denoted ρ.
c The curvature at the point is κ = ρ1 .
d The centre of the circle of curvature is called centre of curvature at
the point.
♦
These definitions are illustrated in the figure below. It shows (part of) the
osculating circle at the point P . The point C is the centre of curvature.
1 Circles are good for studying “curvature”, because, unlike parabolas for example, the
rate at which a circle curves is uniform over the entire circle.

2 “Osculare” is the Latin verb “to kiss”. The German mathematician Gottfried Wilhelm
(von) Leibniz (1646--1716) named the circle the “circulus osculans”.

C
Note that when the curvature κ is large, the radius of curvature ρ is small
and we have a very curvy curve. On the other hand when the curvature κ is
small, the radius of curvature ρ is large and our curve is almost straight. In
particular, straight lines have curvature exactly zero.
We are now going to determine how to find the circle of curvature, starting
by figuring out what its radius should be. We’ll first look at curves3 that lie
in the xy-plane and then move on to curves in 3d. Consider the black curve in
the figure below.
ds
ρ dθ
That figure also contains a (portion of a) red circle that fits the curve really
well between the two radial lines that are (a very small) angle dθ apart. So the
arclength ds of the part of the black curve between the two radial lines, should
be (essentially) the same as the arc length of the circle between the two radial
lines, which is ρ |dθ|, where ρ is the radius of the circle. (We put in absolute
values to take into account the possibility that dθ could be negative.) Thus
ds = ρ |dθ|. When dθ is a macroscopic angle, this is of course an approximation.
But in the limit as dθ → 0, we should end up with

ds
ρ =
dθ
We now have a formula for the radius of curvature, but not in a very convenient
form, because to evaluate it we would need to know the arc length along the
curve as a function of the angle θ in the rightmost figure below. We’ll now
spend some time developing more convenient formulae for ρ. First consider the
three figures below. They all show the same curve as in the last figure. The
leftmost figure just shows
• the curve of interest, which is the black curve, and
• the (blue) point of interest on the black curve. We want to find the
curvature at that point.
The middle figure shows the same curve and point of interest and also shows
• the red circle of curvature (i.e. best fitting circle) for the black curve at
the blue dot.
• The red dot is the centre of curvature.
3 We’llalso assume that the curves of interest are smooth, with no cusps for example, and
not straight, so that the radius of curvature 0 < ρ < ∞.
The rightmost figure shows the same black curve, blue point of interest and
red circle of curvature (at least part of it) somewhat enlarged.
• The angle θ is the angle between ı̂ı and the radius vector from the red
dot (the centre of curvature) to the blue dot (the point of interest).
• T̂ is the tangent vector to the black curve at the blue dot.
• The angle φ is the angle between ı̂ı and T̂. The vector T̂ is also tangent
to the red circle. As the tangent and radius vectors for circles are per-
pendicular to each other4 , we have that φ = θ + π2 and hence ρ = dφ ds
too.
θ
φ ı̂ı
T̂
ı̂ı
We are now in a position to develop a bunch of formulae for the radius
of curvature ρ and the curvature κ = ρ1 , that are more convenient than κ =
ds −1
. These formulae will use the
dφ
Definition 1.3.2 If r(t) is a parametrized curve, then

dr
• v(t) = dt (t) is the velocity vector at r(t)
d2 r
• a(t) = dt2 (t) is the acceleration vector at r(t)
• T̂(t) is the unit tangent vector to the curve at r(t) that points in the
direction of increasing t.
• N̂(t) is the unit normal vector to the curve at r(t) that points toward the
centre of curvature.
• κ(t) is the curvature at r(t)
• ρ(t) is the radius of curvature at r(t)
♦
Theorem 1.3.3
a Given5 s(φ), i.e. if we know the arc length along the curve as a function
of the angle6 φ = ](ı̂ı, T̂), then
−1
ds ds dφ
ρ = κ = κ =
dφ dφ ds
b Given r(s), i.e. if we have a parametrization of the curve in terms of arc

length, then
dT̂
(s) = κ(s) N̂(s)
ds
where N̂(s) is the unit normal vector to the curve at r(s) that points
toward the centre of curvature.
4 We saw that in Example 1.1.6.
c Given r(t), i.e. if we have a general parametrized curve, then

2
d2 s

dT̂ ds ds ds
= κ N̂ v(t) = (t) T̂(t) a(t) = T̂ + κ N̂
dt dt dt dt2 dt
d Given x(t) , y(t) , (for curves in the xy-plane)

dx d2 y dy d2 x
2 − dt dt2
v(t) × a(t)
κ= dt dt
=

ds
3 2 3/2
dx 2
+ dy

dt dt dt
e Given y(x), (again for curves in the xy-plane)

d2 y
2
dx
κ= 3/2
dy 2
1+ dx
Proof. (a) Given s(φ), then
ds ds −1
ρ= κ=

dφ dφ

ds
As we are assuming that 0 < ρ = dφ < ∞, the inverse function theorem says
that we can invert the function s(φ) (at least locally) to get φ as a function of
s, and that
dφ
κ=

ds
(b) Given r(s), then, by Lemma 1.1.4.c, T̂(s) = r0 (s) is a unit tangent to
the curve at r(s) and
dT̂ dT̂ dφ
= (∗)
ds dφ ds
dφ
Now up to a sign ds is κ, and just because φ = ](ı̂ı, T̂), with T̂ a unit vector,
T̂ = cos φı̂ı + sin φ ̂

dT̂ (∗∗)
=⇒ = − sin φı̂ı + cos φ ̂
dφ
T̂
So ddφ is a unit vector that is perpendiculara to T̂, and hence to the curve at
r(s), and
dT̂
(s) = κ(s) N̂(s) (†)
ds
with N̂(s) a unit normal vector to the curve at r(s). In fact, N̂(s) is the unit
normal vector to the curve at r(s) that points toward the centre of curvature.
To see that, look at the figures belowb , and note that substituting the sign
information from each figure into (∗) gives (†). For example,
5 The equation s = s(φ) is called the “intrinsic equation of the curve”.
6 The notation ](ı̂ı, T̂) means “the angle between ı̂ı and T̂”.
dT̂
dφ T̂
dT̂
dφ T̂
φ
φ
dφ
cos φ ą 0, ds
ą0
dφ
cos φ ą 0, ds
ă0
φ
φ
dT̂
dT̂
T̂ dφ
T̂ dφ
dφ
cos φ ă 0, ds
ă0 cos φ ă 0, dφ
ą0
ds
consider the figure on the lower left. In that figure,
• the x component of T̂ is negative (T̂ is leftward pointing in the figure),

◦ which makes cos φ negative (see (∗∗)),
dT̂
◦ which makes the y component of dφ negative (see (∗∗) again),
dT̂
◦ so dφ is downward pointing,
dT̂
so dφ = −N̂ (the centre of curvature is the red dot above the curve) and
• as s increases (i.e. as you move in the direction of the arrow on the curve),
φ decreases (on the far right hand part of the curve φ ≈ 3π 2 , while on the
far left hand part of the curve φ ≈ π), so ds < 0 and κ = ds = − dφ
dφ dφ

ds .

dT̂ dT̂ dφ
• So by (∗), ds = dφ ds = − N̂)(−κ) = κN̂.
T̂
In each of the three other figures we also end up with dds = κ(s)N̂(s).
Note that if κ(s) = 0, then N̂(s) is not defined. This makes sense: if the
curve is (locally) a straight line, there is no “best fitting circle”.
(c) Given r(t), i.e. if we have a general parametrized curve, we can deter-
mine a unit tangent vector by using Lemma 1.1.4:
dr ds r0 (t)
v(t) = (t) = (t) T̂(t) =⇒ T̂(t) =
dt dt |r0 (t)|
Then we can determine κ and N̂ by differentiating T̂(t) and using the chain
rule:
dT̂ dT̂ ds ds |T̂0 (t)|
= = κ N̂ =⇒ κ(t) = 0
dt ds dt dt |r (t)|
ds
Also, if we differentiate v(t) = dt T̂(t), we get that the acceleration
d2 r d2 s ds dT̂ d2 s ds 2
a(t) = = T̂ + = T̂ + κ N̂
dt2 dt2 dt dt dt2 dt

(d) Given x(t) , y(t) , (for curves in the xy-plane), we can read off the
curvature from
ds d2 s ds 2
v(t) × a(t) = (t) T̂(t) × 2
T̂ + κ N̂
dt dt dt
ds 3
=κ T̂ × N̂ (since T̂ × T̂ = 0
dt
Think of T̂ and N̂ as 3d vectors that whose z-components happen to be zero.

As T̂ and N̂ are mutually perpendicular unit vectors in the xy-plane, the cross- 3
product T̂ × N̂ will be either +k̂ or −k̂. In both cases, |v(t) × a(t) = κ ds dt .

So
v(t) × a(t) dx ı̂ı + dy ̂ × d2 x ı̂ı + d2 y ̂

dt dt dt 2 dt 2
κ= =

ds 3 ds 3

dt dt
dx d2 y − dy d2 x k̂

2 dt dt2
= dt dt

ds 3

dt
dx d2 y dy d2 x
2 − dt dt2

= dt dt 3/2
dx 2 2
+ dy

dt dt
(e) Given y(x), again for curves in the xy-plane, we can parametrize the
curve using x as the parameter:

r(t) = X(t) , Y (t) with X(t) = t and Y (t) = y(t)
Then
dX d2 X dY dy d2 Y d2 y
=1 =0 = =
dt dt2 dt dx dt2 dx2
and dX d2 Y d2 y
dY d2 X

dt dt2 − dt dt2
2
dx
κ= 3/2 = 3/2
dX 2 dY 2 dy 2

dt + dt 1 + dx

a Think about why this should be the case. In particular, sketch T̂ and φ and think about
T̂
what the sketch says about ddφ .
b In each of the four figures, the arrow on the curve specifies the direction of increasing
arc length s and the red dot is the centre of curvature for the curve at the blue dot.
Take another look at Theorem 1.3.3 and note that
d2 s
• the tangential component of acceleration, i.e. dt2 , arises purely from
change in speed while
2
• the normal component of acceleration, i.e. κ dsdt , arises from curvature
and is proportional to the square of the speed ds
dt Think about what you
.
feel when you are driving. That’s why velodromes and (car) race tracks
often have banked corners.
Example 1.3.4 As a warm up example, and also a check that our formulae
make sense, we’ll find the curvature κ, radius of curvature, ρ, unit tangent
vector, T̂, unit normal vector, N̂, and centre of curvature of the parametrized
curve
r(t) = a cos tı̂ı + a sin t ̂
with the constant a > 0. This is, of course, the circle of radius a centred on
the origin. As
dr ds
v(t) = (t) = −a sin tı̂ı + a cos t ̂ =⇒ (t) = |v(t)| = a
dt dt
we have that the unit tangent vector
v(t)
T(t) = = − sin tı̂ı + cos t ̂
|v(t)|
Note, as a check, that this is indeed a vector of length one and is perpendicular
to the radius vector (as expected — the curve is a circle). As
dT̂
(t) = − cos tı̂ı − sin t ̂
dt
we have that
dT̂
(t)
N̂(t) = dt = − cos tı̂ı − sin t ̂
dT̂ (t)
dt
dT̂
(t) 1
dt
κ(t) = ds =
dt (t) a
1
ρ(t) = =a
κ(t)
Now look at the figure.

T̂
rptq
N̂
To get to the centre of curvature we should start from r(t) and walk a
distance ρ(t), which after all is the radius of curvature, in the direction N̂(T ),
which is pointing towards the centre of curvature. So the centre of curvature
is
r(t) + ρ(t)N̂(t) = a cos tı̂ı + a sin t ̂ + a − cos tı̂ı − sin t ̂ = 0
This makes perfectly good sense — the radius of curvature is the radius of the
original circle and the centre of curvature is the centre of the original circle.
One alternative calculation of the curvature, using x(t) = a cos t, y(t) =
a sin t, is
dx d2 y
(t) 2 (t) − dy (t) d2 2x (t)

dt dt dt dt
κ(t) = h
dx
2 dy
2 i3/2
dt (t) + dt (t)

− a sin t − a sin t − a cos t − a cos t
= 2 2 3/2
− a sin t + a cos t
1
=
a
√
Another alternative calculation of the curvature, using y(x) = a2 − x2 (for
the part of the circle with y > 0),
x x
y 0 (x) = − √ =−
a2 − x2 y(x)
y(x) − xy 0 (x) y(x)2 + x2 a2
y 00 (x) = − 2
=− 3
=−
y(x) y(x) y(x)3
is
d2 y a2
2 (x)
dx y(x)3 a2
κ(x) = h i 3/2
=h i3/2
= 3/2
2
dy
1 + dx (x) x2
1 + y(x) 2
y(x)2 + x2
1
=
a

Example 1.3.5 As a more computationally involved example, we’ll analyze

r(t) = cos t + t sin t ı̂ı + sin t − t cos t ̂ t>0
v(t) = t cos tı̂ı + t sin t ̂

a(t) = cos t − t sin t ı̂ı + sin t + t cos t ̂
We can read off from v(t), that
ds
(t) = |v(t)| = t
dt
d2 s
(t) = 1
dt2
v(t)
T(t) = = cos tı̂ı + sin t ̂
|v(t)|
Next, from a(t), we read off that

a(t) = cos t − t sin t ı̂ı + sin t + t cos t ̂ and
2
d2 s

ds
a(t) = 2 (t) T̂(t) + κ(t) (t) N̂(t)
dt dt
(by Theorem 1.3.3 .c)
= cos tı̂ı + sin t ̂ + t2 κ(t)N̂(t)

=⇒ t2 κ(t)N̂(t) = −t sin tı̂ı + t cos t ̂
so that t2 κ(t) is the length of −t sin tı̂ı + t cos t ̂, which is t. Thus
1 −t sin tı̂ı + t cos t ̂

κ(t) = and N̂(t) = = − sin tı̂ı + cos t ̂
t t2 κ(t)
As an alternative calculation of the curvature, we have
|v(t) × a(t)|
κ(t) =
( ds (t))3
dt
t cos tı̂ı + t sin t ̂ × cos t − t sin t ı̂ı + sin t + t cos t ̂
=
( ds
dt (t))
3

t cos t sin t + t cos t − t sin t cos t − t sin t k̂
=
( ds
dt (t))
3
|t2 k̂| 1
= 3
=
t t
It pays to think before you calculate!
1.3.1 Exercises
There are a lot of constants in this chapter that might be new to you.
They can take a little getting used to. Questions 1.3.1.1-1.3.1.5 provide practice
working with and interpreting these constants and their relations to each other.
1. Sketch the curve r(t) = (3 sin t, 3 cos t). At the point (0, 3), label T̂
and N̂. Give the values of κ and ρ at this point as well.
2. Consider the circle r(t) = (3 sin t, 3 cos t). Find T̂(t) and T̂(s). Then,
use parts (b) and (c) of Theorem 1.3.3 to find N̂(t) and N̂(s).
3. The functon r(t) = (t cos t, t sin t), t ≥ 0, defines a spiral centred at
the origin. Using only geometric intuition (no calculation), predict
lim κ(t).
t→∞
ds
4. Let r(t) = (et , 3t, sin t). What is dt ?
5. In Question 1.2.1.5 of Section 1.2, we found that the spiral
r(t) = et (cos t, sin t)
parametrized in terms of arclength is

s
s s
R(s) = √ cos ln √ , sin ln √ .
2 2 2
dT̂ dT̂
Find ds and dt for this curve.
6. In this exercise, we make more precise the sense in which the osculating
circle is the circle which best approximates a plane curve at a point.
• By translating and rotating our coordinate system, we can al-
ways arrange that the point is (0, 0) and that the curve is y =
f (x) with f 0 (0) = 0 and f 00 (0) > 0. (We are assuming that the
curvature at the point is nonzero.)
• Let y = g(x) be the bottom half of the circle of radius r which

is centred at (0, r).
Show that if f (x) and g(x) have the same second order Taylor ap-
proximation at x = 0, then r is the radius of curvature of y = f (x) at
x = 0.
7. Given a curve r(t) = (et , t2 + t), compute the following quantities:
A v(t)
B a(t)
ds
C
dt
D T̂(t)
E κ(t)
8. Find the curvature κ(t) of r(t) = (cos t + sin t, sin t − cos t).
9. Find the minimum and maximum values for the curvature of the el-
lipse x(t) = a cos t, y(t) = b sin t. Here a > b > 0.
10. ∗.
a Find the curvature of y = ex at (0, 1).
b Find the equation of the circle best fitting y = ex at (0, 1).
11. ∗. Consider the motion of a thumbtack stuck in the tread of a tire
which is on a bicycle moving at constant speed. This motion is given
by the parametrized curve

r(t) = t − sin t , 1 − cos t
with t > 0.
a Sketch the curve in the xy-plane for 0 < t < 4π.
b Find and simplify the formula for the curvature κ(t).
c Find the radius of curvature of the osculating circle to r(t) at

t = π.
d Find the equation of the osculating circle to r(t) at t = π.
12. Find the curvature κ as a function of arclength s (measured from
(0, 0)) for the curve
Z θ Z θ
1 2 1 2

x(θ) = cos 2 πt dt y(θ) = sin 2 πt dt
0 0
3
13. ∗. Let C be the curve in R2 given by the graph of the function y = x3 .
Let κ(x) be the curvature of C at the point (x, x3 /3). Find all points
where κ(x) attains its maximal values, or else explain why such points
do not exist. What are the limits of κ(x) as x → ∞ and x → −∞?
1.4 Curves in Three Dimensions

So far, we have developed formulae for the curvature, unit tangent vector,
etc., at a point r(t) on a curve that lies in the xy-plane. We now extend our
discussion to curves in R3 . Fix any t. For t0 very close to t, r(t0 ), will, by the
Taylor expansion to second order, be very close to r(t)+r0 (t) (t0 −t)+ 21 r0 (t) (t0 −
t)2 , so that r(t0 ) almost lies in the plane through r(t) that is determined by
the two vectors r0 (t) and r0 (t). Thus, if we restrict our attention to a very
small part of the curve near the point of interest r(t), the curve will, to a very
good approximation lie in some plane. So we can still define, for example, the
osculating circle to the curve at r(t) to be the circle in that plane that fits the
curve best near r(t). And we still have the formulae1
dr ds
v= = T̂
dt dt
dT̂
= κN̂
ds
dT̂ ds
= κ N̂
dt dt
d2 r d2 s ds 2
a = 2 = 2 T̂ + κ N̂
dt dt dt
ds 3
v×a =κ T̂ × N̂
dt
The only 2 difference is that v, a, T̂ and N̂ are now three component vectors
rather than two component vectors.
If we are lucky and our curve happens to lie completely in a single plane,
the vectors T̂(s) and N̂(s) are mutually perpendicular unit vectors that lie in
the same plane, so that their cross product B̂(s) = T̂(s) × N̂(s) is a unit vector
that is perpendicular to the plane. By continuity, B̂(s) has to be a constant
vector, i.e. be independent of s.
If, on the other hand, B̂(s) is not constant, then our curve doesn’t lie in a
single plane, and we can use the derivative
dB̂ d dT̂ dN̂

= T̂ × N̂ = × N̂ + T̂ ×
ds ds ds ds
dN̂ dT̂
= T̂ × since is parallel to N̂
ds ds
as a measure
• of how badly the curve fails to lie in a plane,
• i.e. how much the plane that fits the curve best near r(s) twists as s
increases,
B̂ N̂ B̂
The cross product in dds = T̂ × dds implies that dds is perpendicular to T̂. In
dB̂
addition, ds must be perpendicular to B̂ because
d h i dB̂
|B̂| = 1 =⇒ 1 = B̂ · B̂ =⇒ 0 = B̂ · B̂ = 2B̂ ·
ds ds
dB̂
So ds (s) must be parallel to N̂(s).
Definition 1.4.1
a The binormal vector at r(s) is B̂(s) = T̂(s) × N̂(s). The normal
vector N̂(s) is sometimes called the unit principal normal vector to
distinguish it from the binormal vector.
b We define the torsion τ (s) by
dB̂
(s) = −τ (s)N̂(s)
ds
1 The arguments in the proof of Theorem 1.3.3 that we used to verify these formulae work
in any plane, not just the xy-plane. Just choose ı̂ı and ̂ to be any two mutually perpendicular
unit vectors in the plane.
2 However this can be a significant difference.
The negative sign is included so that τ (s) > 0 indicates “right handed
twisting”. There will be an explanation of what this means in Example
1.4.4 below.
c The osculating plane at r(s) (the plane that fits the curve best at r(s))
is the plane through r(s) with normal vector B̂(s). The equation of the
plane is
B̂(s) · (x, y, z) − r(s) = 0
♦
For each s, T̂(s), N̂(s) and B̂(s) are mutually perpendicular unit vectors.
They form an orthonormal basis for R3 , just as ı̂ı, ̂ and k̂ form an orthonormal
basis for R3 . Furthermore both (T̂(s) , N̂(s) , B̂(s)) and (ı̂ı , ̂ , k̂) are “right
handed triples”3 , meaning that B̂(s) = T̂(s) × N̂(s) and k̂ = ı̂ı × ̂.
k̂ B̂
̂
N̂
ı̂ı T̂
dT̂ dB̂
We have already computed ds and ds . It is now an easy matter to compute
dN̂ d
= B̂(s) × T̂(s)
ds ds

= −τ (s)N̂(s) × T̂(s) + B̂(s) × κ(s)N̂(s)
= τ (s)B̂(s) − κ(s)T̂(s)
To see that N̂(s) × T̂(s) = −B̂(s) and B̂(s) × N̂(s) = −T̂(s), just look at the
right hand figure above.
Now suppose that we have a curve that is parametrized by t rather than s.
How do we find the torsion τ ? The most obvious method is to
3 3
• recall that v × a = κ dsdt T̂ × N̂ = κ dsdt B̂ and that B̂(t) is a unit
vector. So
v(t) × a(t)
B̂(t) =
|v(t) × a(t)|
dB̂
• Having found B(t) we can differentiate it and use ds (s) = −τ (s)N̂(s)
and the chain rule to give
dB dB ds ds
= = −τ N̂
dt ds dt dt
ds
from which we can read off τ , provided we know dt and N̂.
There is another, often more efficient, method to find the torsion τ that
uses
da d d2 s ds 2
= T̂ + κ N̂
dt dt dt2 dt
d3 s d2 s ds d ds 2 ds 3
= 3 T̂ + 2 κN̂ + κ N̂ + κ τ B̂ − κT̂
dt dt dt dt dt dt
3 We shall stick to “right handed triples” to make it easier to get various signs right.
While this looks a little complicated, notice that, with just one exception,
3
namely κ dsdt τ (s)B̂(s), every term on the right hand side is either in the
direction T̂ or in the direction N̂ and so is perpendicular to B̂. So, dotting
3
with v × a = κ ds dt B̂ gives
da ds 6
v×a · = κ2 τ = |v × a|2 τ
dt dt
and hence
v × a · da

dt
τ=
|v × a|2
If the curvature4 κ(s) > 0 and the torsion τ (s) are known, then the system
of equations5
Equation 1.4.2 Frenet–Serret Formulae.
dT̂
(s) = κ(s) N̂(s)
ds
dN̂
(s) = τ (s) B̂(s) − κ(s) T̂(s)
ds
dB̂
(s) = −τ (s) N̂(s)
ds
is a first order linear system of ordinary differential equations

    
T̂(s) 0 κ(s) 0 T̂(s)
d 
N̂(s) = −κ(s) 0 τ (s) N̂(s)
ds
B̂(s) 0 −τ (s) 0 B̂(s)
for the 9 component vector valued function (T̂(s) , N̂(s) , B̂(s)).

Any first order linear initial value problem
d
x(s) = M (s)x(s) x(0) = x0
ds
where x is an n-component vector and M (s) is an n×n matrix with continuous
entries, has exactly one solution. If n = 1, so that x(s) and M (s) are just
functions, this is easy to see. Just let M(s) be the antiderivative of M (s) that
obeys M(0) = 0. Then
d d
x(s) = M (s)x(s) ⇐⇒ e−M(s) x(s) − M (s)e−M(s) x(s) = 0
ds ds
d −M(s)
⇐⇒ e x(s) = 0
ds
by the product rule. So e−M(s) x(s) is a constant independent of s. In partic-
ular e−M(s) x(s) = e−M(0) x(0) = x0 so that x(s) = x0 eM(s) . This argument
4 As in two dimensions, if κ(s) = 0, then N̂(s) is not defined. This makes even more sense
in three dimensions than in two dimensions: if the curve is a straight line, there are infinitely
many unit vectors perpendicular to it and there is no way to distinguish between them.
5 The equations are named after the two French mathematicians who independently dis-
covered them: Jean Frédéric Frenet (1816–1900, the son of a wig maker), in his thesis of 1847
(actually he only gave two of the three equations), and Joseph Alfred Serret (1819–1885) in
1851.
can be generalized to any natural number n. But that is beyond the scope of
this book.
Since the Frenet-Serret formulae constitute a first order system of ordinary
differential equations for the vector (T̂(s) , N̂(s) , B̂(s)) and since any first
order linear initial value problem has a exactly one solution,
• the vector valued function (T̂(s) , N̂(s) , B̂(s)) is determined by the func-
tions κ(s) and τ (s) (assuming that they are continuous) together with
the initial condition (T̂(0) , N̂(0) , B̂(0)).
• Furthermore, once you know T̂(s), then r(s) is determined by r(0) and
dr
ds (s) = T̂(s).
• So any smooth curve r(s) is completely determined by r(0), (T̂(0) , N̂(0) , B̂(0)),
κ(s) and τ (s).
• That is, up to translations (you can move between any two possible
choices of r(0) by a translation) and rotations (you can move between
any two possible choices of (T̂(0) , N̂(0) , B̂(0)) by a rotation) a curve is
completely determined by the curvature κ(s) > 0 and the torsion τ (s).
This result is called “The fundamental theorem of space curves”.
Theorem 1.4.3 The Fundamental Theorem of Space Curves. Let
κ(s) > 0 and τ (s) be continuous. Then up to translations and rotations, there
is a unique curve with curvature κ(s) and torsion τ (s).
Example 1.4.4 Right circular helix. The right circular helix is the curve
r(t) = a cos tı̂ı + a sin t ̂ + bt k̂
with a, b > 0 as in the figure on the left below.

z
t“ 5π
2
t“ 3π
2
π
t“ 2
y
t“0
x
Here is why it is called a right helix rather than a left helix. If the helix
is the thread of a bolt that you are screwing into a nut, and you turn the bolt
in the direction of the (curled) fingers of your right hand (as in the figure6 on
the right above), then it moves in the direction of your thumb (as in the long
straight arrow of the figure on the right above).
To determine the curvature and torsion of this curve we compute
v(t) = −a sin tı̂ı + a cos t ̂ + b k̂

a(t) = −a cos tı̂ı − a sin t ̂
da
(t) = a sin tı̂ı − a cos t ̂
dt
From v(t) we read off
ds p 2
= a + b2
dt
a a b
T̂(t) = − √ sin tı̂ı + √ cos t ̂ + √ k̂
2
a +b 2 2
a +b 2 a + b2
2
2 2
From a = ddt2s T̂ + κ ds
dt N̂ = κ(a2 + b2 )N̂, we read off that
a
κ(t) = N̂(t) = − cos tı̂ı − sin t ̂
a2 + b2
From
 
ı̂ı ̂ k̂
v(t) × a(t) = det  −a sin t a cos t b  = ab sin tı̂ı − ab cos t ̂ + a2 k̂
−a cos t −a sin t 0
|v(t) × a(t)|2 = a2 b2 + a4 = a2 (a2 + b2 )
we read off
v(t) × a(t) b b a
B̂(t) = =√ sin tı̂ı − √ cos t ̂ + √ k̂
|v(t) × a(t)| a2 + b2 a2 + b2 a2 + b2
and
v × a · da

dt a2 b b
τ (t) = 2
= 2 2 2
= 2
|v × a| a (a + b ) a + b2
Note that, for the right handed helix, τ > 0. Finally the centre of curvature is
1 a2 + b2 a2 + b2
r(t) + N̂(t) = a − cos tı̂ı + a − sin t ̂ + bt k̂
κ(t) a a
b2 b2
= − cos tı̂ı − sin t ̂ + bt k̂
a a
which is another helix. In the figure below, the red curve is the original helix
and the blue curve is the helix traced by the centre of curvature.
z
x

6 This figure is a variant of this picture.
1.4.1 Exercises
1. In the sketch below of a three-dimensional curve and its osculating
circle at a point, label T̂ and N̂. Will B̂ be pointing out of the paper
towards the reader, or into the paper away from the reader?
2. In the formula
ds
(t) = |v(t)| = |r0 (t)|
dt
does s stand for speed, or for arclength?
3. Which curve (or curves) below have positive torsion, which have neg-
ative torsion, and which have zero torsion? The arrows indicate the
direction of increasing t.
z
a(t) = (cos t, −2 sin t, t/2)

b(t) = (cos t, 2 sin t, −t/2)
c(t) = (0, t/2 sin t, t cos t)

4. Consider a curve that is parametrized by arc length s.
a Show that if the curve has curvature κ(s) = 0 for all s, then the
curve is a straight line.
b Show that if the curve has curvature κ(s) > 0 and torsion τ (s) =
0 for all s, then the curve lies in a plane.
c Show that if the curve has curvature κ(s) = κ0 , a strictly posi-
tive constant, and torsion τ (s) = 0 for all s, then the curve is a
circle.
5. ∗. The surface z = x2 + y 2 is sliced by the plane x = y. The resulting
curve is oriented from (0, 0, 0) to (1, 1, 2).
a Sketch the curve from (0, 0, 0) to (1, 1, 2).
b Sketch T̂, N̂ and B̂ at 21 , 12 , 12 .

c Find the torsion at 12 , 12 , 12 .

6. ∗. Let C be the space curve
r(t) = et − e−t ı̂ı + et + e−t ̂ + 2t k̂

a Find r0 , r00 and the curvature of C.

b Find the length of the curve between r(0) and r(1).
7. Find the torsion of r(t) = (t, t2 , t3 ) at the point (2, 4, 8).
8. Find the unit tangent, unit normal and binormal vectors and the
curvature and torsion of the curve
t2 t3
r(t) = tı̂ı +
̂ + k̂
2 3
9. For some constant c, define r(t) = (t3 , t, ect ). For which value(s) of c
is τ (5) = 0? For each of those values of c, find an equation for the
plane containing the osculating circle to the curve at t = 5.
10. ∗.
a Consider the parametrized space curve
r(t) = t2 , t, t3

Find an equation for the plane passing through (1, 1, 1) with

normal vector tangent to r at that point.
b Find the curvature of the curve from (a) as a function of the
parameter t.

11. ∗. Let C be the osculating circle to the helix r(t) = cos t , sin t , t
at the point where t = π/6. Find:
a the radius of curvature of C
b the center of C
c the unit normal to the plane of C
12. ∗.
a Consider the parametrized space curve
r(t) = (cos(t), sin(t), t2 )
Find a parametric form for the tangent line at the point corre-
sponding to t = π.
b Find the tangential component aT (t) of acceleration, as a func-
tion of t, for the parametrized space curve r(t).
13. ∗. Suppose, in terms of the time parameter t , a particle moves along
the path r(t) = (sin t − t cos t)ı̂ı + (cos t + t sin t) ̂ + t2 k̂, 1 ≤ t < ∞.
a Find the speed of the particle at time t.
b Find the tangential component of acceleration at time t.

c Find the normal component of acceleration at time t.
d Find the curvature of the path at time t.
14. ∗. Assume the paraboloid z = x2 + y 2 and the plane 2x + z = 8
intersect in a curve C. C is traversed counter-clockwise if viewed
from the positive z-axis.
a Parametrize the curve C.
b Find the unit tangent vector T̂, the principal normal vector
N̂, the binormal vector B̂ and the curvature κ all at the point
(2, 0, 4).
15. ∗. Consider the curve C given by
1 3 1
r(t) = t ı̂ı + √ t2 ̂ + t k̂, −∞ < t < ∞.
3 2
a Find the unit tangent T̂(t) as a function of t.
b Find the curvature κ(t) as a function of t.

8
√
c Determine the principal normal vector N̂ at the point 3 , 2 2, 2 .
2 2
16. ∗. Suppose the curve C is the intersection of the cylinder x + y = 1
with the plane x + y + z = 1.
a Find a parameterization of C.
b Determine the curvature of C.
c Find the points at which the curvature is maximum and deter-

mine the value of the curvature at these points.
17. ∗. Let
r(t) = t2 ı̂ı + 2t ̂ + ln t k̂
Compute the unit tangent and unit normal vectors T̂(t) and N̂(t).
Compute the curvature κ(t). Simplify whenever possible!
18. ∗.
2 3
a Find the length of the curve r(t) = 1, t2 , t3 for 0 ≤ t ≤ 1.
b Find the principal unit normal vector N̂ to r(t) = cos(t)ı̂ı +

sin(t) ̂ + t k̂ at t = π/4.
c Find the curvature of r(t) = cos(t)ı̂ı + sin(t) ̂ + t k̂ at t = π/4.
19. ∗. A particle moves along a curve with position vector given by
r(t) = t + 2 , 1 − t , t2 /2

for −∞ < t < ∞.

a Find the velocity as a function of t.
b Find the speed as a function of t.
c Find the acceleration as a function of t.
d Find the curvature as a function of t.
e Recall that the decomposition of the acceleration into tangential
and normal components is given by the formula
d2 s ds 2
r00 (t) = 2 T̂(t) + κ(t) N̂(t)
dt dt
Use this formula and your answers to the previous parts of this
question to find N̂(t), the principal unit normal vector, as a
function of t.
f Find an equation for the osculating plane (the plane which best
fits the curve) at the point corresponding to t = 0.
g Find the centre of the osculating circle at the point correspond-

ing to t = 0.
20. ∗. Consider the curve C given by
t3 t2
r(t) = ı̂ı + √ ̂ + t k̂ −∞<t<∞
3 2
a Find the unit tangent T̂(t) as a function of t.

b Find the curvature κ(t) as a function of t.
c Evaluate κ(t) at t = 0.
d Determine the principal normal vector N̂(t) at t = 0.
e Compute the binormal vector B̂(t) at t = 0.

21. ∗. A curve in R3 is given by r(t) = (t2 , t , t3 ).
a Find the parametric equations of the tangent line to the curve
at the point (1, −1, −1).
b Find an equation for the osculating plane of the curve at the
point (1, 1, 1).
22. ∗. A curve in R3 is given by
r(t) = (sin t − t cos t)ı̂ı + (cos t + t sin t) ̂ + t2 k̂, 0≤t<∞
a Find the length of the curve r(t) from r(0) = (0, 1, 0) to r(π) =
(π, −1, π 2 ).
b Find the curvature of the curve at time t > 0.

23. ∗. At time t = 0, NASA launches a rocket which follows a trajectory
so that its position at any time t is
√ √
4 2 3/2 4 2 3/2
x= t , y= t , z = t(2 − t)
3 3
a Assuming that the flight ends when z = 0, find out how far the
rocket travels.
b Find the unit tangent and unit normal to the trajectory at its
highest point.
c Also, compute the curvature of the trajectory at its highest
point.
24. ∗. Consider a particle travelling in space along the path parametrized
by
x = cos3 t, y = sin3 t, z = 2 sin2 t
a Calculate the arc length of this path for 0 ≤ t ≤ π/2.

b Find the vectors T̂, N̂, B̂ for the particle at t = π/6.
25. Suppose that the curve C is the intersection of the cylinder x2 +y 2 = 1
with the surface z = x2 − y 2 .
a Find a parameterization of C.
√ √
b Determine the curvature of C at the point 1/ 2 , 1/ 2 , 0 .
√ √
c Find the osculating plane to C at the point 1/ 2 , 1/ 2 , 0 .
In general, the osculating plane to a curve r(t) at the point r(t0 )
is the plane which fits the curve best at r(t0 ). It passes through
r(t0 ) and has normal vector B̂(t0 ).
d Find the radius
√ and√the centre
of the osculating circle to C at
the point 1/ 2 , 1/ 2 , 0 .
26. ∗. Under the influence of a force field F, a particle of mass 2 kg
is moving with constant speed 3 m/s along the path given as the
intersection of the plane z = x and the parabolic cylinder z = y 2 , in
the direction of increasing y. Find F at the point (1, 1, 1). (Length is
measured in m along the three coordinate axes.)
27. ∗. Consider the curve C in 3 dimensions given by
√
r(t) = 2tı̂ı + t2̂ + 3t2 k̂
for t ∈ R.
a Compute the unit tangent vector T(t).
b Compute the unit normal vector N(t).
c Show that the binormal vector B to this curve does not depend
on t and is one of the following vectors:
       
1/2
√ √0 √0 0
(1) − 3/2 (2)  3/2 (3) − 3/2 (4) √−1/2
0 1/2 1/2 3/2
This implies that C is a plane curve.

d According to your choice of vector (1), (2), (3) or (4), give the
equation of the plane containing C.
e Compute the curvature κ(t) of the curve.
f Are there point(s) where the curvature is maximal? If yes, give
the coordinates of the point(s). If no, justify your answer.
g Are there point(s) where the curvature is minimal? If yes, give
the coordinates of the point(s). If no, justify your answer.
h Let √ √
u := 2ı̂ı, v := ̂ + 3 k̂ w := − 3 ̂ + k̂
(i) Express ı̂ı, ̂, k̂ in terms of u, v, w.
(ii) Using (i), write r(t) in the form
a(t)u + b(t)v + c(t)w
where a(t), b(t) and c(t) are functions you have to deter-
mine. You should find that one of these functions is zero.

(iii) Draw the curve given by a(t), b(t) in the xy-plane.
(iv) Is the drawing consistent with parts (f) and (g)? Explain.
28. ∗. Recall that if T̂ is the unit tangent vector to an oriented curve

with arclength parameter s, then the curvature κ and the principle
normal vector N̂ are defined by the equation
dT̂
= κ N̂
ds
Moreover, the torsion τ and the binormal vector B̂ are defined by the
equations
dB̂
B̂ = T̂ × N̂, = −τ N̂
ds
Show that
dN̂
= −κ T̂ + τ B̂
ds
p
29. ∗. A skier descends the hill z = 4 − x2 − y 2 along a trail with
parameterization
π
x = sin(2θ), y = 1 − cos(2θ), z = 2 cos θ, 0≤θ≤
2
Let P denote the point on the trail where x = 1.
a Find the vectors T̂, N̂, B̂ and the curvature κ of the ski trail at
the point P .
√
b The skier’s acceleration at P is a = (−2, 3, −2 2). Find, at P ,
(i) the rate of change of the skier’s speed and
(ii) the skier’s velocity (a vector).
30. ∗. A particle moves so that its position vector is given by r(t) =
cos t , sin t , c sin t , where t > 0 and c is a constant.
a Find the velocity v(t) and the acceleration a(t) of the particle.
b Find the speed v(t) = |v(t)| of the particle.

c Find the tangential component of the acceleration of the parti-
cle.
d Show that the trajectory of this particle lies in a plane.
31. ∗. A race track between two hills is described by the parametric curve
1
r(θ) = 4 cos θ , 2 sin θ , cos(2θ) , 0 ≤ θ ≤ 2π
4
− 4, 0, 14 .

a Compute the curvature of the track at the point
b Compute the radius of the circle that best approximates the
bend at the point\\ − 4, 0, 14 (that is, the radius of the oscu-

lating circle at that point).
c A car drives down the track so that its position at time t is given
by r(t2 ). (Note the relationship between t and θ is θ = t2 ).
Compute the following quantities.
(i) The speed at the point − 4, 0, 14 .

(ii) The acceleration at the point − 4, 0, 14 .

(iii) The magnitude of the normal component of the accelera-

tion at the point
1
− 4, 0, .
4
1.5 A Compendium of Curve Formula

In the following r(t) = x(t) , y(t) , z(t) is a parametrization of some curve.
The vectors T̂(t), N̂(t), and B̂(t) are the unit tangent, normal and binormal
vectors, respectively, at r(t). The tangent vector points in the direction of
travel (i.e. direction of increasing t) and the normal vector points toward the
centre of curvature. The arc length from time 0 to time t is denoted s(t).
The binormal B̂(t) = T̂(t) × N̂(t) is perpendicular to the plane that fits the
curve best at r(t). Some formulae use an arc length parametrization, which is
denoted r(s).
dr ds
the velocity v(t) = (t) = (t) T̂(t)
dt dt
v(t)
the unit tangent vector T̂(t) = |v(t)| (general parametrization)
dr
T̂(s) = ds (s) (arc length parametrization)
d2 r d2 s ds 2
the acceleration a(t) = 2
(t) = 2
(t) T̂(t) + κ(t) (t) N̂(t)
dt dt dt
ds dr
the speed (t) = |v(t)| = (t)
dt dt
Z T Z Tp
ds
the arc length s(T ) = (t) dt = x0 (t)2 +y 0 (t)2 +z 0 (t)2 dt
0 dt 0
|v(t) × a(t)|
the curvature κ(t) = ddtT̂ (t)/ ds
dt (t) =
( ds
dt (t))
3
dφ dT̂
κ(s) = ds (s) = ds (s)
dT̂ dT̂ dT̂

the unit normal vector N̂(t) = (t)/ (t) N̂(s) = (s)/κ(s)
dt dt ds
1
the radius of curvature ρ(t) =
κ(t)
the centre of curvature r(t) + ρ(t)N̂(t)
v(t) × a(t) · da

dt (t)
the torsion τ (t) =
|v(t) × a(t)|2
v(t) × a(t)
the binormal B̂(t) = T̂(t) × N̂(t) =
|v(t) × a(t)|
dr
Under arclength parametrization (i.e. if t = s) we have T̂(s) = ds (s) and
the Frenet-Serret formulae
dT̂
(s) = κ(s) N̂(s)
ds
dN̂
(s) = τ (s) B̂(s) − κ(s) T̂(s)
ds
dB̂
(s) = −τ (s) N̂(s)
ds
which in matrix form is

    
T̂(s) 0 κ(s) 0 T̂(s)
d 
N̂(s) = −κ(s) 0 τ (s) N̂(s)
ds
B̂(s) 0 −τ (s) 0 B̂(s)
When the curve lies entirely in the xy-plane the curvature is given by
(t) d22y (t) − dy (t) d2 2x (t)

dx
dt dt dt dt
κ(t) = h
dx
2 dy
2 i3/2
dt (t) + dt (t)
When the curve lies entirely in the xy-plane and the y-coordinate is given as
a function, y(x), of the x-coordinate, the curvature is
d2 y
2 (x)
dx
κ(x) = h
dy
2 i3/2
1 + dx (x)
dx d2 x
Notice that this follows from the previous formula since dx = 1 and dx2 = 0.
1.6 Integrating Along a Curve

Suppose that we have a curve C that is parametrized as r(t) with a ≤ t ≤ b.
Suppose further that C is actually a piece of wire and that the density (i.e.
mass per unit length) of the wire at the point r is ρ(r). How do we figure out
the mass of C? Of course we use the standard Calculus divide and conquer
strategy. We select a natural number n and
• divide the interval a ≤ t ≤ b into n equal subintervals, each of length
∆t = b−an . We denote by t` = a + `∆t the right hand end of interval
number `.

• Then we approximate
of the part of the curve between r t`−1
the length
and r t` by r t` − r t`−1 and
the mass
of thepart of the curve
between r t`−1 and r t` by ρ r(t` ) r t` − r t`−1 .

rptℓ´1 q rptℓ q
• This gives us, as an approximate mass for C of

n n
r t` − r t`−1

X X
ρ r(t` ) r t` − r t`−1 =
ρ r(t` )
∆t
t` − t`−1
`=1 `=1
Then we take the limit as n → ∞. Assuming1 that r(t) is continuously differ-

entiable and that ρ(r) is continuous we get
Z b
dr
Mass of C = ρ r(t) (t) dt

a dt
which we take to be a definition.
Definition 1.6.1

a For a parametrized curve x(t), y(t), z(t) , a ≤ t ≤ b, in R3 that we call
C, and for a function f (x, y, z), we define
Z Z b p
f (x, y, z) ds = f x(t), y(t), z(t) x0 (t)2 + y 0 (t)2 + z 0 (t)2 dt
C a
In this notation the subscript C specifies the curve, and ds signifies arc
length.
b For a curve y = f (x), a ≤ x ≤ b, in R2 that we call C, and for a function
g(x, y), we define
Z Z b p
g(x, y) ds = g x, f (x) 1 + f 0 (x)2 dx
C a
♦
2
Example 1.6.2 Suppose that we have a helical wire

r(t) = x(t) , y(t) , z(t) = a cos t , a sin t , bt 0 ≤ t ≤ 2π
and that this wire has constant mass density ρ. Let’s find the centre of mass of
the wire. Recall that the centre of mass is x̄, ȳ, z̄) with, for example, x̄ being
the weighted average
R R
xρds xds
x̄ = R = R (since ρ is constant)
ρds ds
R R
yds zds
of x over the wire. Similarly ȳ = R and z̄ = R . For the given curve
ds ds

x(t) , y(t) , z(t) = a cos t , a sin t , bt
x0 (t) , y 0 (t) , z 0 (t) = − a sin t , a cos t , b

ds p
(t) = x0 (t)2 + y 0 (t)2 + z 0 (t)2
dt p
= a2 sin2 t + a2 cos2 t + b2
p
= a2 + b2
so that
R R 2π √ R 2π
xds 0
x(t) a2 + b2 dt a cos(t) dt
x̄ = R = R 2π √ = 0 =0
ds a2 + b2 dt 2π
0
R R 2π √ R 2π
yds 0
y(t) a2 + b2 dt a sin(t) dt
ȳ = R = R 2π √ = 0 =0
ds 2 2
a + b dt 2π
0
1 We could relax these conditions somewhat by instead assuming that r0 (t) and ρ(t) are
bounded and are continuous except at a finite number of points. (r0 (t) need not exist at all
at those points.)
R R 2π √ R 2π
zds 0
z(t) a2 + b2 dt bt dt b h t2 i2π
z̄ = R = R 2π √ = 0 = = bπ
ds a2 + b2 dt 2π 2π 2 0
0
So the centre of mass is right on the axis of the helix, half way up, which makes
perfect sense.
1.6.1 Exercises
1. Give an equation for arclength of a curve C as a line integral.
2.
R
a Show that the integral C f (x, y) ds along the curve C given in
polar coordinates by r = r(θ), θ1 ≤ θ ≤ θ2 , is
s 2
Z θ2

2
dr
f r(θ) cos θ, r(θ) sin θ r(θ) + (θ) dθ
θ1 dθ
b Compute the arc length of r = 1 + cos θ, 0 ≤ θ ≤ 2π. You may

use the formula
θ
1 + cos θ = 2 cos2
2
to simplify the computation.
√ 2
3. Calculate C xy 2 3
R
z ds, where C is the curve 3 t , 3t , 3t from t = 1
to t = 2.
4. A hoop of radius 1 traces out the curve x2 + y 2 = 1, where x and y are
measured in metres. At a point (x, y), its density is x2 kg per metre.
What is the mass of the hoop?
R
5. Compute C (xy + z)ds where C is the straight line from (1, 2, 3) to
(2, 4, 5).
R
6. Evaluate the path integral C f (x, y, z) ds for
a f (x, y, z) = x cos z, C : r(t) = tı̂ı + t2̂, 0 ≤ t ≤ 1.
b f (x, y, z) = x+y 2 3/2

y+z , C : r(t) = t, 3 t , t , 1 ≤ t ≤ 2.
R √
7. Evaluate C sin x ds, where C is the curve (arcsec(t), ln t), 1 ≤ t ≤ 2.
8. ∗. A particle of mass m = 1 has position r(0) = ̂ and velocity
v0 = ı̂ı + k̂ at time t = 0. The particle moves under a force
F(t) = ̂ − sin t k̂
where t denotes time.

a Find the position r(t) of the particle as a function of t.
b Find the position r(t1 ) of the particle when it crosses the plane
x = π/2 for the first time at t1 .
c Determine the work done by F in moving the particle from r(0)

to r(t1 ).
2 For example, your favourite solenoid or spring or slinky.
∗. Evaluate the line integral C F · n̂ ds where F(x, y) = xy 2 ı̂ı + yex ̂
R
9.
, C is the boundary of the rectangle R: 0 ≤ x ≤ 3, −1 ≤ y ≤ 1,
and n̂ is the unit vector, normal to C, pointing to the outside of the
rectangle.
10. ∗. Let C be the curve given by
r(t) = t cos tı̂ı + t sin t ̂ + t2 k̂, 0≤t≤π
a Find the unit tangent T̂ to C at the point (−π, 0, π 2 ).

b Calculate the line integral
Z p
x2 + y 2 ds
C
c Find the equation of a smooth surface in 3-space containing the

curve C.
d Sketch the curve C.

11. A wire traces out a path C described by the curve (t + 12 t2 , t −
1 2 4 3/2
2 t , 3 t ), 0 ≤ t ≤ 4. Its density at the point (x, y, z) is ρ(x, y, z) =
x+y
2 . Find its centre of mass.
1.7 Sliding on a Curve

We are going to investigate the motion of a particle of mass m sliding on a
frictionless1 , smooth curve that lies in a vertical plane. We will consider three
scenarios:
• First, to set things up we’ll look at a bead sliding on a stiff wire.
• Then, we will imagine that we are skiing straight downhill and ask
“Where on the hill can we become airborne?”.
• Then we will imagine that we are skateboarding in a culvert (a large
pipe) and ask “When is it safe?”.
1.7.1 The Sliding Bead

First, consider a bead of mass m that is sliding, without friction, on a stiff
wire. According to Newton’s law of motion
ma = F
where F is the net force being applied to the bead. The bead is subject
to two forces. The gravitational force is −mg̂. By definition, absence of
friction means that the wire is does not apply any force that is in the direction
tangential to the wire. But, because it is stiff, the wire never changes shape
and instead applies just the right amount of force, in the direction normal to
1 We are mathematicians — we like idealized situations.
the wire, that is needed to keep the bead on the wire2 without bending the
wire. Call this normal force W N̂.
̂ wire
W N̂
px, yq ı̂ı
bead
T̂ ´mg̂ (gravity)
So, by Newton’s law,
m a = −mg ̂ + W N̂
We’ll analyse this equation by splitting it into its tangential and normal com-
ponents.
To extract the tangential component of Newton’s law, we dot it with v =
|v|T̂. Since T̂ · N̂ = 0 this kills all normal components.
dv
mv · = −mg̂ · v + W N̂ · v
dt
1 d dy
m (v · v) = −mg
2 dt dt
Here we have used
• Theorem 1.1.3.c on the left hand side and
• that ̂ · v is just the y component of v and
• that N̂ and v = |v|T̂ are perpendicular.
Moving everything to the left hand side of the equation gives

d 1 2
m|v| + mgy = 0
dt 2
and we conclude that the quantity
Equation 1.7.1 Conservation of Energy.
1
E=m|v|2 + mgy
2
is a constant, independent of time. This is, of course,
q the principle of
2E
conservation of energy. It determines the speed |v| = m − 2gy of the bead
as a function of the height y (and of the energy E, which is determined by the
initial conditions).
To extract the normal component of Newton’s law, we dot it with N̂:
ma · N̂ = −mg̂ · N̂ + W
Since
d2 s ds 2 d2 s
a= T̂ + κ N̂ = T̂ + κ|v|2 N̂
dt2 dt dt2
and T̂ and N̂ are perpendicular, this gives, after a little rearrangement,
2 This force is required to keep the bead from either passing through the wire or flying off
the wire.
Equation 1.7.2 Normal Force.
W = mκ|v|2 + mg̂ · N̂ = 2κ(E − mgy) + mg̂ · N̂
1.7.2 The Skier

The difference between the bead on the wire and the skier on the hill is that
while the hill is capable of applying an upward normal force (i.e. it can push
you upward to keep you from falling to the centre of the Earth), it is not
capable of applying a downward normal force. That is the hill cannot pull
down on you to keep you on the hill. Only gravity can keep you grounded.
There are two main possibilities3 .
N̂
T̂
N̂ T̂
• If the hill is concave downward as in the figure on the left above, then
N̂ points downward and the hill is allowed to have W ≤ 0 (which corre-
sponds to the normal force W N̂ pushing upward). If ever W > 0, the hill
would have to pull on you to keep you on hill. It can’t, so you become
airborne. Since ̂ · N̂ < 0, this happens whenever
r
2 g
W > 0 ⇐⇒ mκ|v| + mg̂ · N̂ > 0 ⇐⇒ |v| > |̂ · N̂|
κ
• If the hill is concave upward as in the figure on the right above, then N̂
points upward and the hill is allowed to have W ≥ 0 (which corresponds
to the normal force W N̂ pushing upward). Since ̂ · N̂ > 0 we always
have W = mκ|v|2 + mg̂ · N̂ > 0. You never become airborne. On the
other hand your knees may complain.
1.7.3 The Skate Boarder

So far, Equations 1.7.1 and 1.7.2 apply to any stiff frictionless “wire”. We
now specialize to the special case of a skateboarder inside a circular culvert of
radius a. Let’s put the bottom of the circle at the origin (0, 0), so that the
centre of the circle is at (0, a).
a ´ y φ aN̂
px, yq
3 We assume that you are going downhill and that the curvature κ > 0.
In this case the curvature is κ = a1 and ̂ · N̂ = cos φ = a−y

a so 1.7.1 and
1.7.2 simplify to
r s
2 E
|v| = (E − mgy) = 2g −y
m mg
2 mg 3mg 2 a
W = (E − mgy) + (a − y) = E+ −y
a a a 3mg 3
Imagine now that you start at the bottom of the culvert, that is at y = 0,
with energy E > 0. As time progresses, y increases and consequently |v| and
W both decrease, as, of course, they should. This continues until one of the
following three things happen.
(i) |v| hits 0, in which case you stop rising and start descending. The speed
E
|v| is zero when y = yS = mg . (The subscript “S” stands for “stop”.)
Physicists say that when you reach yS all of your kinetic energy ( 12 m|v|2 )
has been converted into potential energy (mgy).
(ii) W hits zero. When you get higher than this, W becomes negative and
the culvert would have to pull on you to keep your feet on the culvert.
As the culvert can only push on you, you become airborne. The normal
E
force W is zero when y = yA = 23 mg + a3 . (The subscript “A” stands for
“airborne”.)
(iii) y hits 2a. This is the summit of the culvert. You descend on the other
side.
Which case actually happens is determined by the relative sizes of yS , yA and
2a.
2 E 1 E 2 E
• Comparing yS = 3 mg + 3 mg and yA = 3 mg + a3 , we see that yS ≤
E
yA ⇐⇒ mg ≤ a.
2 E a
• Comparing yA = 3 mg + 3 and a = 2
3a + a3 , we see that yA ≤ a ⇐⇒
E
mg ≤ a.
2 E a
• Comparing yA = 3 mg + 3 and 2a = 35 a + a3 , we see that yA ≤ 2a ⇐⇒
E 5
mg ≤ 2 a.
So the conclusions are:
• If 0 ≤ mgE
≤ a then 0 ≤ yS ≤ yA ≤ a . In this case you just oscillate
between heights 0 and yS ≤ a in the bottom half of the culvert, as in the
figure on the left below.
• If a ≤ mgE
≤ 52 a then a ≤ yA ≤ yS , 2a . In this case you make it more
than half way to the top. But you become airborne at y = yA which is
somewhere between the half way mark y = a and the top y = 2a. At
this point our model breaks down because you are no longer in contact
with the culvert. You just freely follow a parabolic arc until you crash
back into the culvert, as in the figure in the centre below.
• If 25 a < mg
E
then 2a < yA < yS . In this case you successfully go
all the way around the culvert, looping the loop, as in the figure on the
E
right below. Note that, as mg > 52 a > 2a, this requires significantly more
energy than that required to reach the top, i.e. to reach height 2a.
happy unhappy thrilled!
1.7.4 Exercises
You may assume the acceleration due to gravity is g = 9.8 m/s2 . You may
also assume that the systems described function as they do in the book: so
tracks are frictionless, etc., unless otherwise mentioned.
1. The figure below represents a bead sliding down a wire. Sketch vectors
representing the normal force the wire exerts on the bead, and the
force of gravity.
Assume the top of the page is “straight up.”

2. In the definition E = 21 m|v|2 + mgy, v is the derivative of position
with respect to what quantity?
3. A bead slides down a wire with the shape shown below, x < 0.
y
N̂
−̂ T̂
Let W N̂ be the normal force exerted by the wire when the bead
is at position x. Note W > 0. Is dW
dx positive or negative?
4. A skateboarder is rolling on a frictionless, very tall parabolic ramp
with cross-section described by y = x2 . Given a boarder of mass
m with system energy E, what is the highest elevation the skater
reaches? How does this compare to a circular culvert?
5. A skateboarder of mass 100 kg is freely rolling in a frictionless circular
culvert of radius 5 m. If the skateboarder oscillates between vertical
heights of 0 and 3 m, what is the energy E of the system?
6. A skateboarder is rolling on a frictionless circular culvert of radius 5
m. What should their speed be when they’re at the bottom of the
culvert (y = 0) for them to make it all the way around?
7. A ball of mass 1 kg rolls down a track with the shape r(θ) = (3 cos θ, 5 sin θ, 4+
4 cos θ) for 0 ≤ θ ≤ π2 . Coordinates are measured in metres, and the
z axis is vertical (so the force due to gravity is −mg k̂.)

When θ = π/4, the particle has instantaneous velocity |v(t)| = 5
m/s. What is the normal force exerted by the track at that time?
Give your answer as a vector.
1
8. A bead of mass 9.8 kg slides down a wire in the shape of the curve
r(θ) = (sin θ, sin θ − θ), θ ≥ 0, with coordinates measured in metres.
The bead will break off the wire when the wire exerts a force of 100
N on the bead.
y
r(θ) = (sin θ, sin θ − θ)
13π
If the bead breaks off the wire at θ = 3 , how fast is the bead
moving at that point?
9. A skier is gliding down a hill. The hill can be described as r(t) =
(ln t, 1 − t), 1/e ≤ t ≤ e, with coordinates measured in kilometres.
How fast would the skier have to be moving in order to catch air?
10. A wire follows the arclength-parametrized path r(s) = (x(s), y(s)). A
bead, equipped with a jet pack, slides down the wire. The jet pack
can exert a variable force in a direction tangent to the wire,
U T̂.
Assuming the bead slides with constant speed dr dt
= c dr = c, find
ds
a simplified equation for U , the signed magnitude of the force exerted
by the jet pack.
Let the acceleration due to gravity be g, and let the mass of the
bead with its jet pack be m. Give U as a function of s.
Remark: most beads this author has seen did not have jet packs.
However, in modelling a frictionful4 system, friction acts as a force
that is directly opposing the direction of motion — much like our jet
pack.
11. A snowmachine is cautiously descending a hill in low gear. Its engine
provides a force M T̂ parallel to the direction of motion. The engine
provides whatever force is necessary to keep the snowmachine moving
at a constant speed, |v|. Its treads do not slip.
a Give a formula for M in terms of the mass m of the snowma-
chine, the acceleration due to gravity g, and the tangent vector
T̂ to the hill.
b Let T̂ point in the downhill direction. Do you expect M to be

positive or negative as the snowmachine moves downhill?
c Find M for the hill of shape y = 1 + cos x (measured in metres)
4 Frictionated? Frictiony? Befrictioned?
3π
at the point x = 4 for a snowmachine of mass 200 kg.
12. A skateboarder rolls along a culvert with elliptical cross-section de-
scribed by
r(θ) = (4 cos θ, 3(1 + sin θ)), 0 ≤ θ ≤ 2π,
with coordinates measured in metres.

a Give the height yS (in terms of m, g, and E) where the skater’s
speed is zero.
b Write an equation relating E, m, g, and yA , where yA is the
y-value where the skater would become airborne, i.e. where
W = 0. (You do not have to solve for yA explicitly.)
c Suppose the skater has speed 11 m/s at the bottom of the cul-
vert. Which of the following describes their journey: they make
it all the way around; they roll back and forth in the bottom
half; or they make it onto the ceiling, then fall off?
13. A frictionless roller-coaster track has the form of one turn of the cir-
cular helix with parametrization (a cos θ, a sin θ, bθ). A car leaves the
point where θ = 2π with zero velocity and moves under gravity to
the point where θ = 0. By Newton’s law of motion, the position r(t)
of the car at time t obeys
mr00 (t) = N r(t) − mg k̂

Here m is the mass of the car, g is a constant, −mg k̂ is the force due
to gravity and N r(t) is the force that the roller-coaster track applies
to the car
to keep the car on the track. Since the track is frictionless,
N r(t) is always perpendicular to v(t) = dr dt (t).
a Prove that E(t) = 12 m|v(t)|2 + mgr(t) · k̂ is a constant, indepen-

dent of t. (This is called “conservation of energy”.)
b Prove that the speed |v| at the point θ obeys |v|2 = 2gb(2π −θ).
c Find the time it takes to reach θ = 0.
1.8 Optional — Polar Coordinates

So far we have always written vectors in two dimensions in terms of the basis
vectors ı̂ı and ̂. This is not always convenient. For example, when working in
polar coordinates it is often convenient to use basis vectors r̂(θ), θ̂θ (θ) which
depend on the value of the current polar coordinate θ — though one usually
just writes r̂, θ̂θ , suppressing the dependence on θ from the notation. When one
is at the point with polar coordinates (r, θ), these basis vectors are defined by
Equation 1.8.1
θ̂θ r̂
̂ θ
r̂(θ) = cos θ ı̂ı + sin θ ̂
r
θ̂θ (θ) = − sin θ ı̂ı + cos θ ̂
θ
ı̂ı
p0, 0q
Note that this basis has two very nice properties.
1. |r̂(θ)| = |θ̂θ (θ)| = 1, r̂(θ) ⊥ θ̂θ (θ) (orthonormality)

dr̂ θ
dθ̂
2. dθ (θ) = θ̂θ (θ), dθ (θ) = −r̂(θ)
dr̂
That dθ (θ) is some scalar multiple of θ̂θ (θ) follows just from the fact that
|r̂(θ)| = 1.
|r̂(θ)| = 1 =⇒ r̂(θ) · r̂(θ) = 1

dr̂ 1 d
=⇒ r̂(θ) · (θ) = r̂(θ) · r̂(θ) = 0
dθ 2 dt
dr̂ dr̂
=⇒ (θ) ⊥ r̂(θ) =⇒ (θ) k θ̂θ (θ)
dθ dθ
θ
dθ̂
Similarly, that dθ (θ) is some scalar multiple of r̂(θ) follows just from the fact
that |θ̂ (θ)| = 1.
θ
Lemma 1.8.2 If we parametrize a curve by giving its polar coordinates1 r(t) , θ(t) ,

then
a the position vector of the point at time t is

r(t) = r(t) r̂ θ(t)
b and the velocity vector of the point at time t is

dr dθ
v(t) = (t) r̂ θ(t) + r(t) (t) θ̂θ θ(t)
dt dt
c and the acceleration vector of the point at time t is

2 dθ 2
d r
a(t) = 2
(t)−r(t) (t) r̂ θ(t)
dt dt
d2 θ

dr dθ
+ r(t) 2 (t)+2 (t) (t) θ̂θ θ(t)
dt dt dt
It is standard to suppress the arguments t and θ(t) and write, for example,
dr dθ
v= r̂ + r θ̂θ
dt dt
But it is important to remember that the arguments really are there.
Proof. The vector from the origin to the point whose polar coordinates are (r, θ)
is r = r r̂(θ). So if we parametrize a curve by giving the polar coordinates at
1 As usual r is the distance from the origin to the point and θ is angle between the
x-axis and the vector from the origin to the point. The symbols r, θ are the standard
mathematics symbols for the polar coordinates. Appendix A.7 gives another set of symbols
that is commonly used in the physical sciences and engineering.
time t,

r(t) = r(t) r̂ θ(t)
dr dr̂ dθ
v(t) = (t) r̂ θ(t) + r(t) θ(t) (t)
dt dθ dt
dr dθ
= (t) r̂ θ(t) + r(t) (t) θ̂θ θ(t)
dt dt
d2 r dr dr̂ dθ dr dθ d2 θ dθ 2 dθ̂θ
a(t) = 2 r̂ + + θ̂θ + r 2 θ̂θ + r
dt dt dθ dt dt dt dt dt dθ
h d2 r dθ 2 i h d2 θ dr dθ i
= −r r̂ + r 2 + 2 θ̂θ
dt2 dt dt dt dt

Example 1.8.3 As an example, consider a bead that is sliding on a frictionless
rod that has one end fixed at the origin and that is rotating about the origin
at a constant Ω rad/sec.
Ωt
Because the rod is frictionless, it is incapable of applying to the bead any
force parallel to the rod. So under Newton’s law, ma = F, the radial2 com-
ponent of the acceleration of the particle is exactly
zero. So, if the polar
coordinates of the bead at time t are r(t), θ(t) , then, by Lemma 1.8.2.c,
d2 r dθ 2
− r =0
dt2 dt
dθ
As the rod is rotating at Ω rad/sec, dt = Ω and
d2 r
− Ω2 r = 0
dt2
The general solution to this constant coefficient second order ordinary differ-
ential equation is3
r(t) = AeΩ t + Be−Ω t
where A and B are arbitrary constants that are determined by initial condi-
tions. Just as an example, if r(0) = 1 and r0 (0) = 0, then A + B = 1 and
AΩ − BΩ = 0, so that A = B = 21 and
1 Ωt
e + e−Ω t

r(t) =
2
If, again for example, θ(0) = 0, then θ(t) = Ωt and the bead follows the polar
coordinate curve
1
r(θ) = eθ + e−θ

2
Observe that r(θ) is 1 when θ = 0, increases as θ increases, and tends to ∞ as
θ → +∞. The curve is a spiral.

Example 1.8.4 Conic sections in polar coordinates. In this example,
we derive the equation of a general conic section in polar coordinates. A conic
section is the intersection of a plane with a cone. This is illustrated in the
figures below.
circle ellipse parabola

hyperbola
For our current purposes, it is convenient to use the equivalent4 (and often
used) definition that a conic section is the set of points P in the xy-plane
• whose distance from a fixed point F (called the focus of the conic)
• is a constant multiple ε ≥ 0 (called the eccentricity of the conic)
• of the distance from P to a fixed line L (called the directrix of the conic).
Choose a coordinate system with the focus F of the conic being the origin and
with the directrix L being x = p for some p > 0.
2 The θ̂
θ component of the acceleration just tells us how much normal force the rod is
applying to the bead to keep it on the rod.
3 A review of the technique used to find this solution is given in Appendix A.9. In any
2
event, it is easy to check that r(t) = AeΩ t + Be−Ω t really does obey ddt2r − Ω2 r = 0.
y L
P Q
r
θ
F x
p
If P has polar coordinates (r, θ), then P has x-coordinate r cos θ. The point
Q on the line L in the figure above has x-coordinate p. So the distance from P
to L, which is also the distance from P to Q, is p − r cos θ. The distance from
P to F is r. We require that the distance from P to F is ε times the distance
from P to L. So
εp
r = ε p − r cos θ ⇐⇒ r =
1 + ε cos θ
The numerator εp is usually renamed to ` giving the equation
`
r=
1 + ε cos θ

Example 1.8.5 Conic sections in polar coordinates, again. We’ll now
take the equation r = 1+ε`cos θ for a conic section in polar coordinates, from
the last example, and convert it to the more familiar Cartesian coordinates.
Just by the definition of polar coordinates

r 1 + ε cos θ = ` ⇐⇒ r = ` − εx
⇐⇒ x2 + y 2 = `2 − 2ε`x + ε2 x2
⇐⇒ (1 − ε2 )x2 + 2ε`x + y 2 = `2 (C)
Now consider separately four different cases, depending on the value of ε ≥ 0.

• If ε = 0, (C) reduces to
y
ℓ
2 2 2
x +y =`
x
which is of course a circle of radius `.

4 It is outside our scope to prove this equivalence.
• If 0 < ε < 1, completing the square in (C) gives

ε` 2 ε2 `2 `2
(1 − ε2 ) x + + y 2
= `2
+ =
1 − ε2 1 − ε2 1 − ε2
which is equivalent to
y
ε`
2
x+ 1−ε2 y2
`2
+ `2
=1 p´rM , 0q x
(1−ε2 )2 1−ε2
p0, ´rm q
`
and is of course an ellipse with semi-major axis rM = 1−ε2 and semi-
`
minor axis rm = √1−ε 2
.
• If ε = 1, (C) reduces to
y
pℓ{2, 0q
y 2 = `2 − 2`x x
p0, ´ℓq
which is of course a parabola.

• If ε > 1, the same computation as in the 0 < ε < 1 case gives
y
ε`
2
x− ε2 −1 y2
`2
− `2
=1 x
(ε2 −1)2 ε2 −1
and is of course a hyperbola.

1.8.1 Exercises
1. Consider the points
(x1 , y1 ) = (3, 0) (x2 , y2 ) = (1, 1) (x3 , y3 ) = (0, 1)
(x4 , y4 ) = (−1, 1) (x5 , y5 ) = (−2, 0)
For each 1 ≤ i ≤ 5,
• sketch, in the xy-plane, the point (xi , yi ) and
• find the polar coordinates ri and θi , with 0 ≤ θi < 2π, for the
point (xi , yi ).
2.
a Find all pairs (r, θ) such that

(−2, 0) = r cos θ , r sin θ
b Find all pairs (r, θ) such that

(1, 1) = r cos θ , r sin θ
c Find all pairs (r, θ) such that

(−1, −1) = r cos θ , r sin θ
3. Consider the points
(x1 , y1 ) = (3, 0) (x2 , y2 ) = (1, 1) (x3 , y3 ) = (0, 1)

(x4 , y4 ) = (−1, 1) (x5 , y5 ) = (−2, 0)
Also define, for each angle θ, the vectors
êr (θ) = cos θ ı̂ı + sin θ ̂ êθ (θ) = − sin θ ı̂ı + cos θ ̂
a Determine, for each angle θ, the lengths of the vectors êr (θ)
and êθ (θ) and the angle between the vectors êr (θ) and êθ (θ).
Compute êr (θ) × êθ (θ) (viewing êr (θ) and êθ (θ) as vectors in
three dimensions with zero k̂ components).
b For each 1 ≤ i ≤ 5, sketch, in the xy-plane, the point (xi , yi )
and the vectors êr (θi ) and êθ (θi ). In your sketch of the vectors,
place the tails of the vectors êr (θi ) and êθ (θi ) at (xi , yi ).
4. ∗. Match the following equations with the corresponding pictures.
Cartesian coordinates are (x, y) and polar coordinates are (r, θ).
(A) y (B) y
x x
(C) y (D) y
x x
(E) y (F) y
x x
(a) r = 2 + sin(4θ) (b) r = 1 + 2 sin(4θ)

(c) r=1 (d) r = 2 cos(θ), − π2 ≤ θ ≤ π
2
(e) r = eθ/10 + e−θ/10 (f) r=θ
5. Recall that a point with polar coordinates r and θ has x = r cos θ and
y = r sin θ. Let r = f (θ) be the equation of a plane curve in polar
coordinates. Find the curvature of this curve at a general point θ.
6. Find the curvature of the cardioid r = a(1 − cos θ).
1.9 Optional — Central Forces

One of the great triumphs of Newtonian mechanics was the explanation of
Kepler’s laws1 , which said
1. The planets trace out ellipses about the sun as focus.
2. The radius vector r sweeps out equal areas in equal times.
3. The square of the period of each planet is proportional to the cube of the
major axis of the planet’s orbit.
Newton showed that all of these behaviours follow from the assumption that
the acceleration a(t) of each planet obeys the law of motion ma = F where m
is the mass of the planet and
GM m
F=− r
r3
is the “gravitational force” applied on the planet by the sun. Here G is a
constant2 , called the “gravitational constant” or the “universal gravitational
constant”, M is the mass of the sun, r is the vector from the sun to the planet
and r = |r|.
1 The German astronomer Johannes Kepler (1571–1630) developed these laws during the
course of an attempt to relate the five extraterrestrial planets then known to the five Platonic
solids. He based the laws on a great number of careful measurements made by the Danish
Astronomer Tycho Brahe (1546–1601). Then Isaac Newton (English, 1642–1727) provided
the explanation in 1687. Kepler also wrote a paper entitled “On the Six-Cornered Snowflake”.
Tycho Brahe lost his nose in a sword duel and wore a prosthetic nose from then on. The
story is that Brahe died from a burst bladder that resulted from his refusing to leave the
dinner table before his host.
2 Its value is about 6.67408 × 10−11 m3 kg−1 sec−2 .
In this section, we’ll show that some of these properties follow from the
weaker assumption that the acceleration a(t) of each planet obeys the law of
motion ma = F with F being a central force. That is, the assumption that
F is parallel to r. The verification that the other properties follow from the
specific form of the gravitational force, proportional to r−2 , will be delayed
until the optional §1.10.
So, in this section, we assume that we have a parametrized curve r(t) and
that this curve obeys
d2 r
m 2 (t) = F r(t)
dt
where, for all r ∈ R3 , F(r) is parallel to r. We shall show that
1. The path r(t) lies in a plane through the origin and that
2. the radius vector r sweeps out equal areas in equal times.
We’ll start by trying to guess what the plane is. Pretend that we know that
r(t) lies in a fixed plane through the origin. Then v(t) = dr
dt (t) lies in the same
plane and r(t) × v(t) is perpendicular to the plane. If our path really does
lie in a fixed plane, r(t) × v(t) cannot change direction — it must always be
parallel to the normal vector to the plane. So let’s define
Ω (t) = r(t) × v(t)
and check how it depends on time. By the product rule,

Ω
dΩ d
(t) = r(t) × v(t) = v(t) × v(t) + r(t) × a(t)
dt dt
1
= r(t) × F r(t)
m
=0 (A)

because r(t) and F r(t) are parallel. So Ω (t) is3 in fact independent of t. It
is a constant vector that we’ll just denote Ω .
As r(t) × v(t) = Ω , we have that r(t) is always perpendicular to Ω and
r(t) · Ω = 0
• If Ω 6= 0, this is exactly the statement that r(t) always lies in the plane
through the origin with normal vector Ω .
• If Ω = 0, then r(t) is always parallel to v(t) and there is some function
α(t) such that
dr
(t) = v(t) = α(t) r(t)
dt
This is a first order, linear, ordinary differential equation that we can
solve by using an integrating factor. Set
Z t
β(t) = α(t) dt
0
Then
dr dr
(t) = α(t) r(t) ⇐⇒ e−β(t) (t) − α(t)e−β(t) r(t) = 0
dt dt
3 Physicists call m Ω (t) the angular momentum at time t and refer to (A) as (an example
of) conservation of angular momentum. Conservation of angular momentum is exploited in
gyro-compasses and by ice skaters (to spin faster/slower).
d −β(t)
⇐⇒ e r(t) = 0
dt
⇐⇒ e−β(t) r(t) = r(0)
⇐⇒ r(t) = eβ(t) r(0)
so that r(t) lies on a line through the origin. This makes sense — the
particle is always moving parallel to its radius vector.
This completes the verification that r(t) lies in a plane through the origin.
Now we show that the radius vector r(t) sweeps out equal areas in equal
times. In other words, we now verify that the rate at which r(t) sweeps out
area is independent of time. To do so we rewrite the statement that |r(t)×v(t)

is constant in polar coordinates. Writing r(t) = r(t)r̂ θ(t) and then applying
Lemma 1.8.2.b gives that
dr dθ dθ
θ̂θ = r2

constant = r × v = rr̂ × r̂ + r
dt dt dt
since |r̂ × r̂| = 0, |r̂ × θ̂θ | = 1
is constant. It now suffices to observe that r(t)2 dθ

dt (t) is exactly twice the rate
at which r(t) sweeps out area. To see this, just look at the figure below. The
shaded area is essentially a wedge of a circular disk of radius r. (If r(t) were
independent of t, it would be exactly a wedge of a circular disk.) Its area is
dθ
the fraction 2π of the area of the full disk, which is
dθ 1
πr2 = r2 dθ rpt ` dtq
2π 2
dθ
rptq
1.9.1 Exercises
1. ∗. Let r(t) = x(t)ı̂ı + y(t) ̂ + z(t) k̂ be the position of a particle at
time t . Suppose the motion of the particle satisfies the differential
2
equation ddt2r = f (r)r where r = |r| .
a Suppose f (r) is an arbitrary function of r . Prove or disprove
each of the following statements.
(i) The motion of the particle is planar.
(ii) The path of the particle sweeps out equal areas in equal
times.
b Find all forms of f (r) for which the motion of the particle always
lies on a straight line.
c Give a specific form of f (r) for which the motion of the particle
could lie on an ellipse.
2. ∗. An object moves along a curve in the xy-plane having polar equa-
1
tion r = θ+α (where α is a constant) under the influence of a central
force so that the object has no transverse acceleration.
a Verify that r2 θ̇ = h remains constant as the object moves.
b Express the magnitude of the acceleration of the object as a

function of r and h.
1.10 Optional — Planetary Motion

We now return to the claim, made in §1.9 on central forces, that if r(t) obeys
Newton’s inverse square law
d2 r GM GM
2
= − 3 r = − 2 r̂
dt r r
then the curve obeys Kepler’s laws
1. r(t) runs over an ellipse having one focus at the origin and
2. r(t) sweeps out equal areas in equal times and
3. the square of the period is proportional to the cube of the major axis of
the ellipse.
We just showed, in §1.9, that the fact that − GMr 3 r is parallel to r implies that
r(t) lies in a plane through the origin and sweeps out equal area in equal times.
We now verify the remaining Kepler laws.
We start by just rewriting Newton’s laws above in polar coordinates. We
saw in Lemma 1.8.2.c, that if we write r(t) = r(t) r̂(t), then
2 !
d2 r d2 r
2
dθ d θ dr dθ
= −r r̂ + r 2 + 2 θ̂θ
dt2 dt2 dt dt dt dt
GM GM
=− r = − 2 r̂
r3 r
The r̂ and θ̂θ components of this equation are

2
d2 r dθ GM
−r =− 2
dt2 dt r
2
d θ dr dθ
r 2 +2 =0
dt dt dt
The second of these two equations only tells us that
2
d 2 dθ d θ dr dθ dθ
r =r r 2 +2 = 0 =⇒ r2 = h, a constant
dt dt dt dt dt dt
dθ h
which we already knew. Substituting dt = r2 into the first equation gives
Equation 1.10.1
d2 r h2 GM
2
− 3 =− 2
dt r r
This equations contains a lot of 1r ’s. So let’s set u = 1r . Furthermore, for
the first of Kepler’s laws, we really want r as a function of θ rather than t. So
let’s make u a function of θ and write
1
r(t) =
u(θ(t))
Then
dr 1 du dθ du dθ h
= 2 = hu2

(t) = − 2 θ(t) (t) = −h θ(t) since
dt u dθ dt dθ dt r
d2 r d2 u dθ 2
2 d2 u
2
(t) = −h 2
θ(t) (t) = −h u θ(t) 2
θ(t)
dt dθ dt dθ
and our equation becomes
Equation 1.10.2
d2 u d2 u GM
−h2 u2 2
− h2 u3 = −GM u2 or +u= 2
dθ dθ2 h
This is a second order, linear, ordinary differential equation with constant
coefficients. Recall1 that the general solution of such an equation is the sum
of a “particular solution” (i.e. any one solution, which in this case we can take
to be the constant function GM h2 ) plus the general solution of the homogeneous
equation u0 + u = 0, which one often writes as
A cos θ + B sin θ
with A and B arbitrary constants. In this particular application it is more

convenient to write the solution in a different, standard but less commonly
used, form. Namely, we can use the triangle
C B
α
A
to write A = C cos α and B = C sin α so that the general solution of the
homogeneous equation u0 + u = 0 becomes
C cos α cos θ + C sin α sin θ = C cos(θ − α)
with C and α being arbitrary constants. So the general solution to 1.10.2 is

GM
u(θ) = + C cos(θ − α)
h2
and the general solution to 1.10.1 is
1
r(t) = GM
h2 + C cos(θ(t) − α)
The angle α just shifts the zero point of our coordinate θ. By rotating our
coordinate system by α, we can arrange that α = 0 and then
1 ` h2 Ch2
r(t) = GM
= with ` = , ε=
h2 + C cos(θ(t)) 1 + ε cos θ GM GM
As we saw in Example 1.8.4, this is exactly the equation of a conic section with
eccentricity ε.
That leaves only the last of Kepler’s laws, relating the period to the semi-
major axis. As we are talking about planets, whose orbits remain bounded, our
conic section must be a circle or ellipse, rather than a parabola or hyperbola.
1 See Appendix A.9.
Looking back at Example 1.8.5, we see that the semi-major and semi-minor
axes of our ellipse are
` `
a= b= √
1 − ε2 1 − ε2
The period T of our orbit is just the length of time it takes the radius vector
r(t) to sweep out the area of the ellipse2 , which is πab. As the rate at which
h
the radius vector is sweeping out area is 21 r2 dθ
dt = 2 , we have
πab 2 4π 2 a2 b2 4π 2 a2 b2 4π 2 3 b2
T2 = = = = a since ` =
h/2 h2 GM ` GM a
1.11 Optional — The Astroid

Imagine a ball of radius a/4 rolling around the inside of a circle of radius a.
The curve traced by a point P painted on the inner circle (that’s the blue curve
in the figures below) is called an astroid1 . We shall find its equation.
P P
P
Define the angles θ and φ as in the figure in the left below.

2 You probably computed the area of an ellipse in first year calculus. If not, you should
be able to do it now in a few lines.
1 The name “astroid” comes from the Greek word “aster”, meaning star, with the suffix
“oid” meaning “having the shape of”. The curve was first discussed by Johann Bernoulli in
1691–92.
̂ Q ̂
φ θ
θ ı P
ı̂ ı̂ı
O O
That is
• the vector from the centre, O, of the circle of radius a to the centre, Q,
of the ball of radius a/4 is 34 a cos θ, sin θ and
• the vector from the
1
centre, Q, of the ball of radius a/4 to the point P is
4 a cos φ, − sin φ
As θ runs from 0 to π2 , the point of contact between the two circles travels
through one quarter of the circumference of the circle of radius a, which is a
distance 14 (2πa), which, in turn, is exactly the circumference of the inner circle.
Hence if φ = 0 for θ = 0 (i.e. if P starts on the x-axis), then for θ = π2 , P
is back in contact with the big circle at the north pole of both the inner and
π
outer circles. That is, φ = 3π2 when θ = 2 . (See the figure on the right above.)
So φ = 3θ and P has coordinates
3 1 a
a cos θ, sin θ + a cos φ, − sin φ = 3 cos θ + cos 3θ, 3 sin θ − sin 3θ
4 4 4
As, recalling your double angle, or even better your triple angle, trig identities,
cos 3θ = cos θ cos 2θ − sin θ sin 2θ

= cos θ[cos2 θ − sin2 θ] − 2 sin2 θ cos θ
= cos θ[cos2 θ − 3 sin2 θ]
sin 3θ = sin θ cos 2θ + cos θ sin 2θ
= sin θ[cos2 θ − sin2 θ] + 2 sin θ cos2 θ
= sin θ[3 cos2 θ − sin2 θ]
we have
3 cos θ + cos 3θ = cos θ[3 + cos2 θ − 3 sin2 θ] = cos θ[3 + cos2 θ − 3(1 − cos2 θ)]
= 4 cos3 θ
3 sin θ − sin 3θ = sin θ[3 − 3 cos2 θ + sin2 θ] = sin θ[3 − 3(1 − sin2 θ) + sin2 θ]
= 4 sin3 θ
and the coordinates of P simplify to
x(θ) = a cos3 θ y(θ) = a sin3 θ
Oof! As x2/3 + y 2/3 = a2/3 cos2 θ + a2/3 sin2 θ , the path traced by P obeys
the equation
x2/3 + y 2/3 = a2/3
which is surprisingly simple, considering what we went through to get here.

There remains the danger that there could exist points (x, y) obeying the
equation x2/3 +y 2/3 = a2/3 that are not of the form x = a cos3 θ, y = a sin3 θ for
any θ. That is, there is a danger that the parametrized curve x = a cos3 θ, y =
a sin3 θ covers only a portion of x2/3 + y 2/3 = a2/3 . We now show that the
parametrized curve x = a cos3 θ, y = a sin3 θ in fact covers all of x2/3 + y 2/3 =
a2/3 as θ runs from 0 to 2π. √ 2 √ 2
First, observe that x2/3 = 3 x ≥ 0 and y 2/3 = 3 y ≥ 0. Hence,
if (x, y) obeys x2/3 + y 2/3 = a2/3 , then necessarily 0 ≤ x2/3 ≤ a2/3 and so
−a ≤ x ≤ a. As θ runs from 0 to 2π, a cos3 θ takes all values between −a and
a and hence takes all possible values of x. For each x ∈ [−a, a], y takes two
3/2
values, namely ±[a2/3 − x2/3 ] . If x = a cos3 θ0 = a cos3 (2π − θ0 ), the two
corresponding values of y are precisely a sin3 θ0 and −a sin3 θ0 = a sin3 (2π −
θ0 ).
1.12 Optional — Parametrizing Circles

We now discuss a simple strategy for parametrizing circles in three dimensions,
starting with the circle in the xy-plane that has radius ρ and is centred on the
origin. This is easy to parametrize:
z
k̂
r(t) = ρ cos tı̂ı + ρ sin t ̂

y 0 ≤ t < 2π
ρ̂
ρı̂ı
x
Now let’s move the circle so that its centre is at some general point c. To
parametrize this new circle, which still has radius ρ and which is still parallel
to the xy-plane, we just translate by c:
k̂
z
c
ρ̂ r(t) = c + ρ cos tı̂ı + ρ sin t ̂
ρı̂ı 0 ≤ t < 2π
y
x
Finally, let’s consider a circle in general position. The secret to parametriz-
ing a general circle is to replace ı̂ı and ̂ by two new vectors ı̂ı0 and ̂0 which
a are unit vectors,
b are parallel to the plane of the desired circle and

c are mutually perpendicular.
z k̂1
ρ̂1
c
r(t) = c + ρ cos tı̂ı0 + ρ sin t ̂0
ρı̂ı1
0 ≤ t < 2π
y
x
To check that this is correct, observe that
• r(t) − c is parallel to the plane of the desired circle because both ı̂ı0
and ̂0 are parallel to the plane of the desired circle and r(t) − c =
ρ cos tı̂ı0 + ρ sin t ̂0
• r(t) − c is of length ρ for all t because
|r(t) − c |2 = (r(t) − c ) · (r(t) − c )

= (ρ cos tı̂ı0 + ρ sin t ̂0 ) · (ρ cos tı̂ı0 + ρ sin t ̂0 )
= ρ2 cos2 t ı̂ı0 · ı̂ı0 + ρ2 sin2 t ̂0 · ̂0 + 2ρ cos t sin t ı̂ı0 · ̂0
= ρ2 (cos2 t + sin2 t) = ρ2
since ı̂ı0 · ı̂ı0 = ̂0 · ̂0 = 1 (ı̂ı0 and ̂0 are both unit vectors) and ı̂ı0 · ̂0 = 0 (ı̂ı0
and ̂0 are perpendicular).
To find such a parametrization in practice, we need to find the centre c of the
circle, the radius ρ of the circle and two mutually perpendicular unit vectors,
ı̂ı0 and ̂0 , in the plane of the circle. It is often easy to find at least one point
p on the circle. Then we can take ı̂ı0 = |p−c| p−c
. It is also often easy to find a
unit vector, k̂0 , that is normal to the plane of the circle. Then we can choose
̂ 0 = k̂0 × ı̂ı0 . We’ll illustrate this now.
Example 1.12.1 Let C be the intersection of the sphere x2 + y 2 + z 2 = 4 and
the plane z = y.
• The intersection of any plane with any sphere is a circle. The plane in
question passes through the centre of the sphere, so C has the same centre
and same radius as the sphere. So C has radius 2 and centre (0, 0, 0).
• Notice that the point (2, 0, 0) satisfies both x2 +y 2 +z 2 = 4 and z = y and
so is on C. We may choose ı̂ı0 to be the unit vector in the direction from
the centre (0, 0, 0) of the circle towards (2, 0, 0). Namely ı̂ı0 = (1, 0, 0).
• Since the plane of the circle is z − y = 0, the vector ∇ (z − y) = (0, −1, 1)

is perpendicular to the plane of C. So we may take k̂0 = √12 (0, −1, 1).
• Then ̂0 = k̂0 × ı̂ı0 = √1 (0, −1, 1)

2
× (1, 0, 0) = √1 (0, 1, 1).
2
Substituting in c = (0, 0, 0), ρ = 2, ı̂ı0 = (1, 0, 0) and ̂0 = √1 (0, 1, 1)

2
gives
1
r(t) = 2 cos t (1, 0, 0) + 2 sin t √ (0, 1, 1)
2
k̂1
̂1 sin t sin t
= 2 cos t, √ , √
2 2
y 0 ≤ t < 2π
1
ı̂ı
x
√ √
To check this, note that x = 2 cos t, y = 2 sin t, z = 2 sin t satisfies both
x2 + y 2 + z 2 = 4 and z = y.
Example 1.12.2 Let C be the circle that passes through the three points
(3, 0, 0), (0, 3, 0) and (0, 0, 3).
• All three points obey x + y + z = 3. So the circle lies in the plane
x + y + z = 3. We guess, by symmetry, or by looking at the figure below,
that the centre of the circle is at the centre of mass of the three points,
which is 13 [(3, 0, 0)+(0, 3, 0)+(0, 0, 3)] = (1, 1, 1). We must check this and
can do so by checking that (1, 1, 1) is equidistant from the three points:
√
(3, 0, 0) − (1, 1, 1) = (2, −1, −1) = 6
√
(0, 3, 0) − (1, 1, 1) = (−1, 2, −1) = 6
√
(0, 0, 3) − (1, 1, 1) = (−1, −1, 2) = 6
This tells us both that (1, 1, 1) is indeed the centre (as only the centre is
equidistant
√ from any three distinct points on a circle) and that the radius
of C is 6.
• We may choose ı̂ı0 to be the unit vector in the direction from the centre
(1, 1, 1) of the circle towards (3, 0, 0). Namely ı̂ı0 = √16 (2, −1, −1).
• Since the plane of the circle is x + y + z = 3, the vector ∇ (x + y +

z) = (1, 1, 1) is perpendicular to the plane of C. So we may take k̂0 =
√1 (1, 1, 1).
3
• Then
1 1
̂0 = k̂0 × ı̂ı0 = √ (1, 1, 1) × (2, −1, −1) = √ (0, 3, −3)
18 18
1
= √ (0, 1, −1)
2
√
Substituting in c = (1, 1, 1), ρ = 6, ı̂ı0 = √1 (2, −1, −1)
6
and ̂0 = √1 (0, 1, −1)
2
gives
√ 1 √ 1
r(t) = (1, 1, 1) + 6 cos t √ (2, −1, −1) + 6 sin t √ (0, 1, −1)
6 2
√ √
= 1 + 2 cos t, 1 − cos t + 3 sin t, 1 − cos t − 3 sin t
x y
To check this, note that r(0) = (3, 0, 0), r 2π 4π

3 = (0, 3, 0) and r 3 =
√ √
(0, 0, 3) since cos 2π
3 = cos 4π
3 = − 1
2 , sin 2π
3 = 2
3
and sin 4π
3 =− 2 .
3

Chapter 2
Vector Fields
2.1 Definitions and First Examples

In the last chapter, we studied vector valued functions of a single variable, like,
for example, the velocity v(t) of a particle at time t. Suppose however that we
are interested in a fluid. There is a, possibly different, velocity at each point in
the fluid. So the velocity of a fluid is really a vector valued function of several
variables. Such a function is called a vector field.
Definition 2.1.1
a A vector field in the plane is a rule which assigns to each point (x, y) in
a subset, D, of the xy-plane, a two component vector v(x, y).
b A vector field in space is a rule which assigns to each point (x, y, z) in a
subset of R3 , a three component vector v(x, y, z).
♦
Here are two typical applications that naturally involve vector fields.
• If v(x, y, z) is the velocity of a moving fluid at position (x, y, z), then v
is called a velocity field.
• If F(x, y, z) is the force at position (x, y, z), then F is called1 a force
field.
Example 2.1.2 The Point Source. Imagine
• The whole world is filled with an incompressible fluid. Call it water.
• Somehow you find a way to produce still more water at the origin. Say
you create 4πm litres per second.
• This forces the water to flow outward. Let’s suppose that it flows sym-
metrically outward from the origin.
Let’s find the resulting vector field v(x, y, z). As the flow is to be symmetric,
the velocity of the water at the point (x, y, z)
• has to be pointing radially outward from the origin. That is, the direction
of the velocity vector v(x, y, z) has to be the unit radial vector
xı̂ı + y̂ + z k̂
r̂(x, y, z) = p
x2 + y 2 + z 2
1 No, force fields are not only a sci-fi trope. Gravity is an example of a force field.
73
CHAPTER 2. VECTOR FIELDS 74
• The magnitude of the velocity, i.e. the speed |v(x, y, z)| of the water, has
to depend only on the distance from the origin. That is, the speed can
only be some function of
p
r(x, y, z) = x2 + y 2 + z 2
Thus the velocity field is of the form

v(x, y, z) = v r(x, y, z) r̂(x, y, z)
We just have to determine the function v(r). Fix any r > 0 and concentrate
on the sphere x2 + y 2 + z 2 = r2 . It is sketched in red in the figure below.
vprq dt thick
During a very short time interval dt seconds, 4πm dt litres of water is created
at the origin (which is the red dot). As the water is incompressible, 4πm dt
litres of water must exit through the sphere during the same time interval to
make room for the newly created water.
But, at the surface of the sphere the water is flowing radially outward with
speed v(r). So during the time interval in question the water near the surface
of the sphere moves outward a distance v(r) dt, andp in particular the water
that was in the thin spherical shell r − v(r) dt ≤ x2 + 2 2
py + z ≤ r at
the beginning of the time interval exits through the sphere x2 + y 2 + z 2 = r
during the time interval. The shell is sketched in gray in the figure above. The
volume of water in the gray shell is essentially the surface area of the shell,
which is 4πr2 , times the thickness of the shell, which is v(r) dt. So, equating
the volume of water created inside the sphere with the volume of water that
exited the sphere,
4πm m
4πm dt = (4πr2 ) v(r) dt =⇒ v(r) =

2
= 2
4πr r
Thus our vector field is
m
v(x, y, z) = r̂(x, y, z)
r(x, y, z)2
If the world were two, rather than three dimensional2 , and the source created
2πm litres per second, the same argument leads to
2πm m
2πm dt = (2πr) v(r) dt =⇒ v(r) = =
2πr r
and to the vector field
m p xı̂ı + y̂
v(x, y) = r̂(x, y) r(x, y) = x2 + y 2 r̂(x, y) = p
r(x, y) x2 + y 2
To get a mental image of what this field looks like, imagine sketching, for each
m
point (x, y), the vector r(x,y) r̂(x, y) with its tail at (x, y). Note that the vector
m
r(x,y) r̂(x, y)
• points radially outward and

m
• has length r(x,y) which
◦ depends only on r = |(x, y)| and

◦ is very long when (x, y) is near the origin and
1
◦ decreases in length like r as r = |(x, y)| increases.
Here is a sketch of a bunch of such vectors.

y
vector field ~v “ m rr̂

Figure 2.1.3
Note that as |(x, y)| → 0, the magnitude of the velocity |v(x, y)| → ∞.
This is a consequence of our idealized assumption that we are producing water
at a single point (the origin).
Example 2.1.4 The Vortex. In this example, we sketch the vector field

v(x, y) = Ω − yı̂ı + x̂
where Ω is just a strictly positive constant. We give an efficient procedure

for getting a rough sketch, which still provides a pretty realistic picture of the
vector field, and which also generalises to other vector fields. First concentrate
on the horizontal component ı̂ı · v(x, y) of the vector field and determine in
which part of the xy-plane it is zero, in which part it is positive and in which
part it is negative.

= 0 if y = 0

ı̂ı · v(x, y) = −Ωy < 0 if y > 0


> 0 if y < 0
2 You might want to think about what happens in d dimensions for general d.
Next repeat with the vertical component.


= 0 if x = 0

̂ · v(x, y) = Ωx < 0 if x < 0


> 0 if x > 0
This naturally divides the xy-plane into nine parts according to whether each
of the components is positive, 0 or negative —

• ı̂ı · v > 0 and ̂ · v > 0 in (x, y) ∈ R2 y < 0, x > 0

• ı̂ı · v > 0 and ̂ · v = 0 in (x, y) ∈ R2 y < 0, x = 0

• ı̂ı · v > 0 and ̂ · v < 0 in (x, y) ∈ R2 y < 0, x < 0

• ı̂ı · v = 0 and ̂ · v > 0 in (x, y) ∈ R2 y = 0, x > 0
• and so on
Now think of v(x, y) as being the velocity at (x, y) of a flowing fluid.

• Look at the first
bullet point2 above.
It says that in the first of the nine
parts, namely (x, y) ∈ R y < 0, x > 0 , which is the fourth
quadrant, the horizontal component ı̂ı · v > 0 signifying that the fluid is
flowing rightwards. Indicate this in the sketch by drawing a rightward
pointing horizontal arrow at some generic point in the middle of the
fourth quadrant. (It’s the blue arrow in the figure below.) The vertical
component ̂ · v > 0 signifying that the fluid is also moving upwards.
Indicate this in the sketch by drawing an upward pointing vertical arrow
at the same generic point in the fourth quadrant. (It’s the red arrow in
the figure below.)
y
• Next, look at the second bullet

point above.
It says that on the second
of the nine parts, namely (x, y) ∈ R2 y < 0, x = 0 , which is the
bottom half of the y-axis, the horizontal component ı̂ı · v > 0, signifying
that the fluid is moving rightwards. Indicate this in the sketch by drawing
a rightward pointing horizontal arrow at some generic point in the middle
of the bottom half of the y-axis. (It’s the second blue arrow in the figure
below.) The vertical component ̂ · v = 0 signifying that the fluid has
no vertical motion at all. Indicate this in the sketch by not drawing any
vertical arrow on the bottom half of the y-axis.
• and so on
By the time we have looked at all nine regions we will have built up the following
sketch.
y
Figure 2.1.5
From this sketch we see that, for example, in the first quadrant,
• the fluid is moving upwards and to the left and
• the fluid crosses the x-axis vertically (so that close to the x-axis, the
arrows will be almost vertical) and
• the fluid crosses the y-axis horizontally (so that close to the y-axis, the
arrows will be almost horizontal) and
• there is one point, namely (0, 0), where the vector field is exactly zero.
It’s the black dot in the centre of the figure above. Furthermore v(x, y) =
Ω(−yı̂ı + x̂) is smaller when (x, y) is closer to (0, 0) and v(x, y) is larger
when (x, y) is farther from (0, 0),
Putting all of this accumulated wisdom together, we come up with this better
sketch of the vector field.
vector field v “ Ωpýı̂ı ` x̂q

Figure 2.1.6
This shows the field swirling around the origin in a counterclockwise direc-
tion. Hence the name “vortex”.
Example 2.1.7 The Undamped Nonlinear Pendulum. In this example,
we illustrate another way in which vector fields arise. Model a pendulum by a
mass m that is connected to a hinge by an idealized rod that is massless3 and
of fixed length `. Denote by θ the angle
θ
ℓ
τ
mg
between the rod and vertical. The forces acting on the mass are
• gravity and
• the tension in the rod, whose magnitude, τ , automatically adjusts itself
so that the distance between the mass and the hinge is fixed at `.
In the optional4 Section 2.5, we show that the angle θ(t) obeys the second order
nonlinear5 differential equation
d2 θ g
+ sin θ = 0
dt2 `
It is often much more convenient to deal with first order, rather than second
order, differential equations. The second order pendulum equation above may
be reformulated6 as a system of first order ordinary differential equations, by

the simple expedient of defining
x(t) = θ(t) y(t) = θ0 (t)
So x(t) is the angle at time t and y(t) is the angular velocity at time t. Then,
x0 (t) = θ0 (t) = y(t)

g
y 0 (t) = θ00 (t) = − sin x(t)
`
Usually, one does not write in the (t) dependence explicitly.
x0 = y
g
y 0 = − sin x
`
The right hand sides form the vector field
g
v (x, y) = y , − sin x
`
We can sketch this vector field, just as we sketched the vector field of Example
2.1.4. Noting that the horizontal component

= 0 if y = 0

ı̂ı · v(x, y) = y > 0 if y > 0


< 0 if y < 0
and the vertical component.


g = 0 if x = 0, ±π, ±2π, · · ·

̂ · v(x, y) = − sin x > 0 if − π < x < 0, π < x < 2π, etc.
` 

< 0 if 0 < x < π, 2π < x < 3π, etc.
we have
• rightward motion7 when y > 0

• leftward motion when y < 0
• downward motion when 0 < x < π, 2π < x < 3π, · · · and
• upward motion when −π < x < 0, π < x < 2π, · · ·.

This gives us the collection of arrows in the figure
y“0
x “ ´π x“0 x“π x “ 2π
Our full sketch will be less cluttered if we make all arrows the same length.
This gives
y
direction field for x1 “ y, y 1 “ ´2 sin x

which is a sketch of what is called the direction field of our vector field (see
below).
In the next section, we’ll learn how to use vector field sketches to sketch
solution trajectories.
Definition 2.1.8 The direction field of a vector field v(x, y, z) is the vector
field ( v(x,y,z)
if v(x, y, z) 6= 0
V(x, y, z) = |v(x,y,z)|
0 if v(x, y, z) = 0
♦
2.1.1 Exercises
1. Below is a sketch of the vector field v(x, y).
3 While we are idealizing, let’s put everything in a vacuum.
4 In the optional Section 2.5 we also include frictional forces. In this example, we do not,
so set the β of Section 2.5 to zero here.
5 It is common, when considering only small amplitude oscillations, to approximate sin θ
by θ. This converts our nonlinear differential equation into a linear differential equation.
6 This “hack” generalizes easily and is commonly used when generating, by computer,
approximate solutions to higher order ordinary differential equations.

7 Note that this is rightward motion of the point (x, y), not of the pendulum itself.
Find the regions where the x-coordinates and y-coordinates are

positive, negative, and zero:

> 0
 when
v(x, y) · ı̂ı = 0 when


<0 when

> 0
 when
v(x, y) · ̂ = 0 when


<0 when
You may assume that v(x, y) behaves as expected at the points you
don’t see. That is, the samples are representative of a smooth, con-
tinuous vector-valued function. You may also assume the tick marks
on the axes correspond to unit distances.
2. Below is a sketch of the vector field v(x, y).
Find the regions where the x-coordinates and y-coordinates are

positive, negative, and zero:

> 0
 when
v(x, y) · ı̂ı =0 when


<0 when

> 0
 when
v(x, y) · ̂ = 0 when


<0 when
You may assume that the samples shown are representative of the
general behaviour of v(x, y). You may also assume the tick marks on
the axes correspond to unit distances.
3. A platform with many small conveyor belts is aligned on a coordinate
plane. Every conveyor belt moves an object on top of it in the direction
of the origin, and a conveyor belt at position (x, y) causes an object on
top of it to move with speed |y|. Assume the objects do not interfere
with one another.
Give a vector-valued formula for the velocity of an object at posi-
tion (x, y).
4. Let F = P ı̂ı + Q ̂ be the two-dimensional vector field sketched below.
x
∂Q ∂Q
Determine the signs of P , Q, ∂x and ∂y at the point A.
5. Imagine that the vector field v(x, y) = xı̂ı + y ̂
is the velocity field of a moving fluid.
a At time 0 you drop a twig into the fluid at the point (1, 1). What
is the approximate position of the twig at time t = 0.01?
b At time 0 you drop a twig into the fluid at the point (0, 0). What
is the position of the twig at time t = 0.01?
c At time 0 you drop a twig into the fluid at the point (0, 0). What
is the position of the twig at time t = 10?
6. Imagine that the vector field v(x, y) = 2xı̂ı − ̂
is the velocity field of a moving fluid. At time 0 you drop a twig
into the fluid at the point (0, 0). What is the position of the twig at
time t = 10?
7. A platform with many small conveyor belts is aligned on a coordinate
plane. Every conveyor belt moves an object on top of it in the direction
of the origin, and a conveyor belt at position (x, y) causes an object
on top of it to move with speed y. Assume the objects do not interfere
with one another.
Give a vector-valued formula for the velocity of an object at posi-
tion (x, y).
8. Friendly bees fly towards your face from all directions. The speed of
each bee is inversely proportional to its distance from your face. Find
a vector field for the velocity of the swarm.
9. Sketch the vector field v(x, y) = (x2 , y).
p
10. Sketch the direction field of v(x, y) =
p
x2 + y 2 , (x − 1)2 + (y − 1)2 .
11. Sketch the direction field of v(x, y) = (x2 + xy, y 2 − xy).

" #
1/3 1/3
12. Sketch the vector field v(x, y) = p (x, y) + p (x − 1, y) .
x2 + y 2 (x − 1)2 + y 2
13. Sketch each of the following vector fields, by drawing a figure like
Figure 2.1.3.
a v(x, y) = xı̂ı + y ̂.
b v(x, y) = 2xı̂ı − ̂.
c v(x, y) = √y ı̂ı−x
2
̂
2
.
x +y
14. A body of mass M exerts a force of magnitude GM D 2 on a particle of

unit mass distance D away from itself, where G is a physical constant.
The force acts in the direction from the particle to the body.
Suppose a mass of 5 kg sits at position (0, 0), a mass of 3 kg

sits at position (2, 3), and a mass of 7 kg sits at position (4, 0) on a
coordinate plane. Give the vector field f (x, y) of the net gravitational
force exerted on a unit mass at position (x, y).
15.
a. A pole leans against a vertical wall. The pole has length 2, and
it touches the wall at height H = 1. The pole slides down,
still touching the wall, with its height decreasing at a rate of
dH
dt = 0.5.
Find a vector function v : [0, 2] → R2 for the velocity, when

H = 1, of a point on the pole that is p units from the lower end,
using the coordinate system from the sketch above.
b. The frame of an umbrella is constructed by attaching straight,

rigid poles to a common centre. The poles are all the same
length, so they form radii of a circle.
The frame is lifted from the centre of the circle. The edges of
the frame drag on the ground, keeping the frame in the shape
of a right circular cone that is becoming taller and thinner.
Suppose the length of each pole is 2 metres, and the centre of

the frame is being lifted at a rate of 50 cm/s. Give a vector field
for the velocity V(x, y, z) of a point (x, y, z) on the frame when
its centre is 1 metre above the ground.
Let the ground have height z = 0, and let the centre of the frame
sit directly above the origin.
2.2 Optional — Field Lines

Suppose that we drop a tiny stick into a river1 with the velocity field of the
flowing water being v(x, y). We are assuming, for simplicity, that the velocity
field does not depend2 on time t. The stick will move along with the water3 .
When the stick is at r, its velocity will be the same as the velocity of the water
at r, which is v(r). Thus if the stick is at r(t) at time t, we will have
dr
= v r(t)
dt
The stick will trace out a path, parametrized by r(t).
Definition 2.2.1 A path that is parametrized by a function r(t) that obeys
dr
= v r(t)
dt
is called a
• field line or integral curve (for general vector fields) or a
• stream line or flow line (when the vector field v is being thought of as a
velocity field) or a
• line of force (when the vector field v is being thought of as a force field)
of the vector field v. ♦
Example 2.2.2 Flow Line Sketch for the Vortex of Example 2.1.4.
Consider the vortex vector field, v(x, y) = Ω −yı̂ı +x̂ of Example 2.1.4. Once
we sketched the vector field, as in Figure 2.1.6, or even made the “skeleton
sketch” of Figure 2.1.5, we can get rough idea of what the stream lines look
like just by following the arrows. For example, suppose that we start a stream
line (i.e. drop the stick into the stream) on the positive x-axis. Looking at
Figure 2.1.5, which is repeated here,
y
the stick
• starts by moving in the +y direction, i.e. straight upward.
• As it moves farther into the first quadrant it develops a larger and larger
negative x-component of velocity. So it also moves leftwards toward the
y-axis.
• Eventually it crosses the positive y-axis moving in the −x direction, i.e.
1 Think Poohsticks.
2 This is not such an unreasonable assumption. The flow often changes on a larger time
scale.
3 This is also not an unreasonable approximation.
to the left.
• As it moves farther into the second quadrant it develops a larger and
larger negative y-component of velocity. So it also moves downwards
toward the x-axis.
• Eventually it crosses the negative x-axis moving in the −y direction, i.e.
straight downward.
• As it moves farther into the third quadrant it develops a larger and larger
positive x-component of velocity. So it also moves rightward towards the
y-axis.
• Eventually it crosses the negative y-axis moving in the +x direction, i.e.
to the right.
• As it moves farther into the fourth quadrant it develops a larger and

larger positive y-component of velocity. So it also moves upwards toward
the x-axis.
With this type of analysis we cannot tell if the streamline, which is the red line
in the figure above, will return to the x-axis
• exactly at its starting point, forming a closed curve, or

• inside its starting point, spiralling inwards, or
• outside its starting point, spiralling outwards.

While the above procedure is a good way to get a qualitative feel for tra-
jectories, we can develop more precise, detailed descriptions of field lines by
working analytically. As we saw above, thinking of r(t) as the position at time
t of a stick dropped into water whose velocity at (x, y) is v(x, y), the velocity
at time t will be the same as the velocity of the water at r(t), which
of the stick
is v r(t) . Thus r(t) will obey the system of first order differential equations
Equation 2.2.3
dr
(t) = v r(t)
dt
R0 (u) =

Notice that if we reparametrize r(t), say to R(u) = r t(u) , then
0 0 0

r t(u) t (u) is parallel to (though not necessarily equal to) r t(u) = v r (t(u) =
v R(u) . So if we only care about the curve traced out by the stick, and not
about when the stick is at each point of the path, then it suffices to impose the
weaker condition4 that, when the stick is at r(t), its velocity r0 (t) is parallel to
0
(though not
necessarily equal to) v r(t) . In three dimensions, r (t) is parallel
to v r(t) when the cross product is zero:
Equation 2.2.4
r0 (t) × v r(t) = 0

In two dimensions we can still use the cross product by the simple expe-
dient of thinking of r0 (t) and v r(t) as three component vectors whose third
components are zero.
A more convenient way to implement the weaker “just parallel” condition,
involves reparametrizing our streamline. Suppose that we are in two dimen-
dy
sions with r0 (t) = dx

dt (t) , dt (t) and v(r) = v1 (r) , v2 (r) and fix some t0 . If
4 We’ll have a more careful discussion of this in the optional §2.2.1.
dx
is nonzero5 , we can reparametrize the curve (at least near r(t0 )) so as
dt (t0 )
to use x, rather than t as the parameter. To do so, we
• solve x = x(t) for t as a function of x. Call the solution T (x). Then

• the point on the curve which has x-coordinate
x is R(x) = X(x) , Y (x)
with X(x) = x and Y (x) = y T (x) .
Then the condition that R0 (x) = 1, Y 0 (x) is parallel to v R(x) says that

R0 (x) is a scalar multiple ofv R(x) so that there is a nonzero number c(x)
so that R0 (x) = c(x)v R(x) . That is
1, Y 0 (x) = c(x)v1 x, Y (x) , c(x)v2 x, Y (x)

or equivalently

Y 0 (x) c(x) v2 x, Y (x) v2 x, Y (x)
Y 0 (x) = = =
1 c(x) v1 x, Y (x) v1 x, Y (x)
This is exactly the statement that y = Y (x) is a solution of the differential

equation
dy v2 x, y
(x) =
dx v1 x, y
dy
It is conventional to pretend6 that dx is the ratio of dy and dx and rewrite the
7
differential equation as
dx dy
=
v1 (x, y) v2 (x, y)
Here is a summary of the discussion we have just completed. It extends to
three dimensions in an obvious way.
Equation 2.2.5 Use the symbol k to stand for “is parallel to”.
In two dimensions
dx dy
(t) , (t) k v1 (r(t)) , v2 (r(t)
dt dt
dx dy
⇐⇒ (t) , (t) , 0 × v1 (r(t)) , v2 (r(t)) , 0 = 0
dt dt
dx dy
⇐⇒ =
v1 (x, y) v2 (x, y)
and in three dimensions
dx dy dz
(t) , (t) , (t) k v1 (r(t)) , v2 (r(t)) , v3 (r(t))
dt dt dt
dx dy dz
⇐⇒ (t) , (t) , (t) × v1 (r(t)) , v2 (r(t)) , v3 (r(t)) = 0
dt dt dt
5 If dx (t ) = dy
dt 0
0, but (t )
dt 0
6= 0, we should use y rather than x as the parameter. If
dx
(t )
dt 0
= dy (t )
dt 0
= 0, then r(t) = r(t0 ) for all t and the streamline doesn’t move. It is just
a single point.
6 Of course dy is not the ratio of dy and dx. However pretending that it is provides
dx
a simple way to remember the technique that is used to solve the equation. You may
have used this mnemonic device before when you learned how to solve separable differential
equations. Section 2.4 of the CLP-2 text contains a treatment of separable differential
equations, including a justification for the mnemonic device.
7 Here is another nonrigorous, but intuitive way to come up with this equation. Suppose

that our stick is at (x, y) and has velocity dx
dt
(t) , dy
dt
(t) . In a tiny time interval dt the stick

moves by dx
dt
(t) , dy
dt
(t) dt = (dx, dy), which is parallel to v1 (x, y) , v2 (x, y) if dx
v1 (x,y)
=
dy
v2 (x,y)
.
dx dy dz
⇐⇒ = =
v1 (x, y, z) v2 (x, y, z) v3 (x, y, z)
Let us apply this to two examples, in which the stream lines of the vortex
field of Example 2.1.4 are found by two different methods.
Example 2.2.6 Stream lines for the vortex field using r0 (t) k v(r(t)).
In this example we will find the stream lines for the vortex field, v(x, y) =
Ω − yı̂ı + x̂ of Example 2.1.4, by using the requirement that, on a stream
line, the velocity vector r0 (t) must be parallel to v r(t) . By 2.2.5 one way to

express this requirement mathematically is

dx dy
=
−Ωy Ωx
This is a simple separable differential equation. We can solve it by cross mul-
tiplying and integrating both sides. (Recall that Ω is a constant.)
Z Z
Ωx dx = −Ωy dy ⇐⇒ Ω x dx = −Ω y dy
⇐⇒ 1 2
2 Ωx = − 12 Ωy 2 + C 0
2 2
⇐⇒ x + y = C
where C 0 and C = Ω2 C 0 are just arbitrary constants. So the stream lines of the
vortex field are exactly circles centred on the origin.
y
We can come to exactly the same conclusion by using the cross product
formulation of 2.2.4.
dx dy
(t) , (t) , 0 × v1 (r(t)) , v2 (r(t)) , 0 = 0
dt dt
dx dy
⇐⇒ (t)ı̂ı + (t) ̂ × − Ωy(t)ı̂ı + Ωx(t) ̂ = 0
dt dt
dx dy
⇐⇒ Ωx(t) (t) + Ωy(t) (t) k̂ = 0
dt dt
dx dy
⇐⇒ Ωx(t) (t) + Ωy(t) (t) = 0
dt dt
d 1 2 1 2

⇐⇒ Ωx(t) + Ωy(t) =0
dt 2 2
(Go ahead and evaluate the derivative.)

⇐⇒ 21 Ω x(t)2 + y(t)2 = C 0

⇐⇒ x(t)2 + y(t)2 = C

Example 2.2.7 Stream lines for the vortex field using r0 (t) = v(r(t)).
This time we will find the stream lines for the vortex field, v(x, y) = Ω −yı̂ı +x̂
of Example 2.1.4, by using 2.2.3, which is
dx
= −Ωy
dt
dy
= Ωx
dt
We can convert this system of first order linear ordinary differential equations
into a single second order linear constant coefficient differential equation8 , by
2
differentiating the first equation, to get ddt2x = −Ω dy
dt , and then substituting in
the second equation to get
d2 x
+ Ω2 x = 0
dt2
This equation is a special case of the ordinary differential equation treated
in Example A.9.3 of the Appendix A.9, entitled “Review of Linear Ordinary
Differential Equations”. In fact it is exactly (A.9.6) with R = 0, L = C = Ω1 .
So the general solution is (A.9.8) with ρ = 0 and ν = Ω, which is
x(t) = A cos(Ωt − θ)
with A and θ being arbitrary constants9 . Then

1 dx
y(t) = − = A sin(Ωt − θ)
Ω dt
giving us the familiar circular stream lines.

2.2.1 More about r0 (t) × v r(t) = 0
Here is a lemma that gives a more precise version of “if we only care about the
curve traced out by the stick, and not about when the stick is at each point of
the path, then it suffices to impose the weaker condition r0 (t) × v r(t) = 0”.

Lemma 2.2.8 Lat a < b and let v(r) be a vector field. Assume that, for
all a < u < b, R(u) is defined, both R0 (u) and v R(u) are continuous and
nonzero and
R0 (u) × v R(u) = 0

Then R(u) a < u < b is contained in a field line.

Proof. As R0 (u) × v R(u) = 0 and both R0 (u) and v R(u) are nonzero,

there is an a(u) such that
R0 (u) = a(u) v R(u)

0
R (u)·v(R(u))
This a(u) = v(R(u))·v(R(u)) is necessarily nonzero and continuous. Since a(u)
is nonzero and continuous, it never changes sign. That is, either a(u) > 0 for
all u, or a(u) < 0 for all u. Let T (u) be an antiderivative of a(u). Then T (u)
is strictly monotone (and continuous) and hence is invertible. That is, there
is a continuous function U (t) that obeys U T (u) = u for all a < u < b and
8 In Example 2.1.7 we converted a second order ordinary differential equation into a system
of first order ordinary differential equations. We are now just reversing the procedure we
used there.
9 Even if you don’t know how x(t) = A cos(Ωt − θ) was arrived at, you should be able to
easily verify that it really does obey x00 + Ω2 x = 0.


T U (t) = t for all t in the range of U . Differentiating T U (t) = t gives
T 0 U (t) U 0 (t) = 1 and hence U 0 (t) = T 0 (U1 (t)) . Set r(t) = R U (t) . Then
1
r0 (t) = R0 U (t) U 0 (t) = a U (t) v R U (t)

T 0 U (t)
1
= a U (t) v r(t)
a U (t)

= v r(t)
So r(t) is a field line and R(u) = r(T (u)) is a reparametrization of r(t).

Here are a couple of examples that show that bad things can happen if we
drop the requirement that v(R(u)) is nonzero.
Example 2.2.9 Let the vector field v(x, y) be identically zero. Then any field
line x(t) , y(t) must obey
x0 (t) = 0 y 0 (t) = 0
which forces both x(t) and y(t) to be constants. So each field line is just a single
point. On the other hand every nonconstant R(u) obeys R0 (u) × v R(u) = 0

but is not contained in a field line. (As R(u) is not constant, it covers more
than one point, while each field line is just a single point.)
Now here is a more interesting example.
Example 2.2.10 Consider the vector field v(x, y) = xı̂ı. This vector field
takes the value 0 at each point on the y-axis, is a positive multiple of ı̂ı at every
point of the right half-plane and is a negative multiple of ı̂ı at every point of
the left half-plane. Let’s find the field lines. Any field line must obey
dx dy
(t) = x(t) (t) = 0
dt dt
So y(t) must be a constant. We can solve the linear ordinary differential equa-
tion dx
dt (t) = x(t) by moving the x(t) to the left hand side, and multiplying by
the (integrating factor) e−t . This gives
dx
e−t (t) − e−t x(t) = 0
dt
By the product rule, this is the same as
d −t
e x(t) = 0
dt
which forces e−t x(t) to be a constant. So our field lines are Cet , D , with C

and D being arbitrary constants. Note that

• if C = 0, the field line is just the single point (0, D) on the y-axis. It is
illustrated by the black dot in the figure below.
• If C > 0, then as t runs from −∞ to +∞, the field line covers the
horizontal half-line
(x, D) x > 0
in the right half-plane. It is illustrated by the red line in the figure below.
• If C < 0, then as t runs from −∞ to +∞, the field line covers the
horizontal half-line
(x, D) x < 0
in the left half-plane. It is illustrated by the blue line in the figure below
(with a different value of D than for the red line).
y
vector field v “ xı̂ı

On the other hand, fix any constant D and set R(u) = uı̂ı + D̂. Then
R0 (u) × v R(u) = ı̂ı × uı̂ı) = 0

But as u runs from −∞ to +∞, R(u) runs over the full line (x, D) − ∞ <
x < ∞ . It is not contained in any single field line and, in fact, completely
covers three different field lines.
2.2.2 Exercises
1. Suppose that the vector field v(x, y) sketched below represents the
velocity of moving water at the point (x, y) in the first quadrant of
the xy-plane.
y
3
x
1 2 3
Sketch the path followed by a rubber ducky dropped in at the
point
a (0, 2)
b (1, 0)
c (1, 2)
2. Find a vector field v(x, y) for which
x(t) = e−t cos t

y(t) = e−t sin t
is a field line.
3. ∗. Consider the function f (x, y) = xy.
a Explicitly determine the field lines (flow lines) of F(x, y) = ∇ f .
b Sketch the field lines of F and the level curves of f in the same
diagram.
x
4. ∗. Find the field line of the vector field F = 2y ı̂ı + y2 ̂ + ey k̂ that
passes through (1, 1, e).
5. ∗. Find and sketch the field lines of the vector field F = xı̂ı + 3y ̂.
2.3 Conservative Vector Fields

Not all vector fields are created equal. In particular, some vector fields are
easier to work with than others. One important class of vector fields that are
relatively easy to work with, at least sometimes, but that still arise in many
applications are “conservative vector fields”.
Definition 2.3.1
a The vector field F is said to be conservative if there exists a function
ϕ such that F = ∇ ϕ. Then ϕ is called a potential for F. Note that if ϕ
is a potential for F and if C is a constant, then ϕ + C is also a potential
for F.
b If F = ∇ ϕ is a conservative field with potential ϕ and if C is a constant,
then the set of points that obey ϕ(x, y, z) = C is called an equipoten-
tial surface. Similarly, in two dimensions, the set of points that obey
ϕ(x, y) = C is called an equipotential curve.
♦
∇ϕ, then
Warning 2.3.2 In physics, when a vector field is of the form F = −∇
ϕ is called a potential for F. Note the minus1 sign in F = −∇ ϕ.
↑
Example 2.3.3 Potential energy. The “conservative” in “conservative

vector field” has nothing to do with politics. It comes from “conservation of
energy”. Here is how. Suppose that you have a particle of mass m moving in
a force field F that happens to be of the form F = ∇ ϕ for some function ϕ. If
the position of the particle a time t is x(t), y(t), z(t) , then, by Newton’s law
of motion,
dv
ma = F =⇒ m (t) = F x(t), y(t), z(t)
dt
dv
=⇒ m (t) = ∇ ϕ x(t), y(t), z(t)
dt
1 Physicists introduce this minus sign in order to eliminate the minus sign in the next
footnote.
Now dot both sides with v(t).
dv
=⇒ m v(t) · (t) = v(t) · ∇ ϕ x(t), y(t), z(t)
dt
∂ϕ ∂ϕ
= x0 (t) x(t), y(t), z(t) + y 0 (t)

x(t), y(t), z(t)
∂x ∂y
∂ϕ
+ z 0 (t)

x(t), y(t), z(t)
∂z
d
Next use dt v · v = 2v · dv
dt on the left hand side and the chain rule on the right
hand side.
d 1 d
=⇒ mv(t) · v(t) = ϕ x(t), y(t), z(t)
dt 2 dt
d 1
=⇒ mv(t) · v(t) − ϕ x(t), y(t), z(t) = 0
dt 2
1
m|v(t)|2 − ϕ x(t), y(t), z(t) = const

=⇒
2
So 21 m|v(t)|2 − ϕ x(t), y(t), z(t) , which is called the energy2 of the particle

at time t, does not actually depend on time — it is conserved. Let’s call

the initial energy E. That is, E = 21 m|v(0)|2 − ϕ x(0), y(0), z(0) . Then

1 2
2 m|v(t)| − ϕ x(t), y(t), z(t) = E for all t and, in particular
1
ϕ x(t), y(t), z(t) = m|v(t)|2 − E ≥ −E
2

So even without having to find
x(t) ,y(t) , z(t) , we know
that our particle
can never escape the region (x, y, z) ϕ(x, y, z) ≥ −E .
Example 2.3.4 Gravity. The gravitational force that a body of mass M at
the origin exerts on a body of mass m at r = (x, y, z) is
GM m
F(r) = − r
r3
p
where r = |r| = x2 + y 2 + z 2 and G is the gravitational constant. This force
is conservative, with potential ϕ(r) = GMr m . To verify that this is correct,
observe that
∂ ∂ GM m 1 GM m(2x) GM m
ϕ(r) = =− =− x
r3
p 2 2 2 3/2
∂x ∂x x + y + z
2 2 2 2 [x + y + z ]
∂ ∂ GM m 1 GM m(2y) GM m
ϕ(r) = =− =− y
r3
p
∂y ∂y x2 + y 2 + z 2 2 [x2 + y 2 + z 2 ]3/2
∂ ∂ GM m 1 GM m(2z) GM m
ϕ(r) = =− =− z
r3
p
∂z ∂z x2 + y 2 + z 2 2 [x2 + y 2 + z 2 ]3/2

We have already found conservation of energy very helpful a couple of times
in Section 1.7 (Sliding on a Curve). So, at this point, there are probably several
questions gnawing away at you.
• Is every vector field conservative?
• If not, is there an easy way to tell whether or not a vector field is con-
servative?
2 1 m|v(t)|2 is the kinetic energy and −ϕ is the potential energy. See Warning 2.3.2.
2
• If we know that a given vector field is conservative, how do you find a

potential for it?
Have no fear. We will consider those questions in some detail shortly. But
first, a couple of more examples.
Example 2.3.5 In this example we will show that the vector field F(x, y) =
xı̂ı − y ̂ is conservative and find both its potential and its field lines.
a The potential: Our vector field F(x, y) = xı̂ı − y ̂ is conservative if we
can find a ϕ obeying
∂ϕ
(x, y) = x
∂x
∂ϕ
(x, y) = −y
∂y
∂
Recall that, when taking the partial derivative ∂x the coordinate y is
viewed as a constant. So the first of these equations is satisfied if and
only if there is a ψ(y), which does not depend on x, so that
x2
ϕ(x, y) = + ψ(y)
2
For this to also satisfy the second equation, we need
∂ϕ ∂ x2
−y = (x, y) = + ψ(y) = ψ 0 (y)
∂y ∂y 2
which is the case if and only if there is a constant C with
y2
ψ(y) = − +C
2
So, for any choice of the constant C,
x2 y2
− +C
2 2
is a potential. In particular, taking C = 0, one possible potential is
x2 y2
ϕ(x, y) = −
2 2
Some equipotential curves for this potential are sketched in (c) below.
They are the blue curves.
b The field lines (Optional): Recalling (2.2.5), the field lines of the vector
field F(x, y) = xı̂ı − y ̂ are determined by
dx dy
= ⇐⇒ −ydx = xdy
x −y
⇐⇒ xdy + ydx = 0
⇐⇒ d(xy) = 0 by the product rule
⇐⇒ xy = C
for some constant C. If you are not comfortable with the use of the
product rule above, here is another way to write the same computation.
dy y dy
= − ⇐⇒ x = −y
dx x dx
dy
⇐⇒ x +y =0
dx
d
⇐⇒ (xy) = 0 by the product rule
dx
⇐⇒ xy = C
Some field lines are sketched in (c) below. They are the red curves. Note
that they appear to cross the equipotential curves, the blue curves, at
right angles. We shall see in Lemma 2.3.6, below, that this is not a
coincidence. Also note that, while the above computation tells what the
field lines are, it does not give us the direction of motion along the field
lines. We determine the direction of motion next.
c Direction of motion (Optional): The sign data
 
> 0
 if x > 0 

ı̂ı · F(x, y) = x = 0 if x = 0

 

< 0 if x < 0
 
> 0 if y < 0 
 
̂ · F(x, y) = −y = 0 if y = 0

 

< 0 if y > 0
is visually displayed in the figure on the left below. The arrows in the
figure on the left gives us the direction of motion along the field lines (in
red) in the figure on the right below. Some equipotential curves are also
sketched (in blue) in the figure on the right below.
y

We have just seen one example of a conservative vector field for which the
field lines appear to cross the equipotential curves at right angles. Here is a
result which says that that was no accident. The field lines of conservative
fields always cross the equipotential surfaces at right angles.
Lemma 2.3.6 Optional. If F is a conservative vector field, then the field
lines of F are perpendicular to the equipotential surfaces of F.
Proof. Let F = ∇ ϕ. Pick any point r0 and any nonzero vector T that is tangent
to the equipotential surface at r0 . That equipotential
surface is ϕ x, y, z =
ϕ(r0 ). Consider any curve r(t) = x(t), y(t), z(t) that

• lies in the equipotential surface of F through r0 , so that ϕ r(t) = ϕ(r0 )
for all t, and also obeys
• r(0) = r0 and
dr
• dt (0) = T.

Differentiating ϕ r(t) = ϕ(r0 ) with respect to t and applying the chain rule
gives
d
ϕ x(t), y(t), z(t) = 0
dt
or
∂ϕ dx ∂ϕ dy
x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
∂x dt ∂y dt
∂ϕ dz
+ x(t), y(t), z(t) (t) = 0
∂z dt
∂ϕ ∂ϕ ∂ϕ

Notice that the left hand side is exactly the dot product of ∂x , ∂y , ∂z = ∇ϕ
dy dz
with dx
dr
dt , dt , dt = dt . So
dr
∇ ϕ r(t) · (t) = 0
dt
dr
F r(t) · (t) = 0
dt
Then set t = 0 to get
F r0 · T = 0
This says that the vector T, which is tangent to the equipotential surface at
r0 , is perpendicular to the vector field at r0 , which is a tangent vector to the
field line of F through r0 .
Here is another example in which we try to find a potential for a vector
field.
Example 2.3.7 Let’s try to find a potential for the vortex vector field v(x, y) =
Ω − yı̂ı + x̂ of Example 2.1.4. The potential would have to obey
∂ϕ
(x, y) = −Ωy
∂x
∂ϕ
(x, y) = Ωx
∂y
We proceed just as we did in Example 2.3.5. The first of these equations is
satisfied if and only if there is a ψ(y), which does not depend on x, so that
ϕ(x, y) = −Ωxy + ψ(y)
For this to also satisfy the second equation, we need

∂ϕ ∂
Ωx = (x, y) = − Ωxy + ψ(y) = −Ωx + ψ 0 (y) ⇐⇒ ψ 0 (y) = 2Ωx
∂y ∂y
If Ω 6= 0, the right hand side of this equation depends on x while the left
hand side in independent of x, no matter what ψ is. So no ψ can work, and
v(x, y) = Ω − yı̂ı + x̂ is not conservative.
The previous example shows that not all vector fields are conservative.
That answers the first of the questions that we posed just before Example
2.3.5. The next theorem provides a simple screening test for conservativeness,
which partially answers the second question. The easy way to remember the
screening test uses the curl, which we now define.

Definition 2.3.8 The curl of a vector field F(x, y, z) is denoted by ∇ ×
F(x, y, z) and is defined by
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
 
ı̂ı ̂ k̂
∂ ∂ ∂
= det 
 
∂x ∂y ∂z 
F1 (x, y, z) F2 (x, y, z) F3 (x, y, z)
The determinant in the second row is really just a mnemonic device used to
make it easy to remember the expression after the equals sign in the first row.
One must be careful about the signs in this definition — the determinant helps
with that. ♦
Theorem 2.3.9 Screening test for conservative vector fields..
a Assume that F1 (x, y) and F2 (x, y) are continuously differentiable. If the
vector field F1 (x, y)ı̂ı + F2 (x, y)̂ is conservative, then we must have
∂F1 ∂F2
=
∂y ∂x
b Assume that F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) are continuously dif-
ferentiable. If the vector field F1 (x, y, z)ı̂ı + F2 (x, y, z)̂ + F3 (x, y, z)k̂ is
conservative, then
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
Equivalently,
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂ = 0
∂y ∂z ∂z ∂x ∂x ∂y
That is, F is curl free.
Proof. (a) If the vector field F1 (x, y)ı̂ı + F2 (x, y)̂ is conservative, then there is
a potential ϕ(x, y) such that
∂ϕ
(x, y) = F1 (x, y)
∂x
∂ϕ
(x, y) = F2 (x, y)
∂y
∂ ∂
Applying ∂y to the first equation and ∂x to the second equation gives
∂2ϕ ∂F1
=
∂y∂x ∂y
2
∂ ϕ ∂F2
=
∂x∂y ∂x
2 2
∂ ϕ ∂ ϕ
Recall that, for any twice continuously differentiable function, ∂y∂x = ∂x∂y .
So the two left hand sides are equal, and the two right hand sides must also be
equal.
(b) If the vector field F1 (x, y, z)ı̂ı + F2 (x, y, z)̂ + F3 (x, y, z)k̂ is conservative,
then there is a potential ϕ(x, y, z) such that
∂ϕ
(x, y, z) = F1 (x, y, z)
∂x
∂ϕ
(x, y, z) = F2 (x, y, z)
∂y
∂ϕ
(x, y, z) = F3 (x, y, z)
∂z
We proceed just as in (a).
∂ ∂
• Applying ∂y to the first equation and ∂x to the second equation gives
 
∂2ϕ ∂F1

∂y∂x = ∂y
 ∂F1 ∂F2
∂2ϕ ∂F2
=⇒ =
 =  ∂y ∂x
∂x∂y ∂x
∂ ∂
• Applying ∂z to the first equation and ∂x to the third equation gives
∂2ϕ
( )
∂F1
∂z∂x = ∂z ∂F1 ∂F3
∂2ϕ ∂F3
=⇒ =
∂x∂z = ∂x
∂z ∂x
∂ ∂
• Applying ∂z to the second equation and ∂y to the third equation gives
 
∂2ϕ ∂F2

∂z∂y = ∂z
 ∂F2 ∂F3
∂2ϕ ∂F3
=⇒ =
 =  ∂z ∂y
∂y∂z ∂y
Combining the three bullet points gives ∇ × F = 0.

Warning 2.3.10 As always, we have to be careful with the flow of logic3 . The
screening test in Theorem 2.3.9 is a one-way test. If, for example, ∂F ∂F2
∂y 6= ∂x
1
then the vector field F cannot be conservative. But if ∂F ∂F2

∂y = ∂x Theorem
1
2.3.9 does not guarantee that F is conservative. In fact there are fields that
are not conservative but do obey ∂F ∂F2
∂y = ∂x . We’ll see one in Example 2.3.14,
1
below. We shall later find some additional regularity conditions which, when
combined with ∂F ∂F2
∂y = ∂x , do imply conservativeness. See Theorem 2.4.8,
1
below.
Example 2.3.11 Example 2.3.7 revisited. In Example 2.3.7, we attempted
to find a potential for the vector field

v(x, y) = Ω − yı̂ı + x̂
In the end we showed that, if Ω 6= 0, no potential could exist, i.e. v(x, y) is not
conservative. Had we known the screening test of Theorem 2.3.9.a, we could
have concluded that v(x, y) is not conservative by simply observing that
∂v1 ∂
= − Ωy)= −Ω
∂y ∂y
∂v2 ∂
= Ωx) = +Ω
∂x ∂x
are not equal, unless Ω = 0. But Ω = 0 makes a rather boring vector field.
3 Use your favourite search engine to look up a list of common logical errors. One is
“affirming the consequent”. An example would be concluding that because Shakespeare is

dead, Elvis, who is also dead, must also be Shakespeare.
Example 2.3.12 Determine whether or not the vector field
F(x, y, z) = yı̂ı − z̂ + xk̂
is conservative. If it is conservative, find a potential.

Solution. Let’s start by applying the screening test Theorem 2.3.9.b. Since
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = ı̂ − ̂ − k̂
ı 
y −z x
is not 0, the vector field F cannot be conservative.

Example 2.3.13 Determine whether or not the vector field
F(x, y, z) = (y 2 + 2xz 2 − 1)ı̂ı + (2x + 1)y ̂ + (2x2 z + z 3 )k̂
is conservative. If it is conservative, find a potential.

Solution. Again start by applying the screening test Theorem 2.3.9.b. This
time
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
y 2 + 2xz 2 − 1 (2x + 1)y 2x2 z + z 3
= 0ı̂ı − (4xz − 4xz)̂ + (2y − 2y)k̂
=0
So F passes the screening test. Let’s look for a function ϕ(x, y, z) obeying
∂ϕ
(x, y, z) = y 2 + 2xz 2 − 1
∂x
∂ϕ
(x, y, z) = (2x + 1)y (∗)
∂y
∂ϕ
(x, y, z) = 2x2 z + z 3
∂z
∂
The partial derivative ∂x treats y and z as constants. So ϕ(x, y, z) obeys the
first equation if and only if there is a function ψ(y, z) with
ϕ(x, y, z) = xy 2 + x2 z 2 − x + ψ(y, z)
This ϕ(x, y, z) will also obey the second equation if and only if
∂
xy 2 + x2 z 2 − x + ψ(y, z) = (2x + 1)y

∂y
∂ψ
⇐⇒ 2xy + (y, z) = (2x + 1)y
∂y
∂ψ
⇐⇒ (y, z) = y
∂y
y2
⇐⇒ ψ(y, z) = + ζ(z)
2
for some function ζ(z) which depends only on z. At this stage we know that
y2
ϕ(x, y, z) = xy 2 + x2 z 2 − x + ψ(y, z) = xy 2 + x2 z 2 − x + + ζ(z)
2
obeys the first two equations of (∗), for any function ζ(z). Finally to have the
third equation of (∗) also satisfied, we also need to chose ζ(z) to obey
y2

∂ 2 2 2
xy + x z − x + + ζ(z) = 2x2 z + z 3
∂z 2
⇐⇒ 2x2 z + ζ 0 (z) = 2x2 z + z 3
⇐⇒ ζ 0 (z) = z 3
z4
⇐⇒ ζ(z) = +C
4
for any constant C. So one possible potential, namely that with C = 0, is
y2 z4
ϕ(x, y, z) = xy 2 + x2 z 2 − x + +
2 4
Note, as a check4 , that
∇ ϕ(x, y, z) = y 2 + 2xz 2 − 1 ı̂ı + 2xy + y)̂ + 2x2 z + z 3 k̂

as desired.
y x
Example 2.3.14 Optional: First look at + − x2 +y ı
2 ı̂
Now is a 
x2 +y 2 ̂ .
good time to reread Warning 2.3.10. In this example we will show that the
vector field
y x
F(x, y) = − ı̂ı + 2 ̂
x2 + y 2 x + y2
defined for all (x, y) in R2 except (x, y) = (0, 0)
passes the screening test of Theorem 2.3.9.a. We will also begin to see why it
is not conservative on the domain R2 \ {(0, 0)}. To verify the screening test,
we compute
∂ y (x2 + y 2 ) − y(2y) y 2 − x2
− 2 2
=− 2 = 2
∂y x +y (x2 + y 2 ) (x2 + y 2 )
∂ x (x2 + y 2 ) − x(2x) y 2 − x2
= 2 = 2
∂x x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
and observe that the two right hand sides are identical. So the screening test
is passed.
In order for F to be conservative on the domain R2 \ {(0, 0}, there must
exist a function ϕ(x, y), that, together with both partial derivatives ∂ϕ ∂x (x, y)
and ∂ϕ
∂y (x, y), is defined for all (x, y) in R 2
except (x, y) = (0, 0), and obeys
∂ϕ y − xy2 ∂ y
(x, y) = − 2 = 2 = arctan
∂x x + y2 1 + xy ∂x x

1
∂ϕ x x ∂ y
(x, y) = 2 2
= 2 = arctan
∂y x +y 1 + xy ∂y x

by the chain rule, because

∂ y y ∂ y 1
=− 2 =
∂x x x ∂y x x
4 It is always worth doing this check.
It looks like we have found a potential, namely arctan xy . But there is a problem.
Recall that, by definition, arctan xy is an angle θ(x, y) that obeys tan θ(x, y) =
arctan xy ; but for any (x, y) ∈ R2 \ {(0, 0} there are infinitely many angles
having the tangent xy . To define ϕ(x, y) we have to select exactly one such
angle. It is impossible to do so in such a way that ϕ(x, y) is continuous on all
of R2 \ {(0, 0}.
To see why, fix any r > 0, and imagine that you are walking on the circle
x2 + y 2 = r2 in the xy-plane. At time θ, you are at x = r cos θ, y = r sin θ and
then xy = tan θ and you are allowed to define ϕ(x, y) = θ + kπ, for any integer
k.
Suppose that at time θ = 0 you choose k = 0. That is, you choose ϕ(r, 0) =
0. Now start walking, choosing an allowed ϕ(x, y), i.e. choosing a k, for each
point (x, y) that you cross. Because ϕ(x, y) has to vary continuously5 with
(x, y), you have to continue choosing k = 0. But you run off a cliff as θ
approaches 2π, because then
• you are approaching (r, 0) from below, as in the figure below, and
• because you are choosing k = 0, ϕ(x, y) is just a little less than 2π, but
• you have already chosen ϕ(r, 0) = 0, not 2π. So ϕ(x, y) has a jump
discontinuity6 along the positive x-axis.
y
ϕ “ 5π{8 ϕ “ 3π{8
ϕ “ 6π{8 ϕ “ 2π{8
ϕ “ 7π{8 ϕ “ π{8
ϕ“π ϕ“0
x
ϕ “ 9π{8 ϕ “ 15π{8
ϕ “ 10π{8 ϕ “ 14π{8
ϕ “ 11π{8 ϕ “ 13π{8
If you are having trouble following this argument, don’t worry about it. We
will return with a less hand-wavy argument later.
2.3.1 Exercises
1. We’ve seen two calculations of the energy E of a system. Equa-
tion 1.7.1 told us E = 12 m|v| 2
+ mgy, while Example 2.3.3 says
1 2

2 m|v(t)| − ϕ x(t), y(t), z(t) = E.
Consider a force given by F = ∇ ϕ for some differentiable function
ϕ : R3 → R. A particle of mass m is being acted on by F and no
other forces, and its position at time t is given by (x(t), y(t), 0).
True or false: mgy(t) = −ϕ(x(t), y(t), 0).
2. For each of the following fields, decide which of the following holds:
A The screening test for conservative vector fields tells us F is
conservative.
5 If ϕ(x, y) is not continuous, its gradient does not exist, and ϕ cannot be a potential.
6 Those who have taken some complex analysis may recognize this as the branch cut in
ln z.
B The screening test for conservative vector fields tells us F is not

conservative.
C The screening test for conservative vector fields does not tell us
whether F is conservative or not.
(The screening test is Theorem 2.3.9.)
a F = xı̂ı + z̂ + y k̂
b F = y 2 zı̂ı + x2 z̂ + x2 y k̂

1
c F = (yexy + 1)ı̂ı + (xexy + z)̂ + + y k̂
z
d F = y cos(xy)ı̂ı + x sin(xy)̂
3. Suppose F is conservative and let a, b, and c be constants. Find a
potential for F + (a, b, c), OR give a conservative field F and constants
a, b, and c for which F + (a, b, c) is not conservative.
4. Prove, or find a counterexample to, each of the following statements.
a If F is a conservative field and G is a non-conservative field,
then F + G is non-conservative.
b If F and G are both non-conservative fields, then F + G is non-
conservative.
c If F and G are both conservative fields, then F + G is conser-

vative.
5. ∗. Let D be the domain consisting of all (x, y) such that x > 1, and
let F be the vector field
y x
F=− ı̂ı + 2 ̂
x2 + y 2 x + y2
Is F conservative on D? Give reasons for your answer.
6. Find a potential ϕ for F(x, y) = (x + y)ı̂ı + (x − y)̂, or prove none
exists.

7. Find a potential ϕ for F(x, y) = x1 − y1 ı̂ı + yx2 ̂, or prove none
exists.
Find a potential ϕ for F(x, y, z) = x2 yz + xz ı̂ı + 13 x3 z + y ̂ +

8.
1 3 1 2

3 x y + 2 x + y k̂, or prove none exists.
9. Find a potential ϕ(x, y, z) for

x y z
F(x, y, z) = ı̂ı+ 2 ̂+ 2 k̂,
x2 + y 2 + z 2 x + y2 + z2 x + y2 + z2
or prove none exists.
10. Determine whether or not each of the following vector fields are con-
servative. Find the potential if it is.
a F(x, y, z) = xı̂ı − 2y̂ + 3z k̂
xı̂ı − y̂
b F(x, y) =
x2 + y 2
2 2
11. Let F = e(z ) ı̂ı + 2Byz 3 ̂ + Axze(z )
+ 3By 2 z 2 k̂.
a For what values of the constants A and B is the vector field F
conservative on R3 ?
b If A and B have values found in (a), find a potential function

for F.
12. Find the velocity field for a two dimensional incompressible fluid when
there is a point source of strength m at the origin. That is, fluid is
emitted from the origin at area rate 2πm cm2 /sec. Show that this
velocity field is conservative and find its potential.
13. A particle of mass 10 kg moves in the force field F = ∇ ϕ, where
ϕ(x, y, z) = −(x2 + y 2 + z 2 ). When its potential energy is 0, the
particle is at the origin, and it moves with a velocity 2 m/s.
Following Example 2.3.3, give a region the particle can never es-
cape.
14. A particle√with constant mass m = 1/2 moves under a force field
F = ̂ + 3 3 z k̂. At position (0, 0, 0), its speed is 1. What is its speed
at (1, 1, 1)?
(You may assume without proof that the particle does indeed reach
the point (1, 1, 1).)
15. For some differentiable, real-valued functions f, g, h : R → R, we
define
F = 2f (x)f 0 (x)ı̂ı + g 0 (y)h(z)̂ + g(y)h0 (z)k̂.
Verify that F is conservative.
16. Describe the region in R3 where the field
F = xy, xz, y 2 + z

has curl 0.
2.4 Line Integrals

We have already seen, in §1.6, one type of integral along curves. We are
now going to see a second, that turns out to have significant connections to
conservative vector fields. It arose from the concept of “work” in classical
mechanics.
Suppose that we wish to find the work done by a force F(r) moving a
particle along a path r(t). During the “infinitesimal time interval”1 from t to
t+dt the particle moves from r(t) to r(t)+dr with dr = drdt (t) dt. By definition,
the work done during that infinitesimal time interval is
dr
F r(t) · dr = F r(t) · (t) dt
dt
The total work done during the time interval from t0 to t1 is then
Z t1
dr
Work = F r(t) · (t) dt
t0 dt
1 Yes, yes. We should first consider short time intervals ∆t > 0 and then take the limit
∆t → 0 at the end. But you have undoubtedly used this type of argument so many times
before that you would be thoroughly bored by it.
There are some useful shorthand notations for this work.

Definition 2.4.1 Denote by C the parametrized path r(t) with t0 ≤ t ≤ t1 .
Then
Z Z Z t1
dr
F · dr = F1 dx + F2 dy + F3 dz = F r(t) · (t) dt
C C t0 dt
H
If C is a closed path, the notation C F · dr is also used. ♦
In the event that F is conservative, and we know the potential ϕ, the
following theorem provides a really easy way to compute “work integrals”. The
theorem is a generalization of the fundamental theorem of calculus, and indeed
some people call it the fundamental theorem of line integrals.
Theorem 2.4.2 Let F = ∇ϕ be a conservative vector field. Then if C is any
curve that starts at P0 and ends at P1 , we have2
Z
F · dr = ϕ(P1 ) − ϕ(P0 )
C
Proof. Let r(t) = x(t), y(t), z(t) , t0 ≤ t ≤ t1 , be any parametrization of C

with r(t0 ) = P0 and r(t1 ) = P1 . Then, by definition,

Z Z t1 dr
Z t1
dr
F · dr = F r(t) · (t) dt = ∇ ϕ r(t) · (t) dt
C t0 dt t0 dt
Z t1 h
∂ϕ dx ∂ϕ dy
= x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
t0 ∂x dt ∂y dt
∂ϕ dz i
+ x(t), y(t), z(t) (t) dt
∂z dt
Z t1 h
d i
= ϕ x(t), y(t), z(t) dt by the chain rule in reverse
t0 dt

= ϕ r(t1 ) − ϕ r(t0 ) = ϕ(P1 ) − ϕ(P0 )
by the fundamental theorem of calculus.

R Observe that, in Theorem 2.4.2, the value, ϕ(P1 ) − ϕ(P0 ), of the integral
C
F · dr depended only on the endpoints P0 and P1 of the curve, not on the
path that the curve followed to get to P0 from P1 . We shall see, in Theorem
2.4.7, below, that this happens only for conservative vector fields. Here are
several examples of line integrals of vector fields that are not conservative.
Example 2.4.3 Set P0 = (0, 0), P1 = (1, 1) and3
F(x, y) = xy ı̂ı + (y 2 + 1) ̂
We shall consider three curves, all starting at P0 and ending at P1 .

a Let C1 be the straight line from P0 to P1 .
b Let C2 be the path, made from two straight lines, which follows the x-axis
from P0 to (1, 0) and then follows the line x = 1 from (1, 0) to P1 .
c Let C3 be the part of the parabola x = y 2 from P0 to P1 .
2 So ϕ acts a bit like the antiderivative of first year calculus.
y
P1
C3
C2
C1
P0 C2 x
R
We shall calculate the work Ci
F · dr for each of the curves.
a We parametrize C1 by r(t) = tı̂ı + t ̂ with t running from 0 to 1. Then
x(t) = t and y(t) = t so that
dr
F r(t) = t2 ı̂ı + (t2 + 1) ̂

and (t) = ı̂ı + ̂
dt
so that
Z Z 1 dr
Z 1
t2 ı̂ı + (t2 + 1) ̂ · [ı̂ı + ̂] dt

F · dr = F r(t) · (t) dt =
C1 0 dt 0
Z 1
2
= 2t + 1 dt
0
5
=
3
b We split C2 into two parts, C2,x running from P0 to (1, 0) along the x-
axis and then C2,y running from (1, 0) to P1 along the line x = 1. We
parametrize C2,x by r(x) = xı̂ı with x running from 0 to 1 and C2,y by
r(y) = ı̂ı + y ̂ with y running from 0 to 1. Then4
Z Z Z
F · dr = F · dr + F · dr
C2 C2,x C2,y
ı̂ı
z }| {
Z 1 d
2

= (x)(0)ı̂ı + (0 + 1) ̂ · xı̂ı dx
0 dx
̂
z }| {
Z 1 d
2

+ (1)(y)ı̂ı + (y + 1) ̂ · ı̂ı + y ̂ dy
0 dy
Z 1 Z 1
y 2 + 1 dy

= 0 dx +
0 0
4
=
3
c We parametrize C3 by r(t) = t2 ı̂ı + t ̂ with t running from 0 to 1. Then

x(t) = t2 and y(t) = t so that
dr
F r(t) = t3 ı̂ı + (t2 + 1) ̂

and (t) = 2tı̂ı + ̂
dt
so that
Z Z 1
t3 ı̂ı + (t2 + 1) ̂ · [2tı̂ı + ̂] dt

F · dr =
C3 0
Z 1
2t4 + t2 + 1 dt

=
0
2 1 26
= + +1=
5 3 15
Note that, despite the

R fact thatR C1 , C2 and CR3 all start at P0 and all end at P1 ,
the three integrals C1 F · dr, C2 F · dr and C3 F · dr all have different values.

3 The reader should check that this vector field is not conservative.
4 You might like to think about why we can split up the integral like this.
Example 2.4.4 Set5

F(x, y) = 2y ı̂ı + 3x ̂
This time we consider two curves.
a Let C1 be circle x2 + y 2 = 1, traversed once counterclockwise, starting at
(1, 0).
b Let C2 be (trivial) curve which just consists of the single point (1, 0).
R
We shall calculate the work Ci F · dr for each curve.
a We parametrize C1 by r(t) = cos tı̂ı + sin t ̂ with t running from 0 to 2π,
just as we did in Example 1.0.1. Then
I Z 2π

F · dr = 2 sin tı̂ı + 3 cos t ̂ · [− sin tı̂ı + cos t ̂] dt
C1 0
Z 2π
− 2 sin2 t + 3 cos2 t dt

=
0
You could evaluate these integrals using double angle trig identities like
you did in first year calculus. But there is a sneaky, much easier, way.
Because sin2 t and cos2 t are translates
R 2π of each other,
R 2πand both are peri-
odic of period π, the two integrals 0 sin2 t dt and 0 cos2 t dt represent
the same area and so are equal. See the figure below.
y y
y “ sin2 x 1
y “ cos2 x
1
x x
π{2 π 3π{2 2π π{2 π 3π{2 2π
Thus
Z 2π Z 2π Z 2π
2 2 1 2
sin t + cos2 t dt

sin t dt = cos t dt =
0 0 0 2
Z 2π
1
= dt = π
2 0
and
I Z 2π Z 2π
2
F · dr = −2 sin t dt + 3 cos2 t dt = π
C1 0 0
dr
R
b We parametrize C2 by r(t) = ı̂ı for all t. Then dt (t) = 0 and C2
F·dr = 0.
Again, despite theR fact that C1 Rand C2 both start at (1, 0) and end at (1, 0),
the two integrals C1 F · dr and C2 F · dr are different.
Example 2.4.5 Example 2.3.14, again. In Example 2.3.14, we saw that
the vector field
y x
F(x, y) = − 2 ı̂ı + 2 ̂
x + y2 x + y2
defined for all (x, y) in R2 except (x, y) = (0, 0)
passed the screening test of Theorem 2.3.9.a, and yet was not conservative.
In this example, we will see that this F violates the conclusion of Theorem
2.4.2, thereby providing a second proof that F(x, y) is not conservative on
5 Again, the reader should verify that this vector field is not conservative.
R2 with (0, 0) removed. For the curve C, of Theorem 2.4.2, we use the circle
parametrized by x = a cos θ, y = a sin θ, 0 ≤ θ ≤ 2π. Then dx = −a sin θ dθ
and dy = a cos θ dθ so that
Z 2π 2 Z 2π
x dy − y dx a cos2 θ dθ + a2 sin2 θ dθ
Z
1 1 1
= = dθ
2π C x2 + y 2 2π 0 a2 cos2 θ + a2 sin2 θ 2π 0
=1
The curve C has initial point

P0 = (a cos θ, a sin θ)θ=0 = (a, 0)
and final point

P1 = (a cos θ, a sin θ)θ=2π = (a, 0) = P0
So, if F were conservative with potential ϕ, Theorem 2.4.2 would give that
x dy − y dx
Z
1
= ϕ(P1 ) − ϕ(P0 ) = 0
2π C x2 + y 2
Consequently, F can’t be conservative.
2.4.1 Path Independence

This brings us to the following question. Let F be any fixed vector field. When
is it true that, given any two fixed points P0 and P1 , the integrals
Z Z
F · dr = F · dr
C C0
for all curves C, C 0 that start at P0 and end

R at P1 ? When can we ignore the
path taken? If this is the case we say that “ C F · dr is independent of the path
chosen” and we write Z P1 Z
F · dr = F · dr
P0 C
for any path C from P0 to P1 . The point of this section is that there is an
intimate relation between path independence and conservativeness of vector
fields, that we will get to in Theorem 2.4.7.
For simplicity, we will consider only vector fields that are defined and con-
tinuous on all of R2 (i.e. the xy-plane) or R3 (i.e. the usual three dimensional
world). Some discussion about what happens for vector fields that are defined
only on part of R2 or R3 is given in the optional §4.5.
First we show that if there is one pair of (not necessarily distinct) points
P0 , P1 such that Z Z
F · dr = F · dr
C1 C2
for all curves C1 , C2 that start at P0 and end at P1 , then it is also true that,
for any other pair of points P00 , P10
Z Z
F · dr = F · dr
C10 C20
for all curves C10 , C20 that start at P00 and end at P10 . This might seem unlikely
at first, but the idea of the proof is really intuitive.
Theorem 2.4.6 Let F be a vector field that is defined and continuous on all
of R2 (or R3 ). Let P0 , P1 , P00 , P10 be any four points in R2 (or R3 ). Assume
that Z Z
F · dr = F · dr
C1 C2
for all curves C1 , C2 that start at P0 and end at P1 . Then

Z Z
F · dr = F · dr
C10 C20
for all curves C10 , C20 that start at P00 and end at P10 .
Proof. Let C10 and C20 be any two curves that start at P00 and end at P10 .
P1
C11
P01 Cr
Cℓ P11
P0 C21
We start by choosing any two (auxiliary) curves
• C` that starts at P0 and ends at P00 and

• Cr that starts at P10 and ends at P1 .
and then we define the curves
• C1 to be C` , followed by C10 , followed by Cr and

• C2 to be C` , followed by C20 , followed by Cr .
Then both C1 and C2 start at P0 and end at P1 , so that, by hypothesis,
Z Z
F · dr = F · dr
C1 C2
and, from the construction of C1 and C2 ,

Z Z Z Z Z Z
F · dr + F · dr + F · dr = F · dr + F · dr + F · dr
C` C10 Cr C` C20 Cr
Z Z
=⇒ F · dr = F · dr
C10 C20
as desired.
We are now ready for our main theorem on conservative fields.
Theorem 2.4.7 Let F be a vector field that is defined and continuous on all
of R2 (or R3 ). Then the following three statements are equivalent.
a F is conservative. That is, there exists a function ϕ such that F = ∇ ϕ.
b The integral C F · dr = 0 for any closed curve C.
H
c The integral F · dr Ris path independent. That is, for any points P0 , P1
R
we have C1 F · dr = C2 F · dr for all curves C1 , C2 that start at P0 and

R
end at P1 .
That is, if any one of the three statements are true, then all three are true.
Proof. It suffices for us to provea that

• the truth of (a) implies the truth of (b) and
• the truth of (b) implies the truth of (c) and
• the truth of (c) implies the truth of (a).
That’s exactly what we will do.
(a) =⇒ (b): Let C be a closed curve that starts at P0 and then ends back
at P0 . Then, by Theorem 2.4.2 with P1 = P0 ,
I
F · dr = ϕ(P0 ) − ϕ(P0 ) = 0
C
H =⇒ (c): Pick any point P0 and set P1 = P0 . Then we are assuming

(b)
that
R C
F · dr = 0 for all curves that start at P0 and end at P1 . In particular
C
F · dr takes the same value for all curves that start at P0 and end at P1 . So
Theorem 2.4.6 immediately yields property (c).
(c) =⇒ (a): We are to show that F is conservative. We’ll start by guessing
ϕ and then we’ll verify that, for our chosen ϕ, we really do have F = ∇ ϕ. Our
guess for ϕ is motivated by Theorem R 2.4.2. If our F really is conservative, its
potential is going to have to obey C F · dr = ϕ(P1 ) − ϕ(P0 ) for any curve C
that starts at P0 and ends at P1 . Let’s choose P0 = 0. Remembering, from
Definition 2.3.1.a, that adding a constant to a potential always
R yields another
potential, we can always choose ϕ(0) = 0. Then ϕ(P1 ) = C F·dr for any R curve
C that starts at 0 and ends at P1 . So define, for each point x, ϕ(x) = C F · dr
for any curve C that starts at 0 and ends R at x. Note that, since we we are
assuming that (c) is true, the integral C F · dr takes the same value for all
curves C that start at 0 and end at x.
We now verify that, for this chosen ϕ, we really do have F = ∇ ϕ. Fix any
point x and any curve Cx that starts at the origin and ends at x. For any
vector u, let Du be the curve with parametrization
ru (t) = x + tu 0≤t≤1
This curve is a line segment that starts at x at t = 0 and ends at x +u

R at t = 1.
Observe that r 0u (t) = u. Recall that, by assumption, ϕ(x + su) = C F · dr for
any curve C that starts at 0 and ends at x + su. So
Z
ϕ(x + su) = F · dr
Cx +Dsu
where Cx + Dsu is the curve which first follows Cx from the origin to x and
then follows Dsu from x to x + su. We have
Z Z Z
F · dr = F · dr + F · dr
Cx +Dsu Cx Dsu
Z Z1
= F · dr + F(x + tsu) · (su) dt
Cx 0
In the second integral, make the change of variables τ = ts, dτ = sdt. This
gives Z Z s
ϕ(x + su) = F · dr + F(x + τ u) · u dτ
Cx 0
By the fundamental theorem of calculus, applied to the second integral,
d
ϕ(x + su) = F(x + su) · u = F(x) · u

ds s=0 s=0
Applying this with u = ı̂ı, ̂, k̂ gives us

∂ϕ ∂ϕ ∂ϕ
(x) , (x) , (x) = F(x) · ı̂ı , F(x) · ̂ , F(x) · k̂
∂x ∂y ∂z
which is
∇ϕ(x) = F(x)
as desired.
a This is a pretty efficient, and standard, way to structure the proof of the equivalence of
three statements.
Using this result, we can completely characterize conservative fields on R2
and R3 .
Theorem 2.4.8 Let F be a vector field that is defined and has continuous first
order partial derivatives on all of R2 (or R3 ). Then F is conservative if and
only if it passes the screening test ∇ × F = 0, i.e. is curl free.
Warning 2.4.9 Note that in Theorem 2.4.8 we are assuming that F passes
the screening test on all of R2 or R3 . We have already seen, in Example 2.3.14,
that if the screening test fails at even a single point, for example because the
vector field is not defined at that point, then F need not be conservative.
We’ll explore what happens in such cases in the (optional) §4.5. We’ll see that
something can be salvaged.
Proof of Theorem 2.4.8. We’ll give the proof for the R2 case. The proof for
the R3 case is very similar. We have already seen, in Theorem 2.3.9, that if F
is conservative, then it passes the screening test and there is nothing more to
do.
So we now have to assume that F obeys ∂F ∂y (x, y) = ∂x (x, y) on all of R
1 ∂F2 2
and prove that it is conservative. We’ll do so using the strategy of Example

2.3.13 to find a function ϕ(x, y), that obeys
∂ϕ
(x, y) = F1 (x, y)
∂x
∂ϕ
(x, y) = F2 (x, y)
∂y
∂
The partial derivative ∂x treats y as a constant. So ϕ(x, y) obeys the first
equation if and only if there is a function ψ(y) with
Z x
ϕ(x, y) = F1 (X, y) dX + ψ(y)
0
This ϕ(x, y) will also obey the second equation if and only if
∂ϕ
F2 (x, y) = (x, y)
∂y
∂ x
Z
= F1 (X, y) dX + ψ(y)
∂y
Z x 0
∂F1
= (X, y) dX + ψ 0 (y)
0 ∂y
So we have to find a ψ(y) that obeys
Z x
∂F1
ψ 0 (y) = F2 (x, y) − (X, y) dX
0 ∂y
This looks bad — no matter what ψ(y) is, the left hand side is independent
of x, while it looks like the right hand side depends on x. Fortunately our
screening test hypothesis now rides in to the rescuea . (We haven’t used it yet,
and it has to come in somewhere.)
Z x Z x
∂F1 ∂F2
F2 (x, y) − (X, y) dX = F2 (x, y) − (X, y) dX
0 ∂y 0 ∂x
X=x
= F2 (x, y) − F2 (X, y)

X=0
= F2 (0, y)
In going from the first line to the second line we used the fundamental theorem
of calculus. So choosing
Z y
ψ(y) = F2 (0, Y ) dY + C
0
for any constant C, does the trick.

a or bails us out, or saves our bacon, or . . .
2.4.2 Exercises
1. Evaluate C x2 y 2 dx + x3 y dy counterclockwise around the square with
R
vertices (0, 0), (1, 0), (1, 1) and (0, 1).

2. For each of the following fields, decide which of the following holds:
A The characterization of conservative vector fields, Theorem 2.4.8
(with Theorem 2.3.9), tells us F is conservative.
B The characterization of conservative vector fields, Theorem 2.4.8

(with Theorem 2.3.9), tells us F is not conservative.
C The characterization of conservative vector fields, Theorem 2.4.8
(with Theorem 2.3.9), does not tell us whether F is conservative
or not.
a F = xı̂ı + z̂ + y k̂
b F = y 2 zı̂ı + x2 z̂ + x2 y k̂
c F = (yexy + 1)ı̂ı + (xexy + z)̂ + (z + y) k̂

d F = y cos(xy)ı̂ı + x sin(xy)̂
2 2
x +y
3. R ϕ(x, y, z) = e
Let + cos(z 2 ), and define F = ∇ ϕ. Evaluate
F · dr over the closed curve C that is an ellipse traversed clockwise,
C √ √
centred
√ at (1, 2, 3), passing
√ through the points
√ ( √5 − 1, −2, 5 − 3),
(( 5 − 2)/2, −1/2, ( 5 − 6)/2), and (−2, 3 − 2, 3 − 3).
4. Let P1 and P2 be points in R2 . Let A and B be paths from P1 to P2 ,
as shown below.
P1
A B
P2
2
R
R F is a conservative vector field in R with A F · dr = 5.
Suppose
What is B F · dr?

5. ∗. Let F(x, y, z) = ex sin y ı̂ı + Raex cos y + bz ̂ + cx k̂. For which
values of the constants a, b, c is C F · dr = 0 for all closed paths C?
6. Consider the four vector fields sketched below. Exactly one of those
vector fields is conservative. Determine which three vector fields are
not conservative and explain why.
(a) y (b) y
(c) y (d) y
x x
7. ∗. Consider the vector field
x − 2y 2x + y
F(x, y, z) = ı̂ı + 2 ̂ + z k̂
x2 + y 2 x + y2
a Determine the domain of F.

b Compute ∇ × F. Simplify the result.
c Evaluate the line integral
Z
F · dr
C
where C is the circle of radius 2 in the plane z = 3, centered

at (0, 0, 3) and traversed counter-clockwise if viewed from the
positive z-axis, i.e. viewed “from above”.
d Is F conservative?
R
8. Find the work, C F·dr, done by the force field F = (x+y)ı̂ı +(x−z)̂ +
(z − y)k̂ in moving an object from (1, 0, −1) to (0, −2, 3). Does the
work done depend on the path used to get from (1, 0, −1) to (0, −2, 3)?
9. Consider the vector field
V(x, y) = (ex cos y + x2 , x2 y + 3)

R
Evaluate the line integral C V·dr along the oriented curve C obtained
by moving from (0, 0) to (1, 0) to (1, π) and finally to (0, π) along
straight line segments.
R
10. Evaluate C F · dr for
a F(x, y) = xy ı̂ı − x2 ̂ along y = x2 from (0, 0) to (1, 1).
b F(x, y, z) = (x − z)ı̂ı + (y − z) ̂ − (x + y) k̂ along the polygonal
path from (0, 0, 0) to (1, 0, 0) to (1, 1, 0) to (1, 1, 1).
11. ∗. Let C be the part of the curve of intersection of xyz = 8 and x = 2y
which lies between the points (2, 1, 4) and (4, 2, 1). Calculate
Z
F · dr
C
where
F = x2 ı̂ı + (x − 2y) ̂ + x2 y k̂
12. Let F = 6x2 yz 2 ı̂ı +(2x3 z 2 +2y−xz) ̂ +4x3 yz k̂ and let G = yz ı̂ı +xy k̂.
a For what value of the constant λ is the vector field H = F + λG
conservative on 3-space?
b Find a scalar potential φ(x, y, z) for the conservative field H
referred to in part (a).
R
c Find C F · dr if C is the curve of intersection of the two surfaces
z = x and y = exz from the point (0, 1, 0) to the point (1, e, 1).
13. ∗. Find the work done by the force field F(x, y, z) = (x − y 2 , y −
z 2 , z − x2 ) on a particle that moves along the line segment from
(0, 0, 1) to (2, 1, 0).
x y
14. ∗. Let F = x2 +y ı + x2 +y
2 ı̂  + x3 k̂. Let P be the path which starts
2 ̂
at (1, 0, 0), ends at √2 , √2 , 12 ln 2 and follows

1 1

x2 + y 2 = 1 xez = 1
Find the work done in moving a particle along P in the field F.

15. ∗. Let F = yz cos x , z sin x + 2yz , y sin x + y 2 − sin z and
Z let C be
the line segment r(t) = (t, t, t), for 0 ≤ t ≤ π/2. Evaluate F · dr.
C
16. ∗. Let C be the upper half of the unit circle centred on (1, 0) (i.e.
that part of the circle whichR lies above the x-axis), oriented clockwise.
Compute the line integral C xy dy.
17. ∗. Show that the following line integral is independent of path and
evaluate the integral.
Z
(yex + sin y) dx + (ex + sin y + x cos y) dy
C
where C is any path from (1, 0) to (0, π/2).

18. ∗. Evaluate the integral
Z
xy dx + yz dy + zx dz
C
around the triangle with vertices (1, 0, 0), (0, 1, 0), and (0, 0, 1), ori-
ented clockwise as seen from the point (1, 1, 1).
R
19. ∗. Evaluate the line integral C F · dr, where F is the conservative
vector field
F(x, y, z) = y + zex , x + ey sin z, z + ex + ey cos z

and C is the curve given by the parametrization
r(t) = (t, et , sin t), t from 0 to π.

20. ∗.
a For which values of the constants α, β and γ is the vector field
F(x, y, z) = αey ı̂ı + (xey + β cos z) ̂ − γy sin z k̂
conservative?
R
b For those values of α, β and γ found in part (a), calculate C F ·
dr, where C is the curve parametrized by x = t2 , y = et , z = πt,
0 ≤ t ≤ 1.
21. ∗. Consider the vector field F(x, y, z) = (cos x, 2 + sin y, ez ).
a Compute the curl of F.
b Is there a function f such that F = ∇ f ? Justify your answer.
R
c Compute the integral C F · dr along the curve C parametrized
by r(t) = (t, cos t, sin t) with 0 ≤ t ≤ 3π.
22. ∗.
a Consider the vector field
F(x, y, z) = (z + ey , xey − ez sin y, 1 + x + ez cos y)
Find the curl of F. Is F conservative?

R
b Find the integral C F · dr of the field F from (a) where C is the
curve with parametrization
r(t) = (t2 , sin t, cos2 t)
where t ranges from 0 to π.

23. ∗. A physicist studies a vector field F. From experiments, it is known

that F is of the form
F = (x − a)yex ı̂ı + (xex + z 3 ) ̂ + byz 2 k̂
where a and b are some real numbers. From theoretical considerations,

it is known that F is conservative.
a Determine a and b.
b Find a potential f (x, y, z) such that ∇ f = F.

R
c Evaluate the line intgeral C F · dr where C is the curve defined
by
r(t) = t , cos 2t , cos t , 0≤t≤π
d Evaluate the line integral

Z
I= (x + 1)yex dx + (xex + z 3 ) dy + 4yz 2 dz,
C
where C is the same curve as in part (c). [Note: the “4” in the
last term is not a misprint!].
Questions 2.4.2.24 and 2.4.2.25 ask you to evaluate line integrals of vector fields
that are not conservative, but that can be expressed as a sum of a conservative
vector field and another vector field that can be written concisely.
24. ∗. Let
F = y 2 e3z + Axy 3 ı̂ı + (2xye3z + 3x2 y 2 ̂ + Bxy 2 e3z k̂

a Find all values of A and B for which the vector field F is con-
servative.
b If A and B have values found in (a), find a potential function

for F.
2t −t
c Let C be the curve with parametrization r(t) = e ı̂ı + e ̂ +
ln(1 + t) k̂ from (1, 1, 0) to e2 , 1e , ln 2 . Evaluate
Z
(y 2 e3z + xy 3 ) dx + (2xye3z + 3x2 y 2 ) dy + 3xy 2 e3z dz.
C
25. ∗.
(a) For which value(s) of the constants a, b is the vector field
F = 2x sin(πy) − ez ı̂ı + ax2 cos(πy) − 3ez ̂ − x + by ez k̂

conservative?
(b) Let F be a conservative field from part (a). Find all functions
ϕ for which F = ∇ ϕ.
R
(c) Let F be a conservative field from part (a). Evaluate C F · dr
where C is the intersection of y = x and z = ln(1 + x) from
(0, 0, 0) to (1, 1, ln 2).
R
(d) Evaluate C G · dr where
G = (2x sin(πy) − ez ) ı̂ı + πx2 cos(πy) − 3ez ̂ − xez k̂

and C is the intersection of y = x and z = ln(1+x) from (0, 0, 0)

to (1, 1, ln 2).
F(x, y, z) = −2y cos x sin xı̂ı + (cos2 x + (1 + yz)eyz ) ̂ + y 2 eyz k̂
a Find a real valued function f (x, y, z) such that F = ∇ f .

b Evaluate the line integral
Z
F · dr
C

where C is the arc of the curve r(t) = t, et , t2 − π 2 , 0 ≤ t ≤ π,
traversed from (0, 1, −π 2 ) to (π, eπ , 0).
27. ∗. Consider the vector field F(x, y, z) = 2xı̂ı + 2y ̂ + 2z k̂.
a Compute ∇ × F.
b If C is any path Rfrom (0, 0, 0) to (a1 , a2 , a3 ) and a = a1 ı̂ı + a2 ̂ +
a3 k̂, show that C F · dr = a · a.
28. ∗. Let C be the parameterized curve given by
π
r(t) = cos t, sin t, t , 0≤t≤
2
and let
F = eyz , xzeyz + zey , xyeyz + ey

a Compute and simplify ∇ × F.

R
b Compute the work integral C F · dr.
29. ∗.
a Show that the planar vector field
F(x, y) = 2xy cos(x2 ) , sin(x2 ) − sin(y)

is conservative.
b Find a potential function for F.

R
c For the vector field F from above compute C F · dr, where C is
the part of the graph x = sin(y) from y = π/2 to y = π.
30. ∗. Consider the following force field, in which m, n, p, q are constants:
F = (mxyz + z 2 − ny 2 )ı̂ı + (x2 z − 4xy) ̂ + (x2 y + pxz + qz 3 ) k̂

H
a Find all values of m, n, p, q such that C
F·dr = 0 for all piecewise
smooth closed curves C in R3 .
b For every possible choice of m, n, p, q in (a), find the work done
by F in moving a particle from the bottom to the top of the
sphere x2 + y 2 + z 2 = 2z. (The direction of k̂ defines “up”.)
31. Let C be the curve from (0, 0, 0) to (1, 1, 1) along the intersection of
the surfaces y = x2 and z = x3 .
R
a Find C ρ ds if s is arc length along C and ρ = 8x + 36z.
R
b Find C F · dr if F = sin y ı̂ı + (x cos y + z) ̂ + (y + z) k̂.
32. ∗. The curve C is the helix that winds around the cylinder x2 +y 2 = 1
(counterclockwise, as viewed from the positive z-axis, looking down on
the xy-plane). It starts at the point (1, 0, 0), winds around the cylinder
once, and ends at the point (1, 0, 1). Compute the line integral of the
vector field
F(x, y, z) = (−y, x, z 2 )
along C.
33. ∗. Evaluate the line integrals below. (Use any method you like.)
a C (x2 +y) dx+x dy, where C is the arc of the parabola y = 9−x2
R
from (−3, 0) to (3, 0).

b C F · n̂ ds, where F(x, y) = 2x2ı̂ı + yex̂, C is the boundary of
R
the square 0 ≤ x ≤ 1, 0 ≤ y ≤ 1. Here n̂ is the unit normal

vector pointing outward from the square, and s is arc length.
34. ∗. A particle of mass m = 1 has position r0 = ̂ and velocity v0 = ı̂ı + k̂
at time t = 0. The particle moves under a force F(t) = ̂ − sin t k̂,
where t denotes time.
a Find the position r(t) of the particle as a function of t.
b Find the position r1 of the particle when it crosses the plane
x = π/2 for the first time after time t = 0.
c Determine the work done by F in moving the particle from r0

to r1 .
Questions 2.4.2.35 and 2.4.2.36 ask you to find a path that leads to a particular
value of a line integral. Many such paths are possible — you only need to find
one.
35. ∗.

a Consider the vector field F x, y = (3y, x − 1) in R2 . Compute
the line integral Z
F · dr
L
where L is the line segment from (2, 2) to (1, 1).

b Find an oriented path C from (2, 2) to (1, 1) such that
Z
F · dr = 4
C
where F is the vector field from (a).

36. ∗. Let F = (2y + 2)ı̂ı be a vectorR field on R2 . Find an oriented curve
C from (0, 0) to (2, 0) such that C F · dr = 8.
37. ∗. Let
F(x, y) = 1, yg(y)
and suppose that g(y) is a function defined everywhere with every-
where continuous partials. Show that for any curve C whose endpoints
P and Q lie on the x-axis,

Z

distance between P and Q = F · dr
C
38. ∗. Let S be the surface z = 2 + x2 − 3y 2 and let F(x, y, z) = (xz +

axy 2 )ı̂ı +yz̂ +z 2 k̂. Consider the points P1 = (1, 1, 0) and P2 = (0, 0, 2)
on the surface S. R R
Find a value of the constant a so that C1 F · dr = C2 F · dr for
any two curves C1 and C2 on the surface S from P1 to P2 .
39. ∗. Consider the vector field F defined as
2 2

F(x, y, z) = (1 + ax2 )ye3x − bxz cos(x2 z) , xe3x , x2 cos(x2 z)
where a and b are real valued constants.

a Compute ∇ × F.
b Determine for which values a and b the vector field F is conser-
vative.
c For the values of a and b obtained in part (b), find a potential
function f such that ∇ f = F.
d Evaluate the line integral
Z
2 2
ye3x + 2xz cos(x2 z) dx + xe3x dy + x2 cos(x2 z)dz
C
where C is the arc of the curve (t, t, t3 ) starting at the point

(0, 0, 0) and ending at the point (1, 1, 1).
40. ∗. Let C be the curve from (0, 0, 0) to (1, 1, 1) along the intersection
of the surfaces y = x2 and z = x3 .
R
a Find C F · dr if F = (xz − y)ı̂ı + (z + x) ̂ + y k̂.
R
b Find C ρ ds if s is arc length along C and ρ(x, y, z) = 8x + 36z.
R
c Find C F · dr if F = sin y ı̂ı + (x cos y + z) ̂ + (y + z) k̂.

41. ∗. The vector field F(x, y, z) = Ax3 y 2 z ı̂ı + z 3 + Bx4 yz ̂ + 3yz 2 −

x4 y 2 k̂ is conservative on R3 .
a Find the values of the constants A and B.
b Find a potential ϕ such that F = ∇ ϕ on R3 .
c If C is the curve
R y = −x, z = x2 from (0, 0, 0) to (1, −1, 1),
evaluate I = C F · dr.
d Evaluate J = C (z−4x3 y 2 z)dx+(z 3 −x4 yz)dy+(3yz 2 −x4 y 2 )dz,

R
where C is the curve of part (c).

e Let T be the closed triangular path with vertices (1, 0, 0), (0, 1, 0)
and (0, 0, 1), oriented
R counterclockwise as seen from the point
(1, 1, 1). Evaluate T (zı̂ı + F) · dr.
42. ∗. A particle of mass
m=2
is acted on by a force
F = 4t , 6t2 , −4t

At t = 0, the particle has velocity zero and is located at the point

(1, 2, 3).
a Find the velocity vector v(t) for t ≥ 0.
b Find the position vector r(t) for t ≥ 0.
c Find κ(t) the curvature of the path traversed by the particle for
t ≥ 0.
d Find the work done by the force on the particle from t = 0 to
t = T.
43. ∗.√ The position of an airplane at time t is given by x = y =
4 2 3/2
3 t , z = t(2 − t) from take-off at t = 0 to landing at t = 2.
a What is the total distance the plane travels on this flight?
b Find the radius of curvature κ at the apex of the flight, which
occurs at t = 1.
c Two external forces are applied to the plane during the flight:
the force of gravity G = −M g k̂, where M is the mass of the
plane and g is a constant; and a friction force F = −|v|2 v, where
v is the velocity of the plane. Find the work done by each of
these forces during the flight.
d One half-hour later, a bird follows the exact same flight — path
as the plane, travelling at a constant speed v = 3. One can
show
√
that
√
at the apex of the path, i.e. when the bird is at
4 2 4 2

3 , 3 , 1 , the principal unit normal N̂ to the path points in
the −k̂ direction. Find the bird’s (vector) acceleration at that
moment.
2.5 Optional — The Pendulum

Model a pendulum by a mass m that is connected to a hinge by an idealized
rod that is massless and of fixed length `. Denote by θ the angle
̂
θ ı̂ı
ℓ
τ βℓ dθ
dt θ
ℓ cos θ ℓ
mg ℓ sin θ
between the rod and vertical. The forces acting on the mass are
• gravity, which has magnitude mg and direction (0, −1),

• tension in the rod, whose magnitude, τ , automatically adjusts itself so

that the distance between the mass and the hinge is fixed at ` and whose
direction, (− sin θ, cos θ), is always parallel to the rod and
• possibly some frictional forces, like friction in the hinge and air resis-
tance. We shall assume that the total frictional force has magnitude
proportional to the speed1 of the mass and has direction opposite to the
direction of motion of the mass.
If we use a coordinate system centered on the hinge, the (x, y) coordinates of
the mass at time t are ` sin θ(t), − cos θ(t) . Hence its velocity vector is
d dθ
v(t) = ` sin θ(t), − cos θ(t) = ` cos θ(t), sin θ(t) (t)
dt dt
and the total frictional force is −β`(cos θ, sin θ) dθ
dt , for some constant β. The
acceleration vector of the mass is
d d2 θ dθ 2
a(t) = v(t) = `(cos θ, sin θ) 2 + `(− sin θ, cos θ)
dt dt dt
so that Newton’s law of motion, F = ma, now tells us
d2 θ dθ 2
ma(t) = m`(cos θ, sin θ) 2
+ m`(− sin θ, cos θ)
dt dt
dθ
= F = mg(0, −1) + τ (− sin θ, cos θ) − β`(cos θ, sin θ)
dt
To eliminate the (unknown) coefficient τ we dot this equation with (cos θ, sin θ),
which extracts the component parallel to the direction of motion of the mass.
2
Dotting with (cos θ, sin θ) gives m` ddt2θ = −mg sin θ − β` dθ
dt or
d2 θ β dθ g
2
+ + sin θ = 0
dt m dt `
which is the equation of motion of the (nonlinear) pendulum. In general, it
can be hard to analyse nonlinear differential equations. But if the amplitude
of oscillation is small enough that we may approximate sin θ by θ we get the
equation of motion of the linear pendulum2 which is
d2 θ β dθ g
2
+ + θ=0
dt m dt `
These equations may be reformulated as systems of first order ordinary differ-
ential equation, that is as equations for the flow lines of a vector field, by the
simple expedient of defining (as we did in Example 2.1.7)
x(t) = θ(t) y(t) = θ0 (t)

d2 θ β dθ g
Then, for the full, nonlinear, equation dt2 + m dt + ` sin θ = 0
x0 (t) = θ0 (t) = y(t)

g β
y 0 (t) = θ00 (t) = − sin x(t) − y(t)
` m
1 The dependence of air resistance (drag) on the speed v is relatively complex. At low
speed drag tends to be approximately proportional to v, while at high speed it tends to be

approximately proportional to v 2 .
2 When β = 0, this equation reduces to the equation d2 θ + g θ = 0, which occurs in many
dt2 `
different applications, and whose solutions exhibit simple harmonic motion.
The solutions of this first order system of ordinary differential equations are
flow lines for the vector field
g β
V (x, y) = y , − sin x − y
` m
When β = 0, this is exactly the vector field of Example 2.1.7.
Chapter 3
Surface Integrals
3.1 Parametrized Surfaces

For many applications we will need to use integrals over surfaces. One obvious
one is just computing surface areas. Another is computing the rate at which
fluid traverses a surface. The first step is to simply specify surfaces carefully.
There are three common ways to specify a surface in three dimensions.
a Graph of a function: Probably the most common way to specify a surface
is to give its equation in the form
z = f (x, y) (x, y) ∈ D ⊂ R2
Here “(x, y) ∈ D ⊂ R2 ” just means that (x, y) runs over the subset D of
R2 . For example, if the surface is the top half of the sphere of radius one
centred on the origin
p
z = 1 − x2 − y 2 with x2 + y 2 ≤ 1
b Implicitly: We can also specify that the surface is the set of points (x, y, z)
that satisfy the equation G(x, y, z) = 0, or, more generally1 , satisfy the
equation G(x, y, z) = K, with K a constant. For example, the sphere of
radius one centred on the origin is the set of points that obey
x2 + y 2 + z 2 = 1
We shall explore this surface a little more in Example 3.1.2 below.

c Range of a function: Probably the most useful way to specify a surface,
when one needs to integrate over the surface, is as the range of a function
r : D ⊂ R2 → R 3

(u, v) ∈ D 7→ r(u, v) = x(u, v) , y(u, v) , z(u, v)
The upper line means that r is a function which is defined on the subset
D of R2 and which assigns to each point on D a point in R3 . The
second line means that the function r assigns to the element (u, v) of D
the element r(u, v) = x(u, v) , y(u, v) , z(u, v) in R3 . Such a surface is
called a parametrized surface — each point of the surface is labelled by
1 Of course we can always convert the equation G(x, y, z) = K into H(x, y, z) = 0 with
H(x, y, z) = G(x, y, z) − K. But it is often more convenient to use G(x, y, z) = K.
123
CHAPTER 3. SURFACE INTEGRALS 124
the values of the two parameters u and v. Parametrized surfaces are of

course the two parameter analog of parametrized curves. Examples of
parametrized surfaces come next.
Example 3.1.1 One simple, even trivial, way to parametrize the surface which
is the graph
z = f (x, y) (x, y) ∈ D ⊂ R2
is to choose x and y as the parameters. That is, to choose

r(u, v) = u, v, f (u, v) , (u, v) ∈ D

or r(x, y) = x, y, f (x, y) , (x, y) ∈ D

Let’s do something a bit more substantial.
Example 3.1.2 Sphere. The sphere of radius 1 centred on the origin is the
set of points (x, y, z) that obey
G(x, y, z) = x2 + y 2 + z 2 = 1
We cannot express this surface as the graph of a function because, for each
(x, y) with x2 + y 2 < 1, there are two z’s that obey x2 + y 2 + z 2 = 1, namely
p
z = ± 1 − x2 − y 2
On the other hand, locally, this surface is the graph of a function. This means
that, for any point (x0 , y0 , z0 ) on the sphere, all points of the surface that are
sufficiently near (x0 , y0 , z0 ) can be expressed in one of the forms z = f (x, y) or
x = g(y, z),√or y = h(x, z). For example, the part of the sphere that is within
a distance 2 of the point (0, 0, 1) is
√
(x, y, z) x2 + y 2 + z 2 = 1, |(x, y, z) − (0, 0, 1)| < 2

= (x, y, z) x2 + y 2 + z 2 = 1, x2 + y 2 + (z − 1)2 < 2

= (x, y, z) x2 + y 2 + z 2 = 1, x2 + y 2 + z 2 − 2z + 1 < 2

= (x, y, z) x2 + y 2 + z 2 = 1, z > 0

p
= (x, y, z) z = 1 − x2 − y 2 , x2 + y 2 < 1

This is illustrated in the figure below which shows the y = 0 section of the
sphere x2 + y 2 + z 2 =√1 and also the y = 0 section of the set of points that
are within a distance 2 of (0, 0, 1). (They are the points inside the dashed
circle.)
z
x2 ` y 2 ` pz ´ 1q2 “ 2
p0, 0, 1q
p1, 0, 0q x
x2 ` y 2 ` z 2 “ 1
Similarly, as illustrated schematically

√ in the figure below, the part of the
sphere that is within a distance 2 of the point (1, 0, 0) is
√
(x, y, z) x2 + y 2 + z 2 = 1, |(x, y, z) − (1, 0, 0)| < 2

= (x, y, z) x2 + y 2 + z 2 = 1, (x − 1)2 + y 2 + z 2 < 2

= (x, y, z) x2 + y 2 + z 2 = 1, x2 − 2x + 1 + y 2 + z 2 < 2

= (x, y, z) x2 + y 2 + z 2 = 1, x > 0

p
= (x, y, z) x = 1 − y 2 − z 2 , y 2 + z 2 < 1

The figure below shows the y = 0 section of the sphere x2 + y 2 + z 2 = √

1 and
also the y = 0 section of the set of points that are within a distance 2 of
(1, 0, 0). (Again, they are the points inside the dashed circle.)
z
p0, 0, 1q
p1, 0, 0q x
x `y `z “1
2 2 2
px ´ 1q2 ` y 2 ` z 2 “ 2
We can parametrize the unit sphere by using spherical coordinates, which

you should have seen before. As a reminder, here is a figure showing the
definitions of the three spherical coordinates2
ρ = distance from (x, y, z) to (0, 0, 0)

ϕ = angle between the line (0, 0, 0) (x, y, z) and the z axis
θ = angle between the line (0, 0, 0) (x, y, 0) and the x axis
z
p0, 0, zq ρ sin ϕ
px, y, zq
ρ cos ϕ
ϕ ρ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and here are two more figures giving the side and top views of the previous
figure.
z
ρ sin ϕ px, y, zq y
p0, 0, zq
θ ρ sin ϕ
ρ cos ϕ ρ sin ϕ cos θ
ρ
ϕ px, y, 0q
px, 0, 0q
ρ sin ϕ sin θ
px, y, 0q
x
side view top view
From the figure, we see that Cartesian and spherical coordinates are related
by
x = ρ sin ϕ cos θ
y = ρ sin ϕ sin θ
z = ρ cos ϕ
The points on the sphere x2 + y 2 + z 2 = 1 are precisely the set of points with
ρ = 1. So we can use the parametrization

r(θ, ϕ) = sin ϕ cos θ , sin ϕ sin θ , cos ϕ
Here is how to see that as ϕ runs over (0, π) and θ runs over [0, 2π), r(θ, ϕ)
covers the whole sphere x2 + y 2 + z 2 = 1 except for the north pole (ϕ = 0 gives
the north pole for all values of θ) and the south pole (ϕ = π gives the south
pole for all values of θ).
• Fix θ and have ϕ run over the interval 0 < ϕ ≤ π2 . Then r(θ, ϕ) traces
out one quarter of a circle starting at the north pole r(θ, 0) = (0, 0, 1)
(but excluding the north pole itself) and ending at the point r(θ, π2 ) =
(cos θ, sin θ, 0) in the xy-plane.
z
ϕ
y
θ
x
• Keep θ fixed at the same value and extend the interval over which ϕ
runs to 0 < ϕ < π. Now r(θ, ϕ) traces out a semi-circle starting at the
north pole r(θ, 0) = (0, 0, 1), ending at the south pole r(θ, π) = (0, 0, −1)
(but excluding both the north and south poles themselves) and passing
through the point r(θ, π2 ) = (cos θ, sin θ, 0) in the xy-plane.
y
θ
• Finally have θ run over 0 ≤ θ < 2π. Then the semicircle rotates about
the z-axis, sweeping out the full sphere, except for the north and south
poles.
Recall that ϕ is the angle between the radius vector and the z-axis. If you
hold ϕ fixed and increase θ by a small amount dθ, r(θ, ϕ) sweeps out the red
circular arc in the figure on the left below. If you hold ϕ fixed and vary θ from
0 to 2π, r(θ, ϕ) sweeps out a line of latitude. The figure on the right below
gives the lines of latitude (or at least the parts of those lines in the first octant)
π 2π 3π 4π π
for ϕ = 10 , 10 , 10 , 10 and 5π
10 = 2 .
z
z
dθ
ϕ y
y
θ
x
x
On the other hand, if you hold θ fixed and increase ϕ by a small amount
dϕ, r(θ, ϕ) sweeps out the red circular arc in the figure on the left below. If
you hold θ fixed and vary ϕ from 0 to π, r(θ, ϕ) sweeps out a line of longitude.
The figure on the right below gives the lines of longitude (or at least the parts
π 2π 3π 4π π
of those lines in the first octant) for θ = 0, 10 , 10 , 10 , 10 and 5π
10 = 2 .
z
z
ϕ dϕ y
y
θ
x
x

2 2
Example 3.1.3 Cylinder. The surface x + z = 1 is an infinite cylinder.
Part of this cylinder in the first octant is sketched below.
Note that the section of this cylinder that lies in the xz-plane, and in fact
in any plane y = c, is the circle x2 + z 2 = 1. We can of course parametrize
this circle by x = cos θ, z = sin θ. So we can parametrize the whole cylinder
by using θ and y as parameters.

r(θ, y) = cos θ , y , sin θ 0 ≤ θ < 2π, −∞ < y < ∞

Example 3.1.4 Surface of Revolution. In this example, we are going to
parametrize a surface of revolution. In your first integral calculus course, you
undoubtedly encountered many surfaces created by rotating a curve y = f (x)
about the x-axis or the y-axis. In this course, we are used to having the z-axis,
rather than the y-axis, run vertically.
So in this example, we’ll parametrize the surface constructed by rotating
the curve
z = g(y) = ey 0≤y≤1
about the z-axis. Exactly the same procedure can be used to parametrize
surfaces created by rotating about the x-axis or the y-axis too.
We start by just sketching the curve, considering the yz-plane as the plane
x = 0 in R3 . The specified curve is the red curve in the figure below. Concen-
trate on any one point on that curve. It is the blue dot at (0, Y, eY )
2 The symbols ρ, ϕ, θ are the standard mathematics symbols for the spherical coordinates.
Appendix A.7 gives another set of symbols that is commonly used in the physical sciences
and engineering.
p0, Y, eY q
x
in the figure. When our curve is rotated about the z-axis, the blue dot
sweeps out a circle. The circle that the blue dot sweeps out
• lies in the horizontal plane z = eY and
• is centred on the z-axis and
• has radius Y .
We can parametrize the circle swept out in the usual way. Here is a top view
of the circle, with the parameter, named θ, indicated.
p0, Y, eY q
θ y
Y
pY sin θ, Y cos θ, eY q
x top view

The coordinates of the red dot are Y sin θ , Y cos θ , eY . This also gives a
parametrization of the surface of revolution
x(Y, θ) = Y sin θ
y(Y, θ) = Y cos θ
z(Y, θ) = eY
0 ≤ Y ≤ 1, 0 ≤ θ < 2π
Notice, by way of checks, that

• when θ = 0,
x(Y, 0) , y(Y, 0) , z(Y, 0) = (0, Y, eY )

runs over the entire desired curve (namely z = g(y), 0 ≤ y ≤ 1), when Y
runs over 0 ≤ Y ≤ 1 and

• for any fixed 0 ≤ Y ≤ 1, x(Y, θ) , y(Y, θ) , z(Y, θ) runs over the circle
x2 + y 2 = Y 2 , in the plane z = eY , when θ runs over 0 ≤ θ < 2π.
Also notice that

x(Y, θ)2 + y(Y, θ)2 = Y 2
so that p
Y = x(Y, θ)2 + y(Y, θ)2
and
√
x(Y,θ)2 +y(Y,θ)2
z(Y, θ) = eY = e
That is, the surface of revolution is contained in the (infinite) surface

√ 2 2
z = e x +y
Remembering that 0 ≤ Y ≤ 1, we have that 1 ≤ z = eY ≤ e. Thus the surface

of revolution is √ 2 2
z = e x +y 1≤z≤e
Finally here is a sketch of the part of the surface in the first octant, x, y, z ≥ 0.
z
?
z “ e x `y
2 2
x

Example 3.1.5 Torus. In this example, we are going to parametrize a donut
(well, its surface), or an inner tube.
The formal mathematical name for the surface of a donut is a torus. Our
strategy will be to first parametrize the section of the torus in the right half of
the yz-plane, and then built up the full torus by rotating the circle about the
z-axis. The section is a circle, sketched below.
r
θ
p0,R,0q
y
We’ll assume that the centre of the circle is a distance R from the z-axis,
and that the circle has radius r. Then the red dot on the circle is at
x=0
y = R + r cos θ
z = r sin θ
In particular the red dot is a distance r sin θ above the xy-plane and is a
distance R + r cos θ from the z-axis. So when we rotate the section about the
z-axis, the red dot sweeps out a circle which is sketched below.
z
y
x
The circle that the red dot sweeps out
• lies in the plane z = r sin θ and
• is centred on the z-axis and
• has radius ρ = R + r cos θ.
We can parametrize the circle swept out in the usual way. Here is a top view
of the circle, with the parameter, named ψ, indicated.
p0, ρ, r sin θq
y
ψ
pρ cos ψ, ρ sin ψ, r sin θq
x top view
So the parametrization of the circle swept out by the red dot, and also the
parametrization of the torus, is
x = ρ cos ψ = (R + r cos θ) cos ψ

y = ρ sin ψ = (R + r cos θ) sin ψ
z = r sin θ
or
r(θ, ψ) = (R + r cos θ) cos ψ ı̂ı + (R + r cos θ) sin ψ ̂ + r sin θ k̂

0 ≤ θ, ψ < 2π
3.1.1 Exercises
1. Parametrize the surface given by z = ex+1 + xy in terms of x and y.
2. ∗. Let S be the surface given by
r(u, v) = u + v , u2 + v 2 , u − v ,

−2 ≤ u ≤ 2, −2 ≤ v ≤ 2
This is a surface you are familiar with. What surface is it (it may be
just a portion of one of the following)? sphere / helicoid / ellipsoid
/ saddle / parabolic bowl / cylinder / cone / plane
3. ∗. Suppose S is the part of the hyperboloid x2 + y 2 − 2z 2 = 1 that
lies inside the cylinder x2 + y 2 = 9 and above the plane z = 1 (i.e. for
which z ≥ 1).
Which of the following are parameterizations of S?
a The vector function

√
u2 + v 2 − 1
√
r(u, v) = uı̂ı + v ̂ + k̂
2

with domain D = (u, v) 2 ≤ u2 + v 2 ≤ 9 .
b The vector function
r
u2 1
r(u, v) = u sin v ı̂ı − u cos v ̂ + − k̂
2 2
√
with domain D = (u, v) 3 ≤ u ≤ 3, 0 ≤ v ≤ 2π .
c The vector function
p p
r(u, v) = 1 + 2v 2 cos uı̂ı + 1 + 2v 2 sin u ̂ + v k̂

with domain D = (u, v) 0 ≤ u ≤ 2π, 1 ≤ v ≤ 2 .
d The vector function
√ √
r
u
r(u, v) = 1 + u sin v ı̂ı + 1 + u cos v ̂ +
k̂
2

e The vector function

√
√ √
u+1
r(u, v) = u cos v ı̂ı − √ u sin v ̂ +
k̂
2

4. ∗. Suppose the surface S is the part of the sphere x2 + y 2 + z 2 = 2
that lies inside the cylinder x2 + y 2 = 1 and for which z ≥ 0. Which
of the following are parameterizations of S?
a
r(φ, θ) = 2 sin φ cos θ ı̂ı + 2 cos φ ̂ + 2 sin φ sin θ k̂
π
0 ≤ φ ≤ , 0 ≤ θ ≤ 2π
4
b p
r(x, y) = xı̂ı − y ̂ + 2 − x2 − y 2 k̂
x2 + y 2 ≤ 1
c p
r(u, θ) = u sin θ ı̂ı + u cos θ ̂ + 2 − u2 k̂
0 ≤ u ≤ 2, 0 ≤ θ ≤ 2π
d √ √ √
r(φ, θ) = 2 sin φ cos θ ı̂ı + 2 sin φ sin θ ̂ + 2 cos φ k̂
π
0 ≤ φ ≤ , 0 ≤ θ ≤ 2π
4
e p p
r(φ, z) = − 2 − z 2 sin φı̂ı + 2 − z 2 cos φ ̂ + z k̂
√
0 ≤ φ ≤ 2π, 1 ≤ z ≤ 2
5. ∗. Let S be the part of the paraboloid z + x2 + y 2 = 4 lying between
the planes z = 0 and z = 1. For each of the following, indicate whether
or not it correctly parameterizes the surface S.
a r(u, v) = uı̂ı + v ̂ + (4 − u2 − v 2 ) k̂, 0 ≤ u2 + v 2 ≤ 1
√ √
b r(u, v) = ( 4 − u cos v)ı̂ı + ( 4 − u sin v) ̂ + u k̂, 0 ≤ u ≤ 1,
0 ≤ v ≤ 2π
√
c r(u, v) = (u cos v)ı̂ı + (u sin v) ̂ + (4 − u2 ) k̂, 3 ≤ u ≤ 2, 0 ≤
v ≤ 2π
6. ∗. Consider the following surfaces
• S1 is the hemisphere given by the equation x2 + y 2 + z 2 = 4 with
z ≥ 0.
• S2 is the cylinder given by the equation x2 + y 2 = 1.
• S3 is the cone given by the equation z 2 = x2 + y 2 with z ≥ 0.
Consider the following parameterizations:
A
√ √ √
r(θ, φ) = 4 cos θ sin φ , 4 sin θ sin φ , 4 cos φ
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/6
B
√ √ √
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/4
C
√ √ √
0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/3
D
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
0 ≤ θ ≤ 2π, 1≤z≤2
E
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
√
0 ≤ θ ≤ 2π, 2≤z≤2
F
p p
r(θ, z) = 4 − z 2 cos θ , 4 − z 2 sin θ , z
√
0 ≤ θ ≤ 2π, 3≤z≤2
G

r(θ, z) = z cos θ , z sin θ , z
0 ≤ θ ≤ 2π, 0≤z≤1
H

√
0 ≤ θ ≤ 2π, 0 ≤ z ≤ 2
I

√
0 ≤ θ ≤ 2π, 0 ≤ z ≤ 3
J
p
r(x, y) = x , y , x2 + y 2
x2 + y 2 ≤ 1
K
p
r(x, y) = x , y , x2 + y 2
√
x2 + y 2 ≤ 2
L
p
r(x, y) = x , y , x2 + y 2
x2 + y 2 ≤ 2
For each of the following, choose from above all of the valid parame-
terization of each of the given surfaces. Note that there may be one
or more valid parameterization for each surface, and not necessarily
all of the above parameterizations will be used.
a The part of S1 contained inside S2 :
b The part of S1 contained inside S3 :
c The part of S3 contained inside S2 :
d The part of S3 contained inside S1 :

7. Parametrize a solid of rotation about a line not parallel to an axis.
Maybe first show that the plane you’re rotating is normal to that axis.
a Give a parametric equation for the circle of radius 1, centred at
(2, 2, 4), lying in the plane x = y.
b Give a parametrized equation for the surface formed by rotating
the circle from part (a) about the line r(t) = 4ı̂ı + 4̂ + tk̂.
r(t) = (4, 4, t)
z
(2, 2, 4) y
3.2 Tangent Planes

If you are confronted with a complicated surface and want to get some idea of
what it looks like near a specific point, probably the first thing that you will
do is find the plane that best approximates the surface near the point. That
is, find the tangent plane to the surface at the point. In general, a good way
to specify a plane is to supply
• a nonzero vector n (called a normal vector) perpendicular to the plane1
(to determine the orientation of the plane) and
• one point (x0 , y0 , z0 ) on the plane.
If (x, y, z) is any other point on the plane, then the vector
(x, y, z) − (x0 , y0 , z0 ) = (x − x0 , y − y0 , z − z0 )
1 Alternatively, you could find two vectors that are in the plane (and not parallel to each
other), and then construct a normal vector by taking their cross product.
lies entirely in the plane and so is perpendicular to n. This gives the following
very neat the equation for the plane.
pa, b, cq
px, y, zq
n · (x − x0 , y − y0 , z − z0 ) = 0
The following theorem provides formulae for normal vectors n to general

surfaces, assuming first that the surface is parametrized, second that the sur-
face is a graph and finally the surface is given by an implicit equation. The
formulae are developed in the proof of the theorem.
Theorem 3.2.1 Normal vectors to surfaces.
a Let
r : D ⊂ R 2 → R3

(u, v) ∈ D 7→ r(u, v) = x(u, v) , y(u, v) , z(u, v)
be a parametrized surface and let (x0 , y0 , z0 ) = r(u0 , v0 ) be a point on the

surface. Set
∂ ∂x ∂y ∂z
Tu = r(u, v0 ) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )

∂u u=u0 ∂u ∂u ∂u
∂ ∂x ∂y ∂z
Tv = r(u0 , v) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )

∂v v=v0 ∂v ∂v ∂v
Then

∂x ı̂ı ̂ k̂

∂y ∂z

n = Tu × Tv = det ∂u (u0 , v0 ) ∂u (u0 , v0 ) ∂u (u0 , v0 )

∂x (u , v ) ∂y ∂z
∂v (u0 , v0 ) ∂v (u0 , v0 )

∂v 0 0
is normal (i.e. perpendicular) to the surface at (x0 , y0 , z0 ).

b Let (x0 , y0 , z0 ) = f (x0 , y0 ) be a point on the the surface z = f (x, y).
Then,
n = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂
is normal to the surface at (x0 , y0 , z0 ).

c Consider the surface given implicitly by the equation G(x, y, z) = K,
where K is a constant. Let (x0 , y0 , z0) be a point on the surface and
assume that the gradient ∇ G x0 , y0 , z0 6= 0. Then

n = ∇ G x0 , y0 , z0
is normal to the surface at (x0 , y0 , z0 ).
Note that none of the normal vectors n above need be of unit length.
Note that if we apply part (c) to G(x, y, z) = z − f (x, y) we get the normal
vector n = ∇ G x0 , y0 , z0 = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂, which is the same
as the normal vector provided by part (b). Of course they had to be at least
parallel.
Proof. (a) First fix v = v0 and let u vary. Then

u 7→ r(u, v0 ) = x(u, v0 ) , y(u, v0 ) , z(u, v0 )
is a curve on the surface (the red curve in the figure on the right below) that
passes through (x0 , y0 , z0 ) (the black dot in the figure) when u = u0 .
n
Tv
Tu
The tangent vector to this curve at (x0 , y0 , z0 ), which is also a tangent

vector to the surface at (x0 , y0 , z0 ), is
∂ ∂x ∂y ∂z
Tu = r(u, v0 ) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )

∂u u=u0 ∂u ∂u ∂u
It is the red arrow in the figure on the right above.
Next fix u = u0 and let v vary. Then

v 7→ r(u0 , v) = x(u0 , v) , y(u0 , v) , z(u0 , v)
is a curve on the surface (the blue curve in the figure on the right above) that
passes through (x0 , y0 , z0 ) when v = v0 . The tangent vector to this curve at
(x0 , y0 , z0 ), which is also a tangent vector to the surface at (x0 , y0 , z0 ), is
∂ ∂x ∂y ∂z
Tv = r(u0 , v) = (u0 , v0 ) , (u0 , v0 ) , (u0 , v0 )

∂v v=v0 ∂v ∂v ∂v
It is the blue arrow in the figure on the right above.
We now have two vectors, namely Tu and Tv , that are tangent to the
surface at (x0 , y0 , z0 ). So their cross product

∂x ı̂ı ̂ k̂

∂y ∂z

n = Tu × Tv = det ∂u (u0 , v0 ) ∂u (u0 , v0 ) ∂u (u0 , v0 )
∂x (u , v ) ∂y (u , v ) ∂z (u , v )
∂v 0 0 ∂v 0 0 ∂v 0 0
is normal (i.e. perpendicular) to the surface at (x0 , y0 , z0 ). Note however that

this vector need not be normalized. That is, it need not be of unit length.
(b) Next assume that the surface is given by the equation z = f (x, y).
Then, renaming u to x and v to y, we may reuse part (a):

r(x, y) = x, y, f (x, y)

parametrizes the surface and, at x0 , y0 , z0 = f (x0 , y0 ) ,
∂r
Tx = (x0 , y0 ) = 1 , 0 , fx (x0 , y0 )
∂x
∂r
Ty = (x0 , y0 ) = 0 , 1 , fy (x0 , y0 )
∂y
and

ı̂ı ̂ k̂

n = Tx × Ty = det 1 0 fx (x0 , y0 ) = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂
0 1 fy (x0 , y0 )
(c) Finally assume that the surface is given implicitly by the equation
G(x, y, z) = 0 or, more generally by the equation, G(x, y, z) = K, where K is
a constant. If r(t) = x(t) y(t) , z(t) is any curve with r(0) = (x0 , y0 , z0 ) that
lies on the surface, then

G r(t) = K for all t
d
=⇒ G x(t), y(t), z(t) = 0 for all t
dt
Applying the chain rule gives
∂G dx ∂G dy
x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
∂x dt ∂y dt
∂G dz
+ x(t), y(t), z(t) (t) = 0
∂z dt
The left hand side is exactly the dot product of ∂G ∂G ∂G

∂x , ∂y , ∂z = ∇ G with
dx dy dz
dr
dt , dt , dt = dt , so that
∇ G r(t) · r0 (t) = 0

for all t
0
=⇒ ∇ G x0 , y0 , z0 · r (0) = 0
This tell us that ∇ G x0 , y0 , z0 is perpendicular to r0 (0), which is a tangent

vector to G = K at (x0 , y0 , z0 ). This is true for all curves r(t) on G = K and

so is true for all tangent vectors to G = K at (x0 , y0 , z0 ). So ∇G x0 , y0 , z0 is
a normal vector to G(x, y, z) = K at (x0 , y0 , z0 ).
Example 3.2.2 Consider the surface
x = x(u, v) = u cos v
y = y(u, v) = u sin v
z = z(u, v) = u
Observe that
x(u, v)2 + y(u, v)2 = u2 = z(u, v)2
So our surface is also
G(x, y, z) = x2 + y 2 − z 2 = 0
We shall sketch it shortly. But first, let’s find it’s tangent plane at (x0 , y0 , z0 ) =
r(u0 , v0 ). In fact, let’s do it twice. Once using the parametrization and once
using its implicit equation. First, using the parametrization r(u, v) = u cos v ı̂ı +
u sin v ̂ + u k̂, we have
∂r
Tu = (u0 , v0 ) = cos v0 ı̂ı + sin v0 ̂ + k̂
∂u
∂r
Tv = (u0 , v0 ) = −u0 sin v0 ı̂ı + u0 cos v0 ̂
∂v
so that

n = cos v0 ı̂ı + sin v0 ̂ + k̂ × − u0 sin v0 ı̂ı + u0 cos v0 ̂
= − u0 cos v0 , −u0 sin v0 , u0 ) = (−x0 , −y0 , z0 )
Next using the implicit equation G(x, y, z) = x2 + y 2 − z 2 = 0, we have the

normal vector

∇ G x0 , y0 , z0 = (2x0 , 2y0 , −2z0 ) = −2(−x0 , −y0 , z0 )
Of course the two vectors (−x0 , −y0 , z0 ) and −2(−x0 , −y0 , z0 ) are parallel to
each other. Either can be used as a normal vector and the tangent plane to
x2 + y 2 − z 2 = 0 at (x0 , y0 , z0 ) is
0 = n · (x − x0 , y − y0 , z − z0 ) = −x0 (x − x0 ) − y0 (y − y0 ) + z0 (z − z0 )
provided (x0 , y0 , z0 ) 6= 0. In the event that (x0 , y0 , z0 ) = 0 the “tangent plane

equation” reduces to 0 = 0 and there is clearly a problem.
More generally, if Tu × Tv = 0 (or ∇ G(x0 , y0 , z0 ) = 0), then either2
• the surface fails to have a tangent plane at (x0 , y0 , z0 ), or
• our parametrization is screwy3 there. For example, we can parametrize

the xy-plane, z = 0, by r(u, v) = u cos v ı̂ı +u sin v ̂. (This is just polar co-
ordinates.) Then Tu = cos v0 ı̂ı+sin v0 ̂ and Tv = −u0 sin v0 ı̂ı+u0 cos v0 ̂,
so that Tu × Tv = u0 k̂ is 0 when u0 = 0. But the plane z = 0 is its own
tangent plane everywhere.
The surface of current interest is x2 + y 2 = z 2 . The intersection of this surface

with the horizontal plane z = z0 is x2 + y 2 = z02 , which is the circle of radius
|z0 | centred on x = y = 0. So our surface is a stack of circles. The radius of the
circle in the xy-plane is zero. The radius increases linearly as we move away
from the xy-plane. Our surface is a cone. It does not have a tangent plane at
(0, 0, 0).

Example 3.2.3 This time we shall find the tangent planes to the surface
x2 + y 2 − z 2 = 1
As for the cone of the last example, the intersection of this

p surface with the
horizontal plane z = z0 is a circle — the circle of radius 1 + z02 centred on
x = y = 0. Our surface is again a stack of circles. The radius of the circle in
the xy-plane is 1. The radius increases as we move away from the xy-plane.
Here is a sketch of the surface.
2 We saw the same dichotomy when considering what happened for a curve when r0 (t) = 0.
See Example 1.1.10.

3 Of course “screwy” is not a mathematically precise word. One way a parametrization
r(u, v) could be “screwy” is if it failed to give a one-to-one correspondence between parameter

values (u, v) and points on (part of) the surface. For example, polar coordinates r(u, v) =
u cos v ı̂ı + u sin v ̂ give r(0, v) = (0, 0) for all values of v.
It is called a hyperboloid4 of one sheet.

Using the implicit equation G(x, y, z) = x2 + y 2 − z 2 = 1, we have

∇ G x0 , y0 , z0 = (2x0 , 2y0 , −2z0 ) = 2(x0 , y0 , −z0 )
and we may take (x0 , y0 , −z0 ) as a normal vector at (x0 , y0 , z0 ). So the tangent
plane to x2 + y 2 − z 2 = 1 at (x0 , y0 , z0 ) is
0 = n · (x − x0 , y − y0 , z − z0 ) = x0 (x − x0 ) + y0 (y − y0 ) − z0 (z − z0 )
This time n = (x0 , y0 , −z0 ) 6= 0, so that we have a tangent plane, at every point
of the surface. In particular, the vanishing of n = (x0 , y0 , −z0 ) at (x0 , y0 , z0 ) =
(0, 0, 0) is not a problem because (0, 0, 0) is not on the surface.
Example 3.2.4 Optional — Parametrizing the Hyperboloid of One
Sheet. The hyperboloid of one sheet, x2 + y 2 − z 2 = 1, has a symmetry. It is
invariant under rotation about the z-axis. So it is natural to parametrize the
surface using cylindrical coordinates.
z
px, y, zq
x = r cos θ
y = r sin θ z
y
z=z r
θ
px, y, 0q
x
2 2 2
√x +y −z =√
In cylindrical coordinates the surface 1 is r2 − z 2 = 1, and
we could parametrize it by r(θ, z) = 1 + z cos θ ı̂ı + 1 + z 2 sin θ ̂ + z k̂.
2
Alternatively, we can eliminate the square roots in the parametrization by

exploiting the hyperbolic trig functions
1 u 1 u
e − e−u e + e−u

sinh u = cosh u =
2 2
The functions have properties5 that are very similar to those of sin θ and cos θ.
d d
cosh u = sinh u sinh u = cosh u cosh2 u − sinh2 u = 1
du du
We can set r = cosh u, z = sinh u to yield the parametrization
r(θ, u) = cosh u cos θ ı̂ı + cosh u sin θ ̂ + sinh u k̂
As an exercise in working with hyperbolic trig functions, we’ll use this parametriza-
tion to find n̂.
x = cosh u cos θ xu = sinh u cos θ xθ = − cosh u sin θ

4 There are also hyperboloids of two sheets. See Appendix A.8.
y = cosh u sin θ yu = sinh u sin θ yθ = cosh u cos θ

z = sinh u zu = cosh u zθ = 0
So

ı̂ı ̂ k̂
n = Tu × Tθ = det sinh u cos θ sinh u sin θ cosh u
− cosh u sin θ cosh u cos θ 0
= − cosh2 u cos θ , − cosh2 u sin θ , sinh u cosh u

3.2.1 Exercises
1. Is it reasonable to say that the surfaces x2 + y 2 + (z − 1)2 = 1 and
x2 + y 2 + (z + 1)2 = 1 are tangent to each other at (0, 0, 0)?
2. Let the point r0 = (x0 , y0 , z0 ) lie on the surface G(x, y, z) = 0. Assume
that ∇ G(x0 , y0, z0 ) 6= 0. Suppose that the parametrized curve r(t) =
x(t), y(t), z(t) is contained in the surface and that r(t0 ) = r0 . Show
that the tangent line to the curve at r0 lies in the tangent plane to
G = 0 at r0 .
3. Find the parametric equations of the normal line to the surface z =
f (x, y) at the point x0 , y0 , z0 = f (x0 , y0 ) . By definition, the normal
line in question is the line through (x0 , y0 , z0 ) whose direction vector
is perpendicular to the surface at (x0 , y0 , z0 ).
4. Let F (x0 , y0 , z0 ) = G(x0 , y0 , z0 ) = 0 and let the vectors ∇ F (x0 , y0 , z0 )
and ∇ G(x0 , y0 , z0 ) be nonzero and not be parallel to each other. Find
the equation of the normal plane to the curve of intersection of the
surfaces F (x, y, z) = 0 and G(x, y, z) = 0 at (x0 , y0 , z0 ). By definition,
that normal plane is the plane through (x0 , y0 , z0 ) whose normal vector
is the tangent vector to the curve of intersection at (x0 , y0 , z0 ).
5. Let f (x0 , y0 ) = g(x0 , y0 ) and let (fx (x0 , y0 ), fy (x0 , y0 )) 6= (gx (x0 , y0 ), gy (x0 , y0 )).
Find the equation of the tangent line to the curve of intersection of
the surfaces z = f (x, y) and z = g(x, y) at (x0 , y0 , z0 = f (x0 , y0 )).
x2 y
6. ∗. Let f (x, y) = . Find the tangent plane to the surface
x4
+ 2y 2
z = f (x, y) at the point −1 , 1 , 13 .

7. ∗. Find the tangent plane to

27
p =9
x2 + y 2 + z 2 + 3
at the point (2, 1, 1).

8. ∗. Consider the surface z = f (x, y) defined implicitly by the equation
xyz 2 + y 2 z 3 = 3 + x2 . Use a 3--dimensional gradient vector to find the
5 This is no accident: cosh u = cos(iu) and sinh u = −i sin(iu), where i is the usual
complex number that obeys i2 = −1. You can verify these formulae by just checking that
cosh u and cos(iu) have the same Taylor expansions and that sinh u and −i sin(iu) have the
same Taylor expansions.
equation of the tangent plane to this surface at the point (−1, 1, 2).
Write your answer in the form z = ax + by + c, where a, b and c are
constants.
9. ∗. A surface is given by
z = x2 − 2xy + y 2 .
a Find the equation of the tangent plane to the surface at x = a,

y = 2a.
b For what value of a is the tangent plane parallel to the plane
x − y + z = 1?
10. ∗. A surface S is given by the parametric equations
x = 2u2
y = v2
z = u2 + v 3
Find an equation for the tangent plane to S at the point (8, 1, 5).
11. ∗. Let S be the surface given by
r(u, v) = u + v , u2 + v 2 , u − v ,

−2 ≤ u ≤ 2, −2 ≤ v ≤ 2
Find the tangent plane to the surface at the point (2, 2, 0).
12. ∗. Find the tangent plane and normal line to the surface z = f (x, y) =
2y
x2 +y 2 at (x, y) = (−1, 2).
13. ∗. Find all the points on the surface x2 + 9y 2 + 4z 2 = 17 where the

tangent plane is parallel to the plane x − 8z = 0.
14. ∗. Let S be the surface z = x2 + 2y 2 + 2y − 1. Find all points
P (x0 , y0 , z0 ) on S with x0 6= 0 such that the normal line at P contains
the origin (0, 0, 0).
15. ∗. Find all points on the hyperboloid z 2 = 4x2 + y 2 − 1 where the
tangent plane is parallel to the plane 2x − y + z = 0.
16. ∗.
a Find a vector perpendicular at the point (1, 1, 3) to the surface
with equation x2 + z 2 = 10.
b Find a vector tangent at the same point to the curve of inter-
section of the surface in part (a) with surface y 2 + z 2 = 10.
c Find parametric equations for the line tangent to that curve at
that point.
17. ∗. Let P be the point where the curve
r(t) = t3 ı̂ı + t ̂ + t2 k̂, (0 ≤ t < ∞)
intersects the surface

z 3 + xyz − 2 = 0
Find the (acute) angle between the curve and the surface at P .
18. Find all horizontal planes that are tangent to the surface with equation
2
+y 2 )/2
z = xye−(x
What are the largest and smallest values of z on this surface?
3.3 Surface Integrals

We are now going to define two types of integrals over surfaces.
RR
• Integrals that look like S ρ dS are used to compute the area and, when
ρ is, for example, a mass density, the mass of the surface S.
RR
• Integrals that look like S F · n̂ dS, with n̂(x, y, z) being a unit vector
that is perpendicular to S at (x, y, z), are called flux integrals. We shall
see in §3.4, that when v is theRRvelocity field of a moving fluid and ρ is
the density of the fluid, then S ρv · n̂ dS is the rate at which fluid is
crossing the surface S.
3.3.1 Parametrized Surfaces

Suppose that we wish to integrate over part, S, of a surface that is parametrized
by r(u, v). We start by cutting S up into small pieces by drawing a bunch of
curves of constant u (the blue curves in the figure below) and a bunch of curves
of constant v (the red curves in the figure below).
Concentrate on any one the small pieces. Here is a greatly magnified sketch.
u“u0 u“u0 `du
v varying v varying
P2 P3 u varying
v“v0 `dv
P1 u varying
P0 v“v0
For example, the lower red curve was constructed by holding v fixed at
the value v0 , varying u and sketching r(u, v0 ), and the upper red curve was
constructed by holding v fixed at the slightly larger value v0 + dv, varying u
and sketching r(u, v0 + dv). So the four intersection points in the figure are
P2 = r(u0 , v0 + dv) P3 = r(u0 + du, v0 + dv)

P0 = r(u0 , v0 ) P1 = r(u0 + du, v0 )
Now if
R(t) = r(u0 + tdU , v0 + tdV )
(where dU and dV are any small constants) then, by Taylor expansion,

r u0 + dU , v0 + dV = R(1)
≈ R(0) + R0 (0) t − 0 t=1

∂r ∂r
= r(u0 , v0 ) + (u0 , v0 ) dU + (u0 , v0 ) dV
∂u ∂v
Applying this three times, once with dU = du, dV = 0, once with dU = 0
dV = dv, and once with dU = du, dV = dv,
P0 = r(u0 , v0 )
∂r
P1 = r(u0 + du, v0 ) ≈ r(u0 , v0 ) +
(u0 , v0 ) du
∂u
∂r
P2 = r(u0 , v0 + dv) ≈ r(u0 , v0 ) + (u0 , v0 ) dv
∂v
∂r ∂r
P3 = r(u0 + du, v0 + dv) ≈ r(u0 , v0 ) + (u0 , v0 ) du + (u0 , v0 ) dv
∂u ∂v
We have dropped all Taylor expansion terms that are of degree two or higher
in du, dv. The reason is that, in defining the integral, we take the limit
du, dv → 0. Because of that limit, all of the dropped terms contribute exactly
0 to the integral. We shall not prove this. But we shall show, in the optional
§3.3.5, why this is the case.
The small piece of our surface with corners P0 , P1 , P2 , P3 is approximately
a parallelogram with sides
P2 P3
−−−→ −−−→ ∂r
P0 P1 ≈ P2 P3 ≈
∂u
(u0 , v0 ) du ÝÝÝÑ
P0 P2
−−−→ −−−→ ∂r P1
P 0 P2 ≈ P1 P3 ≈
∂v
(u0 , v0 ) dv P0 θÝÝÝÑ
P0 P1
−−−→ −−−→
Denote by θ the angle between the vectors P0 P1 and P0 P2 . The base of the
−−−→ −−−→
parallelogram, P0 P1 , has length P0 P1 , and the height of the parallelogram is
−−−→
P0 P2 sin θ. So the area of the parallelogram is1
−−−→ −−−→ −−−→ −−−→

|P0 P1 | |P0 P2 | sin θ = P0 P1 × P0 P2

∂r ∂r
≈ (u0 , v0 ) × (u0 , v0 )dudv
∂u ∂v
∂r ∂r
Furthermore, ∂u (u0 , v0 ) and ∂v (u0 , v0 ) are tangent vectors to the curves
∂r
r(t , v0 ) and r(u0 , t) respectively. Both of these curves lie in S. So ∂u (u0 , v0 )
∂r
and ∂v (u0 , v0 ) are tangent vectors to S at (u0 , v0 ) and the cross product
∂r ∂r
∂u (u0 , v0 ) × ∂v (u0 , v0 ) is perpendicular to S at (u0 , v0 ). We have found both
dS and n̂ dS, where n̂ is a unit normal vector to the surface.
1 As we mentioned above, the approximation below becomes exact when the limit du, dv →
0 is taken in the definition of the integral. See the optional §3.3.5.

Equation 3.3.1 For the parametrized surface r(u, v),
∂r ∂r
n̂ dS = ± (u , v) × (u , v) dudv
∂u ∂v
∂r ∂r
dS = (u , v) ×
(u , v) dudv
∂u ∂v
The ± sign in 3.3.1 is there because there are two unit normal vectors
at each point of a surface, one on each side of the surface. Typically, the
application itself tells you which of the two normal vectors should be used. We
shall see many examples shortly.
3.3.2 Graphs
The surface which is the graph z = f (x, y) can be parametrized by
r(x, y) = xı̂ı + y ̂ + f (x, y) k̂
As
∂r ∂f ∂r ∂f
= ı̂ı + k̂ and = ̂ + k̂
∂x ∂x ∂y ∂y
we have
 
ı̂ı ̂ k̂
∂r ∂r ∂f 
× = det 1 0 = −fx (x, y)ı̂ı − fy (x, y) ̂ + k̂

∂x 
∂x ∂y ∂f
0 1 ∂y
So, 3.3.1 gives the following.

Equation 3.3.2 For the surface z = f (x, y),

n̂ dS = ± − fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ dxdy
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy
Similarly, for the surface x = g(y, z),

n̂ dS = ± ı̂ı − gy (y, z) ̂ − gz (y, z) k̂ dydz
q
dS = 1 + gy (y, z)2 + gz (y, z)2 dydz
and for the surface y = h(x, z),

n̂ dS = ± − hx (x, z)ı̂ı + ̂ − hz (x, z) k̂ dxdz
p
dS = 1 + hx (x, z)2 + hz (x, z)2 dxdz
Again, in any given application, some care must be taken in choosing the
sign in 3.3.2, so as to get the
pappropriate normal vector.
The formulae like dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy in 3.3.2 have geo-
metric interpretations. The red parallelogram in the sketch
z
n̂ k̂
θ
dx
x dy
p
represents a little piece of our surface. It has area dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy.
The blue parallelogram in the same sketch represents the projection of the red
parallelogram onto the xy-plane. It has area dxdy. The vector n̂ in the sketch
is a unit normal for the red parallelogram. We have seen that it is parallel to
∂r ∂r
× = −fx (x, y)ı̂ı − fy (x, y) ̂ + k̂
∂x ∂y
so that the angle θ between n̂ and k̂ obeys
(−fx (x, y)ı̂ı − fy (x, y) ̂ + k̂) · k̂

cos θ =
− fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ |k̂|
1
=p
1 + fx (x, y)2 + fy (x, y)2
p
The geometric interpretation of dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy is that
the area dS of a little piece of surface is the area of its projection on the xy-plane
times the factor cos1 θ where θ is the angle between n̂ (which is perpendicular
to the surface) and k̂ (which is perpendicular to the xy-plane). Notice that
• when θ is close to zero, which corresponds the f being almost constant
and our surface being almost parallel to the xy-plane, dS reduces to
almost dxdy.
• On the other hand, in the limit θ → π2 , which corresponds to fx and/
or fy becoming infinite and our surface becoming perpendicular to the
xy-plane, dS becomes “infinity times” dxdy. In this case, we should
represent our surface either in the form x = g(y, z) or in the form y =
h(x, z), rather than in the form z = f (x, y).
3.3.3 Surfaces Given by Implicit Equations

Finally suppose that the surface is given by the equation G(x, y, z) = K, with
K a constant. Suppose further that at some point on the surface ∂G ∂z 6= 0.
Then near that point we may solve2 the equation G(x, y, z) = K for z as a
2 This is called the implicit function theorem. We will not prove it. But it is not so hard to
understand why it is true, if one thinks in terms of the Taylor expansion of G about the point.
For simplicity, let’s suppose that the point is (0, 0, 0) and G happens to be exactly equal to
its first order Taylor expansion about (0, 0, 0). That is, G(x, y, z) = A + Bx + Cy + Dz, for
some constants A, B, C, D. Since (0, 0, 0) is on the surface, A = K. As ∂G ∂z
= D 6= 0 we can
1
easily solve G(x, y, z) = K for z as a function of x and y. Namely z = D (−Bx − Cy). The
general proof is based on the fact that, under reasonable hypotheses, the first order Taylor
expansion is a good approximation to G near (0, 0, 0).
function of x and y. That is, the surface also obeys z = f (x, y) for a function
f (x, y) that satisfies
G x, y, f (x, y) = K
near the point. Differentiating this with respect to x and y gives, by the chain
rule,
∂ h i
0= G x, y, f (x, y) = Gx x, y, f (x, y) + Gz x, y, f (x, y) fx (x, y)
∂x
∂ h i
0= G x, y, f (x, y) = Gy x, y, f (x, y) + Gz x, y, f (x, y) fy (x, y)
∂y
which implies

Gx x, y, f (x, y) Gy x, y, f (x, y)
fx (x, y) = − fy (x, y) = −
Gz x, y, f (x, y) Gz x, y, f (x, y)
and

Gx x, y, f (x, y) Gy x, y, f (x, y)
−fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ = ı̂ı + ̂ + k̂
Gz x, y, f (x, y) Gz x, y, f (x, y)

∇ G x, y, f (x, y)
=
Gz x, y, f (x, y)
So, by (3.3.2),
Equation 3.3.3 For the surface G(x, y, z) = K, when Gz (x, y, z) 6= 0,

∇ G x, y, z
n̂ dS = ± dxdy
∇ G x, y, z · k̂

∇ G x, y, z
dS =
dxdy
∇ G x, y, z · k̂
Similarly, for the surface G(x, y, z) = K, when Gx (x, y, z) 6= 0,

∇ G x, y, z
n̂ dS = ± dydz
∇ G x, y, z · ı̂ı

∇ G x, y, z
dS =
dydz
∇ G x, y, z · ı̂ı
and for the surface G(x, y, z) = K, when Gy (x, y, z) 6= 0,

∇ G x, y, z
n̂ dS = ± dxdz
∇ G x, y, z · ̂

∇ G x, y, z
dS = dxdz
∇ G x, y, z · ̂
If, for some point (x0 , y0 , z0 ) we have Gx (x0 , y0 , z0 ) = Gy (x0 , y0 , z0 ) =
Gz (x0 , y0 , z0 ) = 0, we also have a problem! Often this is a sign that our
surface is not smooth at (x0 , y0 , z0 ) and in fact does not have a normal vector
there. For an example of this, see Example 3.2.2.
3.3.4 Examples of
RR
S ρ dS
We’ll start by computing, in several different ways, the surface area of the
hemisphere
x2 + y 2 + z 2 = a2 z≥0
(with a > 0). You probably know, from high school, that the answer is 12 ×
4πa2 = 2πa2 . But you have probably not seen a derivation of this answer. Note
that, since x2 +y 2 = a2 −z 2 on the hemisphere, the
set of (x, y)’s for which there

is a z with (x, y, z) on the hemisphere is exactly (x, y) ∈ R2 x2 + y 2 ≤ a2 .
Example 3.3.4 Area of a hemisphere — using cylindrical coordinates.

Let’s parametrize the hemisphere x2 + y 2 + z 2 = a2 , z ≥ 0, using as parameters
the polar coordinates r, θ of the cylindrical coordinates3
z
px, y, zq
x = r cos θ
y = r sin θ z
y
z=z
r
θ
px, y, 0q
x
and then apply 3.3.1. In cylindrical coordinates the equation x2 + y 2 + z 2 =
a becomes r2 + z 2 = a2 , and the condition x2 + y 2 ≤ a2 is 0 ≤ r ≤ a,
2
0 ≤ θ < 2π.
So the hemisphere can be parametrized by
p
x(r, θ) , y(r, θ) , z(r, θ) = r cos θ , r sin θ , a2 − r2
with 0 ≤ r ≤ a, 0 ≤ θ < 2π
√
Note that we selected the positive solution z = a2 − r2 of r2 + z 2 = a2 in
order to satisfy the condition that z ≥ 0. Since
∂x ∂y ∂z r
, , = cos θ , sin θ , − √
∂r ∂r ∂r a2 − r2
∂x ∂y ∂z
, , = (−r sin θ , r cos θ , 0)
∂θ ∂θ ∂θ
3.3.1 yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , drdθ
∂r ∂r ∂r ∂θ ∂θ ∂θ 
ı̂ı ̂ k̂
= ± det  cos θ sin θ − √a2r−r2  drdθ
 
−r sin θ r cos θ 0
r2 cos θ 2
r sin θ
=± √ , √ , r drdθ
2
a −r 2 a2 − r2
r r
r4 a2 r2 ar
dS = 2 2
+ r2 drdθ = drdθ = √ drdθ
a −r a − r2
2
a2 − r2
So the area of the hemisphere is
Z a Z 2π Z a
ar r
dr dθ √ = 2πa dr √
2
a −r 2 a − r2
2
0 0 0
Z 0
−du/2
= 2πa √
a 2 u
with u = a2 − r2 , du = −2r dr
h √ i0
= 2πa − u
a2
2
= 2πa
as it should be.
3 The
symbols r, θ, z are the standard mathematics symbols for the cylindrical coordinates.
Appendix A.7 gives another set of symbols that is commonly used in the physical sciences
and engineering.
Example 3.3.5 Area of a hemisphere — using an implicit equation.

This time we’ll compute the area of the hemisphere by using that, if (x, y, z) is
on the hemisphere, then G(x, y, z) = a2 with G(x, y, z) = x2 + y 2 + z 2 . Since

∇ G(x, y, z) = 2x , 2y , 2z
3.3.3 yields

∇ G x, y, z
dS = dxdy
∇ G x, y, z · k̂

2x , 2y , 2z
= dxdy
2z
p
x2 + y 2 + z 2
= dxdy
|z|
a
=p dxdy on x2 + y 2 + z 2 = a2
a2 − x2 − y 2
So the area is x2 +y2 ≤a2 √ 2 a 2 2 dxdy. To evaluate this integral, we switch

RR
a −x −y
to polar coordinates, substituting x = r cos θ, y = r sin θ. This gives
ZZ Z a Z 2π
a a
area = p dxdy = dr r dθ √
x2 +y 2 ≤a2
2 2
a −x −y 2
0 0 a − r2
2
Z a
r
= 2πa dr √
0 a2 − r2
We already showed, in Example 3.3.4, that the value of this integral is 2πa2 .

Example 3.3.6 Area of a hemisphere — using spherical coordinates.
Of course “integrating over a sphere” cries out for spherical coordinates. So
this time we parametrize the hemisphere x2 + y 2 + z 2 = a2 , z ≥ 0, using as
parameters the angular coordinates θ, ϕ of the spherical coordinates
z
p0, 0, zq ρ sin ϕ
px, y, zq
x = ρ sin ϕ cos θ ρ cos ϕ
y = ρ sin ϕ sin θ ϕ ρ
z = ρ cos ϕ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and then apply 3.3.1. In spherical coordinates the equation x2 +y 2 +z 2 = a2
becomes just ρ2 = a2 , and the condition z ≥ 0 is 0 ≤ ϕ ≤ π2 , 0 ≤ θ < 2π. So
the hemisphere can be parametrized4 by

x(θ, ϕ) , y(θ, ϕ) , z(θ, ϕ) = a sin ϕ cos θ , a sin ϕ sin θ , a cos ϕ
π
0 ≤ ϕ ≤ , 0 ≤ θ < 2π
2
Since
∂x ∂y ∂z
, , = − a sin ϕ sin θ , a sin ϕ cos θ , 0
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (a cos ϕ cos θ , a cos ϕ sin θ , −a sin ϕ)
∂ϕ ∂ϕ ∂ϕ
3.3.1 yields
∂x∂y ∂z ∂x ∂y ∂z
n̂ dS = ± ,, × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ

= ± − a sin ϕ sin θ, a sin ϕ cos θ, 0 ×(a cos ϕ cos θ, a cos ϕ sin θ, −a sin ϕ) dθdϕ

= ± − a2 sin2 ϕ cos θ , −a2 sin2 ϕ sin θ , −a2 sin ϕ cos ϕ dθdϕ

= ∓a2 sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
and
q
dS = a2 sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos2 ϕ dθdϕ
= a2 sin ϕ dθdϕ
So the area of the hemisphere is

π π
Z 2
Z 2π Z 2 h iπ/2
2 2
a dϕ dθ sin ϕ = 2πa dϕ sin ϕ = 2πa2 − cos ϕ
0 0 0 0
= 2πa2

There is an easier way to do this, using a little geometry.
Example 3.3.7 Area of a hemisphere — using spherical coordinates
again. We are now going to again compute the surface area of the hemisphere
using spherical coordinates. But this time instead of determining dS using the
canned formula 3.3.1, we are going to read it off of a sketch.
Sketch the part of the hemisphere that is in the first octant, x ≥ 0, y ≥ 0,
z ≥ 0. Slice it up into small pieces by drawing in curves of constant θ (the blue
lines in the figure below) and curves of constant ϕ (the red lines in the figure
below).
4 As we have noted before, the spherical coordinate system really breaks down at ϕ = 0,
because ρ = 1, ϕ = 0 gives the same point, namely the north pole (0, 0, 1), for all values
of θ. We should really treat our integral like an improper integral, first integrating over
ε < ϕ ≤ π2 and then taking the limit ε → 0+ . However the breakdown of the spherical
coordinate system at ϕ = 0, just like the breakdown of polar coordinates at r = 0, rarely
causes problem and it is routine to skip the “improper integral” step.
z
a sin ϕ dθ
a dϕ
x
Each piece is approximately a little rectangle. Concentrate on one of them,
like the piece with the thick sides in the figure above. The area, dS, of that
piece is (essentially) the product of its height and its width. Each of the two
sides of the piece is
• a segment of a circle of radius a (a fat blue line in both the figure above
and in the figure on the left below)
• that subtends an angle dϕ
dϕ
• and hence is the fraction 2π of a full circle of radius a and hence is of
length dϕ
2π 2πa = adϕ.
The top of the piece is

• a segment of a circle of radius a sin ϕ (a fat red line in both the figure
above and in the figure on the right below)
• that subtends an angle dθ
dθ
• and hence is the fraction 2π of a full circle of radius a sin ϕ and hence is
dθ
of length 2π 2πa sin ϕ = a sin ϕdθ.
These are drawn in the figure below.
z z
a sin ϕ dθ
dθ
a sin ϕ
a a dϕ
ϕ a
ϕ dϕ y
y
x
x
So the area of our piece is
dS = adϕ a sin ϕdθ = a2 sin ϕ dθdϕ

This is exactly the same formula that we found for dS in Example 3.3.6 so
that we will, yet again, get that the area of a hemisphere of radius a is 2πa2 .
But wait! We can do it again, by yet another method!
(Phew!)
Example 3.3.8 Area of a hemisphere — using z = f (x, y). We’ll

compute the area of the hemisphere one last time5 . This time we’ll use that
the equation of the hemisphere is
p
z = f (x, y) = a2 − x2 − y 2 with (x, y) running over x2 + y 2 ≤ a2
So 3.3.2 yields
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy
s
−x 2 −y 2
= 1+ p + p dxdy
a2 − x2 − y 2 a 2 − x2 − y 2
s
x2 + y 2
= 1+ 2 dxdy
a − x2 − y 2
s
a2
= dxdy
a2 − x2 − y 2
√ a
RR
So the area is x2 +y 2 ≤a2
dxdy. We already found, in Example
a2 −x2 −y 2
3.3.5, that the value of this integral in 2πa2 .
Let’s do some more substantial examples, where the integrand is not 1.
RR 2 2 2
Example 3.3.9 Evaluate S
x y z dS where S is the part of the cone
x2 + y 2 = z 2 with 0 ≤ z ≤ 1.
Solution 1. We can express S as
D2 n̂ “ k̂
V
p
z = f (x, y) = x2 + y 2 x2 +y 2 ≤ 1 S n̂
D1
n̂ “ ´k̂
Now since
x y
fx (x, y) = p fy (x, y) = p
x + y2
2 x2 + y2
3.3.2 gives6
h x2 y 2 i1/2 √
dS = 1 + + dxdy = 2 dxdy
x2 + y 2 x2 + y 2
Our integral is then
ZZ
2 2 2
√ ZZ
x y z dS = 2 x2 y 2 (x2 + y 2 ) dxdy
S x2 +y 2 ≤1
Since we are integrating over a circular domain, let’s convert to polar coordi-
nates.
ZZ
2 2 2
√ Z 2π Z 1
x y z dS = 2 dθ dr r(r cos θ)2 (r sin θ)2 r2
S 0 0
5 We promise!
√ 2π 1
Z Z
2 2 7
= 2 dθ cos θ sin θ dr r
0 0
√ Z 2π √ Z 2π
2 2
= dθ cos2 θ sin2 θ = dθ sin2 (2θ)
8 0 32 0
√ Z 2π
2
= dθ 1 − cos(4θ)
64 0
Remembering7 that the integral of cos(θ), or cos(4θ), over a full period is 0,

we end up with √ √
ZZ
2 2 2 2 π 2
x y z dS = (2π) =
S 64 32
Solution 2. We may parametrize8 the cone by
r(z, θ) = z cos θ ı̂ı + z sin θ ̂ + z k̂ 0 ≤ z ≤ 1, 0 ≤ θ ≤ 2π
Then because
∂r ∂r
= cos θ ı̂ı + sin θ ̂ + k̂ and = −z sin θ ı̂ı + z cos θ ̂
∂z ∂θ
3.3.1 yields9
 
ı̂ı ̂ k̂
n̂ dS = ± det  cos θ sin θ 1  dzdθ
−z sin θ z cos θ 0

= ± − z cos θ ı̂ı − z sin θ ̂ + z k̂ dzdθ
√
dS = 2z dzdθ
So our integral becomes

ZZ √ Z 2π Z 1
x2 y 2 z 2 dS = 2 dθ dz z(z cos θ)2 (z sin θ)2 z 2
S 0 0
√ Z 2π Z 1
= 2 dθ dz z 7 cos2 θ sin2 θ
√ Z0 2π 0
2
= dθ cos2 θ sin2 θ
8 0
We evaluated this integral in Solution 1. So again

ZZ √
2 2 2 π 2
x y z dS =
S 32

Let’s do something more celestial.
Example 3.3.10 Consider a spherical shell of radius a with mass density µ
per unit area. Think of it as a hollow planet10 . We are going to determine
the gravitational force that it exerts on a particle of mass m a distance b away
from its centre. This particle can be either outside the shell (b > a) or inside
6 This answer for dS is a very clean. Think about why. Hint: review the discussion
following 3.3.2.
7 If you have forgotten why, sketch the graph.
8 We did so previously, with different variable names, in Example 3.2.2.
9 Again the formula for dS is very neat. Think about why.
the shell (b < a). We can choose the coordinate system so that the centre
of the shell is at the origin and the particle is at (0, 0, b). By Newton’s law
of gravitation, the force exerted on the particle by a tiny piece of the shell of
surface area dS located at r is
G (µdS) m
(r − (0, 0, b))
|r − (0, 0, b)|3
Here G is the gravitational constant, µdS is the mass of the tiny piece of shell,
m is the mass of the particle
p0, 0, bq
r ´ p0, 0, bq r ´ p0, 0, bq dS
dS p0, 0, bq r
r
and r − (0, 0, b) is the vector from the particle to the piece of shell. If we
work in spherical coordinates, as we did in Example 3.3.6,
dS = a2 sin ϕ dϕdθ
and
r = a sin ϕ cos θ ı̂ı + a sin ϕ sin θ ̂ + a cos ϕ k̂

r − (0, 0, b) = a sin ϕ cos θ ı̂ı + a sin ϕ sin θ ̂ + (a cos ϕ − b) k̂
|r − (0, 0, b)|2 = a2 + b2 − 2ab cos ϕ
The total force is then

Z π Z 2π
a sin ϕ cos θ ı̂ı +a sin ϕ sin θ ̂ +(a cos ϕ−b) k̂
F = Gµma2 dϕ dθ sin ϕ 3/2
0 0 a2 + b2 − 2ab cos ϕ
Note for future reference that the square root in [a2 + b2 − 2ab cos ϕ]3/2 is the
positive square root because [b2 + a2 − 2ab cos ϕ]1/2 is the length of r − (0, 0, b),
which is positive.
This integral is a little different than other integrals that we have encoun-
tered so far in that the integrand is a vector. By definition11 ,
ZZ ZZ ZZ ZZ

G1 ı̂ı + G2 ̂ + G3 k̂ dS = ı̂ı G1 dS + ̂ G2 dS + k̂ G3 dS
S S S S
so we just have to compute the three components separately.

In our case, the ı̂ı and ̂ components
Z π " Z 2π #
a sin ϕ
F · ı̂ı = Gµma2 dϕ sin ϕ 3/2 dθ cos θ
0 a2 + b2 − 2ab cos ϕ 0
Z π " Z 2π #
2 a sin ϕ
F · ̂ = Gµma dϕ sin ϕ 3/2 dθ sin θ
0 a2 + b2 − 2ab cos ϕ 0
R 2π R 2π
are both zero12 because 0
cos θ dθ = 0
sin θ dθ = 0 so that
π 2π
a cos ϕ − b
Z Z
2
F = Gµma k̂ dθ sin ϕ
dϕ 3/2
0 0 a2 + b2 − 2ab cos ϕ
Z π
2 a cos ϕ − b
= 2πGµma k̂ dϕ sin ϕ 3/2
0 a2 + b2 − 2ab cos ϕ
To evaluate this integral we substitute
a2 + b2 − u
u = a2 + b2 − 2ab cos ϕ du = 2ab sin ϕ dϕ cos ϕ =
2ab
When ϕ = 0, u = (a − b)2 and when ϕ = π, u = (a + b)2 , so
(a+b)2 a2 +b2 −u
−b
Z
πGµma 2b
F= k̂ du
b (a−b)2 u3/2
Z (a+b)2 a2 −b2 −u
πGµma 2b
= k̂ du
b (a−b)2 u3/2
πGµma h a2 − b2 u−1/2 1 u1/2 i(a+b)
2
= k̂ −
b 2b −1/2 2b 1/2 (a−b)2
Recalling that u1/2 is the positive square root,
πGµma h b2 − a2 1 a + b b2 − a2 1 |a − b| i
F= k̂ − − +
b b a+b b b |a − b| b
If b > a, so that |a − b| = b − a
πGµma h b − a a + b a + b b − a i G(4πa2 µ)m

F= k̂ − − + =− k̂
b b b b b b2
If b < a, so that |a − b| = a − b
πGµma h b − a a + b a + b a − b i
F= k̂ − + + =0
b b b b b
The moral13 is
• if the particle is inside the shell, it feels no gravitational force at all, and
• if the particle is outside the shell, it feels the same gravitational force as
it would if the entire mass of the shell (4πa2 µ) were concentrated at the
centre of the shell.

14
Example 3.3.11 Optional — Gravity Train. The “Gravity Train”
refers to the following curious, though admittedly not very practical, thought
experiment.
10 A favourite of science fiction and fantasy writers. Plug “subterranean fiction” into your
favourite search engine. While you’re at it, also try “gravity train”. We’ll look at it in the
optional Example 3.3.11. RR RR RR
11 Under this definition we still have (A + B) dS = A dS + B dS.
12 Think about why the ı̂ı and ̂ components should both be zero. Think symmetry.
13 These two results appeared in Isaac Newton’s Principia Mathematica (1687). They are
known as Newton’s “superb theorems”.

• Pretend that the Earth is a perfect sphere of radius R and that it has a
constant mass density ρ.
• Pick any two distinct points on the surface of the Earth. Call them V
and M .
• Bore a tunnel straight through the Earth from V to M .
• Place a train in the tunnel at V . Assume that the only forces acting on
the train are gravity, G, and a normal force, N, that the tunnel imposes
on the train to keep it in the tunnel. In particular, there are no frictional
forces, like air resistance, and the train does not have an engine. Release
the train and assume that it does not melt as it passes through the centre
of the Earth.
N
M V
G
O
What happens?
We’ll simplify our analysis of the motion of the train by picking a convenient
coordinate system.
• First translate our coordinate system so that the centre of the Earth, call
it O, is at the origin, (0, 0, 0).
• Then rotate our coordinate system about the origin so that the origin, V
and M all lie in the xz-plane.
• Then rotate our coordinate system about the y-axis so that V and M
have√the same z-coordinate
Z ≥ 0. So the coordinates
√ of V and M are
± R2 − Z 2 , 0 , Z . Let’s suppose that V is at R2 − Z 2 , 0 , Z and
√
M is at − R2 − Z 2 , 0 , Z . It really doesn’t matter which is which, but
√
we can always arrange that it is V at + R2 − Z 2 , 0 , Z by rotating
around the z-axis by 180◦ if necessary.
z
? N ?
p´ R2 ´ Z 2 , 0 , Zq p R2 ´ Z 2 , 0 , Zq
G
x
p0, 0, 0q
The y- and z-coordinates of the train are always fixed at 0 and Z, respec-
tively. So let’s call the x-coordinate at time t x(t), and look at the x-component
of Newton’s law of motion.
ma = G + N
It is
mx00 (t) = G · ı̂ı
because the normal force N has no ı̂ı component. Recall that Newton’s law of
gravity says that
GM m
G=− r
|r|3
where G is the gravitational constant, r is the vector from O to the train, and
m is the mass of the train. In this case, because of our computation in Example
3.3.10, the train only feels gravity from shells of the Earth that are inside the
train, so that M is the mass of the
z
|r| G
x
part of the Earth whose distance to the centre of the Earth is no more than
|r|. So
4
M = π|r|3 ρ
3
and
Gm 4
mx00 (t) = − 3 π|r|3 ρ r · ı̂ı
|r| 3
so that
4πGρ
x00 (t) + x(t) = 0
3
This is exactly the differential equation of simple harmonic motion. We have
seen it before in Example 2.2.7. Except for the constant 4πGρ
3 , it is identical to
the equation solved in Example A.9.4 of the Appendix A.9, entitled “Review
of Linear Ordinary Differential Equations”. The general solution is
r ! r !
4πGρ 4πGρ
x(t) = C1 cos t + C2 sin t
3 3
with C1 and C2 being

√ arbitrary constants. If we release the √
train, from rest, at
t = 0, then x(0) = R2 − Z 2 and x0 (0) = 0 so that C1 = R2 − Z 2 , C2 = 0
and r !
p
2 2
4πGρ
x(t) = R − Z cos t
3
√
q
4πGρ
The train reaches M when x(t) = − R2 − Z 2 . That is, when cos 3 t =
−1. So the transit time, T , from V to M obeys
r r r
4πGρ 3 3π
T = π =⇒ T = π =
3 4πGρ 4Gρ
Notice that this transit time depends only on the gravitational constant G and
the density of the Earth ρ. In particular it is completely independent of
• where V and M are and, in particular,
• how close together V and M are, and also of
• the radius of the Earth.

In the case of the Earth, the transit time is about 42 minutes.
3.3.5 Optional — Dropping Higher Order Terms in du, dv

In the course of deriving 3.3.1, that is, n̂dS and dS formulae for
u“u0 u“u0 `du
v varying v varying
P2 P3 u varying
v“v0 `dv
P1 u varying
P0 v“v0
we approximated, for example, the vectors

−−−→ ∂r ∂r
P0 P1 = r(u0 + du, v0 ) − r(u0 , v0 )= (u0 , v0 ) du + E1 ≈ (u0 , v0 ) du
∂u ∂u
−−−→ ∂r ∂r
P0 P2 = r(u0 , v0 + dv) − r(u0 , v0 ) = (u0 , v0 ) dv + E2 ≈ (u0 , v0 ) dv
∂v ∂v
where E1 is bounded15 by a constant times du2 and E2 is bounded by a constant
times dv 2 . That is, we assumed that we could just drop E1 and E2 .
So we approximated
−−−→ −−−→ h ∂r i h ∂r i
P0 P1 × P0 P2 = (u0 , v0 ) du + E1 × (u0 , v0 ) dv + E2

∂u ∂v
∂r ∂r
= (u0 , v0 ) du × (u0 , v0 ) dv + E3

∂u ∂v
∂r ∂r
≈ (u0 , v0 ) du × (u0 , v0 ) dv

∂u ∂v
where the length of the vector E3 is bounded by a constant times du2 dv +
du dv 2 . We’ll now see why dropping terms like E3 does not change the value
of the integral at all16 .
14 The British physicist and architect (he was Surveyor to the City of London and chief
assistant to Christopher Wren) Robert Hooke (1635--1703) wrote about the gravity train
idea in a letter to Isaac Newton. A gravity train was used in the 2012 movie Total Recall.
15 Remember the error in the Taylor polynomial approximations.
16 See the optional §1.1.6 of the CLP-2 text for an analogous argument concerning Riemann
sums.
Suppose that our domain of integration consists of all (u, v)’s in a rectangle
of width A and height B, as in the figure below.
v
du
dv
B
A
Subdivide the rectangle into a grid of n × n small subrectangles by drawing
lines of constant v (the red lines in the figure) and lines of constant v (the blue
lines in the figure). Each subrectangle has width du = A n and height dv = n .
B
Now suppose that in setting up the integral we make, for each subrectangle,
an error that is bounded by some constant times
A 2 B A B 2 AB(A + B)
du2 dv + du dv 2 = + =
n n n n n3
Because there are a total of n2 subrectangles, the total error that we have
introduced, for all of these subrectangles, is no larger than a constant times
AB(A + B) AB(A + B)
n2 × =
n3 n
When we define our integral by taking the limit n → 0 of the Riemann sums,
this error converges to exactly 0.
3.3.6 Exercises
1. Let 0 < θ < π2 , and a, b > 0. Denote by S the part of the surface
z = y tan θ with 0 ≤ x ≤ a, 0 ≤ y ≤ b.
a Find the surface area of S without using any calculus.
b Find the surface area of S by using (3.3.2).
2. Let a, b, c > 0. Denote by S the triangle with vertices (a, 0, 0), (0, b, 0)
and (0, 0, c).
a Find the surface area of S in three different ways, each using
(3.3.2).
b Denote by Txy the projection of S onto the xy-plane. (It is the
triangle with vertices (0, 0, 0) (a, 0, 0) and (0, b, 0).) Similarly
use Txz to denote the projection of S onto the xz-plane and Tyz
to denote the projection of S onto the yz-plane. Show that
q
Area(S) = Area(Txy )2 + Area(Txz )2 + Area(Tyz )2
3. Let a, h > 0. Denote by S the part of the cylinder x2 + z 2 = a2 with
x ≥ 0, 0 ≤ y ≤ h and z ≥ 0.
z
S
y
(a, h, 0)
x
a Find the surface area of S without using any calculus.
b Parametrize S by
π
r(θ, y) = a cos θ ı̂ı + y ̂ + a sin θ k̂ 0≤θ≤ , 0≤y≤h
2
Find the surface area of S by using (3.3.1).
4. Let S be the part of the surface z = xy lying inside the cylinder
x2 + y 2 = 3. Find the moment of inertia of S about the z-axis, that
is, ZZ
I= (x2 + y 2 ) dS
S
5. ∗. Find the surface area of the part of the paraboloid z = a2 − x2 − y 2
which lies above the xy--plane.
6. ∗. Find the area of the portion of the cone z 2 = x2 + y 2 lying between
the planes z = 2 and z = 3.
2
7. ∗. Determine the surface area of the surface given by z = 3 x3/2 +

y 3/2 , over the square 0 ≤ x ≤ 1, 0 ≤ y ≤ 1.
8. ∗.
RR of the surface z = f (x, y) above the
a To find the surface area
region D, we integrate D F (x, y) dA. What is F (x, y)?
b Consider a “Death Star”, a ball of radius 2 centred
√ at the origin
with another ball of radius 2 centred at (0, 0, 2 3) cut out of it.
The diagram below shows the slice where y = 0.
z
?
2 3
?
p1, 0, 3q
π
6
2
x
(i) The Rebels want to paint part of the surface of Death Star
hot pink; specifically, the concave part (indicated with a
thick line in the diagram). To help them determine how
much paint is needed, carefully fill in the missing parts of

this integral:
Z Z
surface area = dr dθ
(ii) What is the total surface area of the Death Star?

9. ∗. Find the area of the cone z 2 = x2 + y 2 between z = 1 and z = 16.
p
10. ∗. Find the surface area of that part of the hemisphere z = a2 − x2 − y 2
2 2
which lies within the cylinder x − a2 + y 2 = a2 .

11. The cylinder x2 + y 2 = 2x cuts out a portion S of the upper half of

the cone x2 + y 2 = z 2 . Compute
ZZ
(x4 − y 4 + y 2 z 2 − z 2 x2 + 1) dS
S
12. Find the surface area of the torus obtained by rotating the circle
(x − R)2 + z 2 = r2 (the circle is contained in the xz-plane) about the
z-axis.
13. A spherical shell of radius a is centred at the origin. Find the centroid
(i.e. the centre of mass with constant density) of the part of the sphere
that lies in the first octant.
14. Find the area of that part of the cylinder x2 + y 2 = 2ay lying outside
z 2 = x2 + y 2 .
15. ∗. Let a and b be positive constants, and let S be the part of the
conical surface
a2 z 2 = b2 (x2 + y 2 )
where 0 ≤ z ≤ b. Consider the surface integral
ZZ
I= (x2 + y 2 ) dS.
S
a Express I as a double integral over a disk in the xy-plane.

b Use the parametrization x = t cos θ, y = t sin θ, etc., to express
I as a double integral over a suitable region in the tθ-plane.
c Evaluate I using the method of your choice.

RR
16. Evaluate, for each of the following, the flux S
F · n̂ dS where n̂ is
the outward normal to the surface S.
n
a F = (x2 + y 2 + z 2 ) (xı̂ı + y ̂ + z k̂) and the surface S is the
sphere x2 + y 2 + z 2 = a2 .
b F = xı̂ı + y ̂ + z k̂ and S is the surface of the rectangular box

0 ≤ x ≤ a, 0 ≤ y ≤ b, 0 ≤ z ≤ c.
c F= py ı̂ı + z k̂ and S is the surface of the solid cone 0 ≤ z ≤
1 − x2 + y 2 .
17. ∗. Let S be the part of the surface x2 + y 2 + 2z = 2 that lies above
the square −1 ≤ x ≤ 1, −1 ≤ y ≤ 1.
x2 + y 2
ZZ
a Find p dS.
S 1 + x2 + y 2
b Find the flux of F = xı̂ı + y̂ + z k̂ upward through S.

18. ∗. Let S be the part of the surface z = xy that lies above the square
0 ≤ x ≤ 1, 0 ≤ y ≤ 1 in the xy-plane.
x2 y
ZZ
a Find p dS.
S 1 + x2 + y 2
b Find the flux of F = xı̂ı + y̂ + k̂ upward through S.

19. ∗. Find the area of the part of the surface z = y 3/2 that lies above
0 ≤ x, y ≤ 1.
20. ∗. Let S be spherical cap which consists of the part of the sphere
x2 +y 2 +(z −2)2 = 4 which lies under the plane z = 1. Let f (x, y, z) =
(2 − z)(x2 + y 2 ). Calculate
ZZ
f (x, y, z) dS
S
21. ∗.
a Find a parametrization of the surface S of the cone whose vertex
is at the point (0, 0, 3), and whose base is the circle x2 + y 2 = 4
in the xy-plane. Only the cone surface belongs to S, not the
base. Be careful to include the domain for the parameters.
b Find the z-coordinate of the centre of mass of the surface S from

(a).
22. ∗. Let S be the surface of a cone of height a and base radius a. The
surface S does not include the base of the cone or the interior of the
cone. Find the centre of mass of S.
Locate the cone in a coordinate system so that its base is in the
xy-plane, and its vertex on the z-axis. So the vertex will be the point
(0, 0, a). The base is a circle of radius a in the xy-plane with centre
at the origin. The cone surface is characterized by the fact that for
every point of S, the distance from the z-axis and the distance from
the xy-plane add up to a.
23. ∗. Let S be the portion of the elliptical cylinder x2 + 41 y 2 = 1 lying
between the planes z = 0 and z = 1 and let n̂ denote the outward
normal
RR to S. Let F = xı̂ı + xyz ̂ + zy 4 k̂. Calculate the flux integral
S
F · n̂ dS directly, using an appropriate parameterization of S.
24. ∗. Evaluate the flux integral
ZZ
F · n̂ dS
S
where F(x, y, z) = (x + 1)ı̂ı + (y + 1) ̂ + 2z k̂, and S is the part of the

paraboloid z = 4 − x2 − y 2 that lies above the triangle 0 ≤ x ≤ 1,
0 ≤ y ≤ 1 − x. S is oriented so that its unit normal has a negative
z-component.
25. ∗. Evaluate the surface integral
ZZ
xy 2 dS
S
where
p S is the part of the sphere x2 + y 2 + z 2 = 2 for which x ≥
y + z2.
2
26. ∗. Let S be the surface given by the equation
x2 + z 2 = sin2 y
lying between the planes y = 0 and y = π. Evaluate the integral

ZZ p
1 + cos2 y dS
S
27. ∗. Let S be the part of the paraboloid z = 1 − x2 − y 2 lying above
the xy-plane. At (x, y, z) S has density
z
ρ(x, y, z) = √
5 − 4z
Find the centre of mass of S.
28. ∗. Let S be the part of the plane
x+y+z =2
that lies in the first octant oriented so that n̂ has a positive k̂ com-
ponent. Let
F = xı̂ı + y ̂ + z k̂
Evaluate the flux integral
ZZ
F · n̂ dS
S
RR
29. ∗. Find the net flux S F · n̂ dS of the vector field F(x, y, z) =
(x, y, z) upwards (with respect to the
z-axis) through the surface S
parametrized by r = uv 2 , u2 v , uv for 0 ≤ u ≤ 1, 0 ≤ v ≤ 3.
30. ∗. Let S be the surface obtained by revolving the curve z = ey ,
0 ≤ y ≤ 1, around the y-axis, with the orientation of S having n̂
pointing toward the y-axis.
a Draw a picture of S and find a parameterization of S.
b Compute the integral S ey dS.
RR
RR
c Compute the flux integral S F · n̂ dS where F = (x, 0, z).
31. ∗. Compute the net outward flux of the vector field
r xı̂ı + y ̂ + z k̂
F= =p
|r| x2 + y 2 + z 2
across the boundary of the region between the spheres of radius 1 and
radius 2 centred at the origin.
32. ∗. Evaluate the surface integral S z 2 dS where S is the part of the
RR
cone x2 + y 2 = 4z 2 where 0 ≤ x ≤ y and 0 ≤ z ≤ 1.

RR
33. ∗. Compute the flux integral S F · n̂ dS, where
1 1
F= − x3 − xy 2 , − y 3 , z 2
2 2
and S is the part of the paraboloid z = 5 − x2 − y 2 lying inside the
cylinder x2 + y 2 ≤ 4, with orientation pointing downwards.
34. ∗. Let the thin shell S consist of the part of the surface z 2 = 2xy with
x ≥ 1, y ≥ 1 and z ≤ 2. Find the mass of S if it has surface density
given by ρ(x, y, z) = 3z kg per unit area.
35. ∗. Let S be the portion of the paraboloid x = y 2 + z 2 that satisfies
x ≤ 2y. Its unit normal vector n̂ is so chosen that n̂ · ı̂ı > 0. Find the
flux of F = 2ı̂ı + z ̂ + y k̂ out of S.
36. ∗. Let S denote the portion of the paraboloid z = 1 − 14 x2 − y 2
for which z ≥ 0. Orient S so that its unit normal has a positive k̂
component. Let
F(x, y, z) = (3y 2 + z)ı̂ı + (x − x2 ) ̂ + k̂

RR
Evaluate the surface integral S ∇ × F · n̂ dS.
37. Let S be the boundary of the apple core bounded by the sphere x2 +
y 2 + z 2 =RR16 and the hyperboloid x2 + y 2 − z 2 = 8. Find the flux
integral S
F · n̂ dS where F = xı̂ı + y ̂ + z k̂ and n̂ is the outward
normal to the surface S.
38. ∗.
a Consider the surface S given by the equation
x2 + z 2 = cos2 y
1 π 1

Find an equation for the tangent plane to S at the point 2, 4 , 2 .
b Compute the integral
ZZ
sin y dS
S
where S is the part of the surface from (a) lying between the
planes y = 0 and y = 12 π.
39. ∗. Let f be a function on R3 such that all its
first order
partial deriva-

tives are continuous. Let S be the surface (x, y, z) f (x, y, z) = c
for some c ∈ R. Assume that ∇ f 6= 0 on S. Let F be the gradient
field F = ∇ f .
a Let C be a piecewise smooth curve contained
R in S (not neces-
sarily closed). Must it be true that C F · dr = 0? Explain
why.
b Prove that for any vector field G,
ZZ
(F × G) · n̂ dS = 0.
S
40. ∗.

a Give parametric descriptions of the form r(u, v) = x(u, v) , y(u, v) , z(u, v)
for the following surfaces. Be sure to state the domains of your
parametrizations.
(i) The part of the plane 2x + 4y + 3z = 16 in the first octant

(x, y, z) x ≥ 0, y ≥ 0, z ≥ 0
√
(ii) The cap of the sphere x2 + y 2 + z 2 = 16 for 4/ 2 ≤ z ≤ 4.
(iii) The hyperboloid z 2 = 1 + x2 + y 2 for 1 ≤ z ≤ 10.

b Use your parametrization from part (a) to compute√the surface
area of the cap of the sphere x2 +y 2 +z 2 = 16 for 4/ 2 ≤ z ≤ 4.
41. ∗. Let S be the part of the sphere x2 + y 2 + z 2 = 2 where y ≥ 1,
oriented away from the origin.
a Compute ZZ
y 3 dS
S
b Compute ZZ

xy ı̂ı + xz ̂ + zy k̂ · n̂ dS
S
42. ∗. Let S be the part of the surface (x + y + 1)2 + z 2 = 4 which lies in
the first octant. Find the flux of F downwards through S where
F = xy ı̂ı + (z − xy) ̂
3.4 Interpretation of Flux Integrals

We defined, in §3.3, two types of integrals over surfaces. RR We have seen, in
§3.3.4, some applications that lead to integrals of the typeRR S ρ dS. We now
look at one application that leads to integrals of the type S F · n̂ dS. Recall
that integrals of this type are called flux integrals. Imagine a fluid with
• the density of the fluid (say in kilograms per cubic meter) at position
(x, y, z) and time t being ρ(x, y, z, t) and with
• the velocity of the fluid (say in meters per second) at position (x, y, z)
and time t being v(x, y, z, t).
We are going to determine the rate (say in kilograms per second) at which the
fluid is flowing through a tiny piece dS of surface at (x, y, z). During a tiny
time interval of length dt about time t, fluid near dS moves v(x, y, z, t)dt. The
green line in the figure below is a side view of dS and n̂ = n̂(x, y, z) is a unit
normal vector to dS.
n̂ vdt
vdt
θ
vdt vdt
So during that tiny time interval

• the red line moves to the green line and
• the green line moves to the blue line so that
• the fluid filling the dark grey region below the green line crosses through
dS and moves to light grey region above the green line.
If we denote by θ the angle between n̂ and vdt,

• the volume of fluid that crosses through dS during the time interval dt is
the volume whose side view is the dark grey region below the green line.
This region has base dS and height |vdt| cos θ and so has volume
|v(x, y, z, t)dt| cos θ dS = v(x, y, z, t) · n̂(x, y, z) dt dS
because n̂(x, y, z) has length one.

• The mass of fluid that crosses dS during the time interval dt is then
ρ(x, y, z, t)v(x, y, z, t) · n̂(x, y, z) dt dS
• and the rate at which fluid is crossing through dS is
ρ(x, y, z, t)v(x, y, z, t) · n̂(x, y, z) dS
Integrating dS over a surface S, we conclude that

Lemma 3.4.1 The rate at which fluid mass is crossing through a surface S is
the flux integral
ZZ
ρ(x, y, z, t)v(x, y, z, t) · n̂(x, y, z) dS
S
Here ρ is the density of the fluid, v is the velocity field of the fluid, and n̂(x, y, z)
is a unit normal to S at (x, y, z). If the flux integral is positive the fluid is
crossing in the direction n̂. If it is negative the fluid is crossing opposite to the
direction of n̂. The rate at which volume of fluid is crossing through a surface
S is the flux integral
ZZ
v(x, y, z, t) · n̂(x, y, z) dS
S
3.4.1 Examples of Flux Integrals

Example 3.4.2 Point Source. In Example 2.1.2, we found that the vector
field of a point source1 (in three dimensions) that creates 4πm liters per second
is
m
v(x, y, z) = r̂(x, y, z)
r(x, y, z)2
where
p xı̂ı + y̂ + z k̂
r(x, y, z) = x2 + y 2 + z 2 r̂(x, y, z) = p
x2 + y 2 + z 2
We sketched it in Figure 2.1.3. We’ll now compute the flux of this vector field
across a sphere centred on the origin. Suppose that the sphere has radius R.
n̂
Then the outward2 pointing normal at a point (x, y, z) on the sphere is
xı̂ı + y̂ + z k̂ xı̂ı + y̂ + z k̂

n̂(x, y, z) = r̂(x, y, z) = p =
x2 + y 2 + z 2 R
Note that r̂(x, y, z) · r̂(x, y, z) = 1 and that, on the sphere, r(x, y, z) = R. So

the flux of v outward through the sphere is
ZZ ZZ
m
v · n̂ dS = r̂(x, y, z) · r̂(x, y, z) dS
S S r(x, y, z)2
ZZ
m m
= 2
dS = 2 4πR2
S R R
= 4πm
This is the rate at which volume of fluid is exiting the sphere. In our derivation
of the vector field we assumed that the fluid is incompressible, so it is also the
rate at which the point source is creating fluid.
Example 3.4.3 Vortex. In Figure 2.1.6, we sketched the vector field (in two
dimensions)
v(x, y) = Ω − yı̂ı + x̂
We’ll now compute the flux of this vector field across a circle C centred on the
origin. Suppose that the circle has radius R.
n̂
RBy definition, in two dimensions, the flux of a vector field across a curve C
is C
v · n̂ ds.
This is the natural analog of the flux in three dimensions — the surface S
has been replaced by the curve C, and the surface area dS of a tiny piece of S
has been replaced by the arc length ds of a tiny piece of C.
The outward pointing unit normal at a point (x, y) on our circle C is
xı̂ı + y̂ xı̂ı + y̂
n̂(x, y) = p =
2
x +y 2 R
So
Ω
v(x, y) · n̂(x, y) = − yı̂ı + x̂ · xı̂ı + y̂ = 0
R
and the flux across C is Z
v · n̂ ds = 0
C
This should not be a surprise — no fluid is crossing C at all. This is exactly
what we would expect from looking at the arrows in Figure 2.1.6 or at the
stream lines in Example 2.2.6.
1 You
can imagine that a very small pipe pumps water to the origin.
2 It
doesn’t really matter which unit normal we pick here. We just have to be clear which
one we’re using. With the outward normal, the flux gives the rate at which fluid crosses
the sphere in the outward direction. If we were to use the inward pointing normal, the flux
would give the rate at which fluid crosses the sphere in the inward direction.
RR
F · n̂ dS where
F(x, y, z) = (x + y)ı̂ı + (y + z) ̂ + (x + z) k̂

and S is the boundary of V = (x, y, z) 0 ≤ x2 + y 2 ≤ 9, 0 ≤ z ≤ 5 , and
n̂ is the outward normal3 to S.
Solution. The volume V looks like a tin can of radius 3 and height 5.
n̂ “ k̂
St
n̂
Ss
Sb
n̂ “ ´k̂
It is natural to decompose its surface S into three parts
St = (x, y, z) 0 ≤ x2 + y 2 ≤ 9, z = 5 = the top

Sb = (x, y, z) 0 ≤ x2 + y 2 ≤ 9, z = 0 = the bottom

Ss = (x, y, z) x2 + y 2 = 9, 0 ≤ z ≤ 5 = the side

We’ll compute the flux through each of the three parts separately and then
add them together.
The Top: On the top, the outward pointing normal to S is n̂ = k̂ and
dS = dxdy. This is probably intuitively obvious. But if it isn’t, you can
always derive it by parametrizing the top by r(x, y) = xı̂ı + y ̂ + 5 k̂ with
x2 + y 2 ≤ 9. So the flux through the top is
ZZ ZZ ZZ
F · n̂ dS = 2 2
(x + z) dxdy = (x + 5) dxdy
x +y ≤9
St z=5 x2 +y 2 ≤9
RR
The integral x2 +y2 ≤9 x dxdy = 0 since x is odd and the domain of integration
is symmetric about x = 0. So
ZZ ZZ
F · n̂ dS = 5 dxdy = 5π(3)2 = 45π
St x2 +y 2 ≤9
The Bottom: On the bottom, the outward pointing normal to S is n̂ = −k̂

and dS = dxdy. So the flux through the bottom is
ZZ ZZ ZZ
F · n̂ dS = − 2 2
(x + z) dxdy = − x dxdy = 0
x +y ≤9
Sb z=0 x2 +y 2 ≤9
again since x is odd and the domain of integration is symmetric about x = 0.

The Side: We can parametrize the side by using cylindrical coordinates.

r(θ, z) = 3 cos θ , 3 sin θ , z 0 ≤ θ < 2π, 0 ≤ z ≤ 5
Then, using 3.3.1,

n̂
∂r
= (−3 sin θ , 3 cos θ , 0)
∂θ
∂r r
= (0 , 0 , 1) θ
∂z
∂r ∂r
n̂ dS = × dθ dz
∂θ ∂z
= (3 cos θ , 3 sin θ , 0) dθ dz top view
4
Note that n̂ = (cos θ , sin θ , 0) is outward pointing , as desired. Continuing,

F x(θ, z), y(θ, z), z(θ, z) = 3(cos θ+sin θ)ı̂ı + (3 sin θ+z) ̂ + (3 cos θ+z) k̂
F · n̂ dS = 9 cos2 θ+3 sin θ cos θ+9 sin2 θ+3z sin θ dθ dz

3
= 9 + sin(2θ) + 3z sin θ dθ dz
2
So the flux through the side is
ZZ Z 2π Z 5
3
F · n̂ dS = dθ dz 9 + sin(2θ) + 3z sin θ
Ss 0 0 2
Z 2π Z 5 Z 2π Z 2π
=9 dθ dz since sin θ dθ = sin(2θ) dθ = 0
0 0 0 0
= 9 × 2π × 5 = 90π
and the total flux is

ZZ ZZ ZZ ZZ
F · n̂ dS = F · n̂ dS + F · n̂ dS + F · n̂ dS
S St Sb Ss
= 45π + 0 + 90π = 135π

RR 4ı 2
F · n̂ dS where F(x, y, z) = x ı̂ı + 2y ̂ + z k̂, S
is the half of the surface 41 x2 + 19 y 2 + z 2 = 1 with z ≥ 0, and n̂ is the upward
pointing unit normal.
Solution 1. We start by parametrizing the surface, which is half of an el-

lipsoid. By way of motivation for the parametrization, recall that spherical
coordinates, with ρ = 1, provide a natural way to parametrize the sphere
x2 + y 2 + z 2 = 1. Namely x = cos θ sin ϕ, y = sin θ sin ϕ, z = cos ϕ. The reason
that these spherical coordinates work is that the trig identity cos2 α+sin2 α = 1
3 It is necessary that the problem specify, one way or another, whether n̂ is the inward
RR
pointing normal or the outward pointing normal. Without this, the meaning of F· n̂ dS
S
is ambiguous. Think about where the orientation of the normal vector gets used in your
solution.
4 To check, draw, in your head, a sketch of the top view of the can. “Top view” just means
“ignore the z-coordinate”. The top view of the can is a circle of radius 3. Then, at a generic
point, r = (cos θ, sin θ), on the can, draw the unit normal n̂ = (cos θ , sin θ) with its tail at
r. It is pointing away from the origin, just like r is. That is, n̂ is pointing outward.
implies
x2 + y 2 = cos2 θ sin2 ϕ + sin2 θ sin2 ϕ = sin2 ϕ
and then
x2 + y 2 + z 2 = sin2 ϕ + cos2 ϕ = 1

The equation of our ellipsoid is

x 2 y 2
+ + z2 = 1
2 3
so we can parametrize the ellipsoid by replacing x with x2 and y with y3 in our
parametrization of the sphere. That is, we choose the parametrization
x(θ, ϕ) = 2 cos θ sin ϕ

y(θ, ϕ) = 3 sin θ sin ϕ
z(θ, ϕ) = cos ϕ
with (θ, ϕ) running over 0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤ π/2. Note that

1 1
x(θ, ϕ)2 + y(θ, ϕ)2 + z(θ, ϕ)2 = 1
4 9
as desired.
Then, using 3.3.1,
∂y
∂x ∂z
, , = (−2 sin θ sin ϕ , 3 cos θ sin ϕ , 0)
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (2 cos θ cos ϕ , 3 sin θ cos ϕ , − sin ϕ)
∂ϕ ∂ϕ ∂ϕ
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = − , , × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ
= −(−3 cos θ sin2 ϕ, −2 sin θ sin2 ϕ, −6 sin ϕ cos ϕ)dθdϕ
The extra minus sign in n̂ dS was put there to make the z component of n̂
positive. (The problem specified that n̂ is to be upward unit normal.) As

F x(θ, ϕ) , y(θ, ϕ) , z(θ, ϕ)
= 24 cos4 θ sin4 ϕ ı̂ı + 2 × 32 sin2 θ sin2 ϕ ̂ + cos ϕ k̂
we have
h i
F · n̂ dS = 3 × 24 cos5 θ sin6 ϕ+2 × 2 × 32 sin3 θ sin4 ϕ+6 sin ϕ cos2 ϕ dθdϕ
and the desired integral

π
ZZ Z 2
Z 2π h
F · n̂ dS = dϕ dθ 3 × 24 cos5 θ sin6 ϕ + 2 × 2 × 32 sin3 θ sin4 ϕ
S 0 0
i
+ 6 sin ϕ cos2 ϕ
R 2π R 2π
Since 0
cosm θ dθ = 0
sinm θ dθ = 0 for all odd5 natural numbers m,
ZZ Z π/2 Z 2π Z π/2
F · n̂ dS = dϕ dθ 6 sin ϕ cos2 ϕ = 12π dϕ sin ϕ cos2 ϕ
S 0 0 0
1 hiπ/2
= 12π − cos3 ϕ = 4π
3 0
The integral was evaluated by guessing (and checking) that − 31 cos3 ϕ is an

antiderivative of sin ϕ cos2 ϕ. It can also be done by substituting u = cos ϕ,
du = − sin ϕ dϕ.
Solution 2. This time we’ll parametrize the half-ellipsoid using a variant of
cylindrical coordinates.
x(r, θ) = 2r cos θ
y(r, θ) = 3r sin θ
p
z(r, θ) = 1 − r2
with (r, θ) running over 0 ≤ θ ≤ 2π, 0 ≤ r ≤ 1. Because we built the factors

of 2 and 3 into x(r, θ) and y(r, θ), we have
x(r, θ)2 y(r, θ)2

+ = r2 cos2 θ + r2 sin2 θ = r2
4 9
x(r, θ)2 y(r, θ)2 p 2
=⇒ + + z(r, θ)2 = r2 + 1 − r2 = 1
4 9
as desired. Further z(r, θ) ≥ 0 by our choice of square root in the definition of
z(r, θ).
So, using 3.3.1,
∂x ∂y ∂z
, , = (−2r sin θ, 3r cos θ, 0)
∂θ ∂θ ∂θ
∂x ∂y ∂z r
, , = 2 cos θ, 3 sin θ, − √
∂r ∂r ∂r 1 − r2
∂x ∂y ∂z ∂x ∂y ∂z
n̂dS = − , , × , , dr dθ
∂θ ∂θ ∂θ ∂r ∂r ∂r
3r2 cos θ 2r2 sin θ
=− − √ , −√ , −6r dr dθ
1 − r2 1 − r2
Once again, the extra minus sign in n̂dS was put there to make the z component
of n̂ positive. Continuing,
p
F x(r, θ) , y(r, θ) , z(r, θ) = 24 r4 cos4 θ ı̂ı + 2 × 32 r2 sin2 θ ̂ + 1 − r2 k̂

h r6 r4
F · n̂ dS = 3 × 24 √ cos5 θ + 22 32 √ sin3 θ
1 − r2 1 − r2
p i
+ 6r 1 − r2 dr dθ
R 2π R 2π
Again using that 0 cosm θ dθ = 0 sinm θ dθ = 0 for all odd natural num-
bers m,
Z Z 1 Z 2π p
F · n̂ dS = dr dθ 6r 1 − r2
S 0 0
Z 1
3/2 1
p h 1 i
= 12π dr r 1 − r2 = 12π − (1 − r2 )
0 3 0
= 4π
3/2
The integral was evaluated by guessing (and checking) that − 31 (1 − r2 ) is
√
an antiderivative of r 1 − r2 . It can also be done by substituting u = 1 − r2 ,
du = −2r dr.
Solution 3. The surface is of the form G(x, y, z) = 0 with G(x, y, z) = 41 x2 +

1 2 2
9 y + z − 1. Hence, using 3.3.3,
∇G x
ı̂ı + 2y 
9 ̂ + 2z k̂
n̂dS = dx dy = 2 dx dy
∇ G · k̂ 2z
x y
= ı̂ı + ̂ + k̂ dx dy
4z 9z
x5 2y 3
=⇒ F · n̂ dS = + + z dx dy
4z 9z
It is true that n̂dS, and consequently F · n̂ dS become infinite6
as z → 0. So we should really treat the integral as an improper integral,
first integrating over z ≥ ε and then taking the limit ε → 0+ . But, as we shall
see, the singularity
q is harmless. So it is standard to gloss over this point. On
2 y2 x2 y2
S, z = z(x, y) = 1 − x4 − 9 and 4 + 9 ≤ 1, so
x5 2y 3
Z ZZ
F · n̂ dS = + + z(x, y) dx dy
S
2 x
4
2
+ y9 ≤1 4z(x, y) 9z(x, y)
5 3
x 2y
Both 4z(x,y) and 9z(x,y) are odd under x → −x, y → −y and the domain of
integration is even under x → −x, y → −y, so their integrals are zero and
Z ZZ
F · n̂ dS = 2 2
z(x, y) dx dy
S x
4 + y9 ≤1
r
x2 y2
ZZ
= 1− − dx dy
x2
4
2
+ y9 ≤1 4 9
To evaluate this integral, first make the change of variables 7 x = 2X, dx =

2dX, y = 3Y , dy = 3dY to give
Z ZZ p
F · n̂ dS = 1 − X 2 − Y 2 6 dX dY
S X 2 +Y 2 ≤1
Then switch to polar coordinates, X = r cos θ, Y = r sin θ, dXdY = r drdθ to

give
Z Z 1 Z 2π p Z 1 p
F · n̂ dS = dr 2
dθ 6r 1 − r = 12π dr r 1 − r2
S 0 0 0
1 3/2 1
h i
= 12π − (1 − r2 ) = 4π
3 0
q
2 y2
Solution 4. The surface is of the form z = f (x, y) with f (x, y) = 1 − x4 − 9 .
Hence, using 3.3.2,
 
x y
h ∂f ∂f i ı
ı̂ + 
̂
n̂dS = − ı̂ı − ̂ + k̂ dx dy =  q 4 9
+ k̂ dx dy
∂x ∂y x2
1− 4 − 9 y2
 
5 3
2y
x
r
+ x 2 y 2
=⇒ F · n̂ dS =  q 4 9
+ 1− −  dx dy
1− x2
− y2 4 9
4 9
Note that our unit normal is upward pointing, as required. As in Solution 3,

by the oddness of the x5 and y 3 terms in the integrand,
2y 3
" 5 r #
x
4 + 9 x2 y2
Z ZZ
F · n̂ dS = √ + 1− − dx dy
S x2 y2
4 + 9 ≤1
··· 4 9
r
x2 y2
ZZ
= 1− − dx dy
x2
4
2
+ y9 ≤1 4 9
Now continue as in Solution 3.
3.5 Orientation of Surfaces

One thing that made the flux integrals of the last section possible is that we
could choose sensible unit normal vectors n̂. In this section, we explain this
more carefully.
Consider the sphere x2 + y 2 + z 2 = 1. We can think of this surface as
having two sides — an inside (the side you see when you are living inside
the sphere) and an outside (the side you see when you are living outside the
sphere). Concentrate on one point (x0 , y0 , z0 ) on the sphere. The surface
x2 + y 2 + z 2 = 1 has precisely two unit normal vectors at (x0 , y0 , z0 ), namely
n̂+ = +(x0 , y0 , z0 ) and n̂− = −(x0 , y0 , z0 )
We can view n̂+ as being associated to (or attached to) the outside of the
sphere and n̂− as being associated to (or attached to) the inside of the sphere.
Note that, as we move over the sphere, both n̂+ and n̂− change continuously.
Definition 3.5.1 An oriented surface is a surface together with a continuous
function
N̂ : S → R3
such that, for each point p of S, N̂(p) is a unit normal to S at p. ♦
2
Example 3.5.2 Sphere. One orientation of the sphere S = (x, y, z) x +
2 2
y + z = 1 is
N̂(x, y, z) = (x, y, z)
It associates to each point p of S the outward pointing unit normal to S at p.
We can think of S with this orientation as being
the outer
side of S.
The other orientation of the sphere S = (x, y, z) x2 + y 2 + z 2 = 1 is
N̂(x, y, z) = −(x, y, z)
It associates to each point p of S the inward pointing unit normal to S at p.

We can think of S with this orientation as being the inner side of S.
While this discussion might seem inordinately picky, it turns out that not
all surfaces can be oriented. Our next example exhibits one.
Example 3.5.3 Optional — The Möbius Strip. There are some surfaces
S for which it is not possible to choose a continuous orientation map N̂ :
S → R3 . Such surfaces are said to be non-orientable. The most famous non-
orientable surface is the Möbius1 strip2 , which you can construct as follows.
Take a rectangular strip of paper.
5 Look at the graphs of cosm ϕ and sinm ϕ.
6 That’s because the ellipsoid is becoming vertical as z → 0, so that x and y are not really
good parameters there.
7 The reader interested in general changes of variables in multidimensional integrals should
look up “Jacobian determinant”.

Lay it flat and then introduce a half twist so that the arrow on the right
hand end points upwards, rather than downwards. Then glue the two ends of
the strip together, with the two arrows coinciding. That’s the Möbius strip.
Let’s parametrize it. Think of the strip of paper that we used to construct
it as consisting of a backbone (the horizontal black line in the figure below)
with a bunch of ribs (like the thick blue line in the figure) emanating from it.
v
w
θ
ℓ
When we glue the two ends of the strip together, the black line forms a
circle. If the strip has length `, the circle will have circumference ` and hence
`
radius 2π . We’ll parametrize it as the circle
`
r̂(θ) where r̂(θ) = cos(θ)ı̂ı + sin(θ) ̂
2π
This circle is in the xy-plane. It is the black circle in the figure below. (The
figure only shows the part of the circle in the first octant, i.e. with x, y, z ≥ 0.)
k̂
y
θℓ
2π
r̂pθq
r̂pθq
x
Now we’ll add in the blue ribs. We’ll put the blue rib, that is attached to
`
the backbone at 2π r̂(θ), in the plane that contains the vectors r̂(θ) and k̂.
A side view of the plane that contains the vectors r̂(θ) and k̂ is sketched in
the figure below.
k̂ upv, θ, ϕq
ϕ
ℓ r̂pθq
2π
r̂pθq
To put the half twist into the strip of paper, we want the blue rib to rotate
about the backbone by 180◦ , i.e. π radians, as θ runs from 0 to 2π. That will
be the case if we pick the angle ϕ in the figure to be θ2 . The vector that is
running along the blue rib in the figure is
u(v, θ, ϕ) = v cos(ϕ) r̂(θ) + v sin(ϕ) k̂
where the length, v, of the vector is a parameter. If the width of our original
strip of paper is w, then as the parameter v runs from − w2 to + w2 , the tip of
the vector u(v, θ, ϕ) runs over the entire blue rib. So, choosing ϕ = θ2 , our
parametrization of the Möbius strip is

` θ
r(θ, v) = r̂(θ) + u v, θ,
2π 2

` θ θ
= r̂(θ) + v cos r̂(θ) + v sin k̂
2π 2 2
w w
0 ≤ θ < 2π, − ≤ v ≤
2 2
where r̂(θ) = cos(θ)ı̂ı + sin(θ) ̂.
Now that we have parametrized the Möbius strip, let’s return to the ques-
tion of orientability. Recall, from Definition 3.5.1, that, if the Möbius strip
were orientable, there would exist a continuous function N̂ which assigns to
each point r of the strip a unit normal vector N̂(r) at r. First, we’ll find the
normal vectors to the surface using 3.3.1. The partial derivatives

∂r ` 0 θ 0 v θ v θ
(θ, v) = r̂ (θ) + v cos r̂ (θ) − sin r̂(θ) + cos k̂
∂θ 2π 2 2 2 2 2

∂r θ θ
(θ, v) = cos r̂(θ) + sin k̂
∂v 2 2
are relatively messy, so let’s just consider the case v = 0 (i.e. find the normal
vectors on the backbone). Then
∂r ` 0
(θ, 0) = r̂ (θ)
∂θ 2π
∂r θ θ
(θ, 0) = cos r̂(θ) + sin k̂
∂v 2 2
Since
r̂0 (θ) × r̂(θ) = − sin(θ)ı̂ı + cos(θ) ̂ × cos(θ)ı̂ı + sin(θ) ̂ = −k̂

r̂0 (θ) × k̂ = − sin(θ)ı̂ı + cos(θ) ̂ × k̂ = r̂(θ)

we have

∂r ∂r ` h θ θ i
(θ, 0) × (θ, 0) = − cos k̂ − sin r̂(θ)
∂θ ∂v 2π 2 2
As k̂ and r̂(θ) are mutually perpendicular unit vectors, cos θ2 k̂ − sin θ2 r̂(θ)

has length one, and the two unit normal vectors to the Möbius strip at r(θ, 0)
are
h θ θ i
± cos k̂ − sin r̂(θ)
2 2

So, for each θ, N̂ r(θ, 0) must be either

θ θ h θ θ i
cos k̂ − sin r̂(θ) or − cos k̂ − sin r̂(θ)
2 2 2 2

Imagine walking along the Möbius strip. The normal vector N̂ r(θ, v) is
our body when we are at r(θ, v) — our feet are at the tail of the vector
N̂ r(θ, v) and our head is at the arrow of N̂ r(θ, v) . We start walking at
` `

r(0, 0) = 2π ı̂ı. Our body, N̂ 2π ı̂ı = N̂ r(0, 0) has to be one of ± cos(0) k̂ −

sin(0) r̂(0) = ±k̂. Let’s suppose that N̂ r(0, 0) = +k̂. (We start up-
right.) Now we start walking along the backbone of the Möbius strip, in-
creasing θ. Because N̂ r(θ, 0) has to be continuous, N̂ r(θ, 0) has to be
+ cos θ2 k̂ − sin θ2 r̂(θ) . We keep increasing θ. By continuity, N̂ r(θ, 0)

has to be + cos θ2 k̂ − sin θ2 r̂(θ) for bigger and bigger θ. Eventually we

get to θ = 2π, i.e. to

` ` `
r(2π, 0) = r̂(2π) = ı̂ı = r̂(0) = r(0, 0)
2π 2π 2π
We are back to our starting point. Continuity has forced

h θ θ i
N̂ r(2π, 0) = N̂ r(θ, 0) = + cos k̂ − sin r̂(θ) = −k̂

θ=2π 2 2 θ=2π

So we have arrived back upside down. That’s a problem — N̂ r(2π, 0) =
` `

N̂ 2π ı̂ı and we have already defined N̂ 2π ı̂ı = +k̂, not −k̂. So the Möbius
strip is not orientable. The interested reader should look up M. C. Escher’s
Möbius Strip II (Red Ants).
1 August Ferdinand Möbius (1790--1868) was a German mathematician and astronomer.
He was a descendant of Martin Luther and a student of Gauss.

2 Another famous non-orientable surface is the Klein bottle. You can easily find discussions
of it using your favourite search engine.

Chapter 4
Integral Theorems
4.1 Gradient, Divergence and Curl

“Gradient, divergence and curl”, commonly called “grad, div and curl”, refer
to a very widely used family of differential operators and related notations that
we’ll get to shortly. We will later see that each has a “physical” significance.
But even if they were only shorthand1 , they would be worth using.
For example, one of Maxwell’s equations (relating the electric field E and
the magnetic field B) written without the use of this notation is
∂E ∂E2 ∂E ∂E1 ∂E ∂E1
3 3 2
− ı̂ı − − ̂ + − k̂
∂y ∂z ∂x ∂z ∂x ∂y
1 ∂B1 ∂B2 ∂B3
=− ı̂ı + ̂ + k̂
c ∂t ∂t ∂t
The same equation written using this notation is
1 ∂B
∇×E=−
c ∂t
The shortest way to write (and easiest way to remember) gradient, divergence
∂
and curl uses the symbol “∇∇” which is a differential operator like ∂x . It is
defined by
∂ ∂ ∂
∇ = ı̂ı + ̂ + k̂
∂x ∂y ∂z
and is called “del” or “nabla”. Here are the definitions.
Definition 4.1.1
a The gradient of a scalar-valued function f (x, y, z) is the vector field
∂f ∂f ∂f
grad f = ∇ f = ı̂ı + ̂ + k̂
∂x ∂y ∂z
Note that the input, f , for the gradient is a scalar-valued function, while
∇f , is a vector-valued function.
the output,∇
b The divergence of a vector field F(x, y, z) is the scalar-valued function
∂F1 ∂F2 ∂F3

div F = ∇ · F = + +
∂x ∂y ∂z
1 Good shorthand is not only more brief, but also aids understanding “of the forest by
hiding the trees”.
178
CHAPTER 4. INTEGRAL THEOREMS 179
Note that the input, F, for the divergence is a vector-valued function,

while the output, ∇ · F, is a scalar-valued function.
c The curl of a vector field F(x, y, z) is the vector field

∂F ∂F2 ∂F ∂F1 ∂F ∂F1
3 3 2
curl F = ∇ × F = − ı̂ı − − ̂ + − k̂
∂y ∂z ∂x ∂z ∂x ∂y
Note that the input, F, for the curl is a vector-valued function, and the
output, ∇ × F, is a again a vector-valued function.
d The Laplacian2 of a scalar-valued function f (x, y, z) is the scalar-valued
function
∂2f ∂2f ∂2f
∆f = ∇ 2 f = ∇ · ∇ f = + +
∂x2 ∂y 2 ∂z 2
The Laplacian of a vector field F(x, y, z) is the vector field
∂2F ∂2F ∂2F

∆F = ∇ 2 F = ∇ · ∇ F = + +
∂x2 ∂y 2 ∂z 2
Note that the Laplacian maps either a scalar-valued function to a scalar-
valued function, or a vector-valued function to a vector-valued function.
♦
The gradient, divergence and Laplacian all have obvious generalizations to
dimensions other than three. That is not the case for the curl. It does have
a, far from obvious, generalization, which uses differential forms. Differential
forms are well beyond our scope, but are introduced in the optional §4.7.
Example 4.1.2 As an example of an application in which both the divergence
and curl appear, we have Maxwell’s equations3 4 5 , which form the foundation
of classical electromagnetism.
∇ · E = 4πρ
∇·B=0
1 ∂B
∇×E+ =0
c ∂t
1 ∂B 4π
∇×B− = J
c ∂t c
Here E is the electric field, B is the magnetic field, ρ is the charge density, J
is the current density and c is the speed of light.
2 Pierre-Simon Laplace (1749–1827) was a French mathematician and astronomer. He
is also the Laplace of Laplace’s equation, the Laplace transform, and the Laplace-Bayes
estimator. He was Napoleon’s examiner when Napoleon attended the Ecole Militaire in
Paris.
3 To be picky, these are Maxwell’s equations in the absence of a material medium and in
Gaussian units.
4 One important consequence of Maxwell’s equations is that electromagnetic radiation,
like light, propagate at the speed of light.

5 James Clerk Maxwell (1831–1879) was a Scottish mathematical physicist. In a poll
of prominent physicists, Maxwell was voted the third greatest physicist of all time. Only
Newton and Einstein beat him.
4.1.1 Vector Identities

d
Two computationally extremely important properties of the derivative dx are
linearity and the product rule.
d df dg
af (x) + bg(x) = a (x) + b (x)
dx dx dx
d df dg
f (x) g(x) = g(x) (x) + f (x) (x)
dx dx dx
Gradient, divergence and curl also have properties like these, which indeed
stem (often easily) from them. First, here are the statements of a bunch of
them. (A memory aid and proofs will come later.) In fact, here are a very large
number of them. Many are included just for completeness. Only a relatively
small number are used a lot. They are in red.
Theorem 4.1.3 Gradient Identities.
(a) ∇ (f + g) = ∇ f + ∇ g
(b) ∇ (cf ) = c ∇ f , for any constant c
(c) ∇ (f g) = (∇
∇f )g + f (∇
∇g)
(d) ∇(f /g) = g ∇f − f ∇g /g 2 at points x where g(x) 6= 0.

(e) ∇(F · G) = F × (∇
∇ × G) − (∇
∇ × F) × G + (G · ∇)F + (F · ∇)G
Here6
∂F ∂F ∂F
(G · ∇ )F = G1 + G2 + G3
∂x ∂y ∂z
Theorem 4.1.4 Divergence Identities.
(a) ∇ · (F + G) = ∇ · F + ∇ · G
(b) ∇ · (cF) = c ∇ · F, for any constant c
(c) ∇ · (f F) = (∇
∇f ) · F + f ∇ · F
(d) ∇ · (F × G) = (∇
∇ × F) · G − F · (∇
∇ × G)
Theorem 4.1.5 Curl Identities.
(a) ∇ × (F + G) = ∇ × F + ∇ × G
(b) ∇ × (cF) = c ∇ × F, for any constant c
(c) ∇ × (f F) = (∇
∇f ) × F + f ∇ × F
(d) ∇ × (F × G) = F(∇
∇ · G) − (∇
∇ · F)G + (G · ∇ )F − (F · ∇ )G
Here
∂F ∂F ∂F
(G · ∇ )F = G1 + G2 + G3
∂x ∂y ∂z
Theorem 4.1.6 Laplacian Identities.
(a) ∇ 2 (f + g) = ∇ 2 f + ∇ 2 g
(b) ∇ 2 (cf ) = c ∇ 2 f , for any constant c
(c) ∇ 2 (f g) = f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
6 This is really the only definition that makes sense. For example G · (∇
∇F) does not make
sense because you can’t take the gradient of a vector-valued function.
Theorem 4.1.7 Degree Two Identities.

(a) ∇ · (∇
∇ × F) = 0 (divergence of curl)
(b) ∇ × (∇
∇f ) = 0 (curl of gradient)
(c) ∇ · f {∇

∇g × ∇h} = ∇f · (∇∇g × ∇h)
(d) ∇ · (f∇ ∇f ) = f ∇ 2 g − g ∇ 2 f
∇g − g∇
(e) ∇ × (∇
∇ × F) = ∇ (∇∇ · F) − ∇ 2 F (curl of curl)
Memory Aid. Most of the vector identities (in fact all of them except
Theorem 4.1.3.e, Theorem 4.1.5.d and Theorem 4.1.7) are really easy to guess.
Just combine the conventional linearity and product rules with the facts that
• if the left hand side is a vector (scalar), then the right hand side must
also be a vector (scalar) and
• the only valid products of two vectors are the dot and cross products and
• the product of a scalar with either a scalar or a vector cannot be either
a dot or cross product and
• A × B = −B × A. (The cross product is antisymmetric.)
∇f )·F+f ∇ ·F.
For example, consider Theorem 4.1.4.c, which says ∇ ·(f F) = (∇
• The left hand side, ∇ · (f F), is a scalar, so the right hand side must also
be a scalar.
• The left hand side, ∇ · (f F), is a derivative of the product of f and
F, so, mimicking the product rule, the right hand side will be a sum
of two terms, one with F multiplying a derivative of f , and one with f
multiplying a derivative of F.
• The derivative acting on f must be ∇ f , because ∇ · f and ∇ × f are not
well-defined. To end up with a scalar, rather than a vector, we must take
the dot product of ∇ f and F. So that term is (∇ ∇f ) · F.
• The derivative acting on F must be either ∇ · F or ∇ × F. We also need
to multiply by the scalar f and end up with a scalar. So the derivative
∇ · F}.
must be a scalar, i.e. ∇ · F and that term is f {∇
∇f ) · F + f ∇ · F, which, thankfully, is
• Our final guess is ∇ · (f F) = (∇
correct.
Proof of Theorems 4.1.3, 4.1.4, 4.1.5, 4.1.6 and 4.1.7. All of the proofs (except
for those of Theorem 4.1.7.c,d, which we will return to later) consist of
• writing out the definition of the left hand side and
• writing out the definition of the right hand side and
• observing (possibly after a little manipulation) that they are the same.
For Theorem 4.1.3.a,b, Theorem 4.1.4.a,b, Theorem 4.1.5.a,b and Theorem
4.1.6.a,b, the computation is trivial — one line per identity, if one uses some
efficient notation. Rename the coordinates x, y, zP to x1 , x2 , x3 and the standard
3 ∂
unit basis vectors ı̂ı, ̂, k̂ to ı̂ı1 , ı̂ı2 , ı̂ı3 . Then ∇ = n=1 ı̂ın ∂x n
and the proof of,
for example, Theorem 4.1.4.a is
3
X ∂
∇ · (F + G) = ı̂ın · (F + G)
n=1
∂xn
3 3
X ∂ X ∂
= ı̂ın · F + ı̂ın · G = ∇ · F + ∇ · G
n=1
∂x n n=1
∂x n
For Theorem 4.1.3.c,d, Theorem 4.1.4.c, Theorem 4.1.5.c and Theorem

4.1.6.c, the computation is easy — a few lines per identity. For example,
the proof of Theorem 4.1.5.c is
3 3
X ∂ X ∂
∇ × (f F) = ı̂ın × (f F) = f {ı̂ın × F}
n=1
∂xn n=1
∂xn
3 3
X ∂f X ∂
= ı̂ın × F + f ı̂ın × F
n=1
∂xn n=1
∂xn
∇f ) × F + f ∇ × F
= (∇
In the second line we used Theorem 1.1.3.b. The similar verification of Theo-
rems 4.1.3.c,d, 4.1.4.c, and 4.1.6.c are left as exercises. The latter two are parts
(a) and (c) of Question 3 in Section 4.1 of the CLP-4 problembook.
For Theorem 4.1.4.d, the computation is also easy if one uses the fact that
a · (b × c) = (a × b) · c
which is Lemma 4.1.8.a below. The verification of Theorem 4.1.4.d is part (b)
of Question 3 in Section 4.1 of the CLP-4 problembook.
That leaves the proofs of Theorem 4.1.3.e, Theorem 4.1.5.d, Theorem 4.1.7.a,b,c,d,e,
which we write out explicitly.
Theorem 4.1.3.e: First write out the left hand side as
3 3 3
X ∂ X ∂F X ∂G
∇ (F · G) = ı̂ın (F · G) = ı̂ın ·G + ı̂ın F ·
n=1
∂xn n=1
∂xn n=1
∂xn
Then rewrite a × (b × c) = (c · a)b − (b · a)c, which is Lemma 4.1.8.b below,

as
(c · a)b = a × (b × c) + (b · a)c
∂F ∂G
Applying it once with b = ı̂ın , c = ∂x n
, a = G and once with b = ı̂ın , c = ∂xn ,
a = F gives
3
X ∂F ∂F
∇ (F · G) = G × ı̂ın × + (G · ı̂ın )
n=1
∂xn ∂xn
3
X ∂G ∂G
+ F × ı̂ın × + (F · ı̂ın )
n=1
∂xn ∂xn
∇ × F) + (G · ∇ )F + F × (∇
= G × (∇ ∇ × G) + (F · ∇ )G
Theorem 4.1.5.d: We use the same trick. Write out the left hand side as
3
X ∂
∇ × (F × G) = ı̂ın × (F × G)
n=1
∂xn
3 ∂F 3
X X ∂G
= ı̂ın × ×G + ı̂ın × F ×
n=1
∂xn n=1
∂xn
Applying a × (b × c) = (c · a)b − (b · a)c, which is Lemma 4.1.8.b below,

3 h 3
X ∂F ∂Fn i X h ∂Gn ∂G i
∇ × (F × G) = Gn − G + F − Fn
n=1
∂xn ∂xn n=1
∂xn ∂xn
∇ · F)G + (∇
= (G · ∇ )F − (∇ ∇ · G)F − (F · ∇ )G
Theorem 4.1.7.a: Substituting in

∂F ∂F2 ∂F ∂F1 ∂F ∂F1
3 3 2
∇×F= − ı̂ı − − ̂ + − k̂
∂y ∂z ∂x ∂z ∂x ∂y
gives
∂ ∂F3 ∂F2 ∂ ∂F3 ∂F1 ∂ ∂F2 ∂F1
∇· ∇×F = − − − + −
∂x ∂y ∂z ∂y ∂x ∂z ∂z ∂x ∂y
2 2 2 2 2 2
∂ F3 ∂ F2 ∂ F3 ∂ F1 ∂ F2 ∂ F1
= − − + + −
∂x∂y ∂x∂z ∂y∂x ∂y∂z ∂z∂x ∂z∂y
=0
because the two red terms have cancelled, the two blue terms have cancelled
and the two black terms have cancelled.
Theorem 4.1.7.b: Substituting in
∂f ∂f ∂f
∇f = ı̂ı + ̂ + k̂
∂x ∂y ∂z
gives
∂ ∂f ∂ ∂f ∂ ∂f ∂ ∂f
∇ × ∇f = − ı̂ı − − ̂
∂y ∂z ∂z ∂y ∂x ∂z ∂z ∂x
∂ ∂f ∂ ∂f
+ − k̂
∂x ∂y ∂y ∂x
=0
Theorem 4.1.7.c: By Theorem 4.1.4.c, followed by Theorem 4.1.4.d,

∇g × ∇ h) = ∇ f · (∇
∇ · f (∇ ∇g × ∇ h) + f∇
∇ · (∇
∇g × ∇ h)

∇g × ∇ h) + f (∇
= ∇ f · (∇ ∇ × ∇ g) · ∇ h − ∇ g · (∇
∇ × ∇ h)
By Theorem 4.1.7.b, ∇ × ∇ g = ∇ × ∇ h = 0, so

∇g × ∇ f ) = ∇ f · (∇
∇ · f (∇ ∇g × ∇ h)
Theorem 4.1.7.d: By Theorem 4.1.4.c,
∇g − g∇
∇ · (f∇ ∇f ) = (∇
∇f ) · (∇
∇g) + f ∇ · (∇
∇g) − (∇
∇g) · (∇
∇f ) + g ∇ · (∇
∇f )
= f ∇2g − g ∇2f
Theorem 4.1.7.e:
3 3
X 3
X ∂ ∂ X
∇ × F) =
∇ × (∇ ı̂ı` × ı̂ım × ı̂ın Fn
∂x` m=1
∂xm n=1
`=1
3
X ∂ 2 Fn
= ı̂ı` × ı̂ım × ı̂ın
∂x` ∂xm
`,m,n=1
Using a × (b × c) = (c · a)b − (b · a)c, we have

ı̂ı` × ı̂ım × ı̂ın = (ı̂ı` · ı̂ın )ı̂ım − (ı̂ı` · ı̂ım )ı̂ın = δ`,nı̂ım − δ`,mı̂ın
wherea (
1 if m = n
δm,n =
0 if m 6= n
Hence
3 3
X ∂ 2 Fn X ∂ 2 Fn
∇ × F) =
∇ × (∇ δ`,nı̂ım − δ`,mı̂ın
∂x` ∂xm ∂x` ∂xm
`,m,n=1 `,m,n=1
3 3
X ∂ ∂Fn X ∂ 2 Fn
= ı̂ım − ı̂ın
m,n=1
∂xm ∂xn m,n=1 ∂x2m
∇ · F) − ∇ 2 F
= ∇ (∇

aδ
m,n is called the Kronecker delta function. It is named after the German number
theorist and logician Leopold Kronecker (1823–1891). He is reputed to have said “God made
the integers. All else is the work of man.”
Lemma 4.1.8
a a · (b × c) = (a × b) · c
b a × (b × c) = (c · a)b − (b · a)c
Proof. (a) Here are two proofs. For the first, just write out both sides
a · (b × c) = (a1 , a2 , a3 ) · (b2 c3 − b3 c2 , b3 c1 − b1 c3 , b1 c2 − b2 c1 )
= a1 b2 c3 − a1 b3 c2 + a2 b3 c1 − a2 b1 c3 + a3 b1 c2 − a3 b2 c1
(a × b) · c = (a2 b3 − a3 b2 , a3 b1 − a1 b3 , a1 b2 − a2 b1 ) · (c1 , c2 , c3 )
= a2 b3 c1 − a3 b2 c1 + a3 b1 c2 − a1 b3 c2 + a1 b2 c3 − a2 b1 c3
and observe that they are the same.
For the second proof, we again write out both sides, but this time we express
them in terms of determinants.
 
ı̂ı ̂ k̂
a · b × c = (a1 , a2 , a3 ) · det b1 b2 b3 
c1 c2 c3

b2 b3 b1 b3 b1 b2
= a1 det − a2 det + a3 det
c2 c3 c1 c3 c1 c2
 
a1 a2 a3
= det  b1 b2 b3 
c1 c2 c3
 
ı̂ı ̂ k̂
a × b · c = det a1 a2 a3  · (c1 , c2 , c3 )
b1 b2 b3

a2 a3 a1 a3 a1 a2
= c1 det − c2 det + c3 det
b2 b3 b1 b3 b1 b2
 
c1 c2 c3
= det a1 a2 a3 

b1 b2 b3
Exchanging two rows in a determinant changes the sign of the determinant.
Moving the top row of a 3 × 3 determinant to the bottom row requires two
exchanges of rows. So the two 3 × 3 determinants are equal.
(b) The proof is not exceptionally difficult — just write out both sides and
grind. Substituting in
b × c = (b2 c3 − b3 c2 )ı̂ı − (b1 c3 − b3 c1 )̂ + (b1 c2 − b2 c1 )k̂
gives, for the left hand side,

 
ı̂ı ̂ k̂
a × (b × c) = det  a1 a2 a3 
b2 c3 − b3 c2 −b1 c3 + b3 c1 b1 c2 − b2 c1

= ı̂ı a2 (b1 c2 − b2 c1 ) − a3 (−b1 c3 + b3 c1 )

−̂ a1 (b1 c2 − b2 c1 ) − a3 (b2 c3 − b3 c2 )

+k̂ a1 (−b1 c3 + b3 c1 ) − a2 (b2 c3 − b3 c2 )
On the other hand, the right hand side
(a · c)b − (a · b)c = (a1 c1 + a2 c2 + a3 c3 )(b1ı̂ı + b2̂ + b3 k̂)

− (a1 b1 + a2 b2 + a3 b3 )(c1ı̂ı + c2̂ + c3 k̂)

= ı̂ı a1 b1 c1 + a2 b1 c2 + a3 b1 c3 − a1 b1 c1 − a2 b2 c1 − a3 b3 c1

+̂ a1 b2 c1 + a2 b2 c2 + a3 b2 c3 − a1 b1 c2 − a2 b2 c2 − a3 b3 c2

+k̂ a1 b3 c1 + a2 b3 c2 + a3 b3 c3 − a1 b1 c3 − a2 b2 c3 − a3 b3 c3
= ı̂ı [a2 b1 c2 + a3 b1 c3 − a2 b2 c1 − a3 b3 c1 ]
+̂ [a1 b2 c1 + a3 b2 c3 − a1 b1 c2 − a3 b3 c2 ]
+k̂ [a1 b3 c1 + a2 b3 c2 − a1 b1 c3 − a2 b2 c3 ]
The last formula that we had for the left hand side is the same as the last
formula we had for the right hand side.
Example 4.1.9 Screening tests. We have seen the vector identity Theorem
4.1.7.b before. It says that if a vector field F is of the form F = ∇ ϕ for some
some function ϕ (that is, if F is conservative), then
∇ϕ) = 0
∇ × F = ∇ × (∇
Conversely, we have also seen, in Theorem 2.4.8, that, if F is defined and has
continuous first order partial derivatives on all of R3 , and if ∇ × F = 0, then F
is conservative. The vector identity Theorem 4.1.7.b is our screening test for
conservativeness.
Because its right hand side is zero, the vector identity Theorem 4.1.7.a is
suggestive. It says that if a vector field F is of the form F = ∇ × A for some
some vector field A, then
∇ × A) = 0
∇ · F = ∇ · (∇
When F = ∇ ×A, A is called a vector potential for F. We shall see in Theorem

4.1.16, below, that, conversely, if F(x) is defined and has continuous first order
partial derivatives on all of R3 , and if ∇ · F = 0, then F has a vector potential7 .
The vector identity Theorem 4.1.7.a is indeed another screening test.
As an example, consider the Maxwell’s equations
∇·B=0
1 ∂B
∇×E+ =0
c ∂t
that we saw in Example 4.1.2. The first equation implies that (assuming B
is sufficiently smooth) there is a vector field A, called the magnetic potential,
with B = ∇ × A. Substituting this into the second equation gives
1∂ 1 ∂A
0=∇×E+ ∇×A=∇× E+
c ∂t c ∂t
So E+ 1c ∂A
∂t passes the screening test of Theorem 4.1.7.b and there is a function
ϕ (called the electric potential) with
1 ∂A
E+ = −∇ϕ
c ∂t
We have put in the minus sign just to provide compatibility with the usual
physics terminology.
Example 4.1.10 Let r(x, y, z) = xı̂ı +y ̂ +z k̂ and let ψ(x, y, z) be an arbitrary
function. Verify that
∇ · r × ∇ψ = 0
Solution. By the vector identity Theorem 4.1.4.d,

∇ · r × ∇ ψ = (∇ ∇ × r) · ∇ ψ − r · ∇ × (∇
∇ψ)
By the vector identity Theorem 4.1.7.b, the second term is zero. Now since
∂z ∂y ∂z ∂x ∂y ∂x
∇×r= − ı̂ı − − ̂ + − k̂ = 0
∂y ∂z ∂x ∂z ∂x ∂y

the first term is also zero. Indeed ∇ · r × ∇ψ = 0 holds for any curl free
r(x, y, z).
4.1.2 Vector Potentials

We’ll now further explore the vector potentials that were introduced in Exam-
ple 4.1.9. First, here is the formal definition.
Definition 4.1.11 The vector field A is said to be a vector potential for
the vector field B if
B=∇×A
♦
As we saw in Example 4.1.9, if a vector field B has a vector potential, then
the vector identity Theorem 4.1.7.a implies that ∇ · B = 0. This fact deserves
to be called a theorem.
Theorem 4.1.12 Screening test for vector potentials. If there exists a
vector potential for the vector field B, then
∇·B=0
Of course, we’ll consider the converse soon. Also note that the vector
potential, when it exists, is far from unique. Two vector fields A and Ã are
both vector potentials for the same vector field if and only if
∇ × A = ∇ × Ã ⇐⇒ ∇ × (A − Ã) = 0
That is, if and only if the difference A−Ã passes the conservative field screening
test of Theorems 2.3.9 and 2.4.8. In particular, if A is one vector potential for
8 Does this remind you of Theorem 2.4.8? It should.
a vector field B (i.e. if B = ∇ × A), and if ψ is any function, then

∇ × (A + ∇ ψ) = ∇ × A + ∇ × ∇ ψ = B
by the vector identity Theorem 4.1.7.b. That is, A + ∇ ψ is another vector
potential for B.
ψ so that,∂ψ
To simplify computations, we can always choose for example,
the third component
R of A + ∇ ψ, namely A + ∇ ψ · k̂ = A 3 + ∂z , is zero —
just choose ψ = − A3 dz. We have just proven
Lemma 4.1.13 If the vector field B has a vector potential, then, in particular,
there is a vector potential A for B with8 A3 = 0.
Here is an example which exploits this choice to simplify the computations
used to find a vector potential.
Example 4.1.14 Let
B = yz ı̂ı + zx ̂ + xy k̂
This vector field has been set up carefully to obey
∂ ∂ ∂
∇·B= (yz) + (zx) + (xy) = 0
∂x ∂y ∂z
and so passes the screening test of Theorem 4.1.12.
Let’s try and find a vector potential for B. That is, let’s try and find a
vector field A = A1 ı̂ı + A2 ̂ + A3 k̂ that obeys ∇ × A = B, or equivalently,
∂A3 ∂A2
− = B1 = yz
∂y ∂z
∂A3 ∂A1
− + = B2 = zx
∂x ∂z
∂A2 ∂A1
− = B3 = xy
∂x ∂y
This system is nasty to solve because every equation contains more than one
of the three unknowns, A1 , A2 , A3 . Let us take advantage of our observation
above that, if any vector potential exists, then, in particular, a vector potential
A exists that also obeys A3 = 0. So let’s also require that A3 = 0. Then the
equations above simplify to
∂A2
− = yz
∂z
∂A1
= zx
∂z
∂A2 ∂A1
− = xy
∂x ∂y
This system is much easier because, now that we have chosen A3 = 0, the first
equation contains only a single unknown, namely A2 and we can find all A2 ’s
that obey the first equation simply by integrating with respect to z:
yz 2
A2 = − + N (x, y)
2
∂
Note that, because ∂z treats x and y as constants, the constant of integration
N is allowed to depend on x and y.
9 There is nothing special about the subscript 3 here. By precisely the same argument,
we could come up with another vector potential whose second component is zero, and with
a third vector potential whose first component is zero.
Similarly, the second equation contains only a single unknown, A1 , and is

easily solved by integrating with respect to z. The second equation is satisfied
if and only if
xz 2
A1 = + M (x, y)
2
for some function M .
Finally, the third equation is also satisfied if and only if M (x, y) and N (x, y)
obey
∂ yz 2 ∂ xz 2
− + N (x, y) − + M (x, y) = xy
∂x 2 ∂y 2
which simplifies to
∂N ∂M
(x, y) − (x, y) = xy
∂x ∂y
This is one linear equation in two unknowns, M and N . Typically, we can
easily solve one linear equation in one unknown. So we are free to eliminate
one of the unknowns by setting, for example, M = 0, and then choose any N
that obeys
∂N
(x, y) = xy
∂x
x2 y
Integrating with respect to x gives, as one possible choice, N (x, y) = 2 . So
we have found a vector potential. Namely
xz 2 yz 2 x2 y
A= ı̂ı + − + ̂
2 2 2
One can, and indeed should, quickly check that ∇ × A = B.
Let’s do another.
Example 4.1.15 Let
B = (2x)ı̂ı + (2z − 2x) ̂ + (2x − 2z) k̂
This vector field obeys

∂ ∂ ∂
∇·B= (2x) + (2z − 2x) + (2x − 2z) = 0
∂x ∂y ∂z
and so passes the screening test of Theorem 4.1.12. We’ll now find a vector
potential A = A1 ı̂ı + A2 ̂ + A3 k̂ for B. As in the last example, we’ll simplify
the computations by further requiring9 that A3 = 0.
The requirements that ∇ × A = B and A3 = 0 come down to
∂A2
− = 2x
∂z
∂A1
= 2z − 2x
∂z
∂A2 ∂A1
− = 2x − 2z
∂x ∂y
∂
Because ∂z treats x and y as constants, the first equation is satisfied if and
only if there is a function N (x, y)
A2 = −2xz + N (x, y)
and second equation is satisfied if and only if there is a function M (x, y)
A1 = z 2 − 2xz + M (x, y)
Finally, the third equation is also satisfied if and only if M (x, y) and N (x, y)
obey
∂ ∂ 2
− 2xz + N (x, y) − z − 2xz + M (x, y)) = 2x − 2z
∂x ∂y
∂N ∂M
⇐⇒ −2z + (x, y) − (x, y) = 2x − 2z
∂x ∂y
∂N ∂M
⇐⇒ (x, y) − (x, y) = 2x
∂x ∂y
All of the z’s in this equation have cancelled out10 , and we can choose, for
example, M (x, y) = 0 and N (x, y) = x2 . So we have found a vector potential.
Namely
A = (z 2 − 2xz)ı̂ı + (x2 − 2xz)̂
Again it is a good idea to check that ∇ × A = B.
We can use exactly the strategy of the last examples to prove
Theorem 4.1.16 Let B be a vector field that is defined and has all of its first
order partial derivatives continuous on all of R3 . Then there exists a vector
potential for B if and only if it passes the screening test ∇ · B = 0.
Proof. We already know that the existence of a vector potential implies that
∇ · B = 0. So we just have to assume that ∇ · B = 0 and prove that this implies
the existence of a vector field A that obeys ∇ × A = B. Hence we need to
solve
∂A3 ∂A2
− = B1 (x, y, z)
∂y ∂z
∂A3 ∂A1
− + = B2 (x, y, z)
∂x ∂z
∂A2 ∂A1
− = B3 (x, y, z)
∂x ∂y
We’ll explicitly find such an A using exactly the strategy of Example 4.1.14.
In particular, we’ll look for an A that also has A3 = 0. Then the equations
simplify to
∂A2
− = B1 (x, y, z)
∂z
∂A1
= B2 (x, y, z)
∂z
∂A2 ∂A1
− = B3 (x, y, z)
∂x ∂y
The first equation is satisfied if and only if
Z z
A2 (x, y, z) = − B1 (x, y, z̃) dz̃ + N (x, y)
0
for some function N (x, y). And the second equation is satisfied if and only if
Z z
A1 (x, y, z) = B2 (x, y, z̃) dz̃ + M (x, y)
0
10 Of course, we could equally well pick A1 = 0 or A2 = 0.
11 Ifthe z’s had not cancelled out, no N (x, y) and M (x, y), which after all are independent
of z, could satisfy the equation. That would have been a sure sign of a user error.
So all three equations are satisfied if and only only if we can find M (x, y) and
N (x, y) that obey
A2 (x,y,z)
z Z }| {
z
∂
− B1 (x, y, z̃) dz̃ + N (x, y)
∂x 0
A1 (x,y,z)
zZ }| {
∂ z
− B2 (x, y, z̃) dz̃ + M (x, y) = B3 (x, y, z)
∂y 0
which is the case if and only if

Z z
∂N ∂M ∂B
1 ∂B2
(x, y) − (x, y) = B3 (x, y, z) + (x, y, z̃) + (x, y, z̃) dz̃
∂x ∂y 0 ∂x ∂y
Oof! At first sight, it looks like we have a very big problem here. No matter
what N and M we pick the left hand side will depend on x and y only — not
on z. But it appears like the right hand side depends on z too. Fortunately
the screening test (which we have not used to this point in the proof) rides to
the rescue and ensures that the right hand actually does does not depend on
z. By the screening test,
∂B1 ∂B2 ∂B3
∇·B= + + =0
∂x ∂y ∂z
and we have
∂B1 ∂B2 ∂B3
+ =−
∂x ∂y ∂z
so that the right hand side is
Z z iz̃=z
∂B3 h
B3 (x, y, z) + − (x, y, z̃) dz̃ = B3 (x, y, z) + − B3 (x, y, z̃)
0 ∂z z̃=0
= B3 (x, y, 0)
by the fundamental theorem of calculus. So we just have to choose M and N

to obey
∂N ∂M
(x, y) − (x, y) = B3 (x, y, 0)
∂x ∂y
Rx
For example, M = 0, N (x, y) = 0 B3 (x̃, y, 0) dx̃ work. So not only have we
proven that a vector potential exists, but we have found a formula for it.
Warning 4.1.17 Note that in Theorem 4.1.16 we are assuming that B passes
the screening test on all of R3 . If that is not the case, for example because
the vector field is not defined on all of R3 , then B can fail to have a vector
potential. An example (the point source) is provided in Example 4.4.8.
4.1.3 Interpretation of the Gradient

In this section we’ll develop an interpretation of the gradient ∇ f (r0 ). This
should just be a review of material that you have seen before.
Suppose that you are moving
through space and that your position at time
t is r(t) = x(t), y(t), z(t) . As you move along, you measure, for example,
the temperature. If the temperature at position (x, y, z) is f (x, y, z), then the
temperature that you measure at time t is f x(t), y(t), z(t) . So the rate of
change of temperature that you feel is

d
f x(t), y(t), z(t)
dt
∂f dx ∂f dy
= x(t), y(t), z(t) (t) + x(t), y(t), z(t) (t)
∂x dt ∂y dt
∂f dz
+ x(t), y(t), z(t) (t) (by the chain rule)
∂z dt
= ∇ f r(t) · r0 (t)
= ∇ f r(t) r0 (t) cos θ

where θ is the angle between the gradient vector ∇f r(t) and the velocity
vector r0 (t). This is the rate of change per unit time. We can get the rate
of
change per unit distance travelled by moving with speed one, so that r0 (t) = 1
and then
d
f r(t) = ∇ f r(t) cos θ
dt
If, at a given moment t = t0 , you are at r(t0 ) = r0 , then
d
f r(t) = ∇ f (r0 ) cos θ
dt t=t0
Recall that θ is the angle between our direction of motion and the gradient
vector ∇ f (r0 ). So to maximize the rate of change of temperature that we feel,
as we pass through r0 , we should choose our direction of motion to be the
direction of the the gradient vector ∇ f (r0 ). In conclusion
Equation 4.1.18
direction of maximum rate of change

∇ f (r0 ) has direction
of f at r0
magnitude of maximum rate of change

has magnitude
(per unit distance) of f at r0
4.1.4 Interpretation of the Divergence

In this section we’ll develop an interpretation of the divergence ∇ · v(r0 ) of the
vector field v(r) at the point r0 . We shall do so in two steps.
• First we’ll express ∇ · v(r0 ) in terms of flux integrals.
• Then we’ll use the interpretation of flux integrals given in Lemma 3.4.1
to get an interpretation of ∇ · v(r0 ).
Think of v(x, y, z) as the velocity of a fluid at (x, y, z) and fix any point r0 =
(x0 , y0 , z0 ). Let, for any ε > 0, Sε be the sphere
• centered at r0
• of radius ε.
• Denote by n̂(x, y, z) the outward normal to Sε at (x, y, z).
We shall prove, in Lemma 4.1.20, below, that we can write ∇ · v(r0 ) as the
limit
ZZ
1
∇ · v(x0 , y0 , z0 ) = lim 4 3 v(x, y, z) · n̂(x, y, z) dS
ε→0 πε Sε
3
Once we have that lemma we can use that

4 3
• 3 πεis the volume of the interior of the sphere Sε and
• by Lemma 3.4.1, Sε v(x, y, z) · n̂(x, y, z) dS is the rate11 at which fluid
RR
is exiting Sε
to conclude that
Equation 4.1.19
 rate at which fluid is exiting an


∇ · v(r0 ) = infinitesimal sphere centred

at r0 , per unit time, per unit volume

= strength of the source at r0

Here is the critical computation.
Lemma 4.1.20
ZZ
1
∇ · v(x0 , y0 , z0 ) = lim v(x, y, z) · n̂(x, y, z) dS
ε→0 4 πε3 Sε
3
Proof. (Optional). Here is one proofa of Lemma 4.1.20.
By translating our coordinate system, it suffices to consider r0 = (x0 , y0 , z0 ) =
(0, 0, 0). Then
1
Sε = (x, y, z) |(x, y, z)| = ε n̂(x, y, z) = (x, y, z)
ε
We expand v(x, y, z) in a Taylor expansion in powers of x, y, and z, to first
order, with second order error term.
v(x, y, z) = A + B x + C y + D z + R(x, y, z)
where
∂v ∂v ∂v
A = v(0, 0, 0) B= (0, 0, 0) C= (0, 0, 0) D= (0, 0, 0)
∂x ∂y ∂z
and the error term R(x, y, z) is bounded by a constant timesb x2 + y 2 + z 2 . In

particular there is a constant K so that, on Sε ,
|R(x, y, z)| ≤ Kε2
So
ZZ
v(x, y, z) · n̂(x, y, z) dS
Sε
ZZ
1
= A + B x + C y + D z + R(x, y, z) · (x, y, z) dS
ε Sε
Multiply out the dot product so that the integrand becomes
A · ı̂ı x + A · ̂ y + A · k̂ z
2
+ B · ı̂ı x + B · ̂ xy + B · k̂ xz
+ C · ı̂ı xy + C · ̂ y 2 + C · k̂ yz
+ D · ı̂ı xz + D · ̂ yz + D · k̂ z 2
+ R(x, y, z) · (x, y, z)
11 Lemma 3.4.1 is being applied with the density ρ set equal to one, so, more precisely, the
rate is the number of units of volume of fluid exiting Sε per unit time
That’s a lot of terms. But most of them integrate to zero, simply because the
integral of an odd function over an even domain is zero. Because Sε is invariant
under x → −x and under y → −y and under z → −z we have
ZZ ZZ ZZ ZZ ZZ ZZ
x dS = y dS = z dS = xy dS = xz dS = yz dS = 0
Sε Sε Sε Sε Sε Sε
which is a relief. We are now left with

ZZ ZZ
1
B · ı̂ı x2 + C · ̂ y 2 + D · k̂ z 2 dS

v(x, y, z) · n̂(x, y, z) dS =
Sε ε Sε
ZZ
1
+ R(x, y, z) · (x, y, z) dS
ε Sε
As well Sε is invariantc under the interchange of x and y and also under the
interchange of x and z. Consequently
ZZ ZZ ZZ ZZ
2 2 2 1 2
x + y 2 + z 2 dS

x dS = y dS = z dS =
Sε 3
ZSZε Sε Sε
1
= ε2 dS since x2 + y 2 + z 2 = ε2 on Sε
3 Sε
4
= πε4
3
since the surface area of the sphere Sε is 4πε2 . So far, we have
ZZ
4
v(x, y, z) · n̂(x, y, z) dS = πε3 B · ı̂ı + C · ̂ + D · k̂

Sε 3
ZZ
1
+ R(x, y, z) · (x, y, z) dS
ε Sε
ZZ
4 1
= πε3∇ · v(0) + R(x, y, z) · (x, y, z) dS
3 ε Sε
(review the definitions of B, C, D)
which implies
ZZ
1
lim v(x, y, z) · n̂(x, y, z) dS
ε→0 4 πε3 Sε
3
ZZ
3
= ∇ · v(0) + lim R(x, y, z) · (x, y, z) dS
ε→0 4πε4 Sε
Finally, it suffices to recall that |R(x, y, z)| ≤ Kε2 and, on Sε , |(x, y, z)| = ε,
so that
ZZ ZZ
3 ≤ 3

R(x, y, z) · (x, y, z) dS |R(x, y, z)| |(x, y, z)| dS
4πε4 Sε 4πε4
Sε
ZZ
3 3
≤ Kε3 dS = Kε3 4πε2 )
4πε4 Sε 4πε 4
= 3Kε
converges to zero as ε → 0. So we are left with the desired result.
a There is another, easier to understand, proof of this result given in §4.4.1. We cannot
give that proof here because it uses the divergence theorem, which we will get to later in the
chapter.
b Terms like xy, xz and yz are not needed because, for example, |xy| ≤ 1 (x2 + y 2 ). This
2 2
inequality is equivalent to |x| − |y| ≥ 0.
c Spheres have lots of symmetry!
Example 4.1.21 Here is a sketch of the vector field v(x, y, z) = xı̂ı + y ̂ + z k̂
and a sphere centered on the origin, like Sε .
This velocity field has fluid being created and pushed out through the
sphere. We have
∇ · v(0) = 3
consistent with our interpretation 4.1.19.

Example 4.1.22 Here is a sketch of the vector field v(x, y, z) = −y ı̂ı + x ̂ and
a sphere centered on the origin, like Sε .
This velocity field just has fluid going around in circles. No fluid actually
crosses the sphere. The divergence
∇ · v(0) = 0
consistent with our interpretation 4.1.19.

Example 4.1.23 Here is a sketch of the vector field v(x, y, z) = ı̂ı and a sphere
centered on the origin, like Sε .
This velocity field just has fluid moving uniformly to the right. Fluid enters
the sphere from the left and leaves through the right at precisely the same rate,
so that the net rate at fluid crosses the sphere is zero. The divergence
∇ · v(0) = 0
again consistent with our interpretation 4.1.19.
4.1.5 Interpretation of the Curl

We’ll now develop the interpretation of the curl, or more precisely, of ∇ ×
v(r0 ) · n̂ for any unit vector n̂. As we did in developing the interpretation of
divergence, we’ll
• first express ∇ × v(r0 ) · n̂ as a limit of integrals, and

• then we’ll interpret the integrals.
To specify the integrals involved, let Cε be the circle which
• is centered at r0
• has radius ε
• lies in the plane through r0 perpendicular to n̂
• is oriented in the standard way with respect to n̂. Imagine standing on
the circle with your feet on the plane through r0 perpendicular to n̂,
with the vector from your feet to your head in the same direction as n̂
and with your left arm pointing towards r0 . Then you are facing in the
positive direction for Cε .
n̂
ε
Cε
We shall show in Lemma 4.1.25, below, that
I
1
∇ × v(r0 ) · n̂ = lim 2 v(r) · dr
ε→0 πε Cε
Now let’s work on interpreting

H the right hand side, and in particular on inter-
preting the integral Cε v(r) · dr, which is called the circulation of v around
Cε . Place a tiny paddlewheel in the fluid with its axle running along n̂ and its
paddles along Cε , as in the figure below, except that
n̂
t̂
ε
Cε
the paddlewheel is really expensive and has a lot more than just four pad-
dles. Pretend12 that you are one of the paddles.
• If the paddlewheel is rotating at Ω radians per unit time, then in one
unit of time you sweep out an arc of a circle of radius ε that subtends an
angle Ω. That arc has length Ωε. So you are moving at speed Ωε.
• If you are at r, the component of the fluid velocity in your direction of
dr dr
motion, i.e. tangential to Cε , is v(r) · ds , because t̂ = ds , with s denoting
arc length along the circle, is a unit vector tangential to Cε .
• All paddles have to move at the same speed. So the speed of the paddles,
dr
Ωε, should be the average value of v(r) · ds around the circle.
Thus the rate of rotation, Ω, of the paddlewheel should be determined by
dr
H H
Cε
v(r) · ds ds v(r) · dr
Ωε = H = Cε
Cε
ds 2πε
12 Method acting might help you here.
Consequently, ∇ × v(r0 ) · n̂ is the limit as ε (the radius of the paddlewheel)

tends to zero of
I
1
v(r) · dr = 2Ω
πε2 Cε
That’s our interpretation.
Equation 4.1.24 If a fluid has velocity field v and you place an infinitesimal
paddlewheel at r0 with its axle in direction n, then it rotates at 21 ∇ × v(r0 ) · n̂
radians per unit time. In particular, to maximize the rate of rotation, orient
the paddlewheel so that n̂ k ∇ × v(r0 ).
There will be some examples at the end of this section. First, we show
Lemma 4.1.25
I
1
∇ × v(r0 ) · n̂ = lim v(r) · dr
ε→0 πε2 C
ε
Proof. (Optional). Here is one proofa of Lemma 4.1.25.
Just as we did in the proof of Lemma 4.1.20, we can always translate our
coordinate system so that r0 = (x0 , y0 , z0 ) = (0, 0, 0). We can also rotate our
coordinate system so that n̂ = k̂. Because r0 = (0, 0, 0) and n̂ = k̂, so that Cε
lies in the xy-plane, we can parametrize Cε by
r(t) = ε cos tı̂ı + ε sin t ̂
Again as we did in the proof of Lemma 4.1.20, expand v(x, y, z) in a Taylor

expansion in powers of x, y, and z, to first order, with second order error term.
v(x, y, z) = A + B x + C y + D z + R(x, y, z)
where
∂v ∂v ∂v
A = v(0, 0, 0) B= (0, 0, 0) C= (0, 0, 0) D= (0, 0, 0)
∂x ∂y ∂z
and the error term R(x, y, z) is bounded by a constant times x2 + y 2 + z 2 . In

particular there is a constant K so that, on Cε ,
|R(x, y, z)| ≤ Kε2
So
I
v(r) · dr
Cε
Z 2π
= A + B ε cos t + C ε sin t + R(r(t)) · − ε sin tı̂ı + ε cos t ̂ dt
0
Again, multiply out the dot product so that the integrand becomes
− εA · ı̂ı sin t + εA · ̂ cos t

2
− ε B · ı̂ı sin t cos t + ε2 B · ̂ cos2 t
− ε2 C · ı̂ı sin2 t + ε2 C · ̂ sin t cos t

+ R(r(t)) · − ε sin tı̂ı + ε cos t ̂
Again most of these terms integrate to zero, because

Z 2π Z 2π
sin t dt = cos t dt =0
0 0
Z 2π Z 2π
1
sin t cos t dt = sin(2t) dt = 0
0 2 0
and the sin2 t and cos2 t terms are easily integrated using (see Example 2.4.4)
Z 2π Z 2π Z 2π
1
sin2 t dt = cos2 t dt = sin2 t + cos2 t dt = π

0 0 2 0
So we are left with

I Z 2π
v(r) · dr = πε2 B · ̂ − πε2 C · ı̂ı +

R(r(t)) · − ε sin tı̂ı + ε cos t ̂ dt
Cε 0
which implies that

I
1 ∂v2 ∂v1
lim 2 v(r) · dr = (0, 0, 0) − (0, 0, 0)
ε→0 πε Cε ∂x ∂y
Z 2π
1
+ lim 2 R(r(t)) · − ε sin tı̂ı + ε cos t ̂ dt
ε→0 πε 0

= ∇ × v(0, 0, 0) · k̂
Z 2π
1
+ lim 2 R(r(t)) · − ε sin tı̂ı + ε cos t ̂ dt
ε→0 πε 0
Finally, it suffices to recall that |R(x, y, z)| ≤ Kε2 , so that
1 2π
Z Z 2π
≤ 1

R(r(t)) · − ε sin ı
tı̂ + ε cos t 
̂ dt |R(r(t))| dt
πε2 0 πε
0
Z 2π
1
≤ Kε2 dt
πε 0
1
= Kε2 (2π)
πε
= 2Kε
converges to zero as ε → 0.
a There is another, easier to understand, proof of this result given in §4.4.1. We cannot
give that proof here because it uses Stokes’ theorem, which we will get to later in the chapter.
Here are some examples. We will use the same vector fields as in Examples
4.1.21, 4.1.22 and 4.1.23. In all examples, we shall orient the paddlewheel so
that n̂ = k̂ and sketch the top view, so that the paddlewheel looks like
Example 4.1.26 Here is a sketch of the vector field v(x, y, z) = xı̂ı + y ̂ + z k̂
and a circle centered on the origin, like Cε .
This velocity field has fluid moving parallel to the paddles, so the paddle-
wheel should not rotate at all. The computation
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × v(0) = det  ∂x ∂y ∂z  = 0 =⇒ ∇ × v(0) · k̂ = 0
x y z
is consistent with our interpretation 4.1.24.

Example 4.1.27 Here is a sketch of the vector field v(x, y, z) = −y ı̂ı + x ̂ and
a circle centered on the origin, like Cε .
This velocity field has fluid going around in circles, counterclockwise. So

the paddlewheel should rotate counterclockwise too. That is, it should have
positive angular velocity. Our interpretation 4.1.24 predicts an angular velocity
of half
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × v(0) · k̂ = det  ∂x ∂y ∂z  · k̂ = 2k̂ · k̂ = 2
−y x 0
which is indeed positive13 .

Example 4.1.28 Here is a sketch of the vector field v(x, y, z) = ı̂ı and a circle
centered on the origin, like Cε .
The fluid pushing on the top paddle tries to make the paddlewheel rotate
clockwise. The fluid pushing on the bottom paddle tries to make the paddle-
wheel rotate counterclockwise, at the same rate. So the paddlewheel should
not rotate at all. Our interpretation 4.1.24 predicts an angular velocity of
 
ı̂ı ̂ k̂
1 1 ∂ ∂ ∂ 
· ∇ × v(0) · k̂ = det  ∂x ∂y ∂z  · k̂ = 0 · k̂ = 0
2 2
1 0 0
as expected.
18 Even for small values of 2.
4.1.6 Exercises
1. ∗. Let F = P ı̂ı + Q ̂ be the two dimensional vector field shown below.
a Assuming that the vector field in the picture is a force field, the
work done by the vector field on a particle moving from point
A to B along the given path is:
(A) Positive
(B) Negative
(C) Zero
(D) Not enough information to determine.
R
b Which statement is the most true about the line integral C2
F·
dr:
Z
(A) F · dr > 0
C2
Z
(B) F · dr = 0
C2
Z
(C) F · dr < 0
C2
y
N
C2 B
G
x
c ∇ · F at the point N (in the picture) is:
(A) Positive
(B) Negative
(C) Zero
d Qx − Py at the point Q is:

(A) Positive
(B) Negative
(C) Zero
e Assuming that F = P ı̂ı + Q ̂, which of the following statements
is correct about ∂P
∂x at the point D?
∂P
(A) ∂x = 0 at D.
∂P
(B) ∂x > 0 at D.
∂P
(C) ∂x < 0 at D.
(D) The sign of ∂P∂x at D can not be determined by the given
information.
2. Does ∇ × F have to be perpendicular to F?
3. Verify the vector identities
∇ · F + F · ∇f
a ∇ · (f F) = f∇
∇ × F) − F · (∇
b ∇ · (F × G) = G · (∇ ∇ × G)
c ∇ 2 (f g) = f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
4. Evaluate ∇ · F and ∇ × F for each of the following vector fields.
a F = xı̂ı + y ̂ + z k̂
b F = xy 2ı̂ı − yz 2̂ + zx2 k̂
c F = √xı̂ı+y̂
2

2
(the polar basis vector r̂ in 2d)
x +y
−yı̂ı+x̂
d F= √ 2 2
(the polar basis vector θ̂θ in 2d)
x +y
5. ∗.
a Compute and simplify ∇ · rr for r = (x, y, z) and r = |(x, y, z)|.

Express your answer in terms of r.

b Compute ∇ × yz ı̂ı + 2xz ̂ + exy k̂ .
6. ∗. In the following, we use the notation r = xı̂ı + y ̂ + z k̂, r = |r|,
and k is some number k = 0, 1, −1, 2, −2, . . ..
a Find the value k for which
r
∇ (rk ) = −3
r5
b Find the value k for which
∇ · (rk r) = 5r2
c Find the value k for which

2
∇ 2 (rk ) =
r4
7. ∗. Let r be the vector field r = xı̂ı + y ̂ + z k̂ and let r be the function
r = |r|. Let a be the constant vector a = a1 ı̂ı + a2 ̂ + a3 k̂. Compute
and simplify the following quantities. Answers must be expressed in
terms of a, r, and r. There should be no x’s, y’s, or z’s in your
answers.
a ∇·r
b ∇ (r2 )
c ∇ × (r × a)

d ∇ · ∇ (r)
8. ∗. Let
r = xı̂ı + y ̂ + z k̂, r = |r|
a Compute a where ∇ 1r = −ra r.

b Compute a where ∇ · r r = ar.

c Compute a where ∇ · ∇ (r3 ) = ar.
9. Find, if possible, a vector field A that has k̂ component A3 = 0 and
that is a vector potential for
a F = (1 + yz)ı̂ı + (2y + zx)̂ + (3z 2 + xy)k̂
b G = yzı̂ı + zx̂ + xy k̂
10. ∗. Let
−z x
F= ı̂ı + y ̂ + 2 k̂
x2 + z 2 x + z2
a Determine the domain of F.

b Determine the curl of F. Simplify if possible.
c Determine the divergence of F. Simplify if possible.
d Is F conservative? Give a reason for your answer.

11. ∗. A physicist studies a vector field F in her lab. She knows from
theoretical considerations that F must be of the form F = ∇ × G, for
some smooth vector field G. Experiments also show that F must be
of the form
F(x, y, z) = (xz + xy)ı̂ı + α(yz − xy)̂ + β(yz + xz)k̂
where α and β are constant.

a Determine α and β.
b Further experiments show that G = xyzı̂ı − xyz̂ + g(x, y, z)k̂.
Find the unknown function g(x, y, z).
12. A rigid body rotates at an angular velocity of Ω rad/sec about an
axis that passes through the origin and has direction â. When you
are standing at the head of â looking towards the origin, the rotation
is counterclockwise. Set Ω = Ωâ.

a Show that the velocity of the point r = (x, y, z) on the body is
Ω × r.
b Evaluate ∇ × (Ω
Ω × r) and ∇ · (Ω
Ω × r), treating Ω as a constant.
c Find the speed of the students in a classroom located at latitude

49◦ N due to the rotation of the Earth. Ignore the motion of
the Earth about the Sun, the Sun in the Galaxy and so on. The
radius of the Earth is 6378 km.
13. Suppose that the vector field F obeys ∇ · F = 0 in all of R3 . Let
r(t) = txı̂ı + ty ̂ + tz k̂, 0≤t≤1
be a parametrization of the line segment from the origin to (x, y, z).

Define Z 1
dr
G(x, y, z) = t F r(t) × (t) dt
0 dt
Show that ∇ × G = F throughout R3 .
4.2 The Divergence Theorem

The rest of this chapter concerns three theorems: the divergence theorem,
Green’s theorem and Stokes’ theorem. Superficially, they look quite different
from each other. But, in fact, they are all very closely related and all three are
generalizations of the fundamental theorem of calculus
Z b
df
(t) dt = f (b) − f (a)
a dt
The left hand side of the fundamental theorem of calculus is the integral of the
derivative of a function. The right hand side involves only values of the function
on the boundary of the domain of integration. The divergence theorem, Green’s
theorem and Stokes’ theorem also have this form, but the integrals are in more
than one dimension. So the derivatives are multidimensional, like the curl and
divergence, and the integrands can involve vector fields.
• For the divergence theorem, the integral on the left hand side is over a
(three dimensional) volume and the right hand side is an integral over
the boundary of the volume, which is a surface.
• For Green’s and Stokes’ theorems, the integral on the left hand side is
over a (two dimensional) surface and the right hand side is an integral
over the boundary of the surface, which is a curve.
The divergence theorem is going to relate a volume integral over a solid
V to a flux integral over the surface of V . First we need a couple of defini-
tions concerning the allowed surfaces. In many applications solids, for example
cubes, have corners and edges where the normal vector is not defined. On the
other hand, to be able to compute a flux integral over a surface, we certainly
need that the set of points where the normal vector is not well-defined is small
enough that the existence of the flux integral is not jeopardized. This is the
case for “piecewise smooth” surfaces, which we now define.
Definition 4.2.1
a A surface is smooth if it has a parametrization r(u, v) with continuous
∂r ∂r ∂r ∂r
partial derivatives ∂u and ∂v and with ∂u × ∂v nonzero.
b A surface is piecewise smooth if it consists of a finite number of smooth
pieces that meet along sharp curves and at sharp corners.
♦
Here are sketches of a smooth surface (a sausage) and a piecewise smooth
surface (an ice-cream cone), followed by the divergence theorem1 .
Theorem 4.2.2 Divergence Theorem. Let

• V be a bounded solid with a piecewise smooth surface2 ∂V
• F be a vector field that has continuous first partial derivatives at every
point of V .
Then
ZZ ZZZ
F · n̂ dS = ∇ · F dV
∂V V
where n̂ is the outward unit normal of ∂V .

Like the fundamental theorem of calculus, the divergence theorem expresses
the integral of a derivative of a function (in this case a vector-valued function)
over a region in terms of the values of the function on the boundary of the
region.
Warning 4.2.3 Note that in Theorem 4.2.2 we are assuming that the vector
field F has continuous first partial derivatives at every point of V . If that
is not the case, for example because F is not defined on all of V , then the
conclusion of the divergence theorem can fail. An example is F = |r|r 3 , V =

(x, y, z) x2 + y 2 + z 2 ≤ 1 . See Example 4.2.7.
Proof. We have to show that
ZZ ZZZ
∂F1 ∂F2 ∂F3
F1 ı̂ı + F2 ̂ + F3 k̂ · n̂ dS = + + dV
∂V V ∂x ∂y ∂z
Note that the left hand side is a sum of three terms — one involving F1 , one
involving F2 and one involving F3 — and the right hand side is a sum of three
terms — one involving F1 , one involving F2 and one involving F3 . We’ll just
show that the F3 terms on the left hand side and right hand side are equal,
1 It is also known as Gauss’s theorem. Johann Carl Friedrich Gauss (1777–1855) was a
German mathematician. Throughout the 1990’s Gauss’s portrait appeared on the German
ten-mark banknote. In addition to Gauss’s theorem, the Gaussian distribution (the bell
curve), degaussing and the CGS unit for the magnetic field, and the crater Gauss on the
Moon are named in his honour.
2 We are going to consistently use the notation ∂(thing) to denote the boundary of (thing).
i.e. that
ZZ ZZZ
∂F3
F3 k̂ · n̂ dS = dV
∂V V ∂z
Showing that the F1 terms match and the F2 terms match is done in the same
waya .
Special Geometry
We’ll first assume that the solid has the special form

V = (x, y, z) B(x, y) ≤ z ≤ T (x, y), (x, y) ∈ Rxy
where Rx,y is some subset of the xy-plane. We can further assume that, for
each (x, y) ∈ Rxy , we have B(x, y) ≤ T (x, y). After we’re finished with this
special case, we’ll RR
handle the general case.
Let’s work on ∂V F3 k̂ · n̂ dS first. As in the figure below,
z
z “ T px, yq
S
B
z “ Bpx, yq
y
Rxy
x BRxy
the surface ∂V consists of three pieces — the top, the bottom and the side.
We’ll consider each in turn.

• The top is T = (x, y, z) z = T (x, y), (x, y) ∈ Rxy . By 3.3.2, on T

n̂ dS = + − Tx (x, y)ı̂ı − Ty (x, y) ̂ + k̂ dxdy
As n̂ is to be the outward normal, it must point upwards on T . That’s

why we have chosen, and emphasised, the “+” sign. So k̂ · n̂ dS = dxdy
and ZZ ZZ
F3 k̂ · n̂ dS = F3 (x, y, T (x, y)) dxdy
T Rxy

• The bottom is B = (x, y, z) z = B(x, y), (x, y) ∈ Rxy . By 3.3.2,
on B
n̂ dS = − − Bx (x, y)ı̂ı − By (x, y) ̂ + k̂ dxdy
As n̂ is to be the outward normal, it must point downwards on B. That’s
why we have chosen the “−” sign. So k̂ · n̂ dS = −dxdy and
ZZ ZZ
F3 k̂ · n̂ dS = − F3 (x, y, B(x, y)) dxdy
B Rxy

• The side is S = (x, y, z) (x, y) ∈ ∂Rxy , B(x, y) ≤ z ≤ T (x, y) . It
runs vertically. Hence on S the normal vector to ∂V is parallel to the
xy-plane so that k̂ · n̂ = 0 and
ZZ
F3 k̂ · n̂ dS = 0
S
So all together
ZZ ZZ ZZ ZZ
F3 k̂ · n̂ dS = F3 k̂ · n̂ dS + F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
∂V T B S
ZZ

= F3 (x, y, T (x, y)) − F3 (x, y, B(x, y)) dxdy + 0 (∂V )
Rxy
Now let us examine

ZZZ ZZ Z T (x,y)
∂F3 ∂F3
dV = dxdy dz (x, y, z)
V ∂z Rxy B(x,y) ∂z
ZZ

= F3 (x, y, T (x, y)) − F3 (x, y, B(x, y)) dxdy (V )
Rxy
by the fundamental theorem of calculus. That’s exactly what we had to show.

The integrals (∂V ) and (V ) are equal.
General Geometry
Now we’ll drop the assumption on V that we imposed in the “Special Ge-
ometry” section above. The key idea that makes the proof work is that we
can cut up anyb V into pieces, each of which does obey the special assumption
that we just considered. Consider, for example, the sausage shaped solid in
the figure on the left below.
V1
S1
Sc
S2
V2
Call the sausage V . Cut it into two halves by running a cleaver horizontally
through its centre. This splits the solid V into two halves, V1 and V2 as in the
figure on the right above. It also splits the boundary ∂V of V into two halves
S1 and S2 , also as in the figure on the right above. Note that
• the boundary, ∂V1 , of V1 is the union of S1 and the shaded disk Sc (the
cut introduced by the cleaver). On the cut Sc , the outward pointing
normal to V1 is −k̂.
• The boundary, ∂V2 , of V2 is the union of S2 and the shaded disk Sc . On
the cut Sc , the outward pointing normal to V2 is +k̂.
Now both V1 and V2 do satisfy the assumption of the “Special Geometry”
section above. So
ZZZ ZZZ ZZZ
∂F3 ∂F3 ∂F3
dV = dV + dV
V ∂z V1 ∂z V2 ∂z
ZZ ZZ
= F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
∂V1 ∂V2
ZZ ZZ ZZ
= F3 k̂ · n̂ dS + F3 k̂ · n̂ dS + F3 k̂ · n̂ dS
S1 Sc ↓ S2
ZZ
+ F3 k̂ · n̂ dS
Sc ↑
Here Sc ↓ is the surface Sc with normal vector −k̂ and Sc ↑ is the surface Sc
with normal vector +k̂. So the second and fourth integrals are identical except
that n̂ = −k̂ in the second integral and n̂ = +k̂ in the fourth integral. So they
cancel exactly and
ZZZ ZZ ZZ ZZ
∂F3
dV = F3 k̂ · n̂ dS + F3 k̂ · n̂ dS = F3 k̂ · n̂ dS
V ∂z S1 S2 ∂V
as desired.
a Mutatis mutandis.
b We are assuming that V is “reasonable”.
RR
Example 4.2.4 Evaluate the flux integral S F · n̂ dS where n̂ is the outward
normal to S, which is the surface of the hemispherical region
n̂
V
V =

(x, y, z) x2 +y 2 +z 2 ≤ a2 , z ≥ 0
S
n̂
and
F = xz 2 ı̂ı + (x2 y − z 3 ) ̂ + 2xy + y 2 z + ecos y k̂

Solution. The ecos y in F suggests that a direct evaluation of the integral

is difficult. So we’ll use a little trickery to to evaluate it. Not surprisingly,
considering that we have just proven the divergence theorem, the trick is to
apply the divergence theorem3 . Since
∂F1 ∂F2 ∂F3
∇·F= + +
∂x ∂y ∂z
∂ ∂ 2 ∂
xz 2 + x y − z3 + 2xy + y 2 z + ecos y

=
∂x ∂y ∂z
2 2 2
=z +x +y
The divergence theorem tell us that
ZZ ZZZ
x2 + y 2 + z 2 dV

F · n̂ dS =
S V
Spherical coordinates are perfect for this integral. (See Appendix A.6.3, if you
need to refresh your memory.)
ZZZ Z 2π Z π2 Z a
2 2 2
dρ ρ2 sin ϕ ρ2

x + y + z dV = dθ dϕ
V 0 0 0
Z 2π Z π Z a
2
4
= dθ sin ϕ dϕ ρ dρ
0 0 0
h i π2 ρ5 a
= 2π − cos ϕ
0 5 0
2πa5
=
5

RR
Example 4.2.5 Evaluate the flux integral S F · n̂ dS where n̂ is the outward
normal to S, which is the part of the surface z 2 = x2 + y 2 with 1 ≤ z ≤ 2, and
where
F = 3xı̂ı + (5y + ecos x ) ̂ + z k̂
Solution. Again the ecos x in F suggests that a direct evaluation is difficult4

and again we’ll apply the divergence theorem. But this time S is not the
boundary of a solid V . It is the portion of the cone outlined in red in the figure
on the left below and does not have a top or bottom “cap”.
5 It’s almost as though someone rigged the example with this in mind.
D2 n̂ “ k̂
S S V
n̂ n̂
D1
n̂ “ ´k̂
Fortunately, there is a solid V whose boundary, while not being equal to S,

at least contains S. It is (unsurprisingly)
V = (x, y, z) x2 + y 2 ≤ z 2 , 1 ≤ z ≤ 2

and is sketched in the figure on the right above. The boundary, ∂V , is the
union of S and the two disks
D1 = (x, y, z) x2 + y 2 ≤ z 2 , z = 1

D2 = (x, y, z) x2 + y 2 ≤ z 2 , z = 2

So the divergence theorem gives

ZZZ ZZ
∇ · F dV = F · n̂ dS
V
Z Z∂V ZZ ZZ
= F · n̂ dS + F · n̂ dS + F · n̂ dS
S D1 D2
which implies
ZZ ZZZ ZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS − F · n̂ dS
S V D1 D2
The point of this exercise is that the left hand side, which is not easy to evaluate
directly, is the integral we want, while the three integrals on the right hand side
are all easy to evaluate. We do so now. The outward normal to (the horizontal
disk) D2 is +k̂. So
ZZ ZZ ZZ
F · n̂ dS = F · k̂ dS = z dS
D2 D2 D2
As z = 2 on D2 , and D2 is a disk of radius 2,

ZZ
F · n̂ dS = 2Area(D2 ) = 2π22 = 8π
D1
Similarly, the outward normal to (the horizontal disk) D1 is −k̂. So

ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dS = − z dS
D1 D1 D1
As z = 1 on D1 , and D1 is a disk of radius 1,

ZZ
F · n̂ dS = Area(D1 ) = −π12 = −π
D1
Finally, as ∇ · F = 3 + 5 + 1 = 9
ZZZ
∇ · F dV = 9 Vol(V )
V
The volume of V can be easily computed using the first year technique5 of
slicing V into thin horizontal pancakes like that sketched in the figure below.
z“2
z“1
The pancake at height z has
• thickness dz,
• a circular cross-section of radius z (remember that the outer boundary

of V has equation x2 + y 2 = z 2 ), and hence has
• cross-sectional area πz 2 and
• volume πz 2 dz.
So
2 2
πz 3
ZZZ Z
2 7
∇ · F dV = 9 Vol(V ) = 9 πz dz = 9 =9×π = 21π
V 1 3 1 3
and, all together

ZZ ZZZ ZZ ZZ
S V D1 D2
= 21π − (−π) − 8π = 14π

RR
Example 4.2.6 Evaluate the flux integral S
F · n̂ dS where n̂ is the upward
2 2 2
normal to S, which is the part of z = x + y with 0 ≤ z ≤ 1, and
2
F = x + ey

ı̂ı + (y + cos z) ̂ + k̂
Solution. This integral can be evaluated in much the same way as we eval-
uated the integral of Example 4.2.5. We first define a solid V whose boundary
∂V contains S. A good, and hopefully obvious, choice is
2
(x, y, z) x2 + y 2 ≤ z, 0 ≤ z ≤ 1

V =
The boundary of V is the union of S, with outward pointing normal −n (recall

that the problem specifies that the symbol n̂ refers to the upward pointing
normal) and the disk
2
x2 + y 2

D= (x, y, z) z = 1, ≤1
with outward pointing normal k̂.

6 In fact, it is possible to evaluate this integral directly, if one recognizes that the ugly
part of the integrand is odd under y → −y and integrates to exactly zero.

7 You can review in §1.6 of the CLP-2 text.
k̂
D
z“1
n̂ S

ZZZ ZZ ZZ
∇ · F dV = − F · n̂ dS + F · k̂ dS
V S D
which implies
ZZ ZZZ ZZ
F · n̂ dS = − ∇ · F dV + F · k̂ dS
S
Z Z ZV ZZ D
=− 2 dV + dS
V D
D is a circular disk of radius 1, and so has area π. To evaluate the volume

integral we slice V into horizontal pancakes with the pancake at height z having
1
a circular cross-section of radius z 4 . (Recall that the boundary of V has
2
x2 + y 2 = z.) So
ZZ Z 1 √ 2 π
F · n̂ dS = −2 π z dz + π = −2π × + π = −
S 0 3 3
Again, you can see that the actual integration is quite easy. All of the work
(or at least all of the thinking) happens in the setup.
Example 4.2.7 In Warning 4.2.3 we emphasised that the conclusion of the
divergence Theorem 4.2.2 can fail if the vector field F is not defined at even a
single point of V . Here is an example. Set
r
F= where r = xı̂ı + y ̂ + z k̂
|r|3

and V = (x, y, z) x2 + y 2 + z 2 ≤ 1 . Then, if (x, y, z) 6= 0,
∂ x ∂ y
∇ · F(x, y, z) = +
∂x x2 + y 2 + z 2 3/2 ∂y x2 + y 2 + z 2 3/2
∂ z
+
∂z x2 + y 2 + z 2 3/2
x + y 2 + z 2 − x 23 (2x) x + y 2 + z 2 − y 23 (2y)
2 2
= 5/2 + 5/2
x2 + y 2 + z 2 x2 + y 2 + z 2
x + y 2 + z 2 − z 23 (2z)
2
+ 5/2
x2 + y 2 + z 2
=0

On the other hand, the boundary of V is the unit sphere ∂V = (x, y, z) x2 +
r

y 2 + z 2 = 1 . The outward unit normal to ∂V is n̂ = |r| so that
Z Z Z Z
r r 1
F · n̂ dS = · dS = dS = dS
∂V |r|=1 |r|3 |r| |r|=1 |r|2 |r|=1
= 4π 6= 0
4.2.1 Optional — An Application of the Divergence The-

orem — the Heat Equation
4.2.1.1 Derivation of the Heat Equation
Let T (x, y, z, t) be the temperature at time t at the point (x, y, z) in some
object B. The heat equation6 is the partial differential equation that describes
the flow of heat energy and consequently the behaviour of T . We now use the
divergence theorem to derive the heat equation from two physical “laws”, that
we assume are valid:
• The amount of heat energy required to raise the temperature of an object
by ∆T degrees is CM ∆T where, M is the mass of the object and C is
a positive physical constant determined by the material contained in the
object. It is called the specific heat, or specific heat capacity7 , of the
object.
• Think of heat energy as a moving fluid. We will rig its velocity field so
that heat flows in the direction opposite to the temperature gradient.
Precisely, we choose its velocity field to be −κ∇ ∇T (x, y, z, t). Here κ is
another positive physical constant called the thermal conductivity of the
object. So the rate at which heat is conducted across an element of
surface area dS at (x, y, z) in the direction of its unit normal n̂ is given
by −κn̂ · ∇ T (x, y, z, t) dS at time t. (See Lemma 3.4.1.) For example, in
the figure
n̂ cold
dS
∇T hot
the temperature gradient, which points in the direction of increasing tem-
perature, is opposite n̂. Consequently the flow rate −κn̂·∇∇T (x, y, z, t) dS
is positive, indicating flow in the direction of n̂. This is just what you
would expect — heat flows from hot regions to cold regions. Also the rate
of flow increases as the magnitude of the temperature gradient increases.
This also makes sense (and is reminiscent of Newton’s law of cooling).
6 The heat equation was formulated by the French mathematician and physicist Jean-
Baptiste Joseph Fourier in 1807. He lived from 1768 to 1830, a period which included both
the French revolution and the reign of Napoleon. Indeed Fourier served on his local Revolu-
tionary Committee, was imprisoned briefly during the Terror, and was Napoleon Bonaparte’s
scientific advisor on his Egyptian expedition of 1798. Fourier series and the Fourier transform
are named after him. Fourier is also credited with discovering the greenhouse effect.
7 Heat is now understood to arise from the internal energy of the object. In an earlier
theory, heat was viewed as measuring an invisible fluid, called the caloric. The amount of
caloric that an object could hold was called its “heat capacity” by the Scottish physician
and chemist Joseph Black (1728–1799).
Let V ⊂ B be any three dimensional region in the object and denote by ∂V

the surface of V and by n̂ the outward normal to ∂V . The amount of heat
that enters V across an infinitesimal piece dS of ∂V in an infinitesimal time
interval dt is − − κn̂ · ∇ T (x, y, z, t) dS dt. The amount of heat that enters V
across all of ∂V in the time interval dt is given by the integral
n̂
dS
ZZ BV
κn̂ · ∇T (x, y, z, t) dS dt
∂V V
In this same time interval, the temperature at a point (x, y, z) in V changes

by ∂T∂t (x, y, z, t) dt. If the density of the object at (x, y, z) is ρ(x, y, z), the
amount of heat energy required to increase the temperature of an infinitesimal
volume dV of the object centred at (x, y, z) by ∂T ∂T
∂t (x, y, z, t) dt is C(ρdV ) ∂t (x, y, z, t) dt.
The amount of heat energy required to increase the temperature by ∂T ∂t (x, y, z, t) dt
at all points (x, y, z) in V is then
ZZZ
∂T
Cρ (x, y, z, t) dV dt
V ∂t
Assuming that the object is not generating or destroying8 heat itself, this must
be same as the amount of heat that entered V in the time interval dt. That is
ZZ ZZZ
∂T
κn̂ · ∇ T dS dt = Cρ dV dt
∂V V ∂t
Now we cancel the common factor of dt. We can then rewrite the left hand
side as an integral over V by applying the divergence theorem giving
ZZZ ZZZ
∂T
∇ · ∇ T dV =
κ∇ Cρ dV
V V ∂t
As both integrals are over the same volume V , we have

ZZZ ZZZ
∂T
∇ · ∇ T dV −
κ∇ Cρ dV = 0
V V ∂t
ZZZ
∂T
=⇒ ∇2 T − Cρ
κ∇ dV = 0 (H)
V ∂t
2 2 2
where ∇ 2 = ∇ · ∇ = ∂x∂ ∂ ∂
2 + ∂y 2 + ∂z 2 is the Laplacian. This must be true for
all volumes V in the object and for all times t. We claim that this forces
∂T
∇2 T (x, y, z, t) − Cρ
κ∇ (x, y, z, t) = 0
∂t
for all (x, y, z) in the object and all t.
Suppose that to the contrary there was a point (x0 , y0 , z0 ) in the object and
a time t0 with, for example, κ∇ ∇2 T (x0 , y0 , z0 , t0 ) − Cρ ∂T
∂t (x0 , y0 , z0 , t0 ) > 0. By
continuity, which we are assuming, κ∇ ∇2 T (x, y, z, t0 ) − Cρ ∂T ∂t (x, y, z, t0 ) must
8 The caloric theory of heat was itself destroyed by the cannon boring experiment of 1798.
In this experiment the American/British physicist Benjamin Thompson (1753–1814) boiled

water just using the heat generated by friction during the boring of a cannon.
remain close to κ∇ ∇2 T (x0 , y0 , z0 , t0 ) − Cρ ∂T

∂t (x0 , y0 , z0 , t0 ) when (x, y, z) is close
to (x0 , y0 , z0 ). So we would have
∂T
∇2 T (x, y, z, t0 ) − Cρ
κ∇ (x, y, z, t0 ) > 0
∂t
for all (x, y, z) in some small ball B centered on (x0 , y0 , z0 ). Then, necessarily,
ZZZ h
∂T i
∇ · ∇ T (x, y, z, t0 ) − Cρ
κ∇ (x, y, z, t0 ) dV > 0
B ∂t
which violates (H) for V = B. This completes our derivation of the heat
equation, which is
Equation 4.2.8
∂T
∇2 T (x, y, z, t)
(x, y, z, t) = α∇
∂t
where α = κ
Cρ is called the thermal diffusivity.
4.2.1.2 An Application of the Heat Equation

As an application, we look at the temperature a short distance below the
surface of the Earth. For simplicity, we make the Earth flat9 and we assume
that the temperature, T , depends only on time, t, and the vertical coordinate,
z. Then the heat equation simplifies to
∂T ∂2T
(z, t) = α 2 (z, t) (HE)
∂t ∂z
We choose a coordinate system having the surface of the Earth at z = 0 and
having z increase downward. We also assume that the temperature T (0, t) at
the surface of the Earth is primarily determined by solar heating and is given
by
T (0, t) = T0 + TA cos(σt) + TD cos(δt) (BC)
Here T0 is the long term average of the temperature at the surface of the
Earth, TA cos(σt) gives seasonal temperature variations and TD cos(δt) gives
daily temperature variations.
air
z “ 0, T p0, tq “ T0 ` TA cospσtq ` TD cospδtq
earth
Tt “ αTzz
z
We measure time in days so that δ = 2π and σ = 1 2π 2π

year = 365days . Then
TA cos(σt) has period one year and TD cos(δt) has period one day. The solution
to the initial value problem (HE)+(BC) can be found by separation of variables,
a standard topic in courses on partial differential equations. The solution is
√σ r
σ
− 2α z
T (z, t) = T0 + TA e cos σt − z
2α
√δ r
δ
− 2α z
+ TD e cos δt − z (SLN)
2α
Whether or not you can find this solution, you can, and should, check that
(SLN) satisfies both (HE) and (BC).
9 Insert sarcastic footnote here.
Now let’s see what we can learn from the solution (SLN). For any fixed
z, the time average of T (z, t) is T0 (just because the average value if cosine
is zero), the same as the average temperature at the surface z = 0. That is,
under the hypotheses that we have made, the long term average temperature
at any depth z is is the same as the long term average temperature at the
surface. √σ pσ
The term TA e− 2α z cos σt − 2α z
• oscillates in time with a period of one year, just like TA cos(σt)

√σ
• has an amplitude TA e− 2α z which is TA at the surface and decreases
q
2α
exponentially as z increases. Increasing the depth z by a distance σ
1
causes the amplitude of the oscillation to decrease by a factor of e.
Both
of these first two bullet points are probably very consistent with your
intuition. But this term also has a third property that you may find less
obvious. It has
z
• has a time lag of √2ασ with respect to TA cos(σt). The surface term
TA cos(σt) takes its maximum value when
t =p 0, 2π 4π
σ , σ , · · ·. At depth z,
√σ
the corresponding term TA e− 2α z cos σt− 2α σ
z takes its maximum
pσ z
value when σt − 2α z = 0, 2π, 4π, · · · so that t = √2ασ , 2π
σ +
√ z , 4π + √ z , · · ·.
2ασ σ 2ασ
√δ q
Similarly, the term TD e− 2α z cos δt − 2α δ
z
• oscillates in time with a period of one day, just like TD cos(δt)

• has an amplitude which qis TD at the surface and decreases by a factor of
1 2α
e for each increase of δ in depth.
• has a time lag of √z with respect to TD cos(δt).

2αδ
For water α is approximately 0.012 m2 /day. This α gives

r r
2α 2α z z
≈ 1.2 m ≈ 0.062 m √ ≈ 49 z days √ ≈ 2.6 z days
σ δ 2ασ 2αδ
for z measured in centimeters. So at a depth of a couple of meters, the tem-
perature is pretty constant in time. What variation there is lags the surface
variations by several months.
4.2.2 Variations of the Divergence Theorem

Here are a couple useful variations of the divergence theorem.
Theorem 4.2.9 Variations on the divergence theorem. If V is a solid
with surface ∂V , then
ZZ ZZZ
F · n̂ dS = ∇ · F dV
∂V V
ZZ ZZZ
f n̂ dS = ∇ f dV
Z Z ∂V Z Z ZV
n̂ × F dS = ∇ × F dV
∂V V
where n̂ is the outward unit normal of ∂V .

Memory Aid. All three formulae can be combined into
ZZ ZZZ
n̂ ∗ F̃ dS = ∇ ∗ F̃ dV
∂V V
where ∗ can be either ·, × or nothing. When ∗ = · or ∗ = ×, then F̃ = F.

When ∗ is nothing, F̃ = f .
Proof. The first formula is exactly the divergence theorem and was proven in
Theorem 4.2.2.
To prove the second formula, set F = f a, where a is any constant vector,
and apply the divergence theorem.
ZZ ZZZ
f a · n̂ dS = ∇ · (f a) dV
∂V V
ZZZ

= ∇f ) · a + f ∇
(∇ | {z· a} dV
V =0
ZZZ
= ∇f ) · a dV
(∇
V
To get the second line, we used the vector identity Theorem 4.1.4.c. To get the
third line, we just used that a is a constant, so that all of its derivatives are
zero. Rewrite ZZZ ZZZ
∇f ) · a dV =
(∇ ∇f ) dV
a · (∇
V V
Since a is a constant, we can factor it out of both integrals, so
ZZ ZZZ
a· f n̂ dS = a · ∇ f dV
∂V V
ZZ ZZZ
=⇒ a · f n̂ dS − ∇ f dV = 0
∂V V
In particular,
RR RRRa = ı̂ı, ̂ and k̂, we see that all three components of the
choosing
vector ∂V f n̂ dS − V
∇ f dV are zero. So
ZZ ZZZ
f n̂ dS − ∇ f dV = 0
∂V V
which is what we wanted show.

To prove the third formula, apply the divergence theorem, but with F
replaced by a × F, where a is any constant vector.
ZZ ZZZ
(a × F) · n̂ dS = ∇ · (a × F) dV
∂V V
ZZZ

= ∇ × a) −a · (∇
F · (∇ ∇ × F) dV
V | {z }
=0
ZZZ ZZZ
=− ∇ × F) dV = −a ·
a · (∇ ∇ × F dV
V V
To get the second line, we used the vector identity Theorem 4.1.4.d. To get
the third line, we again used that a is a constant, so that all of its derivatives
are zero. For all vectors (a × b) · c = a · (b × c) (in case you don’t remember
this, it was Lemma 4.1.8.a) so that
(a × F) · n̂ = a · (F × n̂)
and
ZZ ZZZ
a· F × n dS = −a · ∇ × F dV
Z∂V
Z ZZZ V
=⇒ a · F × n dS + ∇ × F dV = 0
∂V V
In particular,
RR = ı̂ı, ̂ and k̂, we see that all three components of the
choosing aRRR
vector ∂V F × n dS + V
∇ × F dV are zero. So
ZZZ ZZ ZZ
∇ × F dV = − F × n dS = n̂ × F dS
V ∂V ∂V
4.2.3 An Application of the Divergence Theorem — Buoy-

ancy
In this section, we use the divergence theorem to show that when you immerse
an object in a fluid the net effect of fluid pressure acting on the surface of the
object is a vertical force (called the buoyant force) whose magnitude equals
the weight of fluid displaced by the object. This is known as Archimedes’
principle10 .
We shall also show that the buoyant force acts through the “centre of
buoyancy” which is the centre of mass of the fluid displaced by the object. The
design of self-righting11 boats exploits the fact that the centre of buoyancy and
the centre of gravity, where gravity acts, need not be the same.
We start by computing the total force due to the pressure of the fluid
pushing on the object. Recall that pressure
• is the force per unit surface area that the fluid exerts on the object
• acts perpendicularly to the surface
• pushes on the object
Thus the force due to pressure that acts on an infinitesimal piece of the ob-
ject’s surface at r = (x, y, z) with surface area dS and outward normal n̂ is
−p(r) n̂dS. The minus sign is there because pressure is directed into the ob-
ject. If the object fills the volume V and has surface ∂V , then the total force
on the object due to fluid pressure, called the buoyant force, is
ZZ
B=− p(r) n̂ dS
∂V
WeRRR
now wish to apply a variant of the divergence theorem to rewrite B =
− V
∇ p dV . But there is a problem with this: p(r) is the fluid pressure at r
and is only defined where there is fluid. In particular, there is no fluid12 inside
the object, so p(r) is not defined for any r in the interior of V .
So we pretend that we remove the object from the fluid and we call P (r)
the fluid pressure at r when there is no object in the fluid. We also make the
assumption that at any point r outside of the object, the pressure at r does
10 The interested reader should do a net search for the story of Archimedes and the golden
crown.
11 The first design of a self-righting boat was entered by William Wouldhave in a lifeboat
design competition organised by South Shield’s Law House committee in 1789.

12 A cup of tea in the galley doesn’t count.
not depend on whether the object is in the fluid or not. In other words, we
assume that (
P (r) if r is not in V
p(r) =
not defined if r is in the V
This assumption is only an approximation to reality, but, in practice, it is a
very good approximation. So, by Theorem 4.2.9,
ZZ ZZ ZZZ
B=− p(r) n̂ dS = − P (r) n̂ dS = − ∇ P (r) dV (4.2.1)
∂V ∂V V
Our next job is to compute ∇ P . Concentrate on an infinitesimal cube of

fluid whose edges are parallel to the coordinate axes. Call the lengths of the
edges dx, dy and dz and the position of the centre of the cube (x, y, z). The
forces applied to the various faces of the cube by the pressure of fluid outside
the cube are illustrated in the figure
´P px, y, z ` dz
2
q dxdy k̂
dy dy
P px, y ´ 2
, zq dxdz ̂ ´P px, y ` 2
, zq dxdz ̂
P px, y, z ´ dz
2
q dxdy k̂
The total force due to the pressure acting on the cube is the sum

dx dx
−P x+ , y, z dydz ı̂ı + P x − , y, z dydz ı̂ı
2 2

dy dy
− P x, y + , z dxdz ̂ + P x, y − , z dxdz ̂
2 2

dz dz
− P x, y, z + dxdy k̂ + P x, y, z − dxdy k̂
2 2
of the forces acting on the six faces. Consider the ı̂ı component and rewrite it
as

dx dx
−P x+ , y, z dydz ı̂ı + P x − , y, z dydz ı̂ı
2 2
dx dx
P (x + 2 , y, z) − P (x − 2 , y, z)
=− ı̂ı dxdydz
dx
∂P
=− (x, y, z)ı̂ı dxdydz
∂x
Doing this for the other components as well, we see that the total force due
to the pressure acting on the cube is
n ∂P ∂P ∂P o
− (x, y, z)ı̂ı + (x, y, z) ̂ + ∇P (x, y, z) dxdydz
(x, y, z) k̂ dxdydz = −∇
∂x ∂y ∂z
We shall assume that the only other force acting on the cube is gravity and
that the fluid is stationary (or at least not accelerating). Hence the total force
acting on the cube is zero. If the fluid has density ρf , then the cube has mass
ρf dxdydz so that the force of gravity is −gρf dxdydz k̂. The vanishing of the
total force now tells us that
∇P (r) dxdydz − gρf dxdydz k̂ = 0 =⇒ ∇ P (r) = −gρf k̂

−∇
Subbing this into (4.2.1) gives

ZZZ
B = g k̂ ρf dV = gMf k̂
V
ρ dV is the mass of the fluid displaced by the object — not

RRR
where Mf = V f
the mass of the object itself. Thus the buoyant force acts straight up and has
magnitude equal to gMf , which is also the magnitude of the force of gravity
acting on the fluid displaced by the object. In other words, it is the weight of
the displaced fluid. This is exactly Archimedes’ principle.
We next consider the rotational motion of our submerged object. The
physical law that determines the rotational motion of a rigid body about a
point r0 is analogous to the familiar Newton’s law, m dv
dt = F, that determines
the translational motion of the object. For the rotational law of motion,
• the mass m is replaced by a physical quantity, characteristic of the object,
called the moment of inertia, and
• the ordinary velocity v is replaced by the angular velocity, which is a
vector whose length is the rate of rotation (i.e. angle rotated per unit
time) and whose direction is parallel to the axis of rotation (with the sign
determined by a right hand rule), and
• the force F is replaced by a vector called the torque about r0 . A force F
applied at r = (x, y, z) produces the torque13 (r − r0 ) × F about r0 .
This is derived in the optional §4.2.4, entitled “Torque”, and is all that we
need to know about rotational motion of rigid bodies in this discussion.
Fix any point r0 . The total torque about r0 produced by force of pressure
acting on the surface of the submerged object is
ZZ ZZ

T= (r − r0 ) × − p(r)n̂ dS = n̂ × P (r) (r − r0 ) dS
∂V ∂V
Recall that in these integrals r = (x, y, z) is the position of the infinitesimal

piece dS of the surface S. Applying the cross product variant of the divergence
theorem in Theorem 4.2.9, followed by the vector identity Theorem 4.1.5.c,
gives
ZZZ

T= ∇ × P (r) (r − r0 ) dV
Z Z ZV

= ∇ P (r) × (r − r0 ) + P (r) ∇ × (r − r0 ) dV
V | {z }
=0
ZZZ
= ∇ P (r) × (r − r0 ) dV
V
since ∇ × r0 = 0, because r0 is a constant, and

 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × r = det  ∂x ∂y ∂z  = 0
x y z
We have already found that ∇ P (r) = −gρf k̂. Substituting it in gives

ZZZ
T=− gρf k̂ × (r − r0 ) dV
V
13 This is what Archimedes was referring to when he said “Give me a lever and a place to
stand and I will move the earth.”
ZZZ
= −g k̂ × ρf (r − r0 ) dV
V
ZZZ ZZZ
= −g k̂ × rρf dV − r0 ρf dV
V V
ZZZ RRR
rρf dV
= −g ρf dV k̂ × RRRV − r0
V ρ dV
V f
RRR
rρf dV
= −B × RRRV − r0
ρ dV
V f
RRR
rρf dV
= RRRV − r0 × B
ρ dV
V f
So the torque generated at r0 by pressure over the entire surface is the same
the torque generated at r0 by a force B applied at the single point
RRR
rρf dV
CB = RRRV
ρ dV
V f
This point is called the centre of buoyancy. It is the centre of mass of the
displaced fluid.
The moral of the above discussion is that the buoyant force, B, on a rigid
body
• acts straight upward,
• has magnitude equal to the weight of the displaced fluid and

• acts at the centre of buoyancy, which is the centre of mass of the displaced
fluid.
As above, denoting by ρb the density of the object, the torque about r0 due to
gravity acting on the object is
RRR ZZZ
rρb dV
ZZZ
V
(r − r0 ) × (−gρb k̂) dV = RRR − r0 × −g ρb dV k̂
V ρ dV
V b V
So the gravitational force, G,

• acts straight down,
RRR
• has magnitude equal to the weight gMb = g V
rρb dV (where ρb is the
density of the object) of the object and
RRR
rρb dV
• acts at the centre of mass, CG = RRRV , of the object.
ρb dV
V
Because the mass distribution of the object need not be the same as the
mass distribution of the displaced fluid, buoyancy and gravity may act at two
different points. This is exploited in the design of self-righting boats.
These boats are constructed with a heavy, often lead (which is cheap and
dense), keel. As a result, the centre of gravity is lower in the boat than the
center of buoyancy, which, because the displaced fluid has constant density, is
at the geometric centre of the boat. As the figure below illustrates, a right side
up configuration of such a boat is stable, while an upside down configuration
is unstable. The boat rotates so as to keep the centre of gravity straight below
the centre of buoyancy. To see this pretend that you are holding on to the boat
with one hand holding the centre of buoyancy and the other hand holding the
centre of gravity. Use your hands to apply forces in the directions of the arrows
and think about how the boat will respond.
B
B
G G
4.2.4 Optional — Torque

In this section, we derive the properties of torque that we used in the last
section. Newton’s law of motion says that the position r(t) of a single particle
moving under the influence of a force F obeys mr00 (t) = F. Similarly, the
positions ri (t), 1 ≤ i ≤ n, of a set of particles moving under the influence of
forces Fi obey mr00i (t) = Fi , 1 ≤ i ≤ n. Very often systems of interest consist
of some small number of rigid bodies. Suppose that we are interested in the
motion of a single rigid body, say a piece of wood. The piece of wood is made
up of a huge number14 of atoms. So the system of equations determining the
motion of all of the individual atoms in the piece of wood is huge. On the other
hand, we shall see that because the piece of wood is rigid, its configuration is
completely determined by the position of, for example, its centre of mass and
its orientation (we won’t get into what precisely is meant by “orientation”, but
it is certainly determined by, for example, the positions of a few of the corners
of the piece of wood). To be precise, we shall extract from the huge system
of equations that determine the motion of all of the individual atoms, a small
system of equations that determine the motion of the centre of mass and the
orientation. We’ll do so now.
Imagine a piece of wood moving in R3 .
Furthermore, imagine that the piece of wood consists of a huge number of

particles joined by a huge number of weightless but very strong15 steel rods.
The steel rod joining particle number one to particle number two just represents
a force acting between particles number one and two. Suppose that
• there are n particles, with particle number i having mass mi ,

• at time t, particle number i has position ri (t),
• at time t, the external force (gravity and the like) acting on particle
number i is Fi (t), and
• at time t, the force acting on particle number i, due to the steel rod
joining particle number i to particle number j is Fi,j (t). If there is
no steel rod joining particles number i and j, just set Fi,j (t) = 0. In
particular, Fi,i (t) = 0.
14 Just 12 grams of carbon contains about 6 × 1023 atoms.
15 Mathematicians and their idealizations! Really the rods just represent the atomic/
chemical forces that hold the wood together.
The only assumptions that we shall make about the steel rod forces are
(A1) for each i 6= j, Fi,j (t) = −Fj,i (t). In words, the steel rod joining particles
i and j applies equal and opposite forces to particles i and j.

i 6= j, there is a function Mi,j (t) such that Fi,j (t) = Mi,j (t) ri (t)−
(A2) for each
rj (t) . In words, the force due to the rod joining particles i and j
acts parallel to the line joining particles
i and j. For (A1) to be true,
that is to have Mi,j (t) ri (t) − rj (t) = −Mj,i (t) rj (t) − ri (t) , we need
Mi,j (t) = Mj,i (t).
Newton’s law of motion, applied to particle number i, now tells us that
n
X
mi r00i (t) = Fi (t) + Fi,j (t) (Ni )
j=1
Adding up all of the equations (Ni ), for i = 1, 2, 3, · · · , n gives

n
X n
X X
mi r00i (t) = Fi (t) + Fi,j (t) (ΣNi )
i=1 i=1 1≤i,j≤n
P
The sum Fi,j (t) contains F1,2 (t) exactly once and it also contains
1≤i,j≤n
F2,1 (t) exactly once and these
P two terms cancel exactly, by assumption (A1).
In this way, all terms in Fi,j (t) with i 6= j exactly cancel. All terms
1≤i,j≤n P
with i = j are assumed to be zero. So Fi,j (t) = 0 and the equation
1≤i,j≤n
(ΣNi ) simplifies to
n
X n
X
mi r00i (t) = Fi (t) (ΣNi )
i=1 i=1
n n
1
P P
Phew! Denote by M = mi the total mass of the body, by R(t) = M mi ri (t)
i=1 i=1
n
the centre of mass16 of the body and by F(t) =
P
Fi (t) the total external force
i=1
acting on the system. In this notation, equation (ΣNi ) can be written as
Equation 4.2.10
M R00 (t) = F(t)
The upshot is that the centre of mass of the system moves just like a single
particle of mass M subject to the total external force. This is why we can
often replace an extended object by a point mass at its centre of mass.
Now take the cross product of ri (t) and equation (Ni ) and sum over i. This
gives
n
X
mi ri (t) × r00i (t)
i=1
n
X X
= ri (t) × Fi (t) + ri (t) × Fi,j (t) (Σri × Ni )
i=1 1≤i,j≤n
By the assumption (A2)

r1 (t) × F1,2 (t) = M1,2 (t) r1 (t) × r1 (t) − r2 (t)
16 Note that this is just the weighted average (no pun intended) of the positions of the
particles.

r2 (t) × F2,1 (t) = M2,1 (t) r2 (t) × r2 (t) − r1 (t)

= −M1,2 (t) r2 (t) × r1 (t) − r2 (t)
so that

r1 (t) × F1,2 (t) + r2 (t) × F2,1 (t) = M1,2 (t) r1 (t) − r2 (t) × r1 (t) − r2 (t) = 0
because the cross product of any two parallel vectors is zero.
P
The last equation says that the i = 1, j = 2 term in ri (t) × Fi,j (t)
1≤i,j≤n
exactly
P cancels the i = 2, j = 1 term. In this way all of the terms in
ri (t) × Fi,j (t) with i 6= j cancel. Each term with i = j is exactly
1≤i,j≤n P
zero because Fii = 0. So ri (t) × Fi,j (t) = 0 and (Σri × Ni ) simplifies
1≤i,j≤n
to
n
X n
X
mi ri (t) × r00i (t) = ri (t) × Fi (t) (Σri × Ni )
i=1 i=1
At this point it makes sense to define vectors
n
X
L(t) = mi ri (t) × r0i (t)
i=1
n
X
T(t) = ri (t) × Fi (t)
i=1
because, in this notation, (Σri × Ni ) becomes

Equation 4.2.11
d
L(t) = T(t)
dt
Equation 4.2.11 plays the role of Newton’s law of motion for rotational
motion. T(t) is called the torque and plays the role of “rotational force”. L(t)
is called the angular momentum (about the origin) and is a measure of the rate
at which the piece of wood is rotating. For example, if a particle of mass m
is travelling in a circle of radius ρ in the xy-plane at ω radians per unit time,
then r(t) = ρ cos(ωt)ı̂ı + ρ sin(ωt)̂ and
mr(t) × r0 (t) = m ρ cos(ωt)ı̂ı + ρ sin(ωt)̂ × − ωρ sin(ωt)ı̂ı + ωρ cos(ωt)̂

= mρ2 ω k̂
is proportional to ω, which is the rate of rotation about the origin and is in
the direction k̂, which is normal to the plane containing the circle.
In any event, in order for the piece of wood to remain stationary, equations
4.2.10 and 4.2.11 force F(t) = T(t) = 0.
Now suppose that the piece of wood is a seesaw17 that is supported on a
fulcrum at p. The forces consist of gravity, −mi g k̂, acting on particle number
i, for each 1 ≤ i ≤ n, and the
Φ
´m1 g k̂ ´m2 g k̂ ´m3 g k̂ ´m4 g k̂

17 Or teeter-totter for those who speak a different English dialect.
force Φ imposed by the fulcrum that is pushing up on the particle at p.

n
P
The total external force is F = Φ − mi g k̂ = Φ − M g k̂. If the seesaw is to
i=1
remain stationary, this must be zero so that Φ = M g k̂.
The total torque (about the origin) is
n
X n
X
T=p×Φ− mi gri × k̂ = g M p − mi ri × k̂
i=1 i=1
If the seesaw is to remain stationary, this must also be zero. This will be the
case if the fulcrum is placed at
n
1 X
p= mi ri
M i=1
which is just the centre of mass of the piece of wood.

More generally, suppose that the external forces acting on the piece of
wood consist of Fi , acting on particle number i, for each 1 ≤ i ≤ n, and
a “fulcrum force” Φ acting on a particle at p. The total external force is
n
P
F = Φ+ Fi . If the piece of wood is to remain stationary, this must be zero
i=1
n
P
so that Φ = − Fi . The total torque (about the origin) is
i=1
n
X n
X
T=p×Φ+ ri × Fi = (ri − p) × Fi
i=1 i=1
If the piece of wood is to remain stationary, this must also be zero. That is, the
torque about point p due to all of the forces Fi , 1 ≤ i ≤ n, must be zero.
4.2.5 Optional — Solving Poisson’s Equation

In this section we shall use the divergence theorem to find a formula for the
solution of Poisson’s equation
∇ 2 ϕ = 4πρ
Here ρ = ρ(r) is a given (continuous) function and ϕ is the unknown function

that we wish to find. This equation arises, for example, in electrostatics, where
ρ is the charge density and ϕ is the electric potential.
The main step in finding this solution formula will be to consider an
• arbitrary (smooth) function ϕ and an
• arbitrary (smooth) region V in R3 and an
• arbitrary point r0 in the interior of V
and to find an auxiliary formula which expresses ϕ(r0 ) in terms of
• ∇ 2 ϕ(r), with r running over V and
• ∇ ϕ(r) and ϕ(r), with r running only over ∂V .
This auxiliary formula, which we shall derive below, is
∇ 2 ϕ(r) 3
ZZZ
r − r0
ZZ
1
ϕ(r0 ) = − d r− ϕ(r) · n̂ dS
4π V |r − r 0 | ∂V |r − r0 |3

∇ ϕ(r)
ZZ
− · n̂ dS (V )
∂V |r − r0 |
When we take the limit as V expands to fill all of R3 then, assuming that ϕ
and ∇ ϕ go to zero sufficiently quickly18 at ∞, the two integrals over ∂V will
converge to zero and we will end up with the formula
∇ 2 ϕ(r) 3
ZZZ
1
ϕ(r0 ) = − d r
4π R3 |r − r0 |
This expresses ϕ evaluated at an arbitrary point, r0 , of R3 in terms of ∇ 2 ϕ(r),

with r running over R3 , which is exactly what we want, since ∇ 2 ϕ = 4πρ for
any solution of Poisson’s equation. So once we have proven (V) we will have
proven19
Theorem 4.2.12 Assume that ρ(r) is continuous and decays sufficiently quickly
as r → ∞. If ϕ obeys ∇ 2 ϕ = 4πρ on R3 , and ϕ and ∇ ϕ decay sufficiently
quickly as r → ∞, then
ZZZ
ρ(r)
ϕ(r0 ) = − d3 r
R3 |r − r0 |
for all r0 in R3 .
Let
r(x, y, z) = xı̂ı + y ̂ + z k̂

r0 = x0 ı̂ı + y0 ̂ + z0 k̂
1
We shall exploit three properties of the function |r−r0 | . The first two properties
are
1 r − r0
∇ =− (P1)
|r − r0 | |r − r0 |3
1 r − r0
∇2 ∇·
= −∇ =0 (P2)
|r − r0 | |r − r0 |3
and are valid for all r 6= r0 . Verification of the first property is a simple one
line computation. Verification of the second property is a simple three line
computation. (See the solution to Question 6 in Section 4.1.)
1
The other property of |r−r 0|
that we shall use is the following. Let Sε be
the sphere of radius ε centered on r0 . Then, for any continuous function ψ(r),
ZZ ZZ ZZ
ψ(r) 1 ψ(r0 )
lim p
dS = lim p ψ(r) dS = lim dS
ε→0+ Sε |r − r0 | ε→0+ ε Sε ε→0+ εp Sε
ψ(r0 )
= lim 4πε2
ε→0+ εp

4πψ(r0 )
 if p = 2
= 0 if p < 2 (P3)


undefined if p > 2
18 Suppose, for example, that, for large |r − r0 |, |ϕ(r)| is bounded by a constant times
1 1
|r−r0 |
∇ϕ(r)| is bounded by a constant times |r−r
and |∇ |2
. Then, if ∂V is the sphere of
0
radius R centred on r0 , ∂V has surface area 4πR2 and the two integrals over ∂V are bounded
1
by a constant times R .
19 Note that the theorem does not claim that the ϕ defined in the theorem obeys ∇ 2 ϕ =
4πρ. It does, but the proof is beyond our scope.

Derivation of (V):
Here is the derivation of (V ). Let Vε be the part of V outside of Sε .
Sε V
r0 Vε
Note that the boundary ∂Vε of Vε consists of two parts — the boundary
∂V of V and the sphere Sε — and that the unit outward normal to ∂Vε on Sε
r−r0
is − |r−r 0|
, because it points towards r0 and hence outside of Vε .
Recall the vector identity Theorem 4.1.7.d, which says
∇f ) = f ∇ 2 g − g ∇ 2 f
∇g − g∇
∇ · (f∇
1
Applying this identity with f = |r−r0 | and g = ϕ gives
=0 by (P2)
2
}| { z
1 1 ∇ ϕ

2 1
∇· ∇
∇ ϕ − ϕ∇ = −ϕ∇
|r − r0 | |r − r0 | |r − r0 | |r − r0 |
∇2ϕ
=
|r − r0 |
which is the integrand of the first integral on the right hand side of (V). So,
by the divergence theorem
∇2 ϕ
ZZZ ZZZ 1 1
dV = ∇· ∇ ϕ − ϕ∇ ∇ dV
Vε |r − r0 | Vε |r − r0 | |r − r0 |
ZZ
1 1
= ∇
∇ ϕ − ϕ∇ · n̂ dS
∂V |r − r0 | |r − r0 |
r − r0
ZZ
1 1
+ ∇ ϕ − ϕ∇ ∇ · − dS (M)
Sε |r − r0 | |r − r0 | |r − r0 |
To see the connection between (M) and the rest of (V), note that,
• by (P1), the first term on the right hand side of (M) is
ZZ
1 1
∇
∇ ϕ − ϕ∇ · n̂ dS
∂V |r − r0 | |r − r0 |
∇ ϕ(r) r − r0
ZZ ZZ
= · n̂ dS + ϕ(r) · n̂ dS (R1)
∂V |r − r0 | ∂V |r − r0 |3
which is 4π times the second and third terms on the right hand side of
(V),
1 r−r0
• and substituting in ∇ |r−r 0|
= − |r−r 0|
3 , from (P1), and applying (P3)
with p = 2, the limit of the second term on the right hand side of (M) is
r − r0
ZZ
1 1
lim ∇ϕ − ϕ∇ ∇ · − dS
ε→0+ Sε |r − r0 | |r − r0 | |r − r0 |
ZZ
1
= − lim ∇ ϕ · (r − r0 ) + ϕ dS
ε→0+ Bε |r − r0 |2
h i
= −4π ∇ ϕ · (r − r0 ) + ϕ
r=r0
= −4πϕ(r0 ) (R2)
So applying20 limε→0+ to (M) and substituting in (R1) and (R2) gives
∇2ϕ ∇ ϕ(r) r − r0
ZZZ ZZ ZZ
dV = · n̂ dS + ϕ(r) · n̂ dS − 4πϕ(r0 )
V |r − r0 | ∂V |r − r0 | ∂V |r − r0 |3
which is exactly equation (V).
4.2.6 Exercises
1. Let V be the cube

V = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, 0 ≤ z ≤ 1
and R be the square

R= (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
and let f (x, y, z) have continuous first partial derivatives.

a Use the fundamental theorem of calculus to show that
ZZZ ZZ ZZ
∂f
(x, y, z) dx dy dz = f (x, y, 1) dx dy − f (x, y, 0) dx dy
V ∂z R R
b Use the divergence theorem to show that

ZZZ ZZ ZZ
∂f
(x, y, z) dx dy dz = f (x, y, 1) dx dy − f (x, y, 0) dx dy
V ∂z R R
2.
a By applying the divergence theorem to F = φ a, where a is an
arbitrary constant vector, show that
ZZZ ZZ
∇ φ dV = φ n̂ dS
V ∂V
b Show that the centroid (x̄, ȳ, z̄) of a solid V with volume |V | is
given by
ZZ
1
(x̄, ȳ, z̄) = (x2 + y 2 + z 2 ) n̂ dS
2|V | ∂V
3. Let S be the unit sphere centered at the origin and oriented by the
outward pointing normal. If
F(x, y, z) = x, y, z 2

evaluate the flux of F through S

a directly and
b by applying the divergence theorem.
20 You ∇2 ϕ
might worry about the singularity in |r−r0 |
when applying limε→0+ to
∇2 ϕ
RRR
dV . That this singularity is harmless may be seen using spherical coordi-
Vε |r−r0 |
nates centred on r0 . Then dV contains a factor of |r − r0 |2 (see §A.6.3), which completely
eliminates the singularity.
RR
4. Evaluate, by two methods, the integral S F · n̂ dS, where F = z k̂,
S is the surface x2 + y 2 + z 2 = a2 and n̂ is the outward pointing unit
normal to S.
a First, by direct computation of the surface integral.
b Second, by using the divergence theorem.
5. Let
• F = zy 3 ı̂ı + yx ̂ + (2z + y 2 )k̂ and
• V be the solid in 3-space defined by
9 − x2 − y 2
0≤z≤
9 + x2 + y 2
and
2 2
−y
• D be the bottom surface of V . Because 9−x 9+x2 +y 2 is positive for
x2 + y 2 < 9 and negative for x2 + y 2 > 9, the bottom surface is
z = 0, x2 + y 2 ≤ 9.
• Let S be the curved portion of the boundary of V . It is z =

9−x2 −y 2 2 2
9+x2 +y 2 , x + y ≤ 9. Here is a sketch of the first octant part
of S and D.
z
9−x2 −y 2
z= 9+x2 +y 2
S
y
D x2 + y 2 = 9
x
Denote by |V | the volume of V and compute, in terms of |V |,
ZZ
a F · n̂ dS with n̂ pointing downward
D
ZZZ
b ∇ · F dV
V
ZZ
c F · n̂ dS with n̂ pointing outward
S
Use the divergence theorem to answer at least one of parts (a), (b)
and (c).
RR
6. Evaluate the integral S F · n̂ dS, where F = (x, y, 1), S is the surface
z = 1 − x2 − y 2 for x2 + y 2 ≤ 1, and n̂ is the upward pointing normal,
by two methods.
a First, by direct computation of the surface integral.
b Second, by using the divergence theorem.
7. ∗.
a Find the divergence of the vector field F = (z+sin y, zy, sin x cos y).
b Find the flux of the vector field F of (a) outward through the
sphere of radius 3 centred at the origin in R3 .

8. The sides of a grain silo are described by the portion of the cylinder
x2 + y 2 = 1 with 0 ≤ z ≤ 1. The top of the silo is given by the portion
of the sphere x2 + y 2 + z 2 = 2 lying within the cylinder and above the
xy-plane. Find the flux of the vector field
V(x, y, z) = (x2 yz , yz + ex z , x2 + y)
out of the silo.

9. Let B be the ball of volume V centered at the point (x0 , y0 , z0 ), and
let S be the sphere that is the boundary of B. Find the flux of
F = x2ı̂ı + xy̂ + (3z − yz)k̂ outward (from B) through S.
10. ∗. Let 1+z 1+z
F(x, y, z) = 1 + z 1+z , 1 + z 1+z

,1
Let S be the portion of the surface
x2 + y 2 = 1 − z 4
which is above the xy-plane. What is the flux of F downward through

S?
11. ∗. Use the divergence theorem to find the flux of xı̂ı +y̂ +2z k̂ through
the part of the ellipsoid
x2 + y 2 + 2z 2 = 2
x2 y2 z2
with z ≥ 0. [Note: the ellipsoid a2 + b2 + c2 = 1 has volume 43 πabc.]
12. ∗. Let F(x, y, z) = r/r3 where r = xı̂ı + y ̂ + z k̂ and r = |r|.
a Find ∇ · F.
b Find the flux of F outwards through the spherical surface x2 +

y 2 + z 2 = a2 .
c Do the results of (a) and (b) contradict the divergence theorem?
Explain your answer.
d Let E be the solid region bounded by the surfaces z 2 − x2 − y 2 +

1 = 0, z = 1 and z = −1. Let σ be the bounding surface of E.
Determine the flux of F outwards through σ.
e Let R be the solid region bounded by the surfaces z 2 − x2 − y 2 +
4y − 3 = 0, z = 1 and z = −1. Let Σ be the bounding surface
of R. Determine the flux of F outwards through Σ.
13. ∗. Consider the ellipsoid S given by
y2 z2
x2 + + =1
4 4
with the unit normal pointing outward.
a Parameterize S.
RR
b Compute the flux S
F · n̂ dS of the vector field
F(x, y, z) = (x, y, z)
c Verify your answer in (b) using the divergence theorem.

RR
14. ∗. Evaluate the flux integral S
F · n̂ dS, where
F(x, y, z) = x3 + cos(y 2 ) , y 3 + zex , z 2 + arctan(xy)

and S is the surface of the solid region bounded by the cylinder x2 +

y 2 = 2 and the planes z = 0 and z = 2x + 3. The surface is positively
oriented (its unit normal points outward).
15. ∗. Find the flux of the vector field (x + y, x + z, y + z) through the
cylindrical surface whose equation is x2 + z 2 = 4, and which extends
from y = 0 to y = 3. (Only the curved part of the cylinder is included,
not the two disks bounding it on the left and right.) The orientation
of the surface is outward, i.e., pointing away from the y-axis.
16. ∗. The surface S is the part above the xy-plane of the surface obtained
by revolving the graph of z = 1 − x4 around the z-axis. The surface
S is oriented such that the normal vector has positive z-component.
The circle with radius 1 and centre at the origin in the xy-plane is the
boundary of S.
Find the flux of the divergenceless vector field F(x, y, z) = (yz, x +
z, x2 + y 2 ) through S.
2 2
17. ∗. Let S be the
p part of the paraboloid z = 2 − x − y contained in
2 2
the cone z = x + y and oriented in the upward direction. Let
√ 2
F = (tan z + sin(y 3 ))ı̂ı + e−x ̂ + z k̂
RR
Evaluate the flux integral S F · n̂ dS.
18. ∗. Evaluate the surface integral
ZZ
F · n̂ dS
S
where F(x, y, z) = cos z+xy 2 , xe−z , sin y+x2 z and S is the bound-

ary of the solid region enclosed by the paraboloid z = x2 + y 2 and the

plane z = 4, with outward pointing normal.
19. ∗. Let S be the part of the sphere x2 + y 2 + z 2 = 4 between the planes
z = 1 and z = 0 oriented away from the origin. Let
F = (ey + xz)ı̂ı + (zy + sin(x)) ̂ + (z 2 − 1) k̂
Compute the flux integral

ZZ
F · n̂ dS.
S
20. ∗. Let B be the solid region lying between the planes x = −1, x = 1,
y = 0, y = 2 and bounded below by the plane z = 0 and above by
the plane z + y = 3. Let S be the surface of B. Find the flux of the
vector field
F(x, y, z) = x2 z + cos πy ı̂ı + yz + sin πz ̂ + (x − y 2 ) k̂

21. ∗. Let S be the hemisphere x2 + y 2 + z 2 = 1, z ≥ 0, oriented with n̂

pointing away from the origin. Evaluate the flux integral
ZZ
F · n̂ dS
S
where
p
F = x + cos(z 2 ) ı̂ı + y + ln(x2 + z 5 ) ̂ + x2 + y 2 k̂

22. ∗. Let E be the solid region between the plane z = 4 and the
paraboloid z = x2 + y 2 . Let
1 2
1
F = − x3 + ez ı̂ı + − y 3 + x tan z ̂ + 4z k̂
3 3
a Compute the flux of F outward through the boundary of E.

b Let S be the part of the paraboloid z = x2 + y 2 lying below
the z = 4 plane oriented so that n̂ has a positive k̂ component.
Compute the flux of F through S.
xı̂ı + y ̂ + z k̂
F(x, y, z) = 3/2
[x2 + y 2 + z 2
a Compute ∇ · F.
b Let S1 be the sphere given by
x2 + (y − 2)2 + z 2 = 9
ZZ
oriented outwards. Compute F · n̂ dS.
S1
c Let S2 be the sphere given by
x2 + (y − 2)2 + z 2 = 1
ZZ
oriented outwards. Compute F · n̂ dS.
S2
d Are your answers to (b) and (c) the same or different? Give a
mathematical explanation of your answer.
24. ∗. Let F be the vector field defined by
2 2
F(x, y, z) = y 3 z + 2x ı̂ı + 3y − esin z ̂ + ex +y + z k̂

RR
Calculate the flux integral S F· n̂ dS where S is the boundary surface
of the solid region
E : 0 ≤ x ≤ 2, 0 ≤ y ≤ 2, 0≤z ≤2+y
with outer normal.

F(x, y, z) = z arctan(y 2 ) , z 3 ln(x2 + 1) , 3z

Let the surface S be the part of the sphere x2 + y 2 + z 2 = 4 that lies

above the plane z = 1 and be oriented downwards.
a Find the divergence of F.
RR
b Compute the flux integral S
F · n̂ dS.
26. ∗. Let S be the sphere x2 + y 2 + z 2 = 3 oriented inward. Compute

the flux integral ZZ
F · n̂ dS
S
where
F = xy 2 + y 4 z 6 , yz 2 + x4 z , zx2 + xy 4

27. ∗. Consider the vector field F(x, y, z) = −2xy ı̂ı + y 2 + sin(xz) ̂ +
(x2 + y 2 ) k̂.
a Calculate ∇ · F.
b Find the flux of F through the surface S defined by
x2 + y 2 + (z − 12)2 = 132 , z ≥ 0
using the outward normal to S.

28. ∗. Let S be the portion of the hyperboloid x2 + y 2 − z 2 = 1 between

z = −1 and z = 1. Find the flux of F = (x+eyz )ı̂ı + 2yz +sin(xz) ̂ +
(xy − z − z 2 ) k̂ out of S (away from the origin).
2 cos y
29. ∗. Let F be the vector field
RR F(x, y, z) = (x − y − 1)ı̂ı + (e + z 3 ) ̂ +
5
(2xz + z ) k̂. Evaluate S ∇ × F · n̂ dS where S is the part of the
ellipsoid x2 + y 2 + 2z 2 = 1 with z ≥ 0.
30. ∗. Let S be the portion of the sphere x2 + y 2 + (z − 1)2 = 4 that lies
2 2
above the xy-plane. Find the flux of F = (x2 + ey )ı̂ı + (ex + y 2 ) ̂ +
(4 + 5x) k̂ outward across S.
31. ∗. Find the flux of F = xy 2ı̂ı + x2 y̂ + k̂ outward through the hemi-
spherical surface
x2 + y 2 + z 2 = 4, z≥0
2 2
32. ∗. Let D be the cylinder x + y ≤ 1, 0 ≤ z ≤ 5. Calculate the flux
of the vector field
F = (x + xyez )ı̂ı + 21 y 2 zez ̂ + (3z − yzez ) k̂
outward through the curved part of the surface of D.

33. Find the flux of F = (y + xz)ı̂ı + (y + yz)̂ − (2x + z 2 )k̂ upward through
the first octant part of the sphere x2 + y 2 + z 2 = a2 .
34. Let F = (x − yz)ı̂ı + (y + xz)̂ + (z + 2xy)k̂ and let
• S1 be the portion of the cylinder x2 + y 2 = 2 that lies inside
the sphere x2 + y 2 + z 2 = 4
• S2 be the portion of the sphere x2 + y 2 + z 2 = 4 that lies
outside the cylinder x2 + y 2 = 2
• V be the solid bounded by S1 and S2
Compute
RR
a S1 F · n̂ dS with n̂ pointing inward
ZZZ
b ∇ · F dV
V
RR
c S2
F · n̂ dS with n̂ pointing outward
Use the divergence theorem to answer at least one of parts (a), (b)
and (c).
35. Let E(r) be the electric field due to a charge configuration that has
density ρ(r). Gauss’ law states that, if V is any solid in R3 with
surface ∂V , then the electric flux
ZZ ZZZ
E · n̂ dS = 4πQ where Q= ρ dV
∂V V
is the total charge in V . Here, as usual, n̂ is the outward pointing

unit normal to ∂V . Show that
∇ · E(r) = 4πρ(r)
for all r in R3 . This is one of Maxwell’s equations. Assume that

∇ · E(r) and ρ(r) are well--defined and continuous everywhere.
36. Let V be a solid in R3 with surface ∂V . Show that
ZZ
r · n̂ dS = 3 Volume(V )
∂V
where r = xı̂ı + y ̂ + z k̂ and, as usual, n̂ is the outer normal to ∂V .

See if you can explain this result geometrically.
37. ∗. Let S be the sphere of radius 3, centered at the origin and with
outward orientation. Given the vector field F(x, y, z) = (0, 0, x + z):
a Calculate (using the definition) the flux of F through S
ZZ
F · n̂ dS
S
That is, compute the flux by evaluating the surface integral di-
rectly.
b Calculate the same flux using the divergence theorem.
38. ∗. Consider the cube of side length 1 that lies entirely in the first
octant (x ≥ 0, y ≥ 0, z ≥ 0) with one corner at the origin and
another corner at point (1, 1, 1). As such, one face lies in the plane
x = 0, one lies in the plane y = 0, and another lies in the plane z = 0.
The other three faces lie in the planes x = 1, y = 1, and z = 1. Denote
S as the open surface that consists of the union of the 5 faces of the
cube that do not lie in the plane z = 0. The surface S is oriented in
such a way that the unit normal vectors point outwards (that is, the
orientation of S is such that the unit normal vectors on the top face
point towards positive z-directions). Determine the value of
ZZ
I= F · n̂ dS
S
where F is the vector field given by

z 2
F = y cos(y 2 ) + z − 1 , + 1 , xyez
x+1
39. ∗.
a Find an upward pointing unit normal vector to the surface z =
xy at the point (1, 1, 1).
b Now consider the part of the surface z = xy, which lies within
the cylinder x2 + y 2 = 9 and call it S. Compute the upward flux
of F = (y, x, 3) through S.
c Find the flux of F = (y, x, 3) through the cylindrical surface
x2 + y 2 = 9 in between z = xy and z = 10. The orientation is
outward, away from the z-axis.
40. ∗.
a Find the divergence of the vector field F = (x + sin y, z + y, z 2 ).
b Find the flux of F through the upper hemisphere x2 + y 2 + z 2 =
25, z ≥ 0, oriented in the positive z-direction.
RR
c Specify an oriented closed surface S, such that the flux S F ·
n̂ dS is equal to −9.
41. ∗. Evaluate the surface integrals. (Use any method you like.)
a S z 2 dS, if S is the part of the cone x2 + y 2 = 4z 2 where
RR
0 ≤ x ≤ y and 0 ≤ z ≤ 1.
RR
b S F · n̂ dS, if F = z k̂ and S is the rectangle with vertices
(0, 2, 0), (0, 0, 4), (5, 2, 0), (5, 0, 4), oriented so that the normal
vector points upward.
c S F · n̂ dS, where F = (y − z 2 )ı̂ı + (z − x2 )̂ + z 2 k̂ and S is the
RR
boundary surface of the box 0 ≤ x ≤ 1, 0 ≤ y ≤ 2, 0 ≤ z ≤ 3,

with the normal vector pointing outward.
42. ∗. Let σ1 be the open surface given by z = 1 − x2 − y 2 , z ≥ 0.
Let σ2 be the open surface given by z = x2 + y 2 − 1, z ≤ 0. Let
σ3 be the planar surface given by z = 0, x2 + y 2 ≤ 1. Let F =
[a(y 2 + z 2 ) + bxz]ı̂ı + [c(x2 + z 2 ) + dyz] ̂ + x2 k̂ where a, b, c, and d
are constants.
a Find the flux of F upwards across σ1 .
b Find all values of the constants a, b, c, and d so that the flux of
F outwards across the closed surface σ1 ∪ σ3 is zero.
c Find all values of the constants a, b, c, and d so that the flux of
F outwards across the closed surface σ1 ∪ σ2 is zero.
43. ∗. Let S be the ellipsoid x2 + 2y 2 + 3z 2 = 16 and n̂ its outward unit
normal.
RR (x, y, z) − (2, 1, 1)
a Find S F·n̂ dS if F(x, y, z) = 3/2 .
(x − 2)2 + (y − 1)2 + (z − 1)2
RR (x, y, z) − (3, 2, 2)
b Find S
G·n̂ dS if G(x, y, z) = 3/2 .
(x − 3) + (y − 2)2 + (z − 2)2
2
44. ∗. Let Ω ⊂ R3 be a smoothly bounded domain, with boundary ∂Ω

and outer unit normal n̂. Prove that for any vector field F which is
continuously differentiable in Ω ∪ ∂Ω,

ZZZ ZZ
∇ × F dV = − F × n̂ dS
Ω ∂Ω
45. ∗. Recall that if S is a smooth closed surface with outer normal field
n̂, then for any smooth function p(x, y, z) on R3 , we have
ZZ ZZZ
pn̂ ds = ∇p dV
S E
where E is the solid bounded by S. Show that as a consequence,

the total force exerted on the surface of a solid body contained in a
gas of constant pressure is zero. (Recall that the pressure acts in the
direction normal to the surface.)
46. ∗. Let F be a smooth 3-dimensional vector field such that the flux of
F out of the sphere x2 + y 2 + z 2 = a2 is equal to π(a3 + 2a4 ) for every
a > 0. Calculate ∇ · F(0, 0, 0).
2
47. ∗. Let F = (x2 + y 2 + z 2 )ı̂ı + (ex + y 2 ) ̂ + (3 + x + z) k̂ and let S
be the part of the surface x2 + y 2 + z 2 = 2az + 3a2 having z ≥ 0,
oriented with normal pointing away from the origin. Here a > 0 is a
constant. Compute the flux of F through S.
48. ∗. Let u = u(x, y, z) be a solution of Laplace’s Equation,
∂2u ∂2u ∂2u

+ 2 + 2 = 0,
∂x2 ∂y ∂z
in R3 . Let R be a smooth solid in R3 .

a Prove that the total flux of F = ∇u out through the boundary
of R is zero.
b Prove that the total flux of G = u∇u out through the boundary
of R equals
Z Z Z h 2 2 2 i
∂u ∂u ∂u
+ + dV
R ∂x ∂y ∂z
49. ∗. Let R be the part of the solid cylinder x2 + (y − 1)2 ≤ 1 satisfying
0 ≤ z ≤ y 2 ; let S be the boundary of R. Given F = x2 ı̂ı + 2y ̂ − 2z k̂,
a Find the total flux of F outward through S.
b Find the total flux of F outward through the (vertical) cylindri-
cal sides of S.
Z π
n − 1 π n−2
Z
Hint: sinn θ dθ = sin θ dθ for n = 2, 3, 4, . . ..
0 n 0
50. ∗. A smooth surface S lies above the plane z = 0 and has as its
boundary the circle x2 + y 2 = 4y in the plane z = 0. This circle
also bounds a disk D in that plane. The volume of the 3-dimensional
region R bounded by S and D is 10 cubic units. Find the flux of
F(x, y, z) = (x + x2 y)ı̂ı + (y − xy 2 )̂ + (z + 2x + 3y)k̂
through S in the direction outward from R.

4.3 Green’s Theorem

Our next variant of the fundamental theorem of calculus is Green’s1 theorem,
which relates an integral, of a derivative of a (vector-valued) function, over a
region in the xy-plane, with an integral of the function over the curve bounding
the region. First we need to define some properties of curves.
Definition 4.3.1
a A curve C with parametrization r(t), a ≤ t ≤ b, is said to be closed if
r(a) = r(b).
b A curve C is said to be simple if it does not cross itself. More precisely, if
r(t), a ≤ t ≤ b, is a parametrization of the curve and if a ≤ t1 , t2 ≤ b obey
t1 6= t2 and {t1 , t2 } =
6 {a, b}, then r(t1 ) 6= r(t2 ). That is, if r(t1 ) = r(t2 ),
then either t1 = t2 or t1 = a, t2 = b, or t1 = b, t2 = a.
c A curve C is piecewise smooth if it has a parametrization r(t) which
• is continuous and which
• is differentiable except possibly at finitely many points with
• the derivative being continuous and nonzero except possibly at finitely
many points.
♦
Here are sketches of some examples.
simple curve simple closed curve not a simple piecewise smooth

curve curve
And here is Green’s theorem.
Theorem 4.3.2 Green’s Theorem. Let
• R be a finite region in the xy-plane,
• the boundary, C, of R consist of a finite number of piecewise smooth,
simple closed curves
◦ that are oriented (i.e. arrows are put on C) consistently with R in
the sense that if you walk along C in the direction of the arrows,
then R is on your left
y y
C
R C
R
x
C
x
• F1 (x, y) and F2 (x, y) have continuous first partial derivatives at every
point of R.
1 George Green (1793–1841) was a British mathematical physicist. He spent much of the
early part of his life working in his father’s bakery and grain mill. He was finally admitted
as an undergraduate to Cambridge in 1832, aged nearly forty.
Then I ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy
C R ∂x ∂y
Warning 4.3.3 Note that in Theorem 4.3.2 we are assuming that F1 and
F2 have continuous first partial derivatives at every point of R. If that is
not the case, for example because F1 or F2 is not defined on all of R, then
y
the conclusion of Green’s theorem can fail. An example is F1 = − x2 +y 2,
x
2 2

F2 = x2 +y2 , R = (x, y) x + y ≤ 1 . See Examples 4.3.7 and 4.3.8.

Here are three notational remarks before we start the proof.
• One way to remember the integrand on the right hand side is to write it
∇ × F) · k̂.
as (∇
• Many people use M instead of F1 and N instead of F . Then Green’s
RR 2∂N
theorem becomes C M (x, y) dx + N (x, y) dy = R ∂x − ∂M
H
∂y dxdy
H R
• The symbol C is just an alternate notation for C that is sometimes
used when C is a closed curve. See Notation 2.4.1.
Proof. We prove the result by reformulating it as a divergence theorem state-
ment. To that end, we define

V = (x, y, z) (x, y) ∈ R, 0 ≤ z ≤ 1
G(x, y, z) = F2 (x, y)ı̂ı − F1 (x, y) ̂
Notice that V is exactly the volume obtained by expanding R vertically upward

by one unit.
z
z“1
V
y
R
x C
The definition of G does not contain a typo — the x-component of G really
is F2 and the y-component of G really is −F1 . (More or less the reverse of
what you would normally write down.)
These definitions have been rigged so that the divergence theorem applied
to G and V , namely
ZZ ZZZ
G · n̂ dS = ∇ · G dV
∂V V
gives us exactly Green’s theorem, as we shall now see.

Since ∇ · G = ∂F ∂F1
∂x − ∂y , the right hand side is just
2
ZZZ ZZ Z 1
∇ · G dV = dxdydz ∇ · G
V R 0
ZZ Z 1
∂F2 ∂F1
= dxdy dz (x, y) − (x, y)
R 0 ∂x ∂y
ZZ
∂F2 ∂F1
= dxdy (x, y) − (x, y)
R ∂x ∂y
because the integrand is independent of z. This is exactly the right hand side
of Green’s theorem.
Now for the left hand side. The boundary, ∂V , of V is the union of the (flat)
bottom, the (flat) top and the (curved) side. The outward unit normal on the
(horizontal, flat) top is +k̂ and the outward unit normal on the (horizontal,
flat) bottom is −k̂ so that
ZZ ZZ ZZ ZZ
G · n̂ dS = G · k̂ dS + G · (−k̂) dS + G · n̂ dS
∂V top bottom side
ZZ
= G · n̂ dS
side
We have used the fact that the k̂ component of G is exactly zero to discard
the integrals over the top and bottom of ∂V . To evaluate the integral over the
side, we’ll parametrize the side. Suppose that r(t) = x(t)ı̂ı + y(t) ̂, a ≤ t ≤ b, is
a parametrization of C, with the arrow in the figure above giving the direction
of increasing t. Then we can use
R(t, z) = r(t) + z k̂ = x(t)ı̂ı + y(t) ̂ + z k̂ a ≤ t ≤ b, 0 ≤ z ≤ 1
as a parametrization of the side. We’ll use (3.3.1) to determine n̂ dS for the

side. Since
∂R
(t, z) = x0 (t)ı̂ı + y 0 (t) ̂
∂t
∂R
(t, z) = k̂
∂z
(3.3.1) gives
∂R ∂R
n̂ dS = (t, z) × (t, z) dtdz
∂t ∂z
= x0 (t)ı̂ı + y 0 (t) ̂ × k̂ dtdz

= − x0 (t) ̂ + y 0 (t)ı̂ı dtdz

Note that with this choice of ± sign (that is, ∂R ∂R ∂R

∂t × ∂z dtdz rather than − ∂t ×
∂z dtdz), the vector n̂ really is the outward pointing normal, as we see from
∂R
the sketch
z
z“1
k̂ V
y
R
x C
BR
n̂
Bt
“ r1
We can now compute the surface integral directly.
ZZ ZZ
G · n̂ dS = G · n̂ dS
∂V side
Z b Z 1
dz G R(t, z) · − x0 (t) ̂ + y 0 (t)ı̂ı

= dt
a 0
Z b Z 1
dz F2 (x(t), y(t))ı̂ı − F1 (x(t), y(t)) ̂ · − x0 (t) ̂ + y 0 (t)ı̂ı

= dt
a 0
Z b
dt F2 (x(t), y(t)) y 0 (t) + F1 (x(t), y(t)) x0 (t)

=
a
since the integrand is independent of z
I

= F1 (x, y) dx + F2 (x, y) dy
C
This is exactly the left hand side of Green’s theorem.

Example 4.3.4 Evaluate
I
(x − xy) dx + (y 3 + 1) dy

C
where C is the curve given in the figure

y p1, 1q p2, 1q
R C
p2, 0q x
p1, 0q

Solution. Let R = (x, y) 1 ≤ x ≤ 2, 0 ≤ y ≤ 1 . By Green’s theorem
I ZZ h
∂ 3 ∂ i
(x − xy) dx + (y 3 + 1) dy =

(y + 1) − (x − xy) dxdy
C R ∂x ∂y
Z 2 Z 1 2 2

x 3
= dx dy x = =2
1 0 2 1

Here is a simple corollary of Green’s theorem that tells how to compute the
area enclosed by a curve in the xy-plane.
Corollary 4.3.5 Let
• R be a finite region in the xy-plane whose boundary
• C consists of a finite number of piecewise smooth, simple closed curves.
• Orient C (i.e. put arrows on C) so that if you walk along C in the
direction of the arrows, then R is on your left.
Then I I
I
1
Area(R) =

xdy = − xdy − ydx
ydx =
C C 2 C
Proof. This is just Green’s theorem applied first with F = x ̂, then with
F = −y ı̂ı and finally with F = 21 − y ı̂ı + x ̂ . For all three of these F’s,

∂F2 ∂F1
− =1
∂x ∂y
so that Green’s theorem gives
I ZZ ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy = dxdy
C R ∂x ∂y R
= Area(R)

Example 4.3.6 In this example we will use Green’s theorem to compute the
area enclosed by the astroid x2/3 + y 2/3 = a2/3 .
y
In Example 1.1.9 we found the parametrization
r(t) = x(t)ı̂ı + y(t) ̂ = a cos3 tı̂ı + a sin3 t ̂ 0 ≤ t ≤ 2π
for the astroid. So, by Corollary 4.3.5,
1 2π
I Z
1
x(t)y 0 (t) − y(t)x0 (t) dt

Area = xdy − ydx =
2 C 2 0
3a2 2π
Z
cos3 t sin2 t cos t + sin3 t cos2 t sin t dt

=
2 0
3a2 2π
Z
cos2 t sin2 t cos2 t + sin2 t dt

=
2 0
3a2 2π
Z
= cos2 t sin2 t dt
2 0
3a2 2π 2 3a2 2π
Z Z
= sin (2t) dt = [1 − cos(4t)] dt
8 0 16 0
3
= a2 π
8

Example 4.3.7 Trick Question. Evaluate
I
B · dr
C
where
−y ı̂ı + x ̂
B=
x2 + y 2
and C is the curve
x(t) = sin(cos t)
y(t) = sin(sin t)
z(t) = 0
with 0 ≤ t ≤ 2π.
Solution. First let’s think about the curve C. If the curve were just X(t) =
cos t, Y (t) = sin t, Z(t) = 0, it would be the unit circle centred on the origin
in the xy-plane, traversed counterclockwise. For − π2 ≤ u ≤ π2 , the function
sin u increases monotonically with u and is of the same sign as u so that, since
| sin t|, | cos t| ≤ 1 < π2 ,
• x(t) = sin cos t) has the same sign as X(t) = cos t and is increasing at
precisely the same t’s as is X(t)
• y(t) = sin sin t) has the same sign as Y (t) = sin t and is increasing at
precisely the same t’s as is Y (t)
So the extra sine in our parametrization of C just distorts the circle, straight-
ening the sides a little as depicted here.
y
It looks like our problem is a straightforward Green’s theorem problem like

Example 4.3.4. Let’s just try using the strategy of Example 4.3.4. Because
∂B2 ∂B1 ∂ x ∂ −y
− = 2 2
−
∂x ∂y ∂x x + y ∂y x + y 2
2
1 2x2 1 2y 2
= 2 − 2 + − 2
x + y2 (x2 + y 2 ) x2 + y 2 (x2 + y 2 )
(x2 + y 2 ) − 2x2 + (x2 + y 2 ) − 2y 2
= 2
(x2 + y 2 )
=0
it looks like Green’s theorem gives us, trivially,

I I ZZ
∂B2 ∂B1
B · dr = B1 dx + B2 dy = − dxdy = 0
C C R ∂x ∂y
where R is the region inside our curve C.

That was easy — but it’s also very wrong! Our next steps are to
H
• verify that C B · dr 6= 0, and
• explain why we got the wrong answer, and

• modify our computation so as to give the correct answer. We’ll do this
in Example 4.3.8.
Verification that C B · dr 6= 0:}
H
Since
x0 (t) = − cos(cos t) sin t

y 0 (t) = cos(sin t) cos t
z 0 (t) = 0
our integral is
I I

B · dr = B1 dx + B2 dy
C C
Z2π
B1 x(t), y(t) x0 (t) + B2 x(t), y(t) y 0 (t) dt

=
0
Z 2π
sin(sin t) cos(cos t) sin t + sin(cos t) cos(sin t) cos t
= dt
0 sin2 (cos t) + sin2 (sin t)
This is a very ugly looking integral2 . But even if we can’t evaluate Hthe integral,
we can see that the integrand is strictly positive, and that forces C B · r > 0.
Because
π
0 ≤ | sin t|, | cos t| ≤ 1 <
2
• cos(cos t) > 0, and sin(sin t) has the same sign as sin t, and sin(sin t) is
zero if and only if sin t = 0. So the first term in the numerator,
cos(cos t) sin(sin t) sin t ≥ 0
and is zero if and only if sin t = 0

• cos(sin t) > 0, and sin(cos t) has the same sign as cos t, and sin(cos t) is
zero if and only if cos t = 0. So the second term in the numerator,
cos(sin t) sin(cos t) cos t ≥ 0
and is zero if and only if cos t = 0.

• There is no t for which both sin t and cos t are simultaneously zero. So
the whole numerator
sin(sin t) cos(cos t) sin t + sin(cos t) cos(sin t) cos t > 0
is strictly positive.
Since the integrand is strictly positive, the integral is strictly positive.
Why we got the wrong answer:
In our initial and wrong calculation above, we assumed that ∂B ∂x (x, y) −
2
∂B1
∂y (x, y) = 0 at all points (x, y) of the region R inside C. That’s not true.
While it is true for most points, it is not true for all points. The vector field
B(x, y) is not defined at (x, y) = (0, 0). So ∂B ∂B1
∂x (x, y) − ∂y (x, y) is also not
2
defined at (x, y) = (0, 0). That’s enough to invalidate Green’s theorem. Read
the statement of Theorem 4.3.2 again carefully.
Example 4.3.8 Example 4.3.7, again.. Evaluate
I
B · dr
C
where
−y ı̂ı + x ̂
B=
x2 + y 2
and C is the curve
x(t) = sin(cos t)
y(t) = sin(sin t)
z(t) = 0
with 0 ≤ t ≤ 2π.
Solution. This is the same integral that weHcomputed incorrectly in Example
4.3.7. We’ll use two ingredients to compute C B · dr correctly.
• Let a > 0 and denote by Ca the clockwise oriented circle in the xy-
plane that is of radius a and is centered on the origin. We can explicitly
2 Indeed!
H
compute Ca B · dr. To do so just parametrize Ca by x(t) = a cos t,
y(t) = a sin t, z(t) = 0. Then x0 (t) = −a sin t, y 0 (t) = a cos t and
I Z 2π h −a sin tı̂ı + a cos t ̂ i
B · dr = 2 2 2 2 · − a sin tı̂ı + a cos t ̂ dt
Ca 0 a cos t + a sin t
Z 2π
= dt = 2π
0
• Pick an a that is small enough that Ca lies entirely inside C and apply
Green’s theorem with the region, Ra , that is between C and Ca .
y
C
Ra
Ća
The curve bounding Ra has two components — C and Ca , but now Ca

is oriented clockwise. (Recall that, in Green’s theorem, when you walk
along a boundary curve in the direction of the arrow, Ra has to be on your
left.). Use −Ca to denote Ca oriented clockwise. ∂B ∂B1
∂x (x, y) − ∂y (x, y)
2
really is zero at all points (x, y) of the region Ra . So Green’s theorem

gives
ZZ I I
∂B2 ∂B1
0= − dxdy = B · dr + B · dr
Ra ∂x ∂y C −Ca
I I
= B · dr − B · dr
C Ca
and so
I I
B · dr = B · dr = 2π
C Ca
4.3.1 Exercises
1. Let R be the square

R= (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1
and let f (x, y) have continuous first partial derivatives.

a Use the fundamental theorem of calculus to show that
ZZ Z 1 Z 1
∂f
(x, y) dx dy = f (x, 1) dx − f (x, 0) dx
R ∂y 0 0
b Use Green’s theorem to show that

ZZ Z 1 Z 1
∂f
(x, y) dx dy = f (x, 1) dx − f (x, 0) dx
R ∂y 0 0
2. Let R be a finite region in the xy-plane, whose boundary, C, consists
of a single, piecewise smooth, simple closed curve that is oriented
counterclockwise. “Simple” means that the curve does not intersect
itself. Use Green’s theorem to show that
ZZ I
∇ · F dx dy = F · n̂ ds
R C
where F = F1 ı̂ı + F2 ̂, n̂ is the outward unit normal to C and s is the

arclength along C.
y
n̂
R
C
x
x dy − y dx
I
1
3. Integrate counterclockwise around
2π C x2 + y 2
a the circle x2 + y 2 = a2
b the boundary of the square with vertices (−1, −1), (−1, 1), (1, 1)
and (1, −1)
c the boundary of the region 1 ≤ x2 + y 2 ≤ 2, y ≥ 0

4. Show that
∂ x ∂ −y
=
∂x x2 + y 2 ∂y x2 + y 2
for all (x, y) 6= (0, 0). Discuss the connection between this result and
the results of Q[4.3.1.3].
5. Evaluate C F · dr where F = x2 y 2 ı̂ı + 2xy ̂ and C is the boundary of
R
the square in the xy-plane having one vertex at the origin and diago-
nally opposite vertex at the point (3, 3), oriented counterclockwise.
I
6. Evaluate (x sin y 2 −y 2 ) dx+(x2 y cos y 2 +3x) dy where C is the coun-
C
terclockwise boundary of the trapezoid with vertices (0, −2), (1, −1), (1, 1)
and (0, 2).
I
1 2 3
x y − x4 y dx + xy 4 + x3 y 2 dy counterclock-

7. ∗. Evaluate I =
C 3 √
wise around the boundary of the half-disk 0 ≤ y ≤ 4 − x2 .
8. ∗. Let C be the counterclockwise boundary of the rectangle with
vertices (1, 0), (3, 0), (3, 1) and (1, 1). Evaluate
I
2 2
3y 2 + 2xey dx + 2yx2 ey dy
C
9. ∗. Consider the closed region enclosed by the curves y = x2 + 4x + 4
and y = 4 − x2 . Let C be its boundary and suppose that C is oriented
counter-clockwise.
a Draw the oriented curve C carefully in the xy-plane.
b Determine the value of

I
xy dx + (ey + x2 )dy
C
10. ∗. Let 2 2
F(x, y) = y 2 − e−y + sin x , 2xye−y + x

Let C be the boundary of the triangle with vertices (0, 0), (1, 0) and
(1, 2), oriented counter-clockwise. Compute
Z
F · dr
C
11. ∗. Suppose the curve C is the boundary of the region enclosed between
the curves y = x2 − 4x + 3 and y = 3 − x2 + 2x. Determine the value
of the line integral
Z p
2xey + 2 + x2 dx + x2 (2 + ey ) dy

C
where C is traversed counter-clockwise.

12. ∗. Let
F(x, y) = 32 y 2 + e−y + sin x ı̂ı + 12 x2 + x − xe−y ̂

R
Find C F · dr, where C is the boundary of the triangle (0, 0), (1, −2),
(1, 2), oriented anticlockwise.
13. ∗.
a Use Green’s theorem to evaluate the line integral
−y
Z
x
2 + y2
dx + 2 dy
C x x + y2
where C is the arc of the parabola y = 41 x2 + 1 from (−2, 2) to

(2, 2).
b Use Green’s theorem to evaluate the line integral

−y
Z
x
2 2
dx + 2 dy
C x +y x + y2
where C is the arc of the parabola y = x2 − 2 from (−2, 2) to

(2, 2).
c Is the vector field
−y x
F= ı̂ı + 2 ̂
x2 +y 2 x + y2
conservative? Provide a reason for your answer based on your
answers to the previous parts of this question.
14. ∗. Suppose the curve C is the boundary of the region enclosed between
the curves y = x2 − 4x + 3 and y = 3 − x2 + 2x. Determine the value
of the line integral

Z √
2xey + 2 + x2 dx + x2 2 + ey )dy

C
where C is traversed counter-clockwise.

15. ∗. Let F(x, y) = P ı̂ı + Q ̂ be a smooth plane vector field defined
for (x, y) 6= (0,R 0), and suppose Qx = Py for (x, y) 6= (0, 0). In the
following Ij = Cj F·dr for integer j, and all Cj are positively oriented
circles. Suppose I1 = π where C1 is the circle x2 + y 2 = 1.
a Find I2 for C2 : (x − 2)2 + y 2 = 1. Explain briefly.
b Find I3 for C3 : (x − 2)2 + y 2 = 9. Explain briefly.
c Find I4 for C4 : (x − 2)2 + (y − 2)2 = 9. Explain briefly.
16. ∗. Consider the vector field F = P ı̂ı + Q ̂, where
x+y y−x
P = , Q=
x2 + y 2 x2 + y 2
a Compute and simplify Qx − Py .

R
b Compute the integral CR F · dr directly using a parameteriza-
tion, where CR is the circle of radius R, centered at the origin,
and oriented in the counterclockwise direction.
c Is F conservative? Carefully explain how your answer fits with
the results you got in the first two parts.
R
d Use Green’s theorem to compute C F·dr where C is the triangle
with vertices (1, 1), (1, 0), (0, 1) oriented in the counterclockwise
direction.
R
e Use Green’s theorem to compute C F·dr where C is the triangle
with vertices (−1, −1), (1, 0), (0, 1) oriented in the counterclock-
wise direction.
17. ∗.
a Evaluate Z p
1 + x3 dx + 2xy 2 + y 2 dy

C
where C is the unit circle x2 +y 2 = 1, oriented counterclockwise.

b Evaluate Z p
1 + x3 dx + 2xy 2 + y 2 dy

C
where C is now the part of the unit circle x2 + y 2 = 1, with

x ≥ 0, still oriented counterclockwise.
18. ∗. Evaluate the line integral
Z
(x2 + yex ) dx + (x cos y + ex ) dy
C
where C is the arc of the curve x = cos y for −π/2 ≤ y ≤ π/2,

traversed in the direction of increasing y.
19. ∗. Use Green’s theorem to establish that if C is a simple closed curve

in the plane, then the area A enclosed by C is given by
I
1
A= x dy − y dx
2 C
Use this to calculate the area inside the curve x2/3 + y 2/3 = 1.
20. ∗. Let F(x, y) = (x+3y)ı̂ı +(x+y) ̂ and G(x, y) = (x+y)ı̂ı +(2x−3y) ̂
be vector fields. Find a number A such that for each circle C in the
plane I I
F · dr = A G · dr
C C
y3 xy 2
21. ∗. Let F(x, y) = − (x, y) 6= (0, 0).
(x2 +y 2 )2
ı̂ı (x2 +y 2 )2
̂,
H
a Compute C F · dr where C is the unit circle in the xy-plane,
positively oriented.
H
b Use (a) and Green’s theorem to find C0 F · dr where C0 is the
x2 y2
ellipse 16 + 25 = 1, positively oriented.
22. ∗. Let C1 be the circle (x − 2)2 + y 2 = 1 and let C2 be the circle
y x
(x − 2)2 + y 2 = 9. Let F = − x2 +y ı + x2 +y
2 ı̂ . Find the integrals
2 ̂
H H
C1
F · dr and C2
F · dr.
23. ∗. Let R be the region in the first quadrant of the xy-plane bounded
by the coordinate axes and the curve y = 1 − x2 . Let C be the
boundary of R, oriented counterclockwise.
R
a Evaluate C x ds.
R
b Evaluate C
F · dr, where F(x, y) = sin(x2 ) − xy ı̂ı + x2 +
cos(y 2 ) ̂.
24. ∗. Let C be the curve defined by the intersection of the surfaces
z = x + y and z = x2 + y 2 .
a Show that C is a simple closed curve.
H
b Evaluate C F · dr where
(i) F = x2 ı̂ı + y 2 ̂ + 3ez k̂.

(ii) F = y 2 ı̂ı + x2 ̂ + 3ez k̂.
25. Find a smooth, simple, closed, counterclockwise oriented curve, C,
in the xy-plane for the which the value of the line integral C (y 3 −
H
y) dx − 2x3 dy is a maximum among all smooth, simple, closed, coun-

terclockwise oriented curves.
4.4 Stokes’ Theorem

Our last variant of the fundamental theorem of calculus is Stokes’1 theorem,
which is like Green’s theorem, but in three dimensions. It relates an integral
1 Sir George Gabriel Stokes (1819–1903) was an Irish physicist and mathematician. In
addition to Stokes’ theorem, he is known for the Navier-Stokes equations of fluid dynamics
and for his work on the wave theory of light. He gave evidence to the Royal Commission on
the Use of Iron in Railway Structures after the Dee bridge disaster of 1847.
over a finite surface in R3 with an integral over the curve bounding the surface.
Theorem 4.4.1 Stokes’ Theorem. Let
• S be a piecewise smooth oriented surface (i.e. a unit normal n̂ has been
chosen at each point of S and this choice depends continuously on the
point)
• the boundary, ∂S, of the surface S consist of a finite number of piecewise
smooth, simple curves that are oriented consistently with n̂ in the sense
that
◦ if you walk along ∂S in the direction of the arrow on ∂S,

◦ with the vector from your feet to your head having direction n̂
◦ then S is on your left hand side.
z
n̂
S
BS
x
• F be a vector field that has continuous first partial derivatives at every
point of S.
Then I ZZ
F · dr = ∇ × F · n̂ dS
∂S S
Note that
• in Stokes’ theorem, S must be an oriented surface. In particular, S may
not be a Möbius strip. (See Example 3.5.3.)
• If S is part of the xy-plane, then Stokes’ theorem reduces to Green’s the-
orem. Our proof of Stokes’ theorem will consist of rewriting the integrals
so as to allow an application of Green’s theorem.
• If ∂S is a simple closed curve and
◦ when you look at ∂S from high on the z-axis, it is oriented coun-
terclockwise (look at the figure in Theorem 4.4.1), then
◦ n̂ is upward pointing, i.e. has positive z-component, at least near
∂S.
Proof. Write F = F1 ı̂ı + F2 ̂ + F3 k̂. Both integrals involve F1 terms and F2
terms and F3 terms. We shall show that the F1 terms in the two integrals
agree. In other words, we shall assume that F = F1ı̂ı. The proofs that the
F2 and F3 terms also agree are similar. For simplicity, we’ll assumea that the
boundary of S consists of just a single curve, and that we can
• pick a parametrization of S with
S = r(u, v) = x(u, v), y(u, v), z(u, v) (u, v) in R ⊂ R2

∂r
and with r(u, v) orientation preserving in the sense that n̂ dS = + ∂u ×
∂r
∂v du dv. Also

• pick a parametrization of the curve, ∂R, bounding R as u(t), v(t) , a ≤
t ≤ b, in such a way that when you walk along ∂R in the direction of
increasing t, then R is on your left.

Then the curve ∂S bounding S can be parametrized as R(t) = r u(t), v(t) ,
a ≤ t ≤ b.
z n̂
v
rpu, vq
S R
BS pu, vq
y BR
u
x
The orientation of R(t):
We’ll now verify that the direction of increasing t for the parametrization
R(t) of ∂S is the direction of the arrow on ∂S in the figure on the left above.
By continuity, it suffices to check the orientation at a single point.
Find a point (u0 , v0 ) on ∂R where the forward pointing tangent vector is
a positive multiple of ı̂ı. The horizontal arrow on ∂R in the figure on the left
below is at such a point. Suppose that t = t0 at this point — in other words,
suppose that (u0 , v0 ) = u(t0 ), v(t0 ) . Because the forward pointing tangent
vector to ∂R at (u0 , v0 ), namely u0 (t0 ), v 0 (t0 ) , is a positive multiple of ı̂ı, we
have u0 (t0 ) > 0 and v 0 (t0 ) = 0. The tangent vector to ∂S at R(t0 ) = r u0 , v0 ,

pointing in the direction of increasing t, is

d ∂r ∂r
R0 (t0 ) = r u(t), v(t) t=t0 = u0 (t0 ) (u0 , v0 ) + v 0 (t0 ) (u0 , v0 )

dt ∂u ∂v
0 ∂r
= u (t0 ) (u0 , v0 )
∂u
∂r
and so is a positive multiple of ∂u (u0 , v0 ). See the figure on the right below.
If we now walk along a path in the uv-plane which starts at (u0 , v0 ), holds u
fixed at u0 and increases v, we move into the interior of R starting at (u0 , v0 ).
Correspondingly, if we walk along the path, r(u0 , v), in R3 with v starting at
v0 and increasing, we move into the interior of S. The forward tangent to this
∂r
new path, ∂v (u0 , v0 ), points from r(u0 , v0 ) into the interior of S. It’s the blue
arrow in the figure on the right below.
v rpu, vq z
Br Br
Bu
ˆ Bv
Br
R Bv
Br
Bu
1
BR pu0 , v0 q pu pt0 q, 0q
BS Rpt0 q “ rpu0 , v0 q
y
u
x
Now imagine that you are walking along ∂S in the direction of increasing t.
At time t0 you are at R(t0 ). You point your right arm straight ahead of you. So
∂r
it is pointing in the direction ∂u (u0 , v0 ). You point your left arm out sideways
∂r
into the interior of S. It is pointing in the direction ∂v (u0 , v0 ). If the direction
of increasing t is the same as the forward direction of the orientation of ∂S, then
∂r ∂r
the vector from our feet to our head, which is ∂u (u0 , v0 ) × ∂v (u0 , v0 ), should
∂r ∂r
be pointing in the same direction as n̂. And since n̂ dS = + ∂u × ∂v du dv, it
is.
Now, with our parametrization and orientation sorted out, we can examine
the integrals.
The surface integral:
Since F = F1 ı̂ı, so that
 
ı̂ı ̂ k̂
∂ ∂ ∂  ∂F1 ∂F1
∇ × F = det  ∂x ∂y ∂z  = 0, ,−
∂z ∂y
F1 0 0
and
 
ı̂ı ̂ k̂
∂r ∂r ∂x ∂y ∂z 
n̂ dS = × du dv = det  ∂u ∂u ∂u
∂u ∂v ∂x ∂y ∂z
∂v ∂v ∂v
∂y ∂z ∂z ∂y ∂z ∂x ∂x ∂z
= − ı̂ı + − ̂
∂u ∂v ∂u ∂v ∂u ∂v ∂u ∂v

∂x ∂y ∂y ∂x
+ − k̂
∂u ∂v ∂u ∂v
and
ZZ ZZ
∂F1 ∂F1 ∂r ∂r
∇ × F · n̂ dS = 0, ,− · × du dv
S R ∂z ∂y ∂u ∂v
ZZ
∂F1 ∂z ∂x ∂x ∂z ∂F1 ∂x ∂y ∂y ∂x
= − − − du dv
R ∂z ∂u ∂v ∂u ∂v ∂y ∂u ∂v ∂u ∂v
Now we examine the line integral and show that it equals this one.
The line integral:
I Z b d
F · dr = F r u(t), v(t) · r u(t), v(t) dt
∂S a dt
Z b
h ∂r du ∂r dv i
= F r u(t), v(t) · u(t), v(t) (t) + u(t), v(t) (t) dt
a ∂u dt ∂v dt
We can write this as the line integral
I
M (u, v) du + N (u, v) dv
∂R
Z b du
h dv i
= M u(t), v(t) (t) + N u(t), v(t)) (t) dt
a dt dt
around ∂R, if we choose
∂r ∂x
M (u, v) = F r(u, v) · (u, v) = F1 x(u, v), y(u, v), z(u, v) (u, v)
∂u ∂u
∂r ∂x
N (u, v) = F r(u, v) · (u, v) = F1 x(u, v), y(u, v), z(u, v) (u, v)
∂v ∂v
Finally, we show that the surface integral equals the line integral:
By Green’s Theorem, we have

I I
F · dr = M (u, v) du + N (u, v) dv
∂S ∂R
ZZ
∂N ∂M
= − dudv
R ∂u ∂v
∂2x
ZZ
∂ ∂x
= F1 x(u, v), y(u, v), z(u, v) + F1
R ∂u ∂v ∂u∂v
∂2x

∂ ∂x
− F1 x(u, v), y(u, v), z(u, v) − F1 dudv
∂v ∂u ∂v∂u
∂2x
Z Z
∂F1 ∂x ∂F1 ∂y ∂F1 ∂z ∂x
= + + +F1
R ∂x ∂u ∂y ∂u ∂z ∂u ∂v ∂u∂v
∂2x
∂F ∂x ∂F ∂y ∂F ∂z ∂x
1 1 1
− + + −F1 dudv
∂x ∂v ∂y ∂v ∂z ∂v ∂u ∂v∂u
Z Z
∂F1 ∂y ∂F1 ∂z ∂x ∂F1 ∂y ∂F1 ∂z ∂x o
= + − + du dv
R ∂y ∂u ∂z ∂u ∂v ∂y ∂v ∂z ∂v ∂u
ZZ
= ∇ × F · n̂ dS
S
which is the conclusion that we wanted.

a Otherwise, decompose S into simpler pieces, analogously to what we did in the proof of
the divergence theorem.
Before we move on to some examples, here are a couple of remarks.
H RR
• Stokes’ theorem says that C F · dr = S ∇ × F · n̂ dS for any (suitably
oriented) surface whose boundary is C. So if S1 and S2 are two different
(suitably oriented) surfaces having the same boundary curve C, then
ZZ ZZ
∇ × F · n̂ dS = ∇ × F · n̂ dS
S1 S2
For example, if C is the unit circle
C = (x, y, z) x2 + y 2 = 1, z = 0

oriented counterclockwise when viewed from above, then both
S1 = (x, y, z) x2 + y 2 ≤ 1, z = 0

S2 = (x, y, z) z ≥ 0, x2 + y 2 + z 2 = 1

pointing unit normalRRvectors, have boundary C. So Stokes’

with upward RR
tells us that S1 ∇ × F · n̂ dS = S2 ∇ × F · n̂ dS.
n̂
S2
n̂
S1
C
RR RR
It should not be a surprise that S1 ∇ × F · n̂ dS = S2 ∇ × F · n̂ dS,
for the following reason. Let
V = (x, y, z) x2 + y 2 + z 2 ≤ 1, z ≥ 0

be the solid between S1 and S2 . The boundary ∂V of V is the union of

S1 and S2 .
N̂
S2
V
S1
N̂
But beware that the outward pointing normal to ∂V (call it N̂) is +n̂
on S2 and −n̂ on S1 . So the divergence theorem gives
ZZ ZZ
∇ × F · n̂ dS − ∇ × F · n̂ dS
S2 S1
ZZ ZZ
= ∇ × F · N̂ dS + ∇ × F · N̂ dS
S2 S1
ZZ
= ∇ × F · N̂ dS
Z Z∂V
Z

= ∇ · ∇ × F dV
V
by the divergence theorem
=0
by the vector identity Theorem 4.1.7.a.

• As a second remark, suppose that the vectorH field F obeys ∇ × F = 0
everywhere. Then Stokes’ theorem forces C F · dr = 0 are around all
closed curves C, which implies that F is conservative, by Theorem 2.4.7.
So Stokes’ theorem provides another proof of Theorem 2.4.8.
Here is an easy example which shows that Stokes’ can be very useful when
∇ × F simplifies.

Example 4.4.2 Evaluate C F · dr where F = 2z + sin x146 ı̂ı − 5z ̂ − 5y k̂
H
and the curve C is the circle x2 + y 2 = 4, z = 1, oriented counterclockwise

when viewed from above.
Solution. The x146 in F will probably make a direct evaluation of the integral
difficult. So we’ll use Stokes’ theorem. To do so we need a surface S with
∂S = C. The simplest is just the flat disk
S = (x, y, z) x2 + y 2 ≤ 4, z = 1

z
n̂
S
C
y
x
Since
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
2z + sin x146 −5z −5y
" #
∂ ∂ ∂ ∂

= ı̂ı det ∂y ∂z
− ̂ det ∂x ∂z
−5z −5y 2z + sin x146 −5y
" #
∂ ∂
+ k̂ det ∂x ∂y
146
2z + sin x −5z
= 2̂
and the normal to S is k̂, Stokes’ theorem gives

I ZZ ZZ
F · dr = ∇ × F · n̂ dS = (2̂) · k̂ dS = 0
C S S

Now we’ll repeat the last example with a harder curve.

Example 4.4.3 Evaluate C F · dr where F = 2z + sin x146 ı̂ı − 5z ̂ − 5y k̂
H
and the curve C is the intersection of x2 + y 2 + z 2 = 4 and z = y, oriented

counterclockwise when viewed from above.
Solution. The surface x2 + y 2 + z 2 = 4 is the sphere of radius 2 centred
on the origin and z = y is a plane which contains the origin. So C, being
the intersection of a sphere with a plane through the centre of the sphere, is
a circle, with centre (0, 0, 0) and radius 2. The part of the circle in the first
octant is sketched on the left below.
z z
n̂
C y S C y
x x
146
The x in F will probably make a direct evaluation of the integral difficult.
So we’ll use Stokes’ theorem. To do so we need a surface S with ∂S = C. The
simplest is the flat disk
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y

The first octant of S is shown in the figure on the right above. We saw in the
last Example 4.4.2 that
∇ × F = 2̂
So Stokes’ theorem gives
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = 2 ̂ · n̂ dS
C S S
RR
We’ll evaluate the integral 2 S ̂ · n̂ dS in two ways. The first way is more
efficient, but also requires more insight. Since ∇ (z −y) = k̂−̂, the upward unit
normal to the plane z − y = 0, and hence to S, is n̂ = √1 (k̂ −̂

). Consequently
2
the integrand
−̂ + k̂ 1
̂ · n̂ = ̂ · √ = −√
2 2
is a constant and we do not need a formula for n̂ dS:
I ZZ √ ZZ √ √
F · dr = 2 ̂ · n̂ dS = − 2 dS = − 2Area(S) = − 2π 22
C
√S S
= −4 2π
RR
Alternatively, we can evaluate the integral S ̂ · n̂ dS using our normal
protocol. As S is part of the plane z = f (x, y) = y,

n̂ dS = ± − fx ı̂ı − fy ̂ + k̂ dxdy = ±(−̂ + k̂) dxdy
To get the upward pointing normal pointing normal, we take the + sign so
that n̂ dS = (−̂ + k̂) dxdy. As (x, y, z) runs over
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y

= (x, y, z) x2 + 2y 2 ≤ 4, z = y

2 2
= (x, y, z) x4 + y2 ≤ 1, z = y

2 2
(x, y) runs over the elliptical disk R = (x, y) x4 + y2 ≤ 1 . The part of

this ellipse in the first octant is the shaded region in the figure below.
z
n̂
y
x
√ √
This ellipse has semiaxes a = 2 and b = 2 and hence area πab = 2 2π.
So
I ZZ ZZ ZZ
F · dr = 2 ̂ · n̂ dS = 2 ̂ · (−̂ + k̂) dxdy = −2 dxdy
C S R R
= −2Area(R)
√
= −4 2π

H 2
Example 4.4.4 Evaluate C F·dr where F = (x+y)ı̂ı +2(x−z) ̂ +(y +z) k̂ and
C is the oriented curve obtained by going from (2, 0, 0) to (0, 3, 0) to (0, 0, 6)
and back to (2, 0, 0) along straight line segments.
C2
C3
y
C1
x
Solution 1. In this first solution, we’ll evaluate the integral directly. The
first line segment (C1 in the figure above) may be parametrized as

r(t) = (2, 0, 0) + t (0, 3, 0) − (2, 0, 0) = 2 − 2t , 3t , 0 0≤t≤1
So the integral along this segment is
Z 1 Z 1
dr
F(r(t)) · dt = (2 + t , 2(2 − 2t) , (3t)2 ) · (−2 , 3 , 0) dt
0 dt 0
Z 1
= (8 − 14t) dt
0
h i1
= 8t − 7t2 =1
0
The second line segment (C2 in the figure above) may be parametrized as

r(t) = (0, 3, 0) + t (0, 0, 6) − (0, 3, 0) = 0 , 3 − 3t , 6t 0 ≤ t ≤ 1.
So the integral along this segment is
Z 1 Z 1
dr
3(1 − t) , −12t , 9(1 − t)2 + 6t · (0, −3, 6) dt

F(r(t)) · dt =
0 dt 0
Z 1
= [36t + 54(1 − t)2 + 36t] dt
0
h i1
= 18t2 − 18(1 − t)3 + 18t2
0
= 54
The final line segment (C3 in the figure above) may be parametrized as

r(t) = (0, 0, 6) + t (2, 0, 0) − (0, 0, 6) = (2t , 0 , 6 − 6t) 0≤t≤1
So the line integral along this segment is
Z 1 Z 1
dr
F(r(t)) · dt = 2t , 4t − 12(1 − t) , 6(1 − t) · (2, 0, −6) dt
0 dt 0
Z 1 h i1
= [4t − 36(1 − t)] dt = 2t2 + 18(1 − t)2 = −16
0 0
The full line integral is

I
F · dr = 1 + 54 − 16 = 39
C
Solution 2. This time we shall apply Stokes’ Theorem. The curl of F is

 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
x + y 2(x − z) y 2 + z
= (2y + 2)ı̂ı − (0 − 0)̂ + (2 − 1)k̂

= 2(y + 1)ı̂ı + k̂
The curve C is a triangle and so is contained in a plane. Any plane has an

equation of the form Ax + By + Cz = D. Our plane does not pass through the
origin (look at the figure above) so the D must be nonzero. Consequently we
may divide Ax + By + Cz = D through by D giving an equation of the form
ax + by + cz = 1.
• Because (2, 0, 0) lies on the plane, a = 21 .
• Because (0, 3, 0) lies on the plane, b = 31 .
• Because (0, 0, 6) lies on the plane, c = 16 .
So the triangle is contained in the plane x2 + y3 + z6 = 1. It is the boundary of
the surface S that consists of the portion of the plane x2 + y3 + z6 = 1 that obeys
x ≥ 0, y ≥ 0 and z ≥ 0. Rewrite the equation of the plane as z = 6 − 3x − 2y.
For this surface
n̂ dS = (3ı̂ı + 2̂ + k̂) dx dy
by 3.3.2, and we can write

S = (x, y, z) x ≥ 0, y ≥ 0, z ≥ 0, z = 6 − 3x − 2y

= (x, y, z) x ≥ 0, y ≥ 0, 6 − 3x − 2y ≥ 0, z = 6 − 3x − 2y
As (x, y, z) runs over S, (x, y) runs over the triangle

R = (x, y, z) x ≥ 0, y ≥ 0, 3x + 2y ≤ 6

= (x, y, z) x ≥ 0, 0 ≤ y ≤ 23 (2 − x)

z “ 6 ´ 3x ´ 2y
S
y
R
3x ` 2y “ 6
x z“0
Using horizontal strips as in the figure on the left below,
I ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
= [2(y + 1)ı̂ı + k̂] · [3ı̂ı + 2̂ + k̂] dx dy
Z ZR
= [6y + 7] dx dy
R
1
Z 3 Z 3 (6−2y)
= dy dx [6y + 7]
0 0
Z 3
1
= dy [6y + 7][6 − 2y]
0 3
Z 3
1
= dy [−12y 2 + 22y + 42]
3 0
1h i3
= − 4y 3 + 11y 2 + 42y
3 0
= − 4 × 9 + 11 × 3 + 42 = 39
y y
3 3
x “ 13 p6 ´ 2yq y “ 32 p2 ´ xq
2 x 2 x
Alternatively, using vertical strips as in the figure on the right above,
I ZZ
F · dr = [6y + 7] dx dy
C R
3
Z 2 Z 2 (2−x)
= dx dy [6y + 7]
0 0
Z 2
h 32 3 i
= dx 3 2 (2 − x)2 + 7 (2 − x)
0 2 2
h 27 1 21 1 i2
= − (2 − x)3 − (2 − x)2
4 3 2 2 0
9 21
= 8+ 4 = 39
4 4

H
Example 4.4.5 Evaluate C F·dr where F = (cos x+y+z)ı̂ı +(x+z) ̂ +(x+y) k̂
and C is the intersection of the surfaces
y2 z2
x2 + + =1 and z = x2 + 2y 2
2 3
oriented counterclockwise when viewed from above.
2 2
Solution. First, let’s sketch the curve. x2 + y2 + z3 = 1 is an ellipsoid centred
on the origin and z = x2 + 2y 2 is an upward opening paraboloid that passes
through the origin. They are sketched in the figure below. The paraboloid is
red.
z “ x2 ` 2y 2
y2 z2
x2 ` 2
` 3
“1
Their intersection, the curve C, is the blue curve in the figure. It looks like
a deformed2 circle.
One could imagine parametrizing C. For example, substituting x2 = z−2y 2
2
into the equation of the ellipsoid gives − 32 y 2 + 13 (z+ 32 = 74 . This can be solved
to give y as a function of z and then x2 = z − 2y 2 also gives x as a function of
z. However this would clearly yield, at best, a really messy integral. So let’s
try Stokes’ theorem.
In fact, since
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det 

∂x ∂y ∂z 
cos x + y + z x + z x + y

= ı̂ı 1 − 1 − ̂ 1 − 1 + k̂ 1 − 1
=0

This FHis conservative! (In fact F = ∇ sin x + xy + xz + yz .) As C is a closed
curve, C F · dr = 0.
RR
Example 4.4.6 Evaluate S G·n̂ dS where G = (2x)ı̂ı +(2z−2x) ̂ +(2x−2z) k̂
and
S = (x, y, z) z = 1 − x2 − y 2 (1 − y 3 ) cos x ey , x2 + y 2 ≤ 1

with upward pointing normal

Solution 1. The surface S is sketched below. It is a pretty weird surface.
About the
z “ p1 ´ x2 ´ y 2qp1 ´ y 3qey cos x
x2 ` y 2 “ 1, z “ 0
only simple thing about it is that its boundary, ∂S, is the circle x2 +y 2 = 1,
z = 0. It is clear that we should not try to evaluate the integral directly3 . In
this solution we will combine the divergence theorem with the observation that
∂ ∂ ∂
∇·G= (2x) + (2z − 2x) + (2x − 2z) = 0
∂x ∂y ∂z
to avoid ever having work with the surface S. Here is an outline of what we
will do.
• We first select a simple surface S 0 whose boundary ∂S 0 is also the circle
x2 + y 2 = 1, z = 0. A nice simple choice of S 0 , and the surface that we
will use, is the disk
S 0 = (x, y, z) x2 + y 2 = 1, z = 0

• Then we define V to be the solid whose top surface is S and whose bottom
surface is S 0 . So the boundary of V is the union of S and S 0 .
3 By Salvador Dali?
n̂ V
n̂
S1
• For S 0 , we will use the upward pointing normal n̂ = k̂, which is minus
the outward pointing normal to ∂V on S 0 . So the divergence theorem
says that
ZZZ ZZ ZZ
∇ · G dV = G · n̂ dS − G · n̂ dS
V S S0
The left hand side is zero because, as we have already seen, ∇ · G = 0.

So ZZ ZZ
G · n̂ dS = G · n̂ dS
S S0
RR
• Finally, we compute S0
G · n̂ dS.
We saw an argument like this (with G = ∇ × F) in the first remark following
the proof of Theorem 4.4.1.
So all that we have to do now is compute
ZZ ZZ ZZ ZZ
G · n̂ dS = G · n̂ dS = G · k̂ dS = 2 2
(2x − 2z) dxdy
x +y ≤1
S S0 S0 z=0
ZZ
= (2x) dxdy
x2 +y 2 ≤1
z=0
=0
simply because the integrand is odd under x → −x.

Solution 2. In this second solution we’ll use Stokes’ theorem instead of the
divergence theorem. To do so, we have to express G in the form ∇ × F. So
the first thing to do is to check if G passes the screening test, Theorem 4.1.12,
for the existence of vector potentials. That is, to check if ∇ · G = 0. It is. We
saw this in Solution 1 above.
Next, we have to find a vector potential. In fact, we have already found, in
Example 4.1.15, that
F = (z 2 − 2xz)ı̂ı + (x2 − 2xz)̂
is a vector potential for G, which we can quickly check.

Parametrizing C by r(t) = cos tı̂ı + sin t ̂, 0 ≤ t ≤ 2π, Stokes’ theorem gives

(recalling that z = 0 on C so that F r(t) = x2 ̂ = cos2 t)

x=cos t
ZZ ZZ I Z 2π dr
G · n̂ dS = ∇ × F · n̂ dS = F · dr = F r(t) · dt
S S C 0 dt
Z 2π
cos2 t (cos t) dt

=
0
Of course this integral can be evaluated by using that one antiderivative of the
integrand cos3 t = 1 − sin2 t cos t is sin t − 31 sin3 t and that this antiderivative
is zero at t = 0 and at t = 2π. But it is easier to observe that the integral of
any odd power of sin t or cos t over any full period is zero. Look, for example,
at the graphs of sin3 x and cos3 x, below.
y y
y “ sin3 x y “ cos3 x
1 1
x x
π{2 π 3π{2 2π π{2 π 3π{2 2π
´1 ´1
Either way
ZZ
G · n̂ dS = 0
S

H
Example 4.4.7 In this example we compute, in three different ways, C
F · dr
where
F = (z − y)ı̂ı − (x + z) ̂ − (x + y) k̂
and C is the curve x2 + y 2 + z 2 = 4, z = y oriented counterclockwise when
viewed from above.
Solution 1. Direct Computation: H

In this first computation, we parametrize the curve C and compute C F·dr
directly. The plane z = y passes through the origin, which is the centre of the
sphere x2 + y 2 + z 2 = 4. So C is a circle which, like the sphere, has radius 2
and centre (0, 0, 0). We use a parametrization of the form
r(t) = c + ρ cos tı̂ı0 + ρ sin t ̂0 0 ≤ t ≤ 2π
where
• c = (0, 0, 0) is the centre of C,
• ρ = 2 is the radius of C and
• ı̂ı0 and ̂0 are two vectors that
a are unit vectors,
b are parallel to the plane z = y and
4 That way lies pain.
c are mutually perpendicular.

z
k̂1
̂1
y
ı̂ı1
p2, 0, 0q
x
The trickiest part is finding suitable vectors ı̂ı0 and ̂0 :
• The point (2, 0, 0) satisfies both x2 + y 2 + z 2 = 4 and z = y and so is
on C. We may choose ı̂ı0 to be the unit vector in the direction from the
centre (0, 0, 0) of the circle towards (2, 0, 0). Namely ı̂ı0 = (1, 0, 0).
• Since the plane of the circle is z − y = 0, the vector ∇ (z − y) = (0, −1, 1)

is perpendicular to the plane of C. So k̂0 = √12 (0, −1, 1) is a unit vector
normal to z = y. Then ̂0 = k̂0 ×ı̂ı0 = √1 (0, −1, 1) × (1, 0, 0)
2
= √1 (0, 1, 1)
2
0 0 0
is a unit vector that is perpendicular to ı̂ı and k̂ . Since ̂ is perpendicular
to k̂0 , it is parallel to z = y.
Substituting in c = (0, 0, 0), ρ = 2, ı̂ı0 = (1, 0, 0) and ̂0 = √1 (0, 1, 1)
2
gives
1 sin t sin t
r(t) = 2 cos t (1, 0, 0) + 2 sin t √ (0, 1, 1) = 2 cos t, √ , √
2 2 2
0 ≤ t ≤ 2π
√
To check
√ that this parametrization is correct, note that x = 2 cos t, y = 2 sin t,
z = 2 sin t satisfies both x2 + y 2 + z 2 = 4 and z = √ y.
At t = 0, r(0) = (2, 0, 0). As t increases,
√ √ z(t) = 2 sin t increases and r(t)
moves upwards towards r π2 = (0, 2, 2). This is the desired counterclock-
wise direction (when viewed from above). Now that we have a parametrization,
we can set up the integral.
√ √
r(t) = 2 cos t, 2 sin t, 2 sin t
√ √
r 0 (t) = − 2 sin t, 2 cos t, 2 cos t

F r(t) = z(t) − y(t), −x(t) − z(t), −x(t) − y(t)
√ √ √ √
= 2 sin t − 2 sin t, −2 cos t − 2 sin t, −2 cos t − 2 sin t
√ √
= − 0, 2 cos t + 2 sin t, 2 cos t + 2 sin t
√
F r(t) · r 0 (t) = − 4 2 cos2 t + 4 cos t sin t

√ √
= − 2 2 cos(2t) + 2 2 + 2 sin(2t)
by the double angle formulae sin(2t) = 2 sin t cos t and cos(2t) = 2 cos2 t − 1.
So
I Z 2π
F r(t) · r 0 (t) dt

F · dr =
C 0
2π √ √
Z

= − 2 2 cos(2t) + 2 2 + 2 sin(2t) dt
0
h√ √ i2π
= − 2 sin(2t) + 2 2t − cos(2t)
0
√
= −4 2π
Oof! Let’s do it an easier way.

Solution 2. Stokes’ Theorem:
To apply Stokes’ theorem we need to express C as the boundary ∂S of a
surface S. As
C = (x, y, z) x2 + y 2 + z 2 = 4, z = y

is a closed curve, this is possible. In fact there are many possible choices of S
with ∂S = C. Three possible S’s (sketched below) are
S = (x, y, z) x2 + y 2 + z 2 ≤ 4, z = y

S 0 = (x, y, z) x2 + y 2 + z 2 = 4, z ≥ y

S 00 = (x, y, z) x2 + y 2 + z 2 = 4, z ≤ y

S1
S
S2
The first of these, which is part of a plane, is likely to lead to simpler
computations than the last two, which are parts of a sphere. So we choose
what looks like the simpler way.
In preparation for application of Stokes’ theorem, we compute ∇ × F and
n̂ dS. For the latter, we apply the formula n̂ dS = ±(−fx , −fy , 1) dxdy (of
Equation 3.3.2) to the surface z = f (x, y) = y. We use the + sign to give the
normal a positive k̂ component.
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
z − y −x − z −x − y

= ı̂ı − 1 − (−1) − ̂ − 1 − 1 + k̂ − 1 − (−1)
= 2 ̂
n̂ dS = (0, −1, 1) dxdy
∇ × F · n̂ dS = (0, 2, 0) · (0, −1, 1) dxdy = −2 dxdy
The integration variables are x and y and, by definition, the domain of inte-
gration is
R = (x, y) (x, y, z) is in S for some z
To determine precisely what this domain of integration is, we observe that since
z = y on S, x2 + y 2 + z 2 ≤ 4 is the same as x2 + 2y 2 ≤ 4 on S,
S = (x, y, z) x2 + 2y 2 ≤ 4, z = y =⇒ R = (x, y) x2 + 2y 2 ≤ 4

So the domain
√ of integration is√an ellipse with semimajor axis a = 2, semiminor
axis b = 2 and area πab = 2 2π. The integral is then
I ZZ ZZ √
F · dr = ∇ × F · n̂ dS = (−2) dxdy = −2 Area (R) = −4 2π
C S R
Remark (Limits of integration):

If the integrand were more complicated, we would have to evaluate the
integral over R by expressing it as an iterated integrals with the correct limits
of integration. First suppose that we slice up R using thin vertical
p slices. On
each
p such slice, x is essentially constant and y runs from − (4 − x2 )/2 to
(4 − x2 )/2. The leftmost such slice would have x = −2 and the rightmost
such slice would have x = 2. So the correct limits with this slicing are
ZZ Z 2 Z √ 2 (4−x )/2
f (x, y) dxdy = dx √ dy f (x, y)
R −2 − (4−x2 )/2
y
x2 ` 2y 2 “ 4
If, instead, we slice up R using thin horizontalp slices, then,pon each such
slice, y is essentially constant and x runs from − 4 − 2y 2 to 4 − 2y 2 . The
√
bottom
√ such slice would have y = − 2 and the top such slice would have
y = 2. So the correct limits with this slicing are
ZZ Z √ 2Z √ 2 4−2y
f (x, y) dxdy = √ dy √ dx f (x, y)
R − 2 − 4−2y 2
y
x2 ` 2y 2 “ 4
Note that the integral with limits

y
√
Z 2 Z 2
x √ dy dx f (x, y)
− 2 −2
corresponds to a slicing with x running from −2 to 2 on {\bf every} slice.

This corresponds to a rectangular domain of integration, not what we have
here.
Stokes’ Theorem, Again:
Since the integrand is just a constant (after Stoking — not the original
RR and S is so simple (because we chose it wisely), we can evaluate the
integrand)
integral S ∇ × F · n̂ dS without ever determining dS explicitly and without
ever setting up any limits of integration. We already know that ∇ × F = 2 ̂.
Since S is the level surface z − y = 0, the gradient ∇ (z − y) = −̂ + k̂ is normal
to S. So n̂ = √1 (−̂
 + k̂) and
2
I ZZ ZZ
1
F · dr = ∇ × F · n̂ dS = (2̂) · √ (−̂ + k̂) dS
C 2
Z ZS √ √
S
= − 2 dS = − 2 Area (S)
S
H √
As S is a circle of radius 2, C
F · dr = −4 2π, yet again.
Example 4.4.8 In Warning 4.1.17, we stated that if a vector field fails to
pass the screening test ∇ · B = 0 at even a single point, for example because
the vector field is not defined at that point, then B can fail to have a vector
potential. An example is the point source
r̂(x, y, z)
B(x, y, z) =
r(x, y, z)2
of Example 3.4.2. Here, as usual,

p xı̂ı + y̂ + z k̂
r(x, y, z) = x2 + y 2 + z 2 r̂(x, y, z) = p
x2 + y 2 + z 2
This vector field is defined on all of R3 , except for the origin, and its divergence

∂ x ∂ y
∇·B= +
∂x (x2 + y 2 + z 2 )3/2 ∂y (x2 + y 2 + z 2 )3/2

∂ z
+
∂z (x2 + y 2 + z 2 )3/2
3x2

1
= − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
3y 2

1
+ − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
3z 2

1
+ − 2
(x2 + y 2 + z 2 )3/2 (x + y 2 + z 2 )5/2
2 2 2
3 3(x + y + z )
= 2 2 2 3/2
− 2
(x + y + z ) (x + y 2 + z 2 )5/2
is zero everywhere except at the origin, where it is not defined.

This vector field cannot have a vector
potential on its domain of definition,
i.e. on R3 \ {(0, 0, 0)} = (x, y, z) (x, y, z) 6= (0, 0, 0) . To see this, suppose
to the contrary that it did have a vector potential A. Then its flux through
any closed surface4 (i.e. surface without a boundary) S would be
ZZ ZZ I
B · n̂ dS = ∇ × A · n̂ dS = A · dr = 0
S S ∂S
by Stokes’ theorem, since ∂S is empty. But we found in Example 3.4.2, with

m = 1, that the flux of B through any sphere centred on the origin is 4π.
5 If you are uncomfortable with the surface not having a boundary, poke a very small hole
in the surface, giving it a very small boundary. Then take the limit as the hole tends to zero.
4.4.1 The Interpretation of Div and Curl Revisited

In sections 4.1.4 and 4.1.5 we derived interpretations of the divergence and of
the curl. Now that we have the divergence theorem and Stokes’ theorem, we
can simplify those derivations a lot.
4.4.1.1 Divergence
Let ε > 0 be a tiny positive number, and then let
Bε (x0 , y0 , z0 ) = (x, y, z) (x − x0 )2 + (y − y0 )2 + (z − z0 )2 < ε2

be a tiny ball of radius ε centred on the point (x0 , y0 , z0 ). Denote by
Sε (x0 , y0 , z0 ) = (x, y, z) (x − x0 )2 + (y − y0 )2 + (z − z0 )2 = ε2

its surface. Because Bε (x0 , y0 , z0 ) is really small, ∇ · v is essentially constant

in Bε (x0 , y0 , z0 ) and we essentially have
ZZZ

∇ · v dV = ∇ · v(x0 , y0 , z0 ) Vol Bε (x0 , y0 , z0 )
Bε (x0 ,y0 ,z0 )
Of course we are really making an approximation here, based on the assump-

tion that v(x, y, z) is continuous and so takes values very close to v(x0 , y0 , z0 )
everywhere on the domain of integration. The approximation gets better and
better as ε → 0 and a more precise statement is
RRR
Bε (x0 ,y0 ,z0 )
∇ · v dV
∇ · v(x0 , y0 , z0 ) = lim
ε→0 Vol Bε (x0 , y0 , z0 )
By the divergence theorem, we also have

ZZZ ZZ
∇ · v dV = v · n̂ dS
Bε (x0 ,y0 ,z0 ) Sε (x0 ,y0 ,z0 )
Think of the vector field v as the velocity of a moving fluid which has density
one. We have already seen, in §3.4, that the flux integral for a velocity field
has the interpretation
ZZ
the volume of fluid leaving Bε (x0 , y0 , z0 )
v · n̂ dS =
Sε (x0 ,y0 ,z0 ) through Sε (x0 , y0 , z0 ) per unit time
We conclude that, as we said in 4.1.19,
the rate at which fluid is exiting Bε (x0 , y0 , z0 )

∇ · v(x0 , y0 , z0 ) = lim
ε→0 Vol Bε (x0 , y0 , z0 )

 rate at which fluid is exiting an
= infinitesimal sphere centred at (x0 , y0 , z0 ),

per unit time, per unit volume
= strength of the source at (x0 , y0 , z0 )
If our world is filled with an incompressible fluid, a fluid whose density is

constant and so never expands or compresses, we will have ∇ · v = 0.
4.4.1.2 Curl
Again let ε > 0 be a tiny positive number and let Dε (x0 , y0 , z0 ) be a tiny
flat circular disk of radius ε centred on the point (x0 , y0 , z0 ) and denote by
Cε (x0 , y0 , z0 ) its boundary circle. Let n̂ be a unit normal vector to Dε . It
tells us the orientation of Dε . Give the circle Cε the corresponding orientation
using the right hand rule. That is, if the fingers of your right hand are pointing
in the corresponding direction of motion along Cε and your palm is facing Dε ,
then your thumb is pointing in the direction n̂.
n̂
ε
Cε
Because Dε (x0 , y0 , z0 ) is really small, ∇ × v is essentially constant on
Dε (x0 , y0 , z0 ) and we essentially have
ZZ

∇ × v · n̂ dS = ∇ × v(x0 , y0 , z0 ) · n̂ Area Dε (x0 , y0 , z0 )
Dε (x0 ,y0 ,z0 )
= πε2 ∇ ×v(x0 , y0 , z0 ) · n̂
Again, this is really an approximate statement which gets better and better as
ε → 0. A more precise statement is
RR
Dε (x0 ,y0 ,z0 )
∇ × v · n̂ dS
∇ ×v(x0 , y0 , z0 ) · n̂ = lim
ε→0 πε2
By Stokes’ theorem, we also have
ZZ I
∇ × v · n̂ dS = v · dr
Dε (x0 ,y0 ,z0 ) Cε (x0 ,y0 ,z0 )
Again,
H think of the vector field v as the velocity of a moving fluid. Then
Cε
v · dr is called the circulation of v around Cε .
To measure the circulation experimentally, place a small paddle wheel in
the fluid, with the axle of the paddle wheel pointing along n̂ and each of the
paddles perpendicular to Cε and centred on Cε .
n̂
t̂
ε
Cε
Each paddle moves tangentially to Cε . It would like to move with the same
speed as the tangential speed v· t̂ (where t̂ is the forward pointing unit tangent
vector to Cε at the location of the paddle) of the fluid at its location. But all
paddles are rigidly fixed to the axle of the paddle wheel and so must all move
with the same speed. That common speed will be the average value of v · t̂
around Cε . If ds represents an element of arc length of Cε , the average value
of v · t̂ around Cε is
I I
1 1
vT = v · t̂ ds = v · dr
2πε Cε 2πε Cε
since dr has direction t̂ and length ds so that dr = t̂ds, and since 2πε is the
circumference of Cε . If the paddle wheel rotates at Ω radians per unit time,
each paddle travels a distance Ωε per unit time (remember that ε is the radius
of Cε ). That is, vT = Ωε. Combining all this information,
RR
Dε (x0 ,y0 ,z0 )
∇ × v · n̂ dS
∇ ×v(x0 , y0 , z0 ) · n̂ = lim
ε→0
H πε2
v · dr
= lim Cε 2
ε→0 πε
2πε vT
= lim
ε→0 πε2
2πε (Ωε)
= lim
ε→0 πε2
= 2Ω
so that
Ω = 21 ∇ ×v(x0 , y0 , z0 ) · n̂
The component of ∇ × v(x0 , y0 , z0 ) in any direction n̂ is twice the rate at
which the paddle wheel turns when it is put into the fluid at (x0 , y0 , z0 ) with
its axle pointing in the direction n̂. The direction of ∇ × v(x0 , y0 , z0 ) is the
axle direction which gives maximum rate of rotation and the magnitude of
∇ × v(x0 , y0 , z0 ) is twice that maximum rate of rotation. For this reason,
∇ × v is called the “vorticity”.
4.4.2 Optional — An Application of Stokes’ Theorem —

Faraday’s Law
Magnetic induction refers to a physical process whereby an electric voltage is
created (“induced”) by a time varying magnetic field. This process is exploited
in many applications, including electric generators, induction motors, induc-
tion cooking, induction welding and inductive charging. Michael Faraday5 is
generally credited with the discovery of magnetic induction. Faraday’s law is
the following. Let S be an oriented surface with boundary C. Let E and B be
the (time dependent) electric and magnetic fields and define
B
I
E · dr = voltage around C
C
ZZ
B · n̂ dS = magnetic flux through S S
S
C
Then the voltage around C is the negative of the rate of change of the
magnetic flux through S. As an equation, Faraday’s Law is
I ZZ
∂
E · dr = − B · n̂ dS
C ∂t S
We can reformulate this as a partial differential equation. By Stokes’ Theorem,

I ZZ
E · dr = ∇ × E) · n̂ dS
(∇
C S
5 Michael Faraday (1791–1867) was an English physicist and chemist. He ended up being
an extremely influential scientist despite having only the most basic of formal educations.
so Faraday’s law becomes

ZZ
∂B
∇×E+ · n̂ dS = 0
S ∂t
This is true for all surfaces S. So the integrand, assuming that it is continuous,
must be zero.
To see this, let G = ∇ × E + ∂B ∂t . Suppose that G(x0 ) 6= 0. Pick a unit
vector n̂ in the direction of G(x0 ). Let S be a very small flat disk centered
on x0 with normal n̂ (the vector we picked). Then G(x0 ) · n̂ > 0 and, by
continuity, G(x) ·
n̂ > 0 for all x on S, if we have picked S small enough. Then
RR ∂B
S
∇ × E + ∂t · n̂ dS > 0, which is a contradiction. So G = 0 everywhere
and we conclude that
∂B
∇×E+ =0
∂t
This is one of Maxwell’s electromagnetic field equations6 .
4.4.3 Exercises
1. Each of the figures below contains a sketch H of a surface
RR S and its
boundary ∂S. Stokes’ theorem says that ∂S F · dr = S ∇ × F · n̂ dS
if n̂ is a correctly oriented unit normal vector to S. Add to each sketch
a typical such normal vector.
(a) (b) (c)
S ∂S
S ∂S
∂S S
2. Let
• R be a finite region in the xy-plane,
• the boundary, C, of R consist of a single piecewise smooth, sim-
ple closed curve
◦ that is oriented (i.e. an arrow is put on C) consistently with
R in the sense that if you walk along C in the direction of
the arrow, then R is on your left
y
C
x
• F1 (x, y) and F2 (x, y) have continuous first partial derivatives at
every point of R.
Use Stokes’ theorem to show that
I ZZ
∂F2 ∂F1
F1 (x, y) dx + F2 (x, y) dy = − dxdy
C R ∂x ∂y
i.e. to show Green’s theorem.
6 For the others, see Example 4.1.2
H H
3. Verify the identity C φ∇ ∇ψ · dr = − C ψ∇∇φ · dr for any continuously
differentiable scalar fields φ and ψ and curve C that is the boundary
of a piecewise smooth surface.
4. Let C be the curve of intersection of the cylinder x2 + y 2 = 1 and the
surface z = y 2 oriented in the counterclockwise direction as seen from
(0, 0, 100). Let F = (x2 − y , y 2 + x , 1). Calculate C F · dr
H
a by direct evaluation
b by using Stokes’ Theorem.
Evaluate C F · dr where F = yex ı̂ı + (x + ex ) ̂ + z 2 k̂ and C is the
H
5.
curve
r(t) = (1 + cos t)ı̂ı + (1 + sin t) ̂ + (1 − sin t − cos t) k̂ 0 ≤ t ≤ 2π

RR
6. ∗. Find the value of S ∇ × F · n̂ dS where F = z − y , x , −x and
S is the hemisphere
(x, y, z) ∈ R3 x2 + y 2 + z 2 = 4, z ≥ 0

oriented so the surface normals point away from the centre of the
hemisphere.
2
7. ∗. Let S be the part of the surface z = 16 − (x2 + y 2 ) which lies
above the xy-plane. Let F be the vector field
F = x ln(1 + z)ı̂ı + x(3 + y) ̂ + y cos z k̂
Calculate ZZ
∇ × F · n̂ dS
S
where n̂ is the upward normal on S.
8. ∗. Let C be the intersection of the paraboloid z = 4 − x2 − y 2 with
the cylinder x2 + (y − 1)2 = 1, oriented counterclockwise when
H viewed
from high on the z-axis. Let F = xz ı̂ı + x ̂ + yz k̂. Find C F · dr.
9. Let F = −yez ı̂ı + x3 cos z ̂ + z sin(xy) k̂, and let S be the part of the
surface z = (1 − x2 )(1 − y 2 ) that lies above the square −1 ≤ x ≤ 1,
−1 ≤ y ≤ 1 in the xy-plane. Find the flux of ∇ × F upward through
S.
2
10. Evaluate the integral C F·dr, in which F = (ex −yz , sin y−yz , xz+
H
2y) and C is the triangular path from (1, 0, 0) to (0, 1, 0) to (0, 0, 1)

to (1, 0, 0).
11. ∗. Let F(x, y, z) = −z ı̂ı + x ̂ + y k̂ Hbe a vector field. Use Stokes’
theorem to evaluate the line integral C F · dr where C is the inter-
2 2 2
section of the plane z = y and the ellipsoid x4 + y2 + z2 = 1, oriented
counter-clockwise when viewed from high on the z-axis.
12. ∗. Consider the vector field F(x, y, z) = z 2 ı̂ı + x2 ̂ + y 2 k̂ in R3 .
R
a Compute the line integral I1 = C1 F · dr where C1 is the curve
consisting of three line segments, L1 from (2, 0, 0) to (0, 2, 0),
then L2 from (0, 2, 0) to (0, 0, 2), finally L3 from (0, 0, 2) to
(2, 0, 0).
b A simple closed curve C2 lies on the plane E : x + y + z = 2,

enclosing a region R on the plane of area 3, and oriented in a
counterclockwise direction as observed
R from the positive x-axis.
Compute the line integral I2 = C2 F · dr.
13. ∗. Let C = C1 + C2 + C3 be the curve given by the union of the three
parameterized curves

r1 (t) = 2 cos t, 2 sin t, 0 , 0 ≤ t ≤ π/2

r2 (t) = 0, 2 cos t, 2 sin t , 0 ≤ t ≤ π/2

r3 (t) = 2 sin t, 0, 2 cos t , 0 ≤ t ≤ π/2
a Draw a picture of C. Clearly mark each of the curves C1 , C2 , and

C3 and indicate the orientations given by the parameterizations.
b Find and parameterize an oriented surface S whose boundary is
C (with the given orientations).
R
c Compute the line integral C F · dr where
2

F = y + sin(x2 ) , z − 3x + ln(1 + y 2 ) , y + ez
p
14. ∗. We consider the cone with equation z = x2 + y 2 . Note that its
tip, or vertex, is located at the origin (0, 0, 0). The cone is oriented in
such a way that the normal vectors point downwards (and away from
the z axis). In the parts below, both
S1 and S2 are oriented this way.
Let F = − zy, zx, xy cos(yz) .
a Let S1 be the part of the cone that lies between the planes z = 0
and z = 4. Note that S1 does not include any part of the plane
z = 4. Use Stokes’ theorem to determine the value of
ZZ
∇ × F · n̂ dS
S1
Make a sketch indicating the orientations of S1 and of the con-

tour(s) of integration.
b Let S2 be the part of the cone that lies below the plane z = 4
and above z = 1. Note that S2 does not include any part of
the planes z = 1 and z = 4. Determine the flux of ∇ × F
across S2 . Justify your answer, including a sketch indicating
the orientations of S2 and of the contour(s) of integration.
15. ∗. Consider the curve C that is the intersection of the plane z = x + 4
and the cylinder x2 + y 2 = 4, and suppose C is oriented so that it is
traversed clockwise as seen from above.
Let F(x, y, z) = x3 + 2y , sin(y) + z , x + sin(z 2 ) .H
Use Stokes’ Theorem to evaluate the line integral C F · dr.
16. ∗.
2 2 2 3
a Consider the vector fieldH F(x, y, z) = (z , x , y ) in R . Com-
pute the line integral C F · dr, where C is the curve consisting
of the three line segments, L1 from (2, 0, 0) to (0, 2, 0), then L2
from (0, 2, 0) to (0, 0, 2), and finally L3 from (0, 0, 2) to (2, 0, 0).
b A simple closed curve C lies in the plane x + y + z = 2. The
surface this curve C surrounds inside the plane x + y + z = 2 has

area 3. The curve C is oriented in a counterclockwise direction
as
H observed from the positive x-axis. Compute the line integral
C
F · dr , where F is as in (a).
17. ∗. Evaluate the line integral
Z
1 x
z+ dx + xz dy + 3xy − dz
C 1+z (z + 1)2
where C is the curve parameterized by
r(t) = cos t , sin t , 1 − cos2 t sin t

0 ≤ t ≤ 2π
18. ∗. A simple closed curve C lies in the plane x + y + z = 1. The
surface this curve C surrounds inside the plane x + y + z = 1 has area
5. The curve C is oriented in a clockwise direction as observed from
the positive z-axis looking down at the plane x + y + z = 1.
Compute the line integral of F(x, y, z) = (z 2 , x2 , y 2 ) around C.
19. ∗. Let C be the oriented curve consisting of the 5 line segments
which form the paths from (0, 0, 0) to (0, 1, 1), from (0, 1, 1) to (0, 1, 2),
from (0, 1, 2) to (0, 2, 0), from (0, 2, 0) to (2, 2, 0), and from (2, 2, 0) to
(0, 0, 0). Let
√
F = (−y + ex sin x)ı̂ı + y 4 ̂ + z tan z k̂
R
Evaluate the integral C F · dr.
20. ∗. Suppose the curve C is the intersection of the cylinder x2 + y 2 =
1 with the surface z = xy 2 , traversed clockwise if viewed from the
positive z-axis, i.e. viewed “from above”. Evaluate the line integral
Z
(z + sin z) dx + (x3 − x2 y) dy + (x cos z − y) dz
C
RR
21. ∗. Evaluate S ∇ × F · n̂ dS where S is that part of the sphere
x2 + y 2 + z 2 = 2 above the plane z = 1, n̂ is the upward unit normal,
and
F(x, y, z) = −y 2 ı̂ı + x3 ̂ + ex + ey + z k̂

22. ∗. Let
F = x sin y ı̂ı − y sin x ̂ + (x − y)z 2 k̂
Use Stokes’ theorem to evaluate
Z
F · dr
C
along the path consisting of the straight line segments successively

joining the points P0 = (0, 0, 0) to P1 = (π/2, 0, 0) to P2 = (π/2, 0, 1)
to P3 = (0, 0, 1) to P4 = (0, π/2, 1) to P5 = (0, π/2, 0), and back to
(0, 0, 0).
23. ∗. Let

2z 3z
F= + sin(x2 ) , + sin(y 2 ) , 5(x + 1)(y + 2)
1+y 1+x
Let C be the oriented curve consisting of four line segments from

(0, 0, 0) to (2, 0, 0), from (2, 0, 0) to (0, 0, 2), from (0, 0, 2) to (0, 3, 0),
and from (0, 3, 0) to (0, 0, 0).

a Draw a picture of C. Clearly indicate the orientation on each
line segment.
R
b Compute the work integral C F · dr.
ZZ
24. ∗. Evaluate ∇ × F · n̂ dS where F = y ı̂ı + 2z ̂ + 3x k̂ and S is the
p S
surface z = 1 − x2 − y 2 , z ≥ 0 and n̂ is a unit normal to S obeying
n̂ · k̂ ≥ 0.
25. ∗. Let S be the curved surface below, oriented by the outward normal:
x2 + y 2 + 2(z − 1)2 = 6, z ≥ 0.
(E.g., at the high point of the surface, the unit normal is k̂.)
Define
2
+y 2 +z 2
G = ∇ × F, where F = (xz − y 3 cos z)ı̂ı + x3 ez ̂ + xyzex k̂.
RR
Find S G · n̂dS.
26. ∗. Let C be a circle of radius R lying in the plane x + y + z = 3. Use
Stokes’ Theorem to calculate the value of
I
F · dr
C
where F = z 2ı̂ı + x2̂ + y 2 k̂. (You may use either orientation of the
circle.)
27. Let S be the oriented surface consisting of the top and four sides
of the cube whose vertices are (±1, ±1, ±1), oriented outward. If
F(x, y, z) = (xyz, xy 2 , x2 yz), find the flux of ∇ × F through S.
28. Let S denote the part of the spiral ramp (that is helicoidal surface)
parametrized by
x = u cos v, y = u sin v, z = v 0 ≤ u ≤ 1, 0 ≤ v ≤ 2π
Let C denote the boundary of S with orientation specified by the

upward pointing normal on S. Find
Z
y dx − x dy + xy dz
C
29. Let C be the intersection of x + 2y − z = 7 and x2 − 2x + 4y 2 = 15.
The curve C is oriented counterclockwise when viewed from high on
the z-axis. Let
2
F = ex + yz ı̂ı + cos(y 2 ) − x2 ̂ + sin(z 2 ) + xy k̂

H
Evaluate C F · dr.
30. ∗.

a Find the curl of the vector field F = 2 + x2 + z , 0 , 3 + x2 z .
b Let C be the curve in R3 from the point (0, 0, 0) to the point
(2, 0, 0), consisting of three consecutive line segments connecting
the points (0, 0, 0) to (0, 0, 3), (0, 0, 3) to (0, 1, 0), and (0, 1, 0) to
(2, 0, 0). Evaluate the line integral
Z
F · dr
C
where F is the vector field from (a).

31. ∗.
a Let S be the bucket shaped surface consisting of the cylindrical
surface y 2 + z 2 = 9 between x = 0 and x = 5, and the disc inside
the yz-plane of radius 3 centered at the origin. (The bucket S
has a bottom, but no lid.) Orient S in such a way that the unit
normal points outward. Compute the flux of the vector field
∇ × G through S, where G = (x, −z, y).
b Compute the flux of the vector field F = (2 + z, xz 2 , x cos y)
through S, where S is as in (a).
32. ∗. Let
y 2 2

F(x, y, z) = + x1+x , x2 − y 1+y , cos5 (ln z)
x
a Write down the domain D of F.

b Circle the correct statement(s):
a D is connected.
b D is simply connected.
c D is disconnected.
c Compute ∇ × F.
d Let C be the square with corners (3 ± 1, 3 ± 1) in the plane

z = 2, oriented clockwise (viewed from above, i.e. down z-axis).
Compute Z
F · dr
C
e Is F conservative?
33. ∗. A physicist studies a vector field F(x, y, z). From experiments, it
is known that F is of the form
F(x, y, z) = xz ı̂ı + (axey z + byz) ̂ + (y 2 − xey z 2 ) k̂
for some real numbers a and b. It is further known that F = ∇ × G

for some differentiable vector field G.
a Determine a and b.
b Evaluate the surface integral
ZZ
F · n̂ dS
S
where S is the part of the ellipsoid x2 +y 2 + 14 z 2 = 1 for which z ≥

0, oriented so that its normal vector has a positive z-component.
34. ∗. Let C be the curve in the xy-plane from the point (0, 0) to the point
(5, 5) consisting of the ten line segments consecutively connecting the
points (0, 0), (0, 1), (1, 1), (1, 2), (2, 2), (2, 3), (3, 3), (3, 4), (4, 4), (4, 5),
(5, 5). Evaluate the line integral
Z
F · dr
C
where
F = y ı̂ı + (2x − 10) ̂

35. ∗. Let F = sin x , xz , z 2 . Evaluate C F · dr around the curve C
2
H
of intersection of the cylinder x2 + y 2 = 4 with the surface z = x2 ,

traversed counter clockwise as viewed from high on the z-axis.
36. ∗. Explain how one deduces the differential form
1 ∂H
∇×E=−
c ∂t
of Faraday’s law from its integral form
I ZZ
1 d
E · dr = − H · n̂ dS
C c dt S
37. ∗. Let C be the curve given by the parametric equations:
√
x = cos t, y = 2 sin t, z = cos t, 0 ≤ t ≤ 2π
and let
F = z ı̂ı + x ̂ + y 3 z 3 k̂
Use Stokes’ theorem to evaluate
I
F · dr
C
38. ∗. Use Stokes’ theorem to evaluate
I
z dx + x dy − y dz
C
where C is the closed curve which is the intersection of the plane

x + y + z = 1 with the sphere x2 + y 2 + z 2 = 1. Assume that C is
oriented clockwise as viewed from the origin.
39. ∗. Let S be the part of the half cone
p
z = x2 + y 2 , y ≥ 0,
that lies below the plane z = 1.

a Find a parametrization for S.
b Calculate the flux of the velocity field
v = xı̂ı + y ̂ − 2z k̂
downward through S.
c A vector field F has curl ∇ × F = xı̂ı + y ̂ − 2z k̂. On the xz-
plane, the vector field F is constant with F(x, 0, z) = ̂. Given
this information, calculate

Z
F · dr,
C
where C is the half circle
x2 + y 2 = 1, z = 1, y ≥ 0
oriented from (−1, 0, 1) to (1, 0, 1).

RR
40. Consider S (∇ ∇ × F) · n̂ dS where S is the portion of the sphere x2 +
2 2
y + z = 1 that obeys x + y + z ≥ 1, n̂ is the upward pointing normal
to the sphere and F = (y − z)ı̂ı +RR(z − x)̂ + (x − y)RR
k̂. Find another
surface S 0 with
RRthe property that S
∇
(∇ ×F)· n̂ dS = S0
∇ ×F)· n̂ dS
(∇
and evaluate S 0 (∇ ∇ × F) · n̂ dS.
4.5 Optional — Which Vector Fields Obey ∇ ×

F=0
We already know that if a vector field F passes the screening test ∇ × F = 0
on all of R2 or R3 , then there is a function ϕ with F = ∇ ϕ. That is, F is
conservative. We are now going to take a first look at what happens1 when
F passes the screening test ∇ × F = 0 only on some proper subset D of
Rn , n = 2 or 3. We will just scratch the surface of this topic — there is a
whole subbranch of Mathematics (cohomology theory, which is part of algebraic
topology) concerned with a general form of this question. We shall imagine
that we are given a vector field F that is only defined on D and we shall assume
• that D is a connected, open subset of Rn with n = 2 or n = 3 (see
Definition 4.5.1, below)
• that all first order derivatives of all vector fields and functions that we
consider are continuous and
• that all curves we consider are piecewise smooth. A curve is piecewise
smooth if it is a union of a finite number of smooth curves C1 , C2 , · · · , Cm
with the end point of Ci being the beginning point of Ci+1 for each 1 ≤ i <
m. A curve is smooth2 if it has a parametrization r(t), a ≤ t ≤ b, whose
first derivative r0 (t) exists, is continuous and is nonzero everywhere.
C1 C2
C3
Definition 4.5.1 Let n ≥ 1 be an integer.

a Let a ∈ Rn and ε > 0. The open ball of radius ε centred on a is
Bε (a) = x ∈ Rn |x − a| < ε

Note the strict inequality in |x − a| < ε.

1 Russell Crowe posed a related question in the movie A Beautiful Mind. The movie is
based on the life of the American mathematician John Nash, who won a Nobel Prize in
Economics.
2 The word “smooth” does not have a universal meaning in mathematics. It is used with
different meanings in different contexts. We are here using one of the standard definitions.
Another standard definition requires that all derivatives of all orders are continuous.
b A subset O ⊂ Rn is said to be an “open subset of Rn ” if, for each point

a ∈ O, there is an ε > 0 such that Bε (a) ⊂ O. Equivalently, O is open
if and only if it is a union of open balls.
c A subset D ⊂ Rn is said to be (pathwise) connected if every pair of points
in D can be joined by a piecewise smooth curve in D.
♦
Here are some examples to help explain this definition.
Example 4.5.2

a The open rectangle O = (x, y) ∈ R2 0 < x < 1, 0 < y < 1 is
an open subset of R2 because
any point a = (x0 , y0 ) ∈ O is a nonzero
distance, namely d = min x0 , 1 − x0 , y0 , 1 − y0 away from the boundary
of O. So the open ball Bd/2 (a) is contained in O.
b The closed rectangle C = (x, y) ∈ R2 0 ≤ x ≤ 1, 0 ≤ y ≤ 1 is not

an open subset of R2 . For example, 0 = (0, 0) is a point in C. No matter

what ε > 0 we pick, the open ball Bε (0) is not contained in C because
Bε (0) contains the point (− 2ε , 0), which is not in C.
` ˘
Bε px0 , 0q
C
O
c The x-axis, X = (x, y) ∈ R2 y = 0 , in R2 is not an open subset of

R2 because for any point (x0 , 0) ∈ X and any ε > 0, the ball Bε (x0 , 0)
contains points with nonzero y-coordinates and so is not contained in X .
d The union of open balls

B1 (0, 0) ∪ B1 (2, 0)
= (x, y) ∈ R2 x2 + y 2 < 1 or (x−2)2 + y 2 < 1

is not connected, since any continuous path from, for example, (2, 0) to
(0, 0) must leave the union. In the figure on the left below, an “empty
disk” has been sketched at (1, 0) just to emphasise that the point (1, 0)
is not in the union.
e On the other hand the union of “closed balls”
(x, y) ∈ R2 x2 + y 2 ≤ 1 or (x − 2)2 + y 2 ≤ 1

is connected. For example, the straight line segment from (2, 0) to (0, 0)
remains in the the union.
p0,0q p2,0q p0,0q p2,0q

Many, but not all, of the basic facts that we developed, in §2.4.1, about
conservative fields in Rn also applies (with the same proofs) to fields on D.
Theorem 4.5.3 For a vector field F on D ⊂ Rn ,
F is conservative on D ⇐⇒ F = ∇ ϕ on D, for some function ϕ

Z
⇐⇒ for each P0 , P1 ∈ D, the integral F · dr takes
C
the same value for all curves C from P0 to P1
I
⇐⇒ F · dr = 0 for all closed curves C in D
C
=⇒ ∇ × F = 0 on D
Note that the last line of this theorem contains only a one way implication.
Combining this with Stokes’ Theorem 4.4.1 (when n = 3, or Green’s The-
orem 4.3.2 when n = 2) gives us the following two consequences.
Theorem 4.5.4
a If D has the property that
every closed curve C in D is the boundary

(H)
of a bounded oriented surface, S, in D
then
F is conservative on D ⇐⇒ ∇ × F = 0 on D
b For any D, if ∇ ×F = 0 on D, then F is locally conservative. This means

that for each point x0 ∈ D, there is an ε > 0 and a function ϕ such that
F = ∇ ϕ on Bε (x0 ).
Proof. (a) This is simply because if ∇ × F = 0 on D and if the curve C = ∂S,
with S an oriented surface in D, then Stokes’ theorem gives
Z Z ZZ
F · dr = F · dr = ∇ × F · n̂ dS = 0
C ∂S S
So F is conservative by Theorem 4.5.3.

(b) This is true simply because Bε (x0 ) satisfies property (H).
Example 4.5.5 Here are some examples of D’s that violate (H).

• When D = D1 = (x, y) ∈ R2 0 < |(x, y)| < 3 (an open ball with its
centre removed), then the circle x2 + y 2 = 4 is a curve in D that is not
the boundary of a surface in D. The circle x2 + y 2 = 4 is the boundary
of the disk x2 + y 2 < 4, but the disk x2 + y 2 < 4 is not contained in D
because the point (0, 0) is in the disk and not in D. See the figure on the
left below.

• When D = D2 = (x, y, z) ∈ R3 |(x, y, z)| < 2, |(x, y)| > 0 (an open
ball with the z-axis removed), then the circle x2 + y 2 = 1, z = 0 is a
curve in D that is not the boundary of a surface in D. The circle is the
boundary of many different surfaces in R3 , but each contains a point on
the z-axis and so is not contained in D. See the figure in the centre below.
D1
D2 D3
On the other hand, here is an example which does satisfy (H).

• Let D = D3 = (x, y, z) ∈ R3 0 < |(x, y, z)| < 2 (an open ball with
its centre removed). For example the circle
x2 + y 2 = 1, z = 0 is the
boundary of the surface (x, y, z) ∈ R x + y 2 + z 2 = 1, z > 0 ⊂ D.
3 2
See the figure on the right above.

This leaves the question of what happens when (H) is violated. We’ll just
look at one example, which however gives the flavour of the general theory.
The punctured disk is
(x, y) ∈ R2 0 < |(x, y)| < 1

D=
D
We’ll start by looking at one particular vector field, which passes the screen-
ing test, but which cannot possibly be conservative. The field, which we saw
in Example 2.3.14, is
y x
Θ=− ı̂ı + 2 ̂
x2 + y 2 x + y2
with domain of definition D. We’ll first check that it passes the screening test:
n ∂ x ∂ y o
∇ ×Θ = − − k̂
∂x x2 + y 2 ∂y x2 + y 2
n 1 2x2 1 2y 2 o
= − 2 + − 2 k̂
x2 + y 2 (x2 + y 2 ) x2 + y 2 (x2 + y 2 )
=0
Next we’ll check that it cannot be conservative. Denote by Cε the circle x2 +

y 2 = ε2 , with counterclockwise orientation. Parametrize Cε by r(θ) = ε cos θ ı̂ı +
ε sin θ ̂ with 0 ≤ θ ≤ 2π. Then
Z Z 2π
dr
Θ · dr = Θ r(θ) · (θ) dθ
Cε 0 dθ
Z 2π
1 1
= − sin θ ı̂ı + cos θ ̂ · − ε sin θ ı̂ı + ε cos θ ̂ dθ (E1)
0 ε ε
Z 2π
= dθ
0
= 2π
is not zero. By TheoremR 4.5.3, Θ cannot be conservative on the punctured

disk since the integral Cε Θ · dr around the closed curve Cε is nonzero.
Next we’ll check that it is locally conservative. That is, it can be written
in the form ∇ θ(x, y) near any point (x0 , y0 ) in its domain. Define θ(x, y) to
be the polar angle of (x, y) with, for example, −π < θ < π. This θ is defined
on all of D, except for the negative real axis. The domain of definition, Dπ , is
sketched on the left below.
Dπ D0
If (x0 , y0 ) happens to lie on the negative real axis, just replace −π < θ < π
by a different interval of length 2π, like 0 < θ < 2π. The domain of definition
of θ would then change to the D0 , sketched on the right above.
It’s now a simple matter to check that ∇ θ(x, y) = Θ (x, y) on the domain
of definition of θ. If x 6= 0, then, from the figure below,
y px, yq
?
x2 `y 2
y
θpx, yq
x x
we have that tan θ(x, y) = xy , and cos θ(x, y) = √ x
, so that
x2 +y 2
∂ y h ∂ i y
tan θ(x, y) = − 2 =⇒ θ(x, y) sec2 θ(x, y) = − 2
∂x x ∂x x
∂ y
=⇒ θ(x, y) = − 2 cos2 θ(x, y)
∂x x
y x2 y
=− 2 2 2
=− 2
x x +y x + y2
∂ 1 h∂ i 1
tan θ(x, y) = =⇒ θ(x, y) sec2 θ(x, y) =
∂y x ∂y x
∂ 1
=⇒ θ(x, y) = cos2 θ(x, y)
∂y x
1 x2 x
= 2 2
= 2
x x +y x + y2
If x = 0, then we must have y 6= 0 (since (0, 0) is not in the domain of definition
to θ), and we can use cot θ(x, y) = xy instead and arrive at the same result.
So far we have just looked at one vector field on D. We are now ready to
consider any vector field F on D that passes the screening test ∇ × F = 0 on
D. We claim that there is a function ϕ on D such that
I
1
Θ
F = αF + ϕ ∇ where αF = F · dr (E2)
2π Cε
The significance of this claim is that it says that if a vector field on D passes the
screening test on D, then, either it is conservative (that’s the case if and only
if αF = 0) or, if it fails to be conservative, then it differs from a conservative
field (namely ∇ ϕ) only by a constant (namely αF ) times the fixed vector field
Θ . That is, there is only one nonconservative vector field on D that passes the
screening test, up to multiplication by constants and addition of conservative
fields. This is a nice simple surprise.
Observe that in the definition of αF , we did not specify the radius ε of the
circle Cε to be used for the integration curve. That’s because the answer to the
0
choice of ε.2 To
integral does not depend on the 0<ε <ε<1
0see this, take any
and consider the surface S = (x, y) ∈ R ε < |(x, y)| < ε .
Ćε1
Cε
D
It is completely contained in D. The boundary of S consists of two parts.
The outside part is Cε , oriented counterclockwise as usual. The inside part is
Cε0 , but oriented clockwise. It is usually denoted −Cε0 . So, by Stokes’ theorem,
I I I I I
F · dr − F · dr = F · dr + F · dr = F · dr
Cε Cε0 Cε −Cε0 ∂S
ZZ
= ∇ × F · n̂ dS = 0
S
Finally to verify the claim (E2), we check that the vector

H field G = F−αFΘ
is conservative on D. To do so, it suffices to check that C G · dr = 0 for any
closed curve C in D. In fact we can restrict our attention to curves C that are
simple, closed, counterclockwise oriented curves on D. A curve is called simple
if it does not cross itself. Closed curves which are not simple can be split up
into simple Hclosed subcurves. And changing the orientation of C just changes
the sign of C G · dr = 0, which does not affect whether it is zero or not.
So let C be Ha simple, closed, counterclockwise oriented curve in D. We need
to verify that C G · dr = 0. Any simple closed curve in R2 divides R2 into
three mutually disjoint subsets3 — C itself, the set of points inside C and the
set of points outside C. Since (0, 0) is not on C, it must be either outside C, as
in the figure of the left below, or inside C as in the figure on the right below.
Cε
C C
D D
• Case 1: (0, 0) outside C. In this case C is the boundary of a set, S, which

is completely contained in D, namely all of the points inside C. So, by
3 This, intuitively obvious, but hard to prove, result is called the Jordan curve theorem.
It is named after the French mathematician Camille Jordan (1838–1922), who first proved
it.
Stokes’ theorem,
I I

G · dr = F − αFΘ · dr
C
Z∂S
Z ZZ
= ∇ × F · n̂ dS − αF ∇ × Θ · n̂ dS = 0 − αF 0
S S
=0
• Case 2: (0, 0) inside C. Since (0, 0) is not on C, we can choose ε small

enough that the circle Cε lies completely inside C. Then the curve C − Cε
is the boundary of a set, S, which is completely contained in D, namely
the part of D that is between Cε and C. So, by Stokes’ theorem,
I I I I ZZ
G · dr − G · dr = G · dr = G · dr = ∇ × G · n̂ dS
C Cε C−Cε ∂S S
=0
since ∇ × G = ∇ × F − αF∇ × Θ = 0 on D. Hence

I I I I
G · dr = G · dr = F · dr − αF Θ · dr = 2παF − αF (2π) = 0
C Cε Cε Cε
by the definition, (E2), of αF and (E1).
So G is conservative on D and F is of the form (E2) on D.

The ideas that we have explored here can be generalised quite a bit. For
example, if we had a disk with n > 1 punctures, we could use arguments like
those above to show that any vector field F that passes the screening test has
to be of the form
Xn
F= ϕ+∇ α` Θ`
`=1
with Θ ` simply being the above Θ translated so as to be centered on the `th

puncture.
4.6 Really Optional — More Interpretation of

Div and Curl
We are now going to determine, in much more detail than before1 , what the
divergence and curl of a vector field tells us about the flow of that vector field.
Consider a (possibly compressible) fluid with velocity field v(x, t). Pick
any time t0 and a really tiny piece of the fluid; assume that, at time t0 , it is a
cube with corners at
x0 + n1 εê(1) + n2 εê(2) + n3 εê(3) n1 , n2 , n3 ∈ {0, 1}

εêp1q
εêp2q
εê p3q x0
1 We’ll also use some more mathematics than before. In this section, we’ll use matrix
eigenvalues and eigenvectors and solve some simple systems of ordinary differential equations.
We’ll also need to use a lot of subscripts and superscripts. It only looks intimidating.
Here ε > 0 is the length of each edge of the cube and is assumed to be
really small. The vectors ê(1) , ê(2) and ê(3) are three mutually perpendicular
unit vectors that give the orientation of the edges of the cube. The vectors
from the corner x0 to its three nearest neighbour corners are εê(1) , εê(2) and
εê(3) .
As time progresses, the chunk of fluid moves. In particular, the corners
move. Let us denote by εb(1) (t) the vector, at time t, joining the n1 = n2 =
n3 = 0 corner to the n1 = 1, n2 = n3 = 0 corner. Define εb(2) (t) and εb(3) (t)
similarly. For times very close to t0 we can think of our chunk of fluid as being
essentially a parallelepiped with edges εb(k) (t).
εbp1q ptq
εbp3q ptq
εbp2q ptq
By concentrating on the edges εb(k) (t) of the chunk of fluid, rather than
the corners, we are ignoring any translations that the chunk of fluid might have
undergone. We want, instead, to determine how the size and orientation of the
parallelepiped changes as t increases.
At time t0 , b(k) = ê(k) . The velocities of the corners of the chunk of fluid
at time t0 are
v x0 + n1 εê(1) + n2 εê(2) + n3 εê(3) , t0

In particular, at time t0 , the tail of εb(k) has velocity v(x0 , t0 ) and the head
of εb(k) has velocity v(x0 + εê(k) , t0 ). Consequently (using a Taylor approxi-
mation),
db(k) 3 ∂v (k)
(t0 ) = v x0 + εê(k) , t0 − v x0 , t0 = x0 , t0 êj + O(ε2 )
P
ε ε
dt j=1 ∂xj
and so
db(k) P3 ∂v (k)
(t0 ) = x0 , t0 êj + O(ε)
dt j=1 ∂xj
The notation O(εn ) represents a function that is bounded by a constant times

(k) P3 ∂v (k)
εn for all sufficiently small ε. That is, we are saying that dbdt (t0 ) is j=1 ∂x j
x0 , t0 êj
(k)
plus a small error that is bounded by a constant time ε. The notation êj just
refers to the j th component of the vector ê(k) .
Denote by V the 3 × 3 matrix whose (i, j) matrix element is
∂vi
Vi,j = x0 , t0 1 ≤ i, j ≤ 3 (M)
∂xj
Then we can write the above more compactly:
db(k)
(t0 ) = Vb(k) (t0 ) + O(ε)
dt
Here Vb(k) (t0 ) is the product of the 3 × 3 matrix V and the 3 × 1 column vector
b(k) (t0 ). We study the behaviour of b(k) (t) for small ε and t close to t0 , by
studying the behaviour of the solutions to the initial value problems

db(k)
(t) = Vb(k) (t) b(k) (t0 ) = ê(k) (IVP)
dt
To warm up, we first look at two two-dimensional examples. In both examples,
the velocity field v(x, y) is linear in (x, y). Consequently, in these examples,
P3 ∂v
(k)
v x0 + εê(k) , t0 − v x0 , t0 is exactly j=1 ε ∂x j
x0 , t0 êj and the solution
to (IVP) coincides with the exact b(k) (t). Following each example, we discuss
a broad class of V’s that generate behaviour similar to that example.
Example 4.6.1 v(x, y) = 2xı̂ı + 3y̂. In this example

2 0
V=
0 3
The solution to the initial value problem
b01 (t) = 2b1 (t) b1 (0) = β1

β
b0 (t) = Vb(t) b(0) = 1 or equivalently
β2 b02 (t) = 3b2 (t) b2 (0) = β2
is
b1 (t) = e2t β1
2t
e 0
or equivalently b(t) = b(0)
b2 (t) = e3t β2 0 e3t
If one chooses ê(1) = ı̂ı and ê(2) = ̂, the edges, b(1) (t) = e2t ê(1) and b(2) (t) =
e3t ê(2) , of the chunk of fluid never change direction. But their lengths do
(k)
change. The relative rate of change of length per unit time, | dbdt (t)|/|b(k) (t)|,
is 2 for b(1) and 3 for b(2) . In the figure below, the darker rectangle is the
initial square. That is, the square with edges b(k) (t0 ) = ê(k) . The lighter
rectangle is that with edges b(k) (t) for some t a bit bigger than t0 .

As time increases the initial cube becomes a larger and larger rectangle.
Example 4.6.2 Example 4.6.1, generalized. The behaviour of Example
4.6.1 is typical of V’s that are symmetric matrices, i.e. that obey2 Vi,j = Vj,i
for all i, j. Any d × d symmetric matrix3 (with real entries)
• has d real eigenvalues
• has d mutually orthogonal real unit eigenvectors.
Denote by λk , 1 ≤ k ≤ d, the eigenvalues of V and choose d mutually per-

pendicular real unit vectors, ê(k) , that obey Vê(k) = λk ê(k) for all 1 ≤ k ≤ d.
Then
b(k) (t) = eλk (t−t0 ) ê(k)
obeys
db(k)
(t) = λk eλk (t−t0 ) ê(k) = eλk (t−t0 ) Vê(k) = Vb(k) (t) and b(k) (t0 ) = ê(k)
dt
So b(k) (t) = eλk (t−t0 ) ê(k) satisfies (IVP) for all t and 1 ≤ k ≤ d.
If we start, at time t0 , with a cube whose edges, ê(k) , are eigenvectors of V,
then as time progresses the edges, b(k) (t), of the chunk of fluid never change
direction. But their lengths change with the relative rate of change of length
per unit time being λk for edge number k. This rate of change may be positive
(the edge grows with time) or negative (the edge shrinks in time) depending
on the sign of λk .
The volume of the chunk of fluid at time t is V (t) = eλ1 (t−t0 ) · · · eλd (t−t0 ) .
The relative rate of change of volume per unit time is V 0 (t)/V (t) = λ1 · · · + λd ,
the sum of the d eigenvalues.
Pd The sum of the eigenvalues of any d × d matrix
V is given by its trace i=1 Vi,i . For the matrix (M)
d
V 0 (t0 ) X ∂vi
= x0 , t0 = ∇ · v x0 , t0
V (t0 ) i=1
∂xi
So, at least when the matrix V defined in (M) is symmetric, the divergence
∇ · v x0 , t0 gives the relative rate of change of volume per unit time for our
tiny chunk of fluid at time t0 and position x0 . Thus when ∇ · v = 0 the volume
is fixed. In particular, this is the case when the fluid is incompressible.
In fact we can relax the symmetry condition.
Example 4.6.3 Example 4.6.1, generalized yet again. For any d × d
matrix V, the solution of
b0 (t) = Vb(t) b(t0 ) = e
is
b(t) = eV(t−t0 ) e
where the exponential of a d × d matrix B is defined by the power series
1 3 ∞ 1
eB = 1 + B + 21 B 2 + Bn
P
B + ··· =
3! n=0 n!
with 1 denoting the d × d identity matrix. This sum converges4

for all d × d matrices B. Furthermore it easy to check, using the power
series, that eV(t−t0 ) obeys dtd V(t−t0 )
e = VeV(t−t0 ) and is the identity matrix
V(t−t0 )
when t = t0 . So b(t) = e e really does obey b0 (t) = Vb(t) and b(t0 ) = e.
Pick any d vectors e , 1 ≤ k ≤ d, and define b(k) (t) = eV(t−t0 ) e(k) . Also
(k)
let E be the d × d matrix whose k th column is e(k) and E(t) be the d × d matrix
whose k th column is b(k) (t). Then the volume of the parallelepiped with edges
e(k) , 1 ≤ k ≤ d, is V (t0 ) = det E and the volume of the parallelepiped with
edges b(k) (t), 1 ≤ k ≤ d, is
V (t) = det E(t) = det eV(t−t0 ) E = det eV(t−t0 ) det E = det eV(t−t0 ) V (t0 )

Of course now we have to compute the determinant of the exponential of a

matrix. Luckily, there is an easy way to do this. For any d × d matrix B, we
2 ∂vi
∂vj
In terms of our original vector field, this condition is that ∂x x0 , t0 = ∂x x0 , t0 .
j i
So, in three
dimensions, it comes down to the requirement that ∇ × v be zero at the point
x0 , t 0 .
3 This was proven by the French mathematician and physicist Augustin-Louis Cauchy
(1789–1857) in 1829.
have5 det eB = etr B , where tr B, called the trace of the matrix B, is the sum
of the diagonal matrix elements of B. So
d
V 0 (t0 ) X
V (t) = e(t−t0 ) tr V V (t0 ) ⇒ = tr V = Vi,i
V (t0 ) i=1
So, for any matrix V defined

as in (M) and any choice of ê(k) , 1 ≤ k ≤ d, the
divergence ∇ · v x0 , t0 gives the relative rate of change of volume per unit
time for our tiny chunk of fluid at time t0 and position x0 .
Example 4.6.4 v(x, y) = −yı̂ı + x̂. In this example

0 −1
V=
1 0
The solution6 to
b01 (t) = −b2 (t)

β1 b1 (0) = β1
b0 (t) = Vb(t) b(0) = or equivalently
β2 b02 (t) = b1 (t) b2 (0) = β2
is

b1 (t) = β1 cos t − β2 sin t cos t − sin t
b2 (t) = β1 sin t + β2 cos t sin t cos t
Consequently the vector b(t) has the same length as b(0). The angle between
b(t) and b(0) is just t radians. So, in this example, no matter what direction
vectors ê(k) we pick, the chunk of fluid just rotates at one radian per unit time.
In the figure below, the outlined rectangle is the initial square. That is, the
square with edges b(k) (t0 ) = ê(k) . The shaded rectangle is that with edges
b(k) (t) for some t a bit bigger than t0 .

Example 4.6.5 Example 4.6.4, generalized. The behaviour of Example
4.6.4 is typical of V’s that are antisymmetric matrices, i.e. that obey Vi,j =
−Vj,i for all i, j. As we have already observed, for any d × d matrix V, the
solution of b0 (t) = Vb(t), b(0) = e is b(t) = eVt e. We now show that if V
is a 3 × 3 antisymmetric matrix, then eVt is a rotation.
Assuming that V is not the zero matrix (in which case eVt is the identity
matrix for all t), we can find a number Ω > 0 and a unit vector k̂ = (k1 , k2 , k3 )
4 The proof is not so hard, though we’ll only outline it. Just denote by β the magnitude
of the largest matrix element of B. Then use the definition of the matrix product to prove
that the largest matrix element of B n has magnitude at most (dβ)n .
5 Again, we won’t prove this. But for a diagonal matrix, it is easy — just compute both
sides. So for a diagonalizable matrix it is also easy — diagonalize.

6 You can find the solution either by guessing, or by using eigenvalues and eigenvectors.
(not necessarily the standard unit vector parallel to the z-axis) such that
 
0 −Ωk3 Ωk2
V =  Ωk3 0 −Ωk1  (R)
−Ωk2 Ωk1 0
This is easy. Because V isq

antisymmetric, all of the entries on its diagonal must
be zero. Define Ω to be V1,22 + V 2 + V 2 and k = −V
1,3 2,3 1 2,3 /Ω, k2 = V1,3 /Ω,
k3 = −V1,2 /Ω. Also, let ı̂ı be any unit vector orthogonal to k̂ (again, not
necessarily the standard one) and ̂ = k̂ × ı̂ı. So ı̂ı, ̂, k̂ is a right-handed
system of three mutually perpendicular unit vectors.
Observe that, for any vector e = (e1 , e2 , e3 )
    
0 −Ωk3 Ωk2 e1 k2 e3 − k3 e2
Ve =  Ωk3 0 −Ωk1  e2  = Ω k3 e1 − k1 e3  = Ωk̂ × e
−Ωk2 Ωk1 0 e3 k1 e2 − k2 e1
In particular,
Vı̂ı = Ωk̂ × ı̂ı = Ω̂ V̂ = Ωk̂ × ̂ = −Ωı̂ı V k̂ = Ωk̂ × k̂ = 0

2 2 2 2
V ı̂ı = ΩV̂ = −Ω ı̂ı V ̂ = −ΩVı̂ı = −Ω ̂ V 2 k̂ = V0 = 0
V 3ı̂ı = ΩV 2̂ = −Ω3̂ V 3̂ = −ΩV 2ı̂ı = Ω3ı̂ı V 3 k̂ = V 2 0 = 0
V 4ı̂ı = ΩV 3̂ = Ω4ı̂ı V 4̂ = −ΩV 3ı̂ı = Ω4̂ V 4 k̂ = V 3 0 = 0
and so on. For all odd n ≥ 1,
V nı̂ı = (−1)(n−1)/2 Ωn̂ V n̂ = −(−1)(n−1)/2 Ωnı̂ı V n k̂ = 0
and all even n ≥ 2,
V nı̂ı = (−1)n/2 Ωnı̂ı V n̂ = (−1)n/2 Ωn̂ V n k̂ = 0
Hence we can write

∞ 1 P (−1)n/2 P (−1)(n−1)/2
eVtı̂ı = (Vt)nı̂ı = (Ωt)nı̂ı + (Ωt)n̂
P
n=0 n! n even n! n odd n!
= cos(Ωt)ı̂ı + sin(Ωt) ̂
∞ 1 P (−1)n/2 P (−1)(n−1)/2
eVt̂ = (Vt)n̂ = (Ωt)n̂ − (Ωt)nı̂ı
P
n=0 n! n even n! n odd n!
= − sin(Ωt)ı̂ı + cos(Ωt) ̂
∞ 1
eVt k̂ = (Vt)n k̂
P
n=0 n!
= k̂
So eVt is rotation by an angle Ωt about the axis k̂.

Example 4.6.6 Example 4.6.5, continued. Whether or not the matrix V
defined in (M) is antisymmetric, the related matrix with entries
Ai,j = 12 Vi,j − Vj,i

is. When V is antisymmetric, A and V coincide. The matrix A is (to write it

out explicitly)
 
∂v1
∂v2 ∂v1
x0 , t0 − ∂v

0 ∂x x0 , t0 − ∂x x0 , t0 ∂x3 ∂x
3
x0 , t0
1 ∂v1 2 1 1
x0 , t0 + ∂v ∂v2
x0 , t0 − ∂v

− ∂x2 ∂x1 x0 , t0
2
0 ∂x3 ∂x2 x0 , t0 
3
2 ∂v
x0 , t0 + ∂x13 x0 , t0 −∂v
∂v ∂v3

−∂x31 ∂x3 x0 , t0 + ∂x2 x0 , t0
2
0
Comparing this with (R), we see that
Ωk̂ = 21 ∇ × v x0 , t0

So, at least when the matrix V defined in (M) is antisymmetric,

our tiny cube
rotates about the axis with ∇ × v x0 , t0 at rate 12 ∇ × v x0 , t0 .

Remark 4.6.7 In the generalization, Example 4.6.5, of Example 4.6.4, we only
considered dimension 3. It is a nice exercise in eigenvalues and eigenvectors
to handle general dimension. Here are the main facts about antisymmetric
matrices with real entries that are used.
• All eigenvalues of antisymmetric matrices are either zero or pure imagi-
nary.
• For antisymmetric matrices with real entries, the nonzero eigenvalues
come in complex conjugate pairs. The corresponding eigenvectors may
also be chosen to be complex conjugates.
Choose as basis vectors (like ı̂ı, ̂, k̂ above)
• the eigenvectors of eigenvalue 0 (they act like k̂ above)
• the real and imaginary parts of each complex conjugate pair of eigenvec-
tors (they act like ı̂ı, ̂ above)
Resumé so far:
We have now seen that
• when the matrix V defined in (M) is symmetric and the direction vectors
ê(k) of the cube are eigenvectors of V, then, at time t0 , the chunk of
fluid is not changing orientation
but is changing volume at instantaneous
relative rate ∇ · v x0 , t0 and
• when the matrix V defined in (M) is antisymmetric, then, at time t0 , the
chunk of fluid is not changing
shape orsize but is rotating about the axis
∇ × v x0 , t0 at rate 12 ∇ × v x0 , t0 . For this reason, ∇ × v is often
referred to as a “vorticity” meter.
These agree with our earlier interpretations of divergence and curl.
The general case:
Now consider a general matrix V. It can always be written as the sum
V =S+A
of a symmetric matrix S and an antisymmetric matrix A. Just define
Si,j = 21 Vi,j + Vj,i Ai,j = 12 Vi,j − Vj,i

As we have already observed, the solution of
b0 (t) = Vb(t) b(0) = e
is
b(t) = eVt e = e(A+S)t e
If S and A were ordinary numbers, we would have e(A+S)t = eAt eSt . But for
matrices this need not be the case, unless S and A happen to commute7 . For
arbitrary matrices, it is still true that
h in
e(A+S)t = lim eAt/n eSt/n
n→∞
8
This is called the Lie product formula. It shows that our tiny chunk of fluid
mixes together the behaviours of A and S, scaling a bit, then rotating a bit,
then scaling a bit and so on.
Example 4.6.8 v(x, y) = 2yı̂ı. In this example

0 2 0 1 0 1
V= =S+A with S= A=
0 0 1 0 −1 0
The solution to the full flow

b01 (t) = 2b2 (t) b1 (0) = β1

β
b0 (t) = Vb(t) b(0) = 1 or equivalently
β2 b02 (t) = 0 b2 (0) = β2
is
b1 (t) = β1 + 2β2 t 1 2t
b2 (t) = β2 0 1
The solution to the S part of the flow
b0 (t) = b2 (t) b1 (0) = β1

β
b (t) = Sb(t) b(0) = 1
0
or equivalently 10
β2 b2 (t) = b1 (t) b2 (0) = β2
is9

b1 (t) = β1 cosh t + β2 sinh t cosh t sinh t
b2 (t) = β1 sinh t + β2 cosh t sinh t cosh t
The eigenvectors of S are

(1) 1 1 (2) 1 1
ê =√ ê =√
2 1 2 −1
The corresponding eigenvalues are +1 and −1. The eigenvectors obey

St (1) cosh t sinh t (1)
e ê = ê = et ê(1)
sinh t cosh t

cosh t sinh t (2)
eSt ê(2) = ê = e−t ê(2)
sinh t cosh t
Under the S part of the flow ê(1) scales by a factor of et , which is bigger than
one for t > 0 and ê(2) scales by a factor of e−t , which is smaller than one for
t > 0.
The solution to the A part of the flow
b0 (t) = b2 (t) b1 (0) = β1

β
b0 (t) = Ab(t) b(0) = 1 or equivalently 0 1
β2 b2 (t) = −b1 (t) b2 (0) = β2
7 By
definition, the matrices S and A commute when AS = SA.
8 This
formula is named after the Norwegian mathematician Marius Sophus Lie
(1842–1899). In 1870, he was arrested and held in prison in France for a month, because he
was suspected of being a German spy. His mathematics notes were thought to be top secret
coded messages.
is

b1 (t) = β1 cos t + β2 sin t cos t sin t
b2 (t) = −β1 sin t + β2 cos t − sin t cos t
The A part of the flow rotates clockwise about the origin at one radian per
unit time.
Here are some figures to help us visualize this.
• The first shows a square with edges ê(1) , ê(2) and its image under the
full flow t = 0.4 later. Under this full flow the vector ê(k) → e0.4V ê(k) .
The darkly shaded parallelogram has edges e0.4V ê(k) .
• The second shows its image under 0.4 time units of the S-flow (that is,
ê(k) → e0.4S ê(k) ). The lightly shaded rectangle has edges e0.4S ê(k) .
• The third applies 0.4 time units of the A-flow to the shaded rectangle of
the second figure. So the lightly shaded rectangle of the third figure has
edges e0.4S ê(k) and the darkly shaded rectangle has edges e0.4A e0.4S ê(k) .
êpkq and e0.4V êpkq êpkq and e0.4S êpkq
e0.4S êpkq and e0.4A e0.4S êpkq

Of course e0.4A e0.4S ê(k) (as in the darkly shaded rectangle of the third fig-
ure) is not a very good approximation for e0.4(A+S) ê(k) (asin the darkly shaded
n
parallelogram of the first figure). It is much better to take e0.4A/n e0.4S/n ê(k)
with n large. Each of the following figures shows two parallelograms. In each,
0.4V (k)
e
the shaded region has edges ê = e0.4(A+S) ê(k) and the outlined region
0.4A/n 0.4S/n n (k)
has edges e e ê .
n“1 n“5 n “ 10
n
So we can see that, as n increases, e0.4A/n e0.4S/n ê(k) becomes a better
and better approximation to e0.4(A+S) ê(k) .
4.7 Optional — A Generalized Stokes’ Theorem

As we have seen, the fundamental theorem of calculus, the divergence theorem,
Greens’ theorem and Stokes’ theorem share a number of common features.
There is in fact a single framework which encompasses and generalizes all of
them, and there is a single theorem of which they are all special cases. We
now give a bare bones introduction to this framework and theorem. A proper
treatment typically takes up a good part of a full course. Here is an outline of
what we shall do:
• First, we will define differential forms. To try and keep things as
simple and concrete as possible, we’ll only define1 differential forms on
R3 — all of our functions will be defined on R3 . Very roughly speaking,
a k-form is what you write after the integral sign of an integral over a k
dimensional object. Here k is one of 0, 1, 2, 3. As a example, a 1-form
is an expression of the form F1 (x, y, z) dx + F2 (x, y, z) dy + F3 (x, y, z) dz.
For k = 0, think of a point as a zero dimensional object and think of
evaluating a function at a point as “integrating the function over the
point”.
• Then we will define some operations on differential forms, so that we can
add them, multiply them, differentiate them and, eventually, integrate
them. The derivative of a k-form ω is a (k + 1)-form that is denoted dω.
It will turn out that
◦ differentiating a 0-form amounts to taking a gradient,

◦ differentiating a 1-form amounts to taking a curl, and
◦ differentiating a 2-form amounts to taking a divergence.
• Finally we will get to the generalized Stokes’ theorem which says that,
if ω is a k-form (with k = 0, 1, 2) and D is a (k + 1)-dimensional domain
of integration, then Z Z
dω = ω
D ∂D
It will turn out that
◦ when k = 0, this is just the fundamental theorem of calculus and
9 Recall 1 t −t
1 t −t

that sinh t = 2 e − e and cosh t = 2 e + e .
1 Ingeneral, a differential form is defined on a manifold, which is an abstract generaliza-
tion of a multi-dimensional surface, like a sphere or a torus.
◦ when k = 1, this is both Green’s theorem and our Stokes’ theorem,

and
◦ when k = 2, this is the divergence theorem.
Now let’s get to work. For simplicity, we will assume throughout this section
that all derivatives of all functions exist and are continuous. Our first task to
define differential forms.
As we said above we will define a 1-form as an expression of the form
F1 (x, y, z) dx+F2 (x, y, z) dy+F3 (x, y, z) dz. When you learned the definition of
the integral the symbol “dx” was not given any mathematical meaning by itself.
A meaning was given only to collections of symbols like the indefinite integral
R Rb
“ f (x) dx” and the definite integral “ a f (x) dx”. Later in this section, we
will give a meaning to dx. We will, in Definition 4.7.9, define a differentiation
operator that we will call d. Then dx will be that differentiation operator
applied to the function f (x) = x. However, until then we will have to treat dx
and dy and dz just as symbols. Their sole role in F1 (x, y, z) dx+F2 (x, y, z) dy+
F3 (x, y, z) dz is to allow us to distinguish2 F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z).
Similarly, we will define a 2-form as an expression of the form F1 (x, y, z) dy∧
dz + F2 (x, y, z) dz ∧ dx + F3 (x, y, z) dx ∧ dy. Once again there is a symbol,
namely “∧”, that we have not yet given a meaning to. We will, in Definition
4.7.3, define a product, called the wedge product, with ∧ as the multiplication
symbol. Then dx ∧ dy will be the wedge product of dx and dy. Until then we
will have to treat dy ∧ dz, dz ∧ dx and dx ∧ dy just as three more meaningless
symbols.
Finally here is the definition.
Definition 4.7.1
a A 0-form is a function f (x, y, z).
b A 1-form is an expression of the form
F1 (x, y, z) dx + F2 (x, y, z) dy + F3 (x, y, z) dz
with F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) being functions of three vari-
ables.
c A 2-form is an expression of the form
F1 (x, y, z) dy ∧ dz + F2 (x, y, z) dz ∧ dx + F3 (x, y, z) dx ∧ dy
with F1 (x, y, z), F2 (x, y, z) and F3 (x, y, z) being functions of three vari-
ables.
d A 3-form is an expression of the form f (x, y, z) dx∧dy∧dz, with f (x, y, z)
being a function of three variables.
At this stage (there’ll be more later), just think of “dx”, “dy”, “dz”, “dx ∧ dy”,
and so on, as symbols. Do not yet attempt to attach any significance to them.
♦
There are four operations involving differential forms — addition, multi-
plication (∧), differentiation (d) and integration. Here are their definitions.
First, addition is defined, and works, just the way that you would expect it to.
2 We could also define, for example, a 1-form as an ordered list
F1 (x, y, z) , F2 (x, y, z) , F3 (x, y, z) of three functions and just view F1 (x, y, z) dx +

F2 (x, y, z) dy + F3 (x, y, z) dz as another notation for F1 (x, y, z) , F2 (x, y, z) , F3 (x, y, z) .
Definition 4.7.2 Addition of differential forms.

a The sum of the 0-forms f and g is the 0-form f + g.
b The sum of two 1-forms is the 1-form

F1 dx + F2 dy + F3 dz

+ G1 dx + G2 dy + G3 dz
=(F1 + G1 ) dx + (F2 + G2 ) dy + (F3 + G3 ) dz
c The sum of two 2-forms is the 2-form

F1 dy ∧ dz + F2 dz ∧ dx + F3 dx ∧ dy

+ G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy
=(F1 + G1 ) dy ∧ dz + (F2 + G2 ) dz ∧ dx + (F3 + G3 ) dx ∧ dy
d The sum of two 3-forms is the 3-form

f dx ∧ dy ∧ dz + g dx ∧ dy ∧ dz = f + g dx ∧ dy ∧ dz
♦
There is one wrinkle in multiplication. It is not commutative, meaning
that α ∧ β need not be the same as β ∧ α. You have already seen some
noncommutative products. If a and b are two vectors in R3 , then a × b =
−b × a. Also, if A and B are two n × n matrices, the matrix product AB need
not be the same as BA.
Definition 4.7.3 Multiplication of differential forms. We now define a
multiplication rule for differential forms. If ω is a k-form and ω 0 is a k 0 -form
then the product will be a (k+k 0 )-form and will be denoted ω∧ω 0 (read “omega
wedge omega prime”). It is determined by the following properties.
a If f is a function (i.e. a 0-form), then

f F1 dx + F2 dy + F3 dz = (f F1 ) dx+(f F2 ) dy+(f F3 ) dz

f F1 dy ∧ dz+F2 dz ∧ dx+F3 dx ∧ dy = (f F1 ) dy ∧ dz + (f F2 ) dz ∧ dx
+ (f F3 ) dx ∧ dy

f g dx ∧ dy ∧ dz = (f g) dx ∧ dy ∧ dz
Traditionally, the ∧ is not written when multiplying a differential form

by a function (i.e. a 0-form).
b ω ∧ ω 0 is linear in ω and in ω 0 . This means that if ω = f1 ω1 + f2 ω2 , where
f1 ,f2 are functions and ω1 , ω2 are forms, then
f1 ω1 + f2 ω2 ∧ ω 0 = f1 (ω1 ∧ ω 0 ) + f2 (ω2 ∧ ω 0 )

Similarly,
ω ∧ f1 ω10 + f2 ω20 = f1 (ω ∧ ω10 ) + f2 (ω ∧ ω20 )

c If ω is a k-form and ω 0 is a k 0 -form then

0
ω ∧ ω 0 = (−1)kk ω 0 ∧ ω
That is, if at least one of k and k 0 is even, then
ω ∧ ω0 = ω0 ∧ ω
(so that the wedge product is commutative) and if k and k 0 are both odd
then
ω ∧ ω 0 = −ω 0 ∧ ω
(so that the wedge product is anticommutative). In particular, if ω is a
d-form with d odd
ω∧ω =0
d The wedge product is associative. This means that
(ω ∧ ω 0 ) ∧ ω 00 = ω ∧ ω 0 ∧ ω 00

♦
So the wedge product obeys most of the usual multiplication rules, with
the one big exception that if ω is k-form and ω 0 is a k 0 -form with k and k 0 both
odd then ω ∧ ω 0 = −ω 0 ∧ ω.
The best way to get a handle on the wedge product is to work through
some examples, like these.
Example 4.7.4 Let ω = F1 dx + F2 dy + F3 dz and ω 0 = G1 dx + G2 dy + G3 dz
be any two 1-forms. Their product is
ω ∧ ω 0 = F1 dx + F2 dy + F3 dz ∧ G1 dx + G2 dy + G3 dz

= F1 dx ∧ G1 dx + F1 dx ∧ G2 dy + F1 dx ∧ G3 dz

+ F2 dy ∧ G1 dx + F2 dy ∧ G2 dy + F2 dy ∧ G3 dz

+ F3 dz ∧ G1 dx + F3 dz ∧ G2 dy + F3 dz ∧ G3 dz
(by linearity, i.e. by part (b) of the last Definition)
= F1 G1 dx ∧ dx + F1 G2 dx ∧ dy + F1 G3 dx ∧ dz
+ F2 G1 dy ∧ dx + F2 G2 dy ∧ dy + F2 G3 dy ∧ dz
+ F3 G1 dz ∧ dx + F3 G2 dz ∧ dy + F3 G3 dz ∧ dz
= F1 G2 − F2 G1 ) dx ∧ dy + F3 G1 − F1 G3 ) dz ∧ dx
+ F2 G3 − F3 G2 ) dy ∧ dz
because
dx ∧ dx = dy ∧ dy = dz ∧ dz = 0
and
dx ∧ dy = −dy ∧ dx dx ∧ dz = −dz ∧ dx dz ∧ dy = −dy ∧ dz

Note that, looking at the last example, if we view F = (F1 , F2 , F3 ) and
G = (G1 , G2 , G3 ) as vectors, we can write the product simply as
Equation 4.7.5

F1 dx + F2 dy + F3 dz ∧ G1 dx + G2 dy + G3 dz
= (F × G)1 dy ∧ dz + (F × G)2 dz ∧ dx + (F × G)3 dx ∧ dy
where we are using (F × G)` to denote the `th component of the cross
product F × G. In the special case that F3 = G3 = 0, we have
Equation 4.7.6

F1 dx + F2 dy ∧ G1 dx + G2 dy = F1 G2 − F2 G1 ) dx ∧ dy

F F2
= det 1 dx ∧ dy
G1 G2
We can now see why in the Definition 4.7.1.c of 2-forms
• there were no dx ∧ dx or dy ∧ dy or dz ∧ dz terms — they are all zero
and
• there were no dy ∧ dx or dz ∧ dy or dx ∧ dz terms — they can all be
rewritten using dx ∧ dy, dy ∧ dz and dz ∧ dx terms (or vice versa).
The reason that we chose to write the Definition 4.7.1.c as
as opposed to in the form, for example,
f1 dx ∧ dy + f2 dx ∧ dz + f3 dy ∧ dz
was to make formulae like 4.7.5 work. The easy way to remember
is to rename (in your head) x, y, z to x1 , x2 , x3 . Then the subscripts in the

three terms of
F1 dx2 ∧ dx3 + F2 dx3 ∧ dx1 + F3 dx1 ∧ dx2
are just 1, 2, 3 and 2, 3, 1 and 3, 1, 2 — the three cyclic permutations of 1, 2, 3.

Example 4.7.7 The product of the (general) 1-form ω = F1 dx+F2 dy+F3 dz
and the (general) 2-form ω 0 = G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy (again
note the numbering of the coefficients in the 2-form) is
ω ∧ ω 0 = F1 dx + F2 dy + F3 dz ∧ G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy

= F1 G1 dx ∧ dy ∧ dz + F2 G2 dy ∧ dz ∧ dx + F3 G3 dz ∧ dx ∧ dy
= F1 G1 + F2 G2 + F3 G3 ) dx ∧ dy ∧ dz
Here we have used that, for 1-forms, α ∧ β = −β ∧ α, so that
dy ∧ dz ∧ dx = −dy ∧ dx ∧ dz = dx ∧ dy ∧ dz
dz ∧ dx ∧ dy = −dx ∧ dz ∧ dy = dx ∧ dy ∧ dz
We have also used that any wedge product of three d{x or y or z}’s with at
least two of the coordinates being the same is zero. For example
dx ∧ dz ∧ dx = −dx ∧ dx ∧ dz = 0
So

F1 dx + F2 dy + F3 dz ∧ G1 dy ∧ dz + G2 dz ∧ dx + G3 dx ∧ dy
= F · G dx ∧ dy ∧ dz

Example 4.7.8 Combining Examples 4.7.4 and 4.7.7, we have the wedge prod-
uct of any three (general) 1-forms F1 dx + F2 dy + F3 dz and G1 dx + G2 dy +
G3 dz and H1 dx + H2 dy + H3 dz is

F1 dx+F2 dy+F3 dz ∧ G1 dx+G2 dy+G3 dz ∧ H1 dx + H2 dy + H3 dz

= F1 dx + F2 dy + F3 dz ∧

(G × H)1 dy ∧ dz + (G × H)2 dz ∧ dx + (G × H)3 dx ∧ dy

= F1 (G × H)1 + F2 (G × H)2 + F3 (G × H)3 dx ∧ dy ∧ dz

= F1 (G2 H3 −G3 H2 )+F2 (G3 H1 −G1 H3 )+F3 (G1 H2 −G2 H1 dx ∧ dy ∧ dz
This can be expressed cleanly in terms of determinants. Recalling the rule for
expanding a determinant along its top row

F1 dx+F2 dy+F3 dz ∧ G1 dx+G2 dy+G3 dz ∧ H1 dx+H2 dy+H3 dz
 
F1 F2 F3
= det G1 G2 G3  dx ∧ dy ∧ dz
H1 H2 H3

Our next operation is a differential operator which unifies and generalizes
gradient, curl and divergence.
Definition 4.7.9 Differentiation of differential forms. If ω is a k-form,
then dω is a k + 1-form, with d being the unique3 such operator that obeys
a d is linear. That is, if ω1 , ω2 are k-forms and a1 , a2 ∈ R, then

d a1 ω1 + a2 ω2 = a1 dω1 + a2 dω2
b d obeys a “graded product rule”. Precisely, if ω (k) is a k-form and ω (`) is

an `-form, then
d ω (k) ∧ ω (`) = dω (k) ∧ ω (`) + (−1)k ω (k) ∧ dω (`)

c If f (x, y, z) is a 0-form, then
∂f ∂f ∂f
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz
∂x ∂y ∂z
= ∇ f (x, y, z) · dr where dr = dxı̂ı + dy ̂ + dz k̂
d For any differential form ω,

d dω = 0
♦
Example 4.7.10
a If f (x, y, z) = x, then
∂x ∂x ∂x
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz = dx
∂x ∂y ∂z
That is, dx really is the operator d applied to the function x. Similarly,
dy really is the operator d applied to the function y and dz really is the
operator d applied to the function z.
3 That d is unique just means that the action of d on any differential form is completely
determined by the four rules (a), (b), (c), (d). We will see in Example 4.7.10.c,d,e, that this
is indeed the case.
b For any k-form ω
d ω ∧ dx = dω ∧ dx + (−1)k ω ∧ d dx

= dω ∧ dx
Similarly

d ω ∧ dy = dω ∧ dy d ω ∧ dz = dω ∧ dz
c For any 1-form

d F1 dx + F2 dy + F3 dz = dF1 ∧ dx + dF2 ∧ dy + dF3 ∧ dz
∂F ∂F1 ∂F1
1
= dx + dy + dz ∧ dx
∂x ∂y ∂z
∂F ∂F2 ∂F2
2
+ dx + dy + dz ∧ dy
∂x ∂y ∂z
∂F ∂F3 ∂F3
3
+ dx + dy + dz ∧ dz
∂x ∂y ∂z
∂F ∂F2 ∂F ∂F3
3 1
= − dy ∧ dz + − dz ∧ dx
∂y ∂z ∂z ∂x
∂F ∂F1
2
+ − dx ∧ dy
∂x ∂y
∇ × F)1 dy ∧ dz + (∇
= (∇ ∇ × F)2 dz ∧ dx + (∇ ∇ × F)3 dx ∧ dy
d For any 2-form

d F1 dy ∧ dz + F2 dz ∧ dx + F3 dx ∧ dy
= dF1 ∧ dy ∧ dz + dF2 ∧ dz ∧ dx + dF3 ∧ dx ∧ dy
∂F ∂F1 ∂F1
1
= dx + dy + dz ∧ dy ∧ dz
∂x ∂y ∂z
∂F ∂F2 ∂F2
2
+ dx + dy + dz ∧ dz ∧ dx
∂x ∂y ∂z
∂F ∂F ∂F
3 3 3
+ dx + dy + dz ∧ dx ∧ dy
∂x ∂y ∂z
∂F ∂F2 ∂F3
1
= + + dx ∧ dy ∧ dz
∂x ∂y ∂z
= ∇ · F dx ∧ dy ∧ dz
e For any 3-form

∂f ∂f ∂f
d f dx ∧ dy ∧ dz = dx + dy + dz ∧ dx ∧ dy ∧ dz
∂x ∂y ∂z
=0

Example 4.7.11 In Definition 4.7.9.c, we defined, for any function f (x, y, z)
of three variables
∂f ∂f ∂f
df = (x, y, z) dx + (x, y, z) dy + (x, y, z) dz
∂x ∂y ∂z
The analogous formulae4 for functions of one or two variables also apply.
df
df (t) = (t) dt
dt
∂f ∂f
df (u, v) = (u, v) du + (u, v) dv
∂u ∂v
a Let F1 (x, y, z) dx + F2 (x, y, z) dy + F3 (x, y, z) dz be a 1-form. Suppose

that we substitute x = x(t), y = y(t) and z = z(t), so that we are
restricting our1-form to a parametrized curve. Then, writing r(t) =
x(t), y(t), z(t) ,

F1 x(t), y(t), z(t) dx(t) + F2 x(t), y(t), z(t) dy(t)

+ F3 x(t), y(t), z(t) dz(t)
dx dy dz
= F1 r(t) (t) dt + F2 r(t) (t) dt + F3 r(t) (t) dt
dt dt dt
dr
= F r(t) · (t) dt
dt
b Let F1 (x, y, z) dy ∧ dz + F2 (x, y, z) dz ∧ dx + F3 (x, y, z) dx ∧ dy be a 2-

form. Suppose that we substitute x = x(u, v), y = y(u, v) and z = z(u, v),
so that we are restricting our 2-form to a parametrized surface. Then,
writing r(u, v) = x(u, v), y(u, v), z(u, v) ,

F1 x(u, v), y(u, v), z(u, v) dy(u, v) ∧ dz(u, v)

+ F2 x(u, v), y(u, v), z(u, v) dz(u, v) ∧ dx(u, v)

+ F3 x(u, v), y(u, v), z(u, v) dx(u, v) ∧ dy(u, v)
∂y ∂y ∂z ∂z
= F1 r(u, v) du + dv ∧ du + dv
∂u ∂v ∂u ∂v
∂z ∂z ∂x ∂x
+ F2 r(u, v) du + dv ∧ du + dv
∂u ∂v ∂u ∂v
∂x ∂x ∂y ∂y
+ F3 r(u, v) du + dv ∧ du + dv
∂u ∂v ∂u ∂v
h ∂y ∂z
∂y ∂z ∂z ∂x ∂z ∂x

= F1 r(u, v) − + F2 r(u, v) −
∂u ∂v ∂v ∂u ∂u ∂v ∂v ∂u
∂x ∂y ∂x ∂y
i
+ F3 r(u, v) − du ∧ dv
∂u ∂v ∂v ∂u
h ∂r ∂r i
= F r(u, v) · (u, v) × (u, v) du ∧ dv
∂u ∂v

Let us summarize what we have seen in the Example 4.7.10.
Lemma 4.7.12
a For any 0-form
df = ∇ f (x, y, z) · dr
b For any 1-form

d F1 dx + F2 dy + F3 dz
= (∇∇ × F)1 dy ∧ dz + (∇
∇ × F)2 dz ∧ dx + (∇
∇ × F)3 dx ∧ dy
4 Indeed, you can view f (t) as a function of three variables that happens to be independent
of two of the three variables. Similarly you can view f (u, v) as a function of three variables
that happens to be independent of one of the three variables.
c For any 2-form

d F1 dy ∧ dz + F2 dz ∧ dx + F3 dx ∧ dy = ∇ · F dx ∧ dy ∧ dz
d For any 3-form

d f dx ∧ dy ∧ dz = 0
Our final operation is integration of differential forms.
Definition 4.7.13 Integration of differential forms.
a Let f (x, y, z) be a 0-form and P = (x0 , y0 , z0 ) ∈ R3 be a point. Then
Z

f = f x0 , y0 , z0
P
More generally if, for each 1 ≤ i ≤ `, Pi = (xi , yi , zi ) ∈ R3 is a point and

ni is an integer, then
Z `
X
f= ni f xi , yi , zi
Σì=1 ni Pi i=1
b Let ω = F(r) · dr = F1 (x, y, z) dx + F2 (x, y, z) dy + F3 (x, y, z) dz be a 1-

form. Let C be a curve that is parametrized by r(t) = x(t) , y(t) , z(t) ,
a ≤ t ≤ b. Then, motivated by Example 4.7.11.a above,
Z Z b dr
Z
ω= F r(t) · (t) dt = F · dr
C a dt C
c Let ω = F1 (x, y, z) dy ∧ dz + F2 (x, y, z) dz ∧ dx + F3 (x, y, z) dx ∧ dy be

a 2-form. Let S be an oriented
surface that is parametrized by r(u, v) =
x(u, v) , y(u, v) , z(u, v) , with (u, v) running over a region R in the uv-
plane. Assume that r(u, v) is orientation preserving in the sense that
∂r ∂r
n̂ dS = + ∂u × ∂v du dv.Then, motivated by Example 4.7.11.b above,
Z ZZ h ZZ
∂r ∂r i
ω= F r(u, v) · (u, v) × (u, v) du ∧ dv = F · n̂ dS
S R ∂u ∂v S
d Let ω = f (x, y, z) dx ∧ dy ∧ dz be a 3-form. Let V be a solid in R3 . Then

Z ZZZ
ω= f (x, y, z) dxdydz
V V
♦
Finally, after all of these definitions, we have a very compact theorem that
simultaneously covers the fundamental theorem of calculus, Green’s theorem.
Stokes’ theorem and the divergence theorem. Had we given all of our definitions
in n dimensions, rather than just three dimensions, it would cover a lot more.
This general theorem is also called Stokes’ theorem.
Theorem 4.7.14 Stokes’ Theorem. If ω is a k-form (with k = 0, 1, 2) and
D is a (k + 1)-dimensional domain of integration, then
Z Z
dω = ω
D ∂D
Here ∂D is the boundary of D (suitably oriented).

To see the connection between the general Stokes’ theorem 4.7.14 and the
Stokes’ and divergence theorems of the earlier part of this chapter, here are
the k = 1 and k = 2 cases of Theorem 4.7.14 again.
• Let ω = F1 dx + F2 dy + F3 dz be a 1-form and let S be a piecewise
smooth oriented surface as in (our original) Stokes’ theorem 4.4.1. Then,
by Lemma 4.7.12.b,
∇ × F)1 dy ∧ dz + (∇
dω = (∇ ∇ × F)2 dz ∧ dx + (∇
∇ × F)3 dx ∧ dy
So, by parts (c) (but with

R F replaced
R by ∇ × F) and (b) of Definition
4.7.13, the conclusion D dω = ∂D ω of (the general) Stokes’ theorem
4.7.14 is
ZZ Z Z Z
∇ × F · n̂ dS = dω = ω= F · dr
S S ∂S ∂S
which is the conclusion of (our original) Stokes’ theorem 4.4.1.

• ω = F1 (x, y, z) dy ∧ dz + F2 (x, y, z) dz ∧ dx + F3 (x, y, z) dx ∧ dy be a
2-form and let V be a solid as in the divergence theorem 4.2.2. Then, by
Lemma 4.7.12.c,
dω = ∇ · F dx ∧ dy ∧ dz
So, by parts
R (d) R(with f = ∇ · F) and (c) of Definition 4.7.13, the con-
clusion D dω = ∂D ω of (the general) Stokes’ theorem 4.7.14 is
ZZZ Z Z ZZ
∇ · F dxdydz = dω = ω= F · n̂ dS
V V ∂V ∂V
which is the conclusion of the divergence theorem 4.2.2.

Chapter 5
True/False and Other Short

Questions
5.1 True/False and Other Short Questions

Here is a collection of True/False and other short answer questions that involve
the material in the main body of this text.
5.2 Exercises
1. ∗. True or false?
a ∇ · (a × r) = 0, where a is a constant vector in R3 , and r is the
vector field r = (x, y, z).
b ∇ × (∇∇f ) = 0 for all scalar fields f on R3 with continuous second
partial derivatives.
∇ · F, for every vector field F in R3 with
c ∇ · (f F) = ∇ (f ) · F + f∇
continuous partial derivatives, and every scalar function f in R3
with continuous partial derivatives.
d Suppose F is a vector field with continuous partial derivatives in the
region D, where D is R3 without the origin. If ∇ · F > 0 throughout
D, then the flux of F through the sphere of radius 5 with center at
the origin is positive.
e If a vector field F is defined and has continuous partial derivatives

everywhere in R3 , and it satisfies ∇ · F = 0, everywhere, then, for
every sphere, the flux out of one hemisphere is equal to the flux into
the opposite hemisphere.
f If r(t) is a twice continuously differentiable path in R2 with constant

curvature κ, then r(t) parametrizes part of a circle of radius 1/κ.

y x
g The vector field F = − x2 +y 2 , x2 +y 2 is conservative in its domain,
which is R2 , without the origin.
h If a vector field FH = (P, Q) in R2 has Q = 0 everywhere in R2 , then

the line integral F · dr is zero, for every simple closed curve in R2 .
299
CHAPTER 5. TRUE/FALSE AND OTHER SHORT QUESTIONS 300
i If the acceleration and the speed of a moving particle in R3 are

constant, then the motion is taking place along a spiral.
2. ∗. True or false?
a ∇ × (a × r) = 0, where a is a constant vector in R3 , and r is the
vector field r = (x, y, z).
∇f ) = 0 for all scalar fields f on R3 with continuous second
b ∇ · (∇
∇ · F) = 0 for every vector field F on R3 with continuous second
c ∇ (∇
d Suppose F is a vector field with continuous partial derivatives in the
region D, where D is R3 without the origin. If ∇ · F = 0, then the
flux of F through the sphere of radius 5 with center at the origin is
0.
e Suppose F is a vector field with continuous partial derivatives in

3
H region D, where D is R without the origin. If ∇ × F = 0 then3
the
C
F · dr is zero, for every simple and smooth closed curve C in R
which avoids the origin.
f If a vector field F is defined and has continuous partial derivatives
everywhere in R3 , and it satisfies ∇ · F > 0, everywhere, then, for
every sphere, the flux out of one hemisphere is larger than the flux
into the opposite hemisphere.
g If r(t) is a path in R3 with constant curvature κ, then r(t) parametrizes
part of a circle of radius 1/κ.

y x
h The vector field F = − x2 +y 2 , x2 +y 2 , z is conservative in its do-
main, which is R3 , without the z-axis.
i If all flow lines of a vector field in R3 are parallel to the z-axis, then
the circulation of the vector field around every closed curve is 0.
j If the speed of a moving particle is constant, then its acceleration is
orthogonal to its velocity.
3. ∗.
a True or false? If r(t) is the position at time t of an object moving
in R3 , and r(t) is twice differentiable, then |r00 (t)| is the tangential
component of its acceleration.
b Let r(t) is a smooth curve in R3 with unit tangent, normal and
binormal vectors T̂(t), N̂(t), B̂(t). Which two of these vectors span
the plane normal to the curve at r(t)?
c True or false? If F = Pı̂ı + Q̂ + Rk̂ is a vector field on R3 such that
P , Q, R have continuous first order derivatives, and if ∇ × F = 0
everywhere on R3 , then F is conservative.
d True or false? If F = Pı̂ı +Q̂ +Rk̂ is a vector field on R3 such that P ,
Q, R have continuous second order derivatives, then ∇ ×(∇ ∇ ·F ) = 0.
e True or false? If F is a vector field on R3 such that |F(x, y, z)| =

1 for all x, y, z, and if S is the sphere x2 + y 2 + z 2 = 1, then
RR
S
F · n̂ dS = 4π.
f True or false? Every closed surface S in R3 is orientable. (Recall

that S is closed if it is the boundary of a solid region E.)
4. ∗.
a In the curve shown below (a helix lying in the surface of a cone), is
the curvature increasing, decreasing, or constant as z increases?
z
x
b Of the two functions shown below, one is a function f (x) and one is
its curvature κ(x). Which is which?
y
x
c Let C be the curve of intersection of the cylinder x2 + z 2 = 1 and
the saddle xz = y. Parametrise C. (Be sure to specify the domain
of your parametrisation.)
d Let H be the helical ramp (also known as a helicoid) which revolves

around the z-axis in a clockwise direction viewed from above, be-
ginning at the y-axis when z = 0, and rising 2π units each time it
makes a full revolution. Let S be the the portion of H which lies
outside the cylinder x2 + y 2 = 4, above the z = 0 plane and below
the z = 5 plane. Choose one of the following functions and give
the domain on which the function you have chosen parametrizes S.
(Hint: Only one of the following functions is possible.)

a r(u, v) = u cos v, u sin v, u

b r(u, v) = u cos v, u sin v, v

c r(u, v) = u sin v, u cos v, u

d r(u, v) = u sin v, u cos v, v
e Write down a parametrized curve of zero curvature and arclength 1.
(Be sure to specify the domain of your parametrisation.)
f If ∇ · F is a constant C on all of R3 , and S is a cube of unit volume

such that the flux outward through each side of S is 1, what is C?
g Let
F(x, y) = ax + by , cx + dy
Give the full set of a, b, c and d such that F is conservative.
h If r(s) has been parametrized by arclength (i.e. s is arclength), what
is the arclength of r(s) between s = 3 and s = 5?
i Let F be a 2D vector field which is defined everywhere except at the
points marked P and Q. Suppose that ∇ × F = 0 everywhere on
the domain of F. Consider the five curves R, S, T , U , and V shown
in the picture.
R U
Q
S
T
P
Which of the following is necessarily true?

Z Z
(1) F · dr = F · dr
ZS ZT Z Z
(2) F · dr = F · dr = F · dr = F · dr = 0
R S T U
Z Z Z Z
(3) F · dr + F · dr + F · dr = F · dr
R S T U
Z Z Z
(4) F · dr = F · dr + F · dr
U R S
Z
(5) F · dr = 0
V
j Write down a 3D vector field F such that for all closed surfaces S,
the volume enclosed by S is equal to
ZZ
F · n̂ dS
S
k Consider the vector field F in the xy-plane shown below. Is the k̂th
component of ∇ × F at P positive, negative or zero?
5. ∗. Say whether the following statements are true or false.

3
RRR , and S1 and
a If F is a 3D vector field defined on all of RR S2 are two
surfaces with the same boundary, but S1 F · n̂ dS 6= S2 F · n̂ dS,
then ∇ · F is not zero anywhere.
b If F is a vector field satisfying ∇ × F = 0 whose domain is not

simply-connected, then F is not conservative.
c The osculating circle of a curve C at a point has the same unit
tangent vector, unit normal vector, and curvature as C at that point.
d A planet orbiting a sun has period proportional to the cube of the

major axis of the orbit.
∇ × F) = 0.
e For any 3D vector field F, ∇ · (∇
f A field whose divergence is zero everywhere in its domain has closed
surfaces S in its domain.
g The gravitational force field is conservative.
h If F is a field defined on all of R3 such that C F · dr = 3 for some
R
curve C, then ∇ × F is non-zero at some point.
i The normal component of acceleration for a curve of constant cur-

vature is constant.
j The curve defined by
r1 (t) = cos(t4 )ı̂ı + 3t4̂, −∞ < t < ∞,
is the same as the curve defined by
r2 (t) = cos tı̂ı + 3t k̂, −∞ < t < ∞

6. ∗. Which of the following statements are true (T) and which are false
(F)?
All real valued functions f (x, y, z) and all vector fields F(x, y, z) have
domain R3 unless specified otherwise.
a If f is a continuous real valued function and S a smooth oriented

surface, then ZZ ZZ
f dS = − f dS
S −S
where `−S’ denotes the surface S but with the opposite orientation.
b Suppose the components

RR of the vector field F have continuous partial
derivatives. If S ∇ × F · n̂ dS = 0 for every closed smooth surface,
then F is conservative.
c Suppose S is a smooth surface bounded by a smooth simple closed

curve C. The orientation of C is determined by that of S as in
Stokes’ theorem. Suppose the real valued function f has continuous
partial derivatives. Then
Z ZZ
∂f ∂f
f dx = ̂ − k̂ · n̂ dS
C S ∂z ∂y
d Suppose the real valued function f (x, y, z) has continuous second

order partial derivatives. Then
∇f ) × (∇
(∇ ∇f ) = ∇ × (∇
∇f )
e The curve parameterized by
r(t) = 2 + 4t3 , −t3 , 1 − 2t3

−∞<t<∞
has curvature κ(t) = 0 for all t.
f If a smooth curve is parameterized by r(s) where s is arc length,

then its tangent vector satisfies
|r0 (s)| = 1
g If S isRRthe sphere x2 + y 2 + z 2 = 1 and F is a constant vector field,

then S F · n̂ dS = 0.
h There exists a vector field F whose components have continuous
second order partial derivatives such that ∇ × F = (x, y, z).
7. ∗. The vector field F = P (x, y)ı̂ı + Q(x, y) ̂ is plotted below.
C1
C2
In the following questions, give the answer that is best supported by

the plot.
a The derivative Py at the point labelled A is (a) positive, (b) negative,
(c) zero, (d) there is not enough information to tell.
b The derivative Qx at the point labelled A is (a) positive, (b) nega-
tive, (c) zero, (d) there is not enough information to tell.
c The curl of F at the point labelled A is (a) in the direction of +k̂ (b)
in the direction of −k̂ (c) zero (d) there is not enough information
to tell.
d The work done by the vector field on a particle travelling from point
B to point C along the curve C1 is (a) positive (b) negative (c) zero
(d) there is not enough information to tell.
e The work done by the vector field on a particle travelling from point
B to point C along the curve C2 is (a) positive (b) negative (c) zero
(d) there is not enough information to tell.
f The vector field F is (a) the gradient of some function f (b) the curl
of some vector field G (c) not conservative (d) divergence free.
(F)?
a The curve defined by
r1 (t) = cos(t2 )ı̂ı + sin(t2 ) ̂ + 2t2 k̂, −∞ < t < ∞

r2 (t) = cos tı̂ı + sin t ̂ + 2t k̂, −∞ < t < ∞
b The curve defined by
r1 (t) = cos(t2 )ı̂ı + sin(t2 ) ̂ + 2t2 k̂, 0≤t≤1
r2 (t) = cos tı̂ı + sin t ̂ + 2t k̂, 0≤t≤1
c If a smooth curve is parameterized by r(s) where s is arc length,

then its tangent vector satisfies
|r0 (s)| = 1
d If r(t) defines a smooth curve C in space that has constant curvature

κ > 0, then C is part of a circle with radius 1/κ.
e If the speed of a moving object is constant, then its acceleration is
orthogonal to its velocity.
f The vector field
−y x
F(x, y, z) = ı̂ı + 2 ̂ + z k̂
x2
+y 2 x + y2
is conservative.
g Suppose the vector field F(x, y, z) is defined on an open domain and
its components have continuous partial derivatives. If ∇ × F = 0,
then F is conservative.

h The region D = (x, y) x2 + y 2 > 1 is simply connected.

i The region D = (x, y) y − x2 > 0 is simply connected.
j If F is a vector field whose components have two continuous partial

derivatives, then ZZ
∇ × F · n̂ dS = 0
S
when S is the boundary of a solid region E in R3 .

(F)?
a If a smooth curve C is parameterized by r(s) where s is arc length,
then the tangent vector r0 (s) satisfies |r0 (s)| = 1.
b If r(t) defines a smooth curve C in space that has constant curvature
κ > 0, then C is part of a circle with radius 1/κ.
c Suppose F is a continuous vector field with open domain D. If
Z
F · dr = 0
C
for every piecewise smooth closed curve C in D, then F is conser-

vative.
d Suppose F is a vector field with open domain D, and the components

of F have continuous partial derivatives. If ∇ × F = 0 everywhere
on D, then F is conservative.
e The curve defined by
r1 (t) = cos(t2 )ı̂ı + sin(t2 ) ̂ + 2t2 k̂, −∞ < t < ∞
r2 (t) = cos tı̂ı + sin t ̂ + 2t k̂, −∞ < t < ∞
f The curve defined by
r1 (t) = cos(t2 )ı̂ı + sin(t2 ) ̂ + 2t2 k̂, 0≤t≤1
r2 (t) = cos tı̂ı + sin t ̂ + 2t k̂, 0≤t≤1
g Suppose F(x, y, z) is a vector field whose components have continu-

∇ × F ) = 0.
ous second order partial derivatives. Then ∇ · (∇
h Suppose the real valued function f (x, y, z) has continuous second
∇f ) = 0.
order partial derivatives. Then ∇ · (∇
2
i The region D = (x, y) x + y 2 > 1 is simply connected.

j The region D = (x, y) y − x2 > 0 is simply connected.
10. ∗. Let F, G be vector fields, and f , g be scalar fields. Assume F, G, f ,
g are defined on all of R3 and have continuous partial derivatives of all
orders everywhere. Mark each of the following as True (T) or False (F).
R
a If C is a closed curve and ∇ f = 0, then C f ds = 0.
b If r(t) is a parametrization of a smooth curve C and the binormal
B(t) is constant then C is a straight line.
c If r(t) is the position of a particle which travels with constant speed,
then r0 (t) · r00 (t) = 0.
R
d If C is a path from points A to B, then the line integral C F×G ·dr
is independent of the path C.
R
e The line integral C f ds does not depend of the orientation of the
curve C.
f If S is a parametric surface r(u, v) then a normal to S is given by
∂r ∂r
×
∂u ∂u
g The surface area of the parametric surface S given by r(u, v) =
x(u, v)ı̂ı + y(u, v) ̂ + z(u, v) k̂, (u, v) ∈ D, is given by
ZZ
∂z 2
1/2
∂z 2
1 + ∂u + ∂v dudv
D
h If F is the velocity field of an incompressible fluid then ∇ · F = 0.

i ∇ · F × G = (∇ ∇ · F)G + (∇ ∇ · G)F
11. ∗. Say whether the following statements are true (T) or false (F). You
may assume that all functions and vector fields are defined everywhere
and have derivatives of all orders everywhere.
a The divergence of ∇ × F is zero, for every F.
R
b In a simply connected region, C F·dr depends only on the endpoints
of C.
c If ∇ f = 0, then f is a constant function.
d If ∇ × F = 0, then F is a constant vector field.

RR
e If ∇ · F = 0, then S F · n̂ dS = 0 for every closed surface S.
R
f If C F · dr = 0 for every closed curve C, then ∇ × F = 0.
g If r(t) is a path in three space with constant speed |v(t)|, then the
acceleration is perpendicular to the tangent vector, i.e. a · T̂ = 0.
h If r(t) is a path in three space with constant curvature κ, then r(t)
parameterizes part of a circle of radius 1/κ.
i Let F be a vector field and suppose that S1 and S2 are oriented
surfaces with the same boundary curve C, and C is given the direc-
tion
RR that is compatible
RR with the orientations of S1 and S2 . Then
S1
F · n̂ dS = S2
F · n̂ dS.
j Let A(t) be the area swept out by the trajectory of a planet from
time t = 0 to time t. The dA
dt is constant.
12. ∗. Find the correct identity, if f is a function and G and F are vector
fields. Select the true statement.
∇ × (F) + (∇
a ∇ · (f F) = f∇ ∇f ) × F
∇ · (F) + F · ∇ f
b ∇ · (f F) = f∇
∇ · (F) + F · ∇ f
c ∇ × (f F) = f∇
d None of the above are true.
13. ∗. True or False. Consider vector fields F and scalar functions f and
g which are defined and smooth in all of three-dimensional space. Let
r = (x, y, z) represent a variable point in space, and let ω = (ω1 , ω2 , ω3 )
be a constant vector. Let Ω be a smoothly bounded domain with outer
normal n̂. Which of the following are identites, always valid under these
assumptions?
a ∇ · ∇f = 0
b F × ∇f = f ∇ · F
c ∇ 2 f = ∇ (∇
∇ · f)
d ∇ × ∇f = 0
∇ × f ) + (∇
e (∇ ∇ × g) = ∇ f × ∇ g
f ∇ ·∇ ×F = 0
r
g ∇· |r|2 = 0 for r 6= 0
h ∇ × (ω
ω × r) = 0
ZZZ ZZZ ZZ
i ∇ · F dV = −
f∇ ∇ f · F dV + f F · n̂ dS
Ω Ω ∂Ω
ZZ ZZZ
j f n̂ dS = − ∇ f dV
∂Ω Ω
14. ∗. Determine if the given statements are True or False. Provide a reason
or a counterexample.
a A constant vector field is conservative on R3 .
b If ∇ · F = 0 for all points in the domain of F then F is a constant

vector field.
c Let r(t) be a parametrization of a curve C in R3 . If r(t) and dr
dt are
orthogonal at all points of the curve C, then C lies on the surface
of a sphere x2 + y 2 + z 2 = a2 for some a > 0.
d The curvature κ at a point on a curve depends on the orientation of

the curve.
e The domain of a conservative vector field must be simply connected.
15. ∗. Provide a short answer to each question.

a Compute ∇ · x2 y ı̂ı + ey sin x ̂ + ezx k̂

b Compute ∇ × (cos x2 ı̂ı − y 3 z ̂ + xz k̂
c Let
x y
F= ı̂ı + 2 ̂ + z 2 k̂
x2 + y 2 x + y2
and let D be the domain of F. Consider the following four statments.
(I) D is connected
(II) D is disconnected
(III) D is simply connected
(IV) D is not simply connected
Choose one of the following:
(i) (II) and (III) are true
(ii) (I) and (III) are true
(iii) (I) and (IV) are true
(iv) (II) and (IV) are true
(v) Not enough information to determine
d True or False? If the speed of a particle is constant then the ac-

celeration of the particle is zero. If your answer is True, provide a
reason. If your answer is False, provide a counter example.
16. ∗. Are each of the following statements True or False? Recall that f ∈ C k
means that all derivatives of f up to order k exist and are continuous.
∇f ) = 0 for all C 2 scalar functions f in R3 .
a ∇ × (f∇
∇ · F for all C 1 scalar functions f and C 1
b ∇ · (f F) = ∇ f · F + f∇
3
vector fields F in R .
c A smooth space curve C with constant curvature κ = 0 must be a

part of a straight line.

d A smooth space curve C with constant curvature κ 6= 0 must be
part of a circle of radius 1/κ.
e RIf f is any smooth function defined in R3 and if C is any circle, then
C
∇ f · dr = 0.
f Suppose F is a smooth vector field in R3 and ∇ · F = 0 everywhere.
Then, for every sphere, the flux out of one hemisphere is equal to
the flux into the opposite hemisphere.
g Let F(x, y, z) be a continuously RR
differentiable vector field which is
defined for every (x, y, z). Then, S ∇ × F · n̂ dS = 0 for any closed
surface S. (A closed surface is a surface that is the boundary of a
solid region.)
17. ∗. True or false (reasons must be given):
a If a smooth vector field on R3 is curl free and divergence free, then
its potential is harmonic. By definition, φ(x, y, z) is harmonic if
∂2 ∂2 ∂2

∂x2 + ∂y 2 + ∂z 2 φ(x, y, z) = 0.
b If F is a smooth conservative vector field on R3 , then its flux through

any smooth closed surface is zero.
18. ∗. The following statements may be true or false. Decide which. If true,
give a proof. If false, provide a counter-example.
3
R f is any smooth function defined in R and if C is any circle, then
a If
C
∇ f · dr = 0.
b There is a vector field F that obeys ∇ × F = xı̂ı + y ̂ + z k̂.

19. ∗. Short answers:
R
a Let S be the level surface f (x, y, z) = 0. Why is C ∇ f · dr = 0 for
any curve C on S?
b A point moving in space with position r(t) at time t satisfies the
condition a(t) = f (t)r(t) for all t for some real valued function f .
Why is v × r a constant vector?
c Why is the trajectory of the point in (b) contained in a plane?
d Is the binormal vector, B̂, of a particle moving in space, always
orthogonal to the unit tangent vector T̂ and unit normal N̂?
e If the curvature of the path of a particle moving in space is constant,
is the acceleration zero when maximum speed occurs?
20. ∗. A region R is bounded by a simple closed curve C. The curve C
is oriented such that R lies to the left of C when walking along C in the
direction of C. Determine whether or not each of the following expressions
is equal to the area of R. You must justify your conclusions.
Z
1
a −y dx + x dy
2 C
Z
1
b −x dx + y dy
2 C
Z
c y dx
C
Z
d 3y dx + 4x dy
C
21. ∗. Say whether each of the following statements is true or false and explain
why.
a A moving particle has velocity and acceleration vectors that satisfy
|v| = 1 and |a| = 1 at all times. Then the curvature of this particle’s
path is a constant.
b If F is any smooth vector field defined in R3 and if S is any sphere,
then ZZ
∇ × F · n̂ dS = 0
S
Here n̂ is the outward normal to S.
I I
3
c If F and G are smooth vector fields in R and if F·dr = G·dr
C C
for every circle C, then F = G.
22. ∗. Three quickies:
a A moving particle with position r(t) = (x(t), y(t), z(t)) satisfies
a = f (r, v)r
for some scalar-valued function f . Prove that r × v is constant.

b Calculate S (xı̂ı − y ̂ + z 2 k̂) · n̂dS, where S is the boundary of any
RR
solid right circular cylinder of radius b with one base in the plane
z = 1 and the other base in the plane z = 3.
c Let F and G be smooth vector
H fields defined
RR in R3 . Suppose that,
for every circle C, we have C F · dr = S G · n̂ dS, where S is the
oriented disk with boundary C. Prove that G = ∇ × F.
Appendix A
Appendices
A.1 Trigonometry
A.1.1 Trigonometry — Graphs
sin θ cos θ tan θ

1 1
−π −π π π 3π 2π −π −π π π 3π 2π −π −π π π 3π 2π
2 2 2 2 2 2 2 2 2
−1 −1
A.1.2 Trigonometry — Special Triangles
From the above pair of special triangles we have

√
π 1 π 1 π 3
sin = √ sin = sin =
4 2 6 2 3 2
√
π 1 π 3 π 1
cos = √ cos = cos =
4 2 6 2 3 2
π π 1 π √
tan = 1 tan = √ tan = 3
4 6 3 3
312
APPENDIX A. APPENDICES 313
A.1.3 Trigonometry — Simple Identities

• Periodicity
sin(θ + 2π) = sin(θ) cos(θ + 2π) = cos(θ)
• Reflection
sin(−θ) = − sin(θ) cos(−θ) = cos(θ)
• Reflection around π/4
sin π2 − θ = cos θ π

cos 2 − θ = sin θ
• Reflection around π/2
sin (π − θ) = sin θ cos (π − θ) = − cos θ
• Rotation by π
sin (θ + π) = − sin θ cos (θ + π) = − cos θ
• Pythagoras
sin2 θ + cos2 θ = 1
tan2 θ + 1 = sec2 θ
1 + cot2 θ = csc2 θ
• sin and cos building blocks

sin θ 1 1 cos θ 1
tan θ = csc θ = sec θ = cot θ = =
cos θ sin θ cos θ sin θ tan θ
A.1.4 Trigonometry — Add and Subtract Angles

• Sine
sin(α ± β) = sin(α) cos(β) ± cos(α) sin(β)
• Cosine
cos(α ± β) = cos(α) cos(β) ∓ sin(α) sin(β)
• Tangent
tan α + tan β
tan(α + β) =
1 − tan α tan β
tan α − tan β
tan(α − β) =
1 + tan α tan β
• Double angle
sin(2θ) = 2 sin(θ) cos(θ)

cos(2θ) = cos2 (θ) − sin2 (θ)
= 2 cos2 (θ) − 1
= 1 − 2 sin2 (θ)
2 tan(θ)
tan(2θ) =
1 − tan2 θ
1 + cos(2θ)
cos2 θ =
2
1 − cos(2θ)
sin2 θ =
2
1 − cos(2θ)
tan2 θ =
1 + cos(2θ)
• Products to sums
sin(α + β) + sin(α − β)
sin(α) cos(β) =
2
cos(α − β) − cos(α + β)
sin(α) sin(β) =
2
cos(α − β) + cos(α + β)
cos(α) cos(β) =
2
• Sums to products
α+β α−β
sin α + sin β = 2 sin cos
2 2
α+β α−β
sin α − sin β = 2 cos sin
2 2
α+β α−β
cos α + cos β = 2 cos cos
2 2
α+β α−β
cos α − cos β = −2 sin sin
2 2
A.1.5 Inverse Trigonometric Functions
arctan x
arcsin x arccos x
Domain: −1 ≤ x ≤ 1 Domain: −1 ≤ x ≤ 1 Domain: all real num-
bers
Range: − π2 ≤ Range: 0 ≤ arccos x ≤ π Range: − π2 <
π π
arcsin x ≤ 2 arctan x < 2
π π
2 π 2
π
−1 1 2
−π
2 −π
2
−1 1
Since these functions are inverses of each other we have

π π
arcsin(sin θ) = θ − ≤θ≤
2 2
arccos(cos θ) = θ 0≤θ≤π
π π
arctan(tan θ) = θ − ≤θ≤
2 2
and also
sin(arcsin x) = x −1 ≤ x ≤ 1
cos(arccos x) = x −1 ≤ x ≤ 1
tan(arctan x) = x any real x
arccsc x arcsec x arccot x

Domain: |x| ≥ 1 Domain: |x| ≥ 1 Domain: all real num-
bers
Range: − π2 ≤ Range: 0 ≤ arcsec x ≤ π Range: 0 < arccot x < π
π
arccsc x ≤ 2
π π
2 π
π
π 2
2
−1 1
−π
2
−1 1
π
arccsc x 6= 0 arcsec x 6=
2
Again
π π
arccsc(csc θ) = θ − ≤ θ ≤ , θ 6= 0
2 2
π
arcsec(sec θ) = θ 0 ≤ θ ≤ π, θ 6=
2
arccot(cot θ) = θ 0<θ<π
and
csc(arccsc x) = x |x| ≥ 1
sec(arcsec x) = x |x| ≥ 1
cot(arccot x) = x any real x
A.2 Powers and Logarithms

A.2.1 Powers
In the following, x and y are arbitrary real numbers, q is an arbitrary constant
that is strictly bigger than zero and e is 2.7182818284, to ten decimal places.
• e0 = 1, q0 = 1
ex qx
• ex+y = ex ey , ex−y = , q x+y = q x q y , q x−y =
ey qy
1 1
• e−x = , q −x =
ex qx
y y
• ex = exy , qx = q xy
d x d g(x) d x
• e = ex , e = g 0 (x)eg(x) , q = (ln q) q x
dx dx dx
• ex dx = ex + C, e dx = a1 eax + C if a 6= 0
R R ax
∞
X xn
• ex =
n=0
n!
• lim ex = ∞, lim ex = 0
x→∞ x→−∞
lim q x = ∞, lim q x = 0 if q > 1
x→∞ x→−∞
lim q x = 0, lim q x = ∞ if 0 < q < 1
x→∞ x→−∞
• The graph of 2x is given below. The graph of q x , for any q > 1, is similar.
y y = 2x
6
2
1
x
−3 −2 −1 1 2 3
A.2.2 Logarithms
In the following, x and y are arbitrary real numbers that are strictly bigger
than 0 (except where otherwise specified), p and q are arbitrary constants that
are strictly bigger than one, and e is 2.7182818284, to ten decimal places. The
notation ln x means loge x. Some people use log x to mean log10 x, others use
it to mean loge x and still others use it to mean log2 x.
• eln x = x, q logq x = x

• ln ex = x, logq q x = x for all −∞ < x < ∞
ln x logp x logp x
• logq x = , ln x = , logq x =
ln q logp e logp q
• ln 1 = 0, ln e = 1
logq 1 = 0, logq q = 1
• ln(xy) = ln x + ln y, logq (xy) = logq x + logq y
x x
• ln = ln x − ln y, logq = logq x − logq y
y y
1 1
• ln = − ln y, logq = − logq y
y y
• ln(xy ) = y ln x, logq (xy ) = y logq x
d 1 d 1
• ln x = , logq x =
dx x dx x ln q
Z Z
x
• ln x dx = x ln x − x + C, logq x dx = x logq x − +C
ln q
• lim ln x = ∞, lim ln x = −∞
x→∞ x→0+
lim logq x = ∞, lim logq x = −∞
x→∞ x→0+
• The graph of log10 x is given below. The graph of logq x, for any q > 1,
is similar.
y
y = log10 x
1.0
0.5
x
1 5 10 15
−0.5
−1.0
A.3 Table of Derivatives

Throughout this table, a and b are constants, independent of x.
F (x) F 0 (x) = dFdx
af (x) + bg(x) af 0 (x) + bg 0 (x)
f (x) + g(x) f 0 (x) + g 0 (x)
f (x) − g(x) f 0 (x) − g 0 (x)
af (x) af 0 (x)
f (x)g(x) f 0 (x)g(x) + f (x)g 0 (x)
f (x)g(x)h(x) f 0 (x)g(x)h(x) + f (x)g 0 (x)h(x) + f (x)g(x)h0 (x)
f (x) f 0 (x)g(x)−f (x)g 0 (x)
g(x) g(x)2
1 g 0 (x)
g(x) − g(x)2
0
0
f g(x) f g(x) g (x)
F (x) F 0 (x) = dFdx

a 0
xa axa−1
g(x)a ag(x)a−1 g 0 (x)
sin x cos x
sin g(x) g 0 (x) cos g(x)
cos x − sin x
cos g(x) −g 0 (x) sin g(x)
tan x sec2 x
csc x − csc x cot x
sec x sec x tan x
cot x − csc2 x
ex ex
eg(x) g 0 (x)eg(x)
ax (ln a) ax
F (x) F 0 (x) = dF
dx
1
ln x x0
g (x)
ln g(x) g(x)
1
loga x x ln a
arcsin x √ 1
1−x2
g 0 (x)
arcsin g(x) √
1−g(x)2
1
arccos x − √1−x 2
1
arctan x 1+x2
g 0 (x)
arctan g(x) 1+g(x)2
arccsc x − |x|√1x2 −1
arcsec x √1
|x| x2 −1
1
arccot x − 1+x 2
A.4 Table of Integrals

Throughout this table, a and b are given constants, independent of x and C is
an arbitrary constant.
R
f (x) F (x) = f (x) dx
R R
af (x) + bg(x) a f (x) dx + b g(x) dx + C
R R
f (x) + g(x) f (x) dx + g(x) dx + C
R R
f (x) − g(x) f (x) dx − g(x) dx + C
R
af (x) a f (x) dx + C
u(x)v 0 (x) u(x)v(x) − u0 (x)v(x) dx + C
R
f y(x) y 0 (x)
R
F y(x) where F (y) = f (y) dy
a ax + C
xa+1
xa a+1 + C if a 6= −1
1
x ln |x| + C
g(x)a+1
g(x)a g 0 (x) a+1 + C if a 6= −1
R
f (x) F (x) = f (x) dx
sin x − cos x + C
g 0 (x) sin g(x) − cos g(x) + C
cos x sin x + C
tan x ln | sec x| + C
csc x ln | csc x − cot x| + C
sec x ln | sec x + tan x| + C
cot x ln | sin x| + C
sec2 x tan x + C
csc2 x − cot x + C
sec x tan x sec x + C
csc x cot x − csc x + C
R
f (x) F (x) = f (x) dx
ex ex + C
eg(x) g 0 (x) eg(x) + C
1 ax
eax a e +C
1
ax ln a a x
+C
ln x x ln x − x + C
√ 1 arcsin x + C
1−x2
g 0 (x)
√ arcsin g(x) + C
1−g(x)2
√ 1
a2 −x2
arcsin xa + C
1
1+x2 arctan x + C
g 0 (x)
1+g(x)2 arctan g(x) + C
1 1 x
a2 +x2 a arctan a + C
√1 arcsec x + C \quad(x > 1)
x x2 −1
A.5 Table of Taylor Expansions

Let n ≥ be an integer. Then if the function f has n + 1 derivatives on an
interval that contains both x0 and x, we have the Taylor expansion
1 00 1
f (x) = f (x0 ) + f 0 (x0 ) (x − x0 ) + f (x0 ) (x − x0 )2 + · · · + f (n) (x0 ) (x − x0 )n
2! n!
1
+ f (n+1) (c) (x − x0 )n+1 for some c between x0 and x
(n + 1)!
The limit as n → ∞ gives the Taylor series
∞
X f (n) (x0 )
f (x) = (x − x0 )n
n=0
n!
for f . When x0 = 0 this is also called the Maclaurin series for f . Here are
Taylor series expansions of some important functions.
∞
X 1 n
ex = x for − ∞ < x < ∞
n=0
n!
1 1 1
= 1 + x + x2 + x3 + · · · + xn + · · ·
2 3! n!
∞
X (−1)n 2n+1
sin x = x for − ∞ < x < ∞
n=0
(2n + 1)!
1 3 1 (−1)n 2n+1
=x− x + x5 − · · · + x + ···
3! 5! (2n + 1)!
∞
X (−1)n 2n
cos x = x for − ∞ < x < ∞
n=0
(2n)!
1 2 1 (−1)n 2n
=1− x + x4 − · · · + x + ···
2! 4! (2n)!
∞
1 X
= xn for − 1 ≤ x < 1
1 − x n=0
= 1 + x + x2 + x3 + · · · + xn + · · ·
∞
1 X
= (−1)n xn for − 1 < x ≤ 1
1+x n=0
= 1 − x + x2 − x3 + · · · + (−1)n xn + · · ·
∞
X 1 n
ln(1 − x) = − x for − 1 ≤ x < 1
n=1
n
1 1
= −x − 12 x2 − x3 − · · · − xn − · · ·
3 n
∞
X (−1)n n
ln(1 + x) = − x for − 1 < x ≤ 1
n=1
n
1 (−1)n n
= x − 12 x2 + x3 − · · · − x − ···
3 n
p(p − 1) 2 p(p − 1)(p − 2) 3
(1 + x)p = 1 + px + x + x + ···
2 3!
p(p − 1)(p − 2) · · · (p − n + 1) n
+ x + ···
n!
A.6 3d Coordinate Systems

A.6.1 Cartesian Coordinates
Here is a figure showing the definitions of the three Cartesian coordinates
(x, y, z)
z
px, y, zq
z
y
x
y px, y, 0q
x
and here are three figures showing a surface of constant x, a surface of
constant x, and a surface of constant z.
z z z
p0, 0, zq
y y y
px, 0, 0q p0, y, 0q
x x x
surface of constant x surface of constant y surface of constant z
(a plane) (a plane) (a plane)
Finally here is a figure showing the volume element dV in cartesian coor-
dinates.
dz
dx
dy
volume element dV “ dx dy dz
A.6.2 Cylindrical Coordinates

Here is a figure showing the definitions of the three cylindrical coordinates
r = distance from (0, 0, 0) to (x, y, 0)

θ = angle between the the x axis and the line joining (x, y, 0) to (0, 0, 0)
z = signed distance from (x, y, z) to the xy-plane
px, y, zq
z
y
θ r
x px, y, 0q
The cartesian and cylindrical coordinates are related by
x = r cos θ y = r sin θ z=z

p y
r = x2 + y 2 θ = arctan z=z
x
Here are three figures showing a surface of constant r, a surface of constant θ,
and a surface of constant z.
z z
z
p0, 0, zq
r y
y y
θ
x
x x
surface of constant r surface of constant z
surface of constant θ
(a cylindrical shell) (a plane)
(a plane)
Finally here is a figure showing the volume element dV in cylindrical coor-
dinates.
r dr
dz
r dθ
volume element dV “ r dr dθ dz
A.6.3 Spherical Coordinates

Here is a figure showing the definitions of the three spherical coordinates
ρ = distance from (0, 0, 0) to (x, y, z)

ϕ = angle between the z axis and the line joining (x, y, z) to (0, 0, 0)
θ = angle between the x axis and the line joining (x, y, 0) to (0, 0, 0)
z
p0, 0, zq ρ sin ϕ
px, y, zq
ρ cos ϕ
ϕ ρ
y
px, 0, 0q θ
px, y, 0q
ρ sin ϕ sin θ
x
and here are two more figures giving the side and top views of the previous
figure.
z
ρ sin ϕ px, y, zq y
p0, 0, zq
θ ρ sin ϕ
ρ cos ϕ ρ sin ϕ cos θ
ρ
ϕ px, y, 0q
px, 0, 0q
ρ sin ϕ sin θ
px, y, 0q
x
side view top view
The cartesian and spherical coordinates are related by
x = ρ sin ϕ cos θ y = ρ sin ϕ sin θ z = ρ cos ϕ

p
p y x2 + y 2
ρ = x2 + y 2 + z 2 θ = arctan ϕ = arctan
x z
Here are three figures showing a surface of constant ρ, a surface of constant θ,
and a surface of constant ϕ.
z z z
ϕ
ρ
y y y
θ
x x x
surface of constant ρ surface of constant θ surface of constant ϕ
(a sphere) (a plane) (a cone)
Here is a figure showing the surface element dS in spherical coordinates
ρ sin ϕ dθ
ρ dϕ
x surface element dS “ ρ2 sin ϕ dθ dϕ

and two extracts of the above figure to make it easier to see how the factors
ρ dϕ and ρ sin ϕ dθ arise.
z z
ρ sin ϕ dθ
dθ
ρ sin ϕ
ρ dϕ
ϕ ϕ ρ
dϕ
y y
x x
Finally, here is a figure showing the volume element dV in spherical coor-
dinates
ρ sin ϕ dθ
ρ dϕ
dρ
x
volume element dV “ ρ2 sin ϕ dρ dθ dϕ
A.7 ISO Coordinate System Notation

In this text we have chosen symbols for the various polar, cylindrical and spher-
ical coordinates that are standard for mathematics. There is another, different,
set of symbols that are commonly used in the physical sciences and engineering.
Indeed, there is an international convention, called ISO 80000-2, that specifies
those symbols1 . In this appendix, we summarize the definitions and standard
properties of the polar, cylindrical and spherical coordinate systems using the
ISO symbols.
A.7.1 Polar Coordinates

In the ISO convention the symbols ρ and φ are used (instead of r and θ) for
polar coordinates.
ρ = the distance from (0, 0) to (x, y)
φ = the (counter-clockwise) angle between the x-axis
and the line joining (x, y) to (0, 0)
y
px, yq
ρ
φ
x
Cartesian and polar coordinates are related by
x = ρ cos φ y = ρ sin φ
1 It specifies more than just those symbols. See https://en.wikipedia.org/wiki/ISO_31-
11 and https://en.wikipedia.org/wiki/ISO/IEC_80000 The full ISO 80000-2 is available at
https://www.iso.org/standard/64973.html — for $$.
p y
ρ= x2 + y 2 φ = arctan
x
The following two figures show a number of lines of constant φ, on the left,
and curves of constant ρ, on the right.
y y
x x
lines of constant φ curves of constant ρ

Note that the polar angle φ is only defined up to integer multiples of 2π.
For example, the point (1, 0) on the x-axis could have φ = 0, but could also
have φ = 2π or φ = 4π. It is sometimes convenient to assign φ negative values.
When φ < 0, the counter-clockwise angle φ refers to the clockwise angle |φ|.
For example, the point (0, −1) on the negative y-axis can have φ = − π2 and
can also have φ = 3π
2 .
y
3π{2
x
π{2
ρ “ 1, φ “ ´π{2, φ “ 3π{2
It is also sometimes convenient to extend the above definitions by saying
that x = ρ cos φ and y = ρ sin φ even when ρ is negative. For example, the
following figure shows (x, y) for ρ = 1, φ = π4 and for ρ = −1, φ = π4 .
y
ρ “ 1, φ “ π{4
π{4
x
ρ “ ´1, φ “ π{4
Both points lie on the line through the origin that makes an angle of 45◦
with the x-axis and both are a distance one from the origin. But they are on
opposite sides of the the origin.
The area element in polar coordinates is
dA = ρ dρ dφ
ρ ρ dφ
dφ dρ
A.7.2 Cylindrical Coordinates

In the ISO convention the symbols ρ, φ and z are used (instead of r, θ and z)
for cylindrical coordinates.
ρ = distance from (0, 0, 0) to (x, y, 0)

φ = angle between the the x axis and the line joining (x, y, 0) to (0, 0, 0)
z = signed distance from (x, y, z) to the xy-plane
z
px, y, zq
z
y
ρ
φ
x px, y, 0q
The cartesian and cylindrical coordinates are related by
x = ρ cos φ y = ρ sin φ z=z

p y
ρ = x2 + y 2 φ = arctan z=z
x
Here are three figures showing a surface of constant ρ, a surface of constant φ,
and a surface of constant z.
z z
z
p0, 0, zq
ρ y y y
φ
x x x
surface of constant ρ surface of constant φ surface of constant z
(a cylindrical shell) (half a plane) (a plane)
Finally here is a figure showing the volume element dV in cylindrical coor-
dinates.
ρ dρ
dz
ρ dφ
volume element dV “ ρ dρ dφ dz
A.7.3 Spherical Coordinates

In the ISO convention the symbols r (instead of ρ), φ (instead of θ) and θ
(instead of φ) are used for spherical coordinates.
r = distance from (0, 0, 0) to (x, y, z)

θ = angle between the z axis and the line joining (x, y, z) to (0, 0, 0)
φ = angle between the x axis and the line joining (x, y, 0) to (0, 0, 0)
z
p0, 0, zq r sin θ
px, y, zq
r cos θ
r
θ
y
px, 0, 0q φ
px, y, 0q
r sin θ sin φ
x
Here are two more figures giving the side and top views of the previous
figure.
z
r sin θ px, y, zq y
p0, 0, zq
φ r sin θ
r cos θ r sin θ cos φ
r
θ px, y, 0q
px, 0, 0q
r sin θ sin φ
px, y, 0q
x
side view top view
The cartesian and spherical coordinates are related by
x = r sin θ cos φ y = r sin θ sin φ z = r cos θ

p
p y x2 + y 2
r = x2 + y 2 + z 2 φ = arctan θ = arctan
x z
Here are three figures showing a surface of constant r, a surface of constant φ,
and a surface of constant θ.
z z z
θ
r
y y y
φ
x x x
surface of constant r surface of constant φ surface of constant θ
(a sphere) (half a plane) (a cone)
Here is a figure showing the surface element dS in spherical coordinates
r sin θ dφ
r dθ
x surface element dS “ r 2 sin θ dθ dφ

r dθ and r sin θ dφ arise.
z z
r sin θ dφ
dφ
r sin θ
r dθ
r
θ dθ θ r
y y
x x
Finally, here is a figure showing the volume element dV in spherical coor-
dinates
r sin θ dφ
r dθ
dr
x
volume element dV “ r 2 sin θ dr dθ dφ
r dθ and r sin θ dφ arise.
z
z
r sin θ dφ
dφ
r sin θ
r dθ
r
θ dθ θ
dr y
y
x
x
A.8 Conic Sections and Quadric Surfaces

A conic section is the curve of intersection of a cone and a plane that does not
pass through the vertex of the cone. This is illustrated in the figures below.
circle ellipse parabola

hyperbola
1
An equivalent (and often used) definition is that a conic section is the set
of all points in the xy-plane that obey Q(x, y) = 0 with
Q(x, y) = Ax2 + By 2 + Cxy + Dx + Ey + F = 0
being a polynomial of degree two2 . By rotating and translating our coordinate

system the equation of the conic section can be brought into one of the forms3
• αx2 + βy 2 = γ with α, β, γ > 0, which is an ellipse (or a circle),

• αx2 − βy 2 = γ with α, β > 0, γ 6= 0, which is a hyperbola,
• x2 = δy, with δ 6= 0 which is a parabola.
The three dimensional analogs of conic sections, surfaces in three dimen-

sions given by quadratic equations, are called quadrics. An example is the
sphere x2 + y 2 + z 2 = 1.
Here are some tables giving all of the quadric surfaces.
name elliptic cylin- parabolic hyperbolic

sphere
der cylinder cylinder
equation in x2 2
x2 y2
+ yb2 = 1 y = ax2 − =1 x2+y 2+z 2 = r2
standard form a2 a2 b2
x =constant
two lines one line two lines circle
cross-section
y =constant
two lines two lines two lines circle
cross-section
z =constant
ellipse parabola hyperbola circle
cross-section
sketch
Figure A.8.1 Table of conic sections

1 It is outside our scope to prove this equivalence.
2 Technically, we should also require that the constants A, B, C, D, E, F , are real numbers,
that A, B, C are not all zero, that Q(x, y) = 0 has more than one real solution, and that
the polynomial can’t be factored into the product of two polynomials of degree one.
3 This statement can be justified using a linear algebra eigenvalue/eigenvector analysis. It
is beyond what we can cover here, but is not too difficult for a standard linear algeba course.
elliptic
name ellipsoid elliptic cone
paraboloid
equation in stan- x2 y2 z2 x2 y2 z x2 y2 z2
a2 + b2 + c2 =1 a2 + b2 = c a2 + b2 = c2
dard form
two lines if x = 0,
x = constant
ellipse parabola hyperbola if x 6=
cross-section
0
two lines if
y = constant
ellipse parabola y = 0,hyperbola
cross-section
if y 6= 0
z = constant
ellipse ellipse ellipse
cross-section
sketch
Figure A.8.2 Table of quadric surfaces-1
name hyperboloid of hyperboloid of hyperbolic

one sheet two sheets paraboloid
equation in stan- x2 y2 z2 x2
2 2 y2 x2
a2 + b2 − c2 =1 a2 + yb2 − zc2 = −1 b2 − a2 = z
c
dard form
x = constant
hyperbola hyperbola parabola
cross-section
y = constant
hyperbola hyperbola parabola
cross-section
two lines if z = 0,
z = constant
ellipse ellipse hyperbola if z = 6
cross-section
0
sketch
Figure A.8.3 Table of quadric surfaces-2

A.9 Review of Linear Ordinary Differential Equa-

tions
Definition A.9.1
a A differential equation is an equation for an unknown function that
contains the derivatives of that unknown function. For example y 00 (t) +
y(t) = 0 is a differential equation for the unknown function y(t).
b A differential equation is called an ordinary differential equation (of-
ten shortened to “ODE”) if only ordinary derivatives appear. That is, if
the unknown function has only a single independent variable. A differen-
tial equation is called a partial differential equation (often shortened
to “PDE”) if partial derivatives appear. That is, if the unknown function
has more than one independent variable. For example y 00 (t) + y(t) = 0 is
2 2
an ODE while ∂∂ tu2 (x, t) = c2 ∂∂ xu2 (x, t) is a PDE.
c The order of a differential equation is the order of the highest derivative
that appears. For example y 00 (t) + y(t) = 0 is a second order ODE.
d An ordinary differential equation that is of the form
a0 (t)y (n) (t) + a1 (t)y (n−1) (t) + · · · + an (t)y(t) = F (t) (A.9.1)
with given coefficient functions a0 (t), · · ·, an (t) and F (t) is said to be

linear. Otherwise, the ODE is said to be nonlinear. For example,
y 0 (t)2 + y(t) = 0, y 0 (t)y 00 (t) + y(t) = 0 and y 0 (t) = ey(t) are all nonlinear.
e The ODE (A.9.1) is said to have constant coefficients if the coefficients

a0 (t), a1 (t), · · ·, an (t) are all constants. Otherwise, it is said to have
variable coefficients. For example, the ODE y 00 (t) + 7y(t) = sin t is
constant coefficient, while y 00 (t) + ty(t) = sin t is variable coefficient.
f The ODE (A.9.1) is said to be homogeneous if F (t) is identically
zero. Otherwise, it is said to be inhomogeneous or nonhomoge-
neous. For example, the ODE y 00 (t) + 7y(t) = 0 is homogeneous, while
y 00 (t) + 7y(t) = sin t is inhomogeneous. A homogeneous ODE always has
the trivial solution y(t) = 0.
g An initial value problem is a problem in which one is to find an un-
known function y(t) that satisfies both a given ODE and given initial
conditions, like y(t0 ) = 1, y 0 (t0 ) = 0. Note that all of the conditions
involve the function y(t) (or its derivatives) evaluated at a single time
t = t0 .
h A boundary value problem is a problem in which one is to find an
unknown function y(t) that satisfies both a given ODE and given bound-
ary conditions, like y(t0 ) = 0, y(t1 ) = 0. Note that the conditions involve
the function y(t) (or its derivatives) evaluated at two different times.
♦
The following theorem gives the form of solutions to the ODE (A.9.1).
Theorem A.9.2 Assume that the coefficients a0 (t), a1 (t), · · ·, an−1 (t), an (t)
and F (t) are continuous functions and that a0 (t) is not zero.
a The general solution to the ODE (A.9.1) is of the form
y(t) = yp (t) + C1 y1 (t) + C2 y2 (t) + · · · + Cn yn (t) (A.9.2)
where
• n is the order of (A.9.1)
• yp (t) is any solution to (A.9.1)
• C1 , C2 , · · ·, Cn are arbitrary constants
• y1 , y2 , · · ·, yn are n independent solutions to the homogenous equa-
tion
a0 (t)y (n) (t) + a1 (t)y (n−1) (t) + · · · + an−1 (t)y 0 (t) + an (t)y(t) = 0
associated to (A.9.1). “Independent” just means that no yi can be

written as a linear combination of the other yj ’s. For example, y1 (t)
cannot be expressed in the form b2 y2 (t) + · · · + bn yn (t).
In (A.9.2), yp is called the “particular solution” and C1 y1 (t) + C2 y2 (t) +
· · · + Cn yn (t) is called the “complementary solution”.
b Given any constants b0 , · · ·, bn−1 there is exactly one function y(t) that
obeys the ODE (A.9.1) and the initial conditions
y(0) = b0 y 0 (0) = b1 ··· y (n−1) (0) = bn−1

Example A.9.3 RLC circuit. As an example of the most commonly used
techniques for solving linear, constant coefficient ODE’s, we consider the RLC
circuit, which is the electrical circuit consisting of a resistor of resistance R, a
coil (or solenoid) of inductance L, a capacitor of capacitance C and a voltage
source arranged in series, as shown below. Here R, L and C are all nonnegative
constants.
R L
`
`
xptq iptq yptq
´ C
´
We’re going to think of the voltage x(t) as an input signal, and the voltage
y(t) as an output signal. The goal is to determine the output signal produced
by a given input signal. If i(t) is the current flowing at time t in the loop as
shown and q(t) is the charge on the capacitor, then the voltages across R, L and
di
C, respectively, at time t are Ri(t), L dt (t) and y(t) = q(t)
C . By the Kirchhoff’s
1
law that says that the voltage between any two points has to be independent
of the path used to travel between the two points, these three voltages must
add up to x(t) so that
di q(t)
Ri(t) + L (t) + = x(t) (A.9.3)
dt C
Assuming that R, L, C and x(t) are known, this is still one differential equation
in two unknowns, i(t) and q(t). Fortunately, there is a relationship between
the two. Namely
dq
i(t) = (t) = Cy 0 (t) (A.9.4)
dt
This just says that the capacitor cannot create or destroy charge on its own;
all charging of the capacitor must come from the current. Substituting (A.9.4)
into (A.9.3) gives
LCy 00 (t) + RCy 0 (t) + y(t) = x(t)
As a concrete example, we’ll take an ac voltage source and choose the origin
of time so that x(0) = 0, x(t) = E0 sin(ωt). Then the differential equation
becomes
LCy 00 (t) + RCy 0 (t) + y(t) = E0 sin(ωt) (A.9.5)
This is a second order, linear, constant coefficient ODE. So we know, from
Theorem A.9.2, that the general solution is of the form yp (t)+C1 y1 (t)+C2 y2 (t),
where
• yp (t), the particular solution, is any one solution to (A.9.5),
• C1 , C2 are arbitrary constants and
• y1 (t), y2 (t) are any two independent solutions of the corresponding ho-
mogeneous equation
LCy 00 (t) + RCy 0 (t) + y(t) = 0 (A.9.6)
So to find the general solution to (A.9.5), we need to find three functions: y1 (t),
y2 (t) and yp (t).
• Finding y1 (t) and y2 (t): The best way to find y1 and y2 is to guess
them. Any solution, yh (t), of (A.9.6) has to have the property that
yh (t), RCyh0 (t) and LCyh00 (t) cancel each other out for all t. We choose
our guess so that yh (t), yh0 (t) and yh00 (t) are all proportional to a single
function of t. Then it will be easy to see if yh (t), RCyh0 (t) and LCyh00 (t)
all cancel. All derivatives of the function ert are again proportional to
ert . Hence we try yh (t) = ert , with the constant r to the determined.
This guess is a solution of (A.9.6) if and only if
LCr2 ert + RCrert + ert = 0 ⇐⇒ LCr2 + RCr + 1 = 0

√ (A.9.7)
−RC ± R2 C 2 − 4LC
⇐⇒ r = ≡ r1,2
2LC
2 2
How we
q proceed depends
q q of R C − 4LC. That is, whether
on the sign
L L L
R > 2 C or R < 2 C or R = 2 C .
q
◦ Finding y1 (t) and y2 (t), when R > 2 C L
: Then R2 C 2 − 4LC >
0, and r1 and r2 are two different real numbers. We may take
y1 (t) = er1 t and y2 (t) = er2 t so that the complimentary solution is
C1 y1 (t) + C2 y2 (t) = C1 er1 t + C2 er2 t .
q
◦ Finding y1 (t) and y2 (t), when R < 2 C L
: Then R2 C 2 − 4LC < 0
and r1 and r2 are the two different complex numbers −ρ ± iν, where
√
R 4LC − R2 C 2
ρ= and ν=
2L 2LC
We may again take C1 er1 t + C2 er2 t as the complimentray solution.
However we can also rewrite C1 er1 t + C2 er2 t in terms of real valued
functions by using that e±iθ = cos θ ± i sin θ:
C1 er1 t + C2 er2 t = e−ρt C1 eiνt + C2 e−iνt

= e−ρt C1 cos(νt) + i sin(νt) + C2 cos(νt) − i sin(νt)

= e−ρt D1 cos(νt) + D2 sin(νt)

where2 D1 = C1 +C2 , D2 = i(C1 −C2 ). So we may also take y1 (t) =

e−ρt cos(νt), y2 (t) = e−ρt sin(νt) in the complementary solution.
There is yet a third useful way to write the complementary solution.
Think of (D1 , D2 ) as a point in the xy-plane. Call the polar coordi-
nates of that point A and θ so that D1 = A cos θ and D2 = A sin θ.
Then, using the trig identity cos(α + β) = cos α cos β − sin α sin β,
with α = νt and β = −θ,
e−ρt D1 cos(νt) + D2 sin(νt)

= e−ρt A cos(νt) cos θ + A sin(νt) sin θ

(A.9.8)
= Ae−ρt cos(νt − θ)
We have, in effect, replaced the two arbitrary constants D1 and D2 ,

whose values would normally be determined by initial conditions,
by two other arbitrary constants, R and θ, whose values would also
normally be determined by initial conditions.
q
◦ Finding y1 (t) and y2 (t), when R = 2 C L
: Then R2 C 2 − 4LC = 0
so that r1 = r2 . We may take y1 = e , but er2 t = er1 t is certainly
r1 t
not a second independent solution. So we still need to find y2 . Here

is a trick (called reduction of order3 ) for finding the other solutions:
look for solutions of the form v(t)e−r1 t . Here e−r1 t is the solution we
have already found and v(t) is to be determined. To save writing, set
R
ρ = 2L so that r1 = r2 = ρ. To save writing also divide ((A.9.5)h )
R2
by LC and substitute that R 1
L = 2ρ and LC = 4L2 = ρ . (Recall
2
2 4L
that we are assuming that R = C .) So ((A.9.5)h ) is equivalent to
yh00 (t) + 2ρ yh0 (t) + ρ2 yh (t) = 0
Substitute in
yh (t) = v(t)e−ρt
yh0 (t) = −ρv(t)e−ρt + v 0 (t)e−ρt
yh00 (t) = ρ2 v(t)e−ρt − 2ρv 0 (t)e−ρt + v 00 (t)e−ρt
So when yh (t) = v(t)e−ρt ,
yh00 (t) + 2ρ yh0 (t) + ρ2 yh (t)

= ρ2 −2ρ2 +ρ2 v(t)e−ρt + − 2ρ+2ρ v 0 (t)e−ρt + v 00 (t)e−ρt

= v 00 (t)e−ρt
Thus v(t)e−ρt is a solution of ((A.9.5)h ) whenever the function

v 00 (t) = 0 for all t. But, for any values of the constants C1 and
C2 , v(t) = C1 + C2 t has vanishing second derivative so C1 +
C2 t e−ρt = C1 + C2 t e−r1 t solves ((A.9.5)h ). This is of the form
C1 y1 (t) + C2 y2 (t) with y1 (t) = e−r1 t , the solution that we found
first, and y2 (t) = te−r1 t , a second independent solution. So we may
take y2 (t) = ter1 t .
• Finding yp (t): The best way to find yp is to guess it. We guess that the
circuit responds to an oscillating input voltage with an output voltage
that oscillates at the same frequency. So we try yp (t) = A sin(ωt − ϕ)

with the amplitude A and phase ϕ to be determined.
For yp (t) to be a solution, we need
LCyp00 (t) + RCyp0 (t) + yp (t) = E0 sin(ωt)
or
− LCω 2 A sin(ωt − ϕ) + RCωA cos(ωt − ϕ) + A sin(ωt − ϕ)

= E0 sin(ωt)
= E0 sin(ωt − ϕ + ϕ)
and hence, applying sin(A+B) = sin A cos B+cos A sin B with A = ωt−ϕ
and B = ϕ,
1 − LCω 2 A sin(ωt − ϕ) + RCωA cos(ωt − ϕ)

= E0 cos(ϕ) sin(ωt − ϕ) + E0 sin(ϕ) cos(ωt − ϕ)
Matching coefficients of sin(ωt − ϕ) and cos(ωt − ϕ) on the left and right

hand sides gives
1 − LCω 2 A = E0 cos(ϕ)

(A.9.9)
RCωA = E0 sin(ϕ) (A.9.10)
It is now easy to solve for A and ϕ
(A.9.10) RCω
=⇒ tan(ϕ) =
(A.9.9) 1 − LCω 2
RCω
=⇒ ϕ = arctan
1 − LCω 2
p q 2
(A.9.9)2 + (A.9.10)2 =⇒ 1−LCω 2 + R2 C 2 ω 2 A = E0
E0
=⇒ A = p
(1−LCω 2 )2 + R2 C 2 ω 2
Naturally, different input frequencies ω give different output amplitudes A.

Here is a graph of A against ω, with all other parameters held fixed.
A
ω
Note that there is a small range of frequencies that give a large amplitude
response. This is the phenomenon of resonance. It is exploited in the design of
radio and television tuning circuitry. It has also been dramatically illustrated
in, for example, the collapse4 of the Tacoma narrows bridge.
1 Gustav Robert Kirchhoff (1824--1887) was a German physicist.
2 Don’t make the mistake of thinking that C1 and C2 have to be real numbers, forcing
D2 to be pure imaginary. In most applications, D1 and D2 will be pure real and C1 and C2
Example A.9.4 Boundary Value Problems. By part (b) of Theorem

A.9.2, an initial value problem consisting of an nth order linear ODE with
reasonable5 coefficients and n initial conditions always has exactly one solution.
We shall now see that a boundary value problem may have no solutions at all.
Or it may have exactly one solution. Or it may have infinitely many solutions.
We shall start by finding all solutions to the ODE
y0 + y = 0 (A.9.11)
We shall then impose various boundary conditions and see what happens.
The function y(t) = ert is a solution to (A.9.11) if and only if
r2 ert + ert = 0 ⇐⇒ r2 + 1 = 0 ⇐⇒ r = ±i
where i (which electrical engineers often denote6 j) is a square root of −1.

Thus the general solution to the second order linear ODE (A.9.11) is y(t) =
C10 eit + C20 e−it , with C10 and C20 arbitrary constants. We may rewrite this
general solution in terms of sin t and cos t by substituting in
eit = cos t + i sin t e−it = cos t − i sin t
This gives
y(t) = C10 cos t + i sin t) + C20 (cos t − i sin t) = C1 cos t + C2 sin t
where C1 = C10 + C20 , and C2 = i(C10 − C20 ). Note that there is nothing
stopping C10 and C20 from being complex numbers. So there is nothing stopping
C1 = C10 + C20 , and C2 = i(C10 − C20 ) from being real numbers.
a Now consider the boundary value problem
y0 + y = 0 y(0) = 0 y(2π) = 1 (A.9.12)
The function y(t) satisfies the ODE if and only if it is of the form
y(t) = C1 cos t + C2 sin t
for some constants C1 and C2 . A function of this form satisfies the

boundary condition y(0) = 0 if and only if
0 = y(0) = C1 cos 0 + C2 sin 0 = C1
A function of this form satisfies the boundary condition y(2π) = 1 if and

only if
1 = y(2π) = C1 cos 2π + C2 sin 2π = C1
The two requirements C1 = 0 and C1 = 1 are incompatible. So the
boundary value problem (A.9.12) has no solution at all.
b Next consider the boundary value problem
π
y0 + y = 0 y(0) = 0 y =0 (A.9.13)
2
will be complex.
3 The modern method of reduction of order was created by the French mathematician,
physicist and music theorist Jean le Rond d’Alembert (1717-1783). The interested reader
can easily search out more about his life.
4 There are videos of the collapse on the web.

0 = y(0) = C1 cos 0 + C2 sin 0 = C1
A function of this form satisfies the boundary condition y π2 = 0 if and

only if π π π

0=y = C1 cos + C2 sin = C2
2 2 2
So we have a solution if and only if C1 = C2 = 0 and the boundary value
problem (A.9.13) has exactly one solution, namely y(t) = 0, which is a
bit dull.
c Finally consider the boundary value problem
y0 + y = 0 y(0) = 0 y(2π) = 0 (A.9.14)

0 = y(0) = C1 cos 0 + C2 sin 0 = C1
A function of this form satisfies the boundary condition y(2π) = 0 if and

only if
0 = y(2π) = C1 cos(2π) + C2 sin(2π) = C1
So we have a solution if and only if C1 = 0 and the boundary value
problem (A.9.14) has infinitely many solutions, namely y(t) = C2 sin t
with C2 being an arbitrary constant.
5 For example, continuous.

6 This is to avoid confusion with the current, which is typically called i.
Appendix B
Hints for Exercises
1 · Curves
1.1 · Derivatives, Velocity, Etc.
1.1.1 · Exercises
1.1.1.1. Hint. Draw sketches. Don’t forget the range that the parameter
runs over.
1.1.1.2. Hint. Find the value of t at which the three points occur on
the curve.
1.1.1.3. Hint. The curve “crosses itself” when (sin t, t2 ) gives the same
coordinate for different values of t. When these crossings occur will depend
on which crossing you’re referring to, so your answers should all depend on
t.
1.1.1.4. Hint. For part (b), find the position of P relative to the centre
of the circle. Then combine your answer with part (a).
1.1.1.5. Hint. We aren’t concerned with x, so we can eliminate it by
solving one equation for x as a function of y and z and plugging the result
into the other equation.
1.1.1.6. Hint. To determine whether the particle is rising or falling, we
only need to consider its z-coordinate.
1.1.1.7. Hint. This is the setup from Lemma 1.1.4. The two quantities
you’re labelling are related, but different.
1.1.1.8. Hint. See the note just before Example 1.1.6.
1.1.1.9. ∗. Hint. To simplify your answer, remember: the cross product
of a and b is a vector orthogonal to both a and b; the cross product of a
vector with itself is zero; and two orthogonal vectors have dot product 0.
d 2
1.1.1.10. Hint. Evaluate dt |r(t)| .
2
1.1.1.11. ∗. Hint. Just compute |v(t)|. Note that eat + e−at =
e2at + 2 + e−2at .
1.1.1.12. Hint. To figure out what the path looks like, first concentrate
on the x- and y-coordinates.
1.1.1.13. ∗. Hint. Review §1.5. The arc length should be positive.
339
APPENDIX B. HINTS FOR EXERCISES 340
1.1.1.14. Hint. From Lemma 1.1.4, we know the arclength from t = 0

to t = 1 will be Z 1
dr
(t) dt
dt
0
The notation looks a little confusing at first, but we can break it down
piece by piece: dr
dt (t) is a vector, whose components are functions of t. If
we take its magnitude, we’ll get one big function of t. That function is
what we integrate. Before integrating it, however, we should simplify as
much as possible.
1.1.1.16. Hint. r(t) is the position of the particle, so its acceleration is
r00 (t).
1.1.1.17. ∗. Hint. Review §1.5.
1.1.1.18. ∗. Hint. Review §1.1.
1.1.1.19. Hint. (a) First parametrize x2 + y 2 = 9.
1.1.1.20. ∗. Hint. If you got the answer 0 in part (b), you dropped
some absolute value signs.
1.1.1.22. ∗. Hint. The integral you get can be evaluated with a simple
substitution. You may want to factor the integrand first.
1
1.1.1.23. Hint. (b) 4x + 1 + x is a perfect square.
(c), (d) Let
• r(x) be the position of the particle when its first coordinate is x,

• R(t) be the position of the particle at time t, and
• x(t) be the x--coordinate of the particle at time t.
Then R(t) = r x(t) . We are told |R0 (t)| = 9 for all t.

1.1.1.24. Hint. Given the position of a particle, you can find its velocity.
1.1.1.25. ∗. Hint. If r(u) is the parametrization of C by u, then the

position of the particle at time t is R(t) = r u(t) .
1.1.1.26. ∗. Hint. By Newton’s law, F = ma.
1.1.1.27. ∗. Hint. Denote by r(x) the parametrization of C by x. If
the x-coordinate of the particle at time t is x(t), then the position of the
particle at time t is R(t) = r x(t) . Also, though the particle is moving at
a constant speed, it doesn’t necessarily have a constant value of dxdt .
1.1.1.28. Hint. The question is already set up as an xy-plane, with the

camera at the origin, so the vector in the direction the camera is pointing is
(x(t), y(t)). Let θ be the angle the camera makes with the positive x-axis
(due east). The tangent function gives a clean-looking relation between
θ(t), x(t), and y(t).
1.1.1.29. Hint. Usng the Theorem of Pappus, the surface area and
volume of this pipe are the same as that of a straight pipe with the same
length and radius.
1.1.1.30. Hint. A helix can be parametrized by r(θ) = a cos θ ı̂ı +
a sin θ ̂ + bθ k̂.
−αt
1.1.1.31. Hint. Define u(t) = eαt dr dr
dt (t) and substitute dt (t) = e u(t)
into the given differential equation to find a differential equation for u.
1.2 · Reparametrization
1.2.1 · Exercises
1.2.1.1. Hint. You’re asked to find the arclength of the curve from s = 1
to s = t.
1.2.1.2. Hint. The arclength will be 0 at P .
1.2.1.3. Hint. a(t0 ) and b(s0 ) describe the same point on R.
1.2.1.4. ∗. Hint. On your way to finding the relationship between t

and arclength, you should realize that the curve has constant speed (with
respect to t), though not constant velocity.
1.2.1.5. ∗. Hint. For which values of t is |r(t)| ≤ 1? Check the domain
of t — we’re not starting at zero.
1.2.1.6. Hint. Be careful with the domain.

√
1.2.1.7. Hint. Remember x2 = |x|. You will need to consider cases
for this one.
1.3 · Curvature
1.3.1 · Exercises
1.3.1.1. Hint. The curve is a circle, so you don’t need to do any calculus.
1.3.1.2. Hint. Because r is a circle, you can parametrize it with respect
to arclength without using an integral. You found κ in Question 1.3.1.1.
1.3.1.3. Hint. When t is large, does the spiral locally look like a circle
of large radius, or small?
1.3.1.4. Hint. ds
dt = |v(t)| = |r0 (t)|
v(t) r0 (t)
1.3.1.5. Hint. T̂ = |v(t)| = |r0 (t)|
1.3.1.7. Hint. You can find the last two quantities by making use of the
first three. Looking ahead, the formula list in Section 1.5 might come in
handy.
|v(t) × a(t)|
1.3.1.8. Hint. We can calculate κ = ds 3 . We can also figure

dt
out what kind of a shape our curve is.
1.3.1.9. Hint. The maximum and minimum values of κ(t) should be
obvious from your formula for κ(t).
1.3.1.11. ∗. Hint. For part (a), determine r(0), r(π), r(2π), r(3π), and
r(4π), to help you map out the motion. Also visualize the thumbtack as
the wheel moves.
For part (d), use the fact that you only care about t = π: where is this
on your sketch? What does that mean about the direction of N̂?
1.3.1.12. Hint. You should find that s = θ!

1.3.1.13. ∗. Hint. Since κ(x) is never negative, κ(x) is maximum when
κ2 (x) is maximum. The latter is easier to compute.
1.4 · Curves in Three Dimensions

1.4.1 · Exercises
1.4.1.1. Hint. Use the right-hand rule to figure out how B̂ is oriented.
1.4.1.2. Hint. Speed is the norm of velocity. Does that fit this equation?
1.4.1.3. Hint. Review Example 1.4.4 and remember that positive tor-
sion indicates “right-handed twisting.” You shouldn’t actually need to cal-
culate anything.
1.4.1.4. Hint. (a) Show that the tangent vector T̂(s) is a constant.
(b) Guess the plane. To do so, first show that the binormal B̂(s) is a
constant. Then show that (r(s) − r(0)) · B̂ is a constant.
1
(c) Guess the circle. To do so, first show that rc (s) = r(s) + κ(s) N̂(s)
is a constant.
1.4.1.5. ∗. Hint. It is not necessary to compute anything.
1.4.1.6. ∗. Hint. Both parts of this question make use of the quantity
ds
dt .
(v(t) × a(t)) · da
dt
1.4.1.7. Hint. τ (t) =
|v(t) × a(t)|2
1.4.1.8. Hint. Review §1.5.
1.4.1.9. Hint. The vector perpendicular to the plane containing the
osculating circle is the binormal vector, B̂.
1.4.1.10. ∗. Hint. (a) The tangent vector of the curve is also a normal
vector for the specified plane.
(b) Review §1.5.
2 2
1.4.1.11. ∗. Hint. Remember a(t) = ddt2s (t) T̂(t) + κ(t) dsdt (t) N̂(t).
Remember also that B̂ is orthogonal to T̂ and N̂, which are in the plane
of C.
1.4.1.12. ∗. Hint. By Theorem 1.3.3, the tangential component of
2
acceleration is aT (t) = ddt2s
1.4.1.13. ∗. Hint. Use your answers to previous parts to calculate (d).
Tangential and normal components of acceleration are defined just before
Example 1.3.4.
1.4.1.14. ∗. Hint. (a) All points on the curve obey an equation that
contains x’s and y’s, but no z’s. There is a standard way to get a nice
parametrization of this equation, that doesn’t involve using square roots.
(b) You don’t need to compute the constants for all points: only the
given point.
1.4.1.15. ∗. Hint. For part (c), you only need to find N̂ at a point,
which is easier than finding it for all t.
1.4.1.16. ∗. Hint. First parametrize x2 + y 2 = 1 in the standard way.

You don’t need calculus for part (c).
1.4.1.17. ∗. Hint. Review §1.5.
1.4.1.18. ∗. Hint. Since 0 ≤ t ≤ 1, you can simplify |t| = t.
1.4.1.19. ∗. Hint. For part (f), remember that you can write the equa-
tion of a plane easily once you know a point it passes through, and a vector
normal to it. The plane should touch the curve when t = 0, and the plane
should contain T̂ and N̂.
1.4.1.20. ∗. Hint. It might be easier to find B̂ before you find N̂, then
use the formula N̂(t) = B̂(t) × T̂(t).
1.4.1.21. ∗. Hint. The osculating plane at r(t0 ) is the plane through
r(t0 ) with normal B̂(t0 ). Also, notice the points for parts (a) and (b) are
not the same.
√
1.4.1.22. ∗. Hint. Since t > 0, we can simplify t2 = |t| = t.
1.4.1.23. ∗. Hint. In this context, “distance travelled” means “ar-
clength.”
1.4.1.24. ∗. Hint. Use T̂ and N̂ to compute B̂.

1.4.1.25. Hint. (a) First find a parametrization x(θ), y(θ) for x2 +y 2 =
1.
1.4.1.26. ∗. Hint. You need to find the acceleration at (1, 1, 1). Think
about what strategies are available for computing the acceleration.
1.4.1.27. ∗. Hint. For part (a), T(t) will be a vector of the form
T(t) = (1,at,bt)
√
1+4t2
where a and b are nonzero constant real numbers.
For part (b), N(t) will be a vector of the form N(t) = (−4t,α,β)
√
2 1+4t2
where
α and β are nonzero constant real numbers.
For part (e), κ(t) will be a function of the form κ(t) = (1+4tγ2 )3/2 , where
γ is a positive constant real number.
1.4.1.28. ∗. Hint. Differentiate N̂ = B̂ × T̂ with respect to s.

The vectors N̂, B̂, and T̂ form a right-handed triple. Sketch them (the
same way you might sketch the x, y, and z axes) to figure out the signs of
their cross products.
1.4.1.29. ∗. Hint. In part (b), note that a is the second derivative with
respect to time (not θ). Exploit a = dv 2
dt T̂ + v κN̂ to find what you’re asked
for.
1.4.1.30. ∗. Hint. For part (d), what is the relationship between the y-
and z-components of the particle’s position? How can you use that to find
a plane containing the particle at all times t?
1.4.1.31. ∗. Hint. Rather than trying to wrangle trig identities, plug
in θ = π as soon as you can for part (a). For part (c), remember that you
need the chain rule if you want to make use of your previous derivatives.
1.6 · Integrating Along a Curve

1.6.1 · Exercises
1.6.1.1. Hint. Your differential is ds, where s is arclength.

1.6.1.2. Hint. (a) You can parametrize the curve by r(θ) = r(θ) cos θ ı̂ı +
r(θ) sin θ ̂, θ1 ≤ θ ≤ θ2 .
xy
1.6.1.3. Hint. Following Definition 1.6.1, set f (x, y, z) = z , x(t) = 23 t3 ,
√
y(t) = 3t2 , and z(t) = 3t.
1.6.1.4. Hint. Parametrize the circle in the usual way.
1.6.1.5. Hint. C can be parametrized as (1+t, 2+2t, 3+2t) for 0 ≤ t ≤ 1.
d √1
1.6.1.7. Hint. Simplify! Also: dt {arcsec t} = |t| t2 −1
.
1.6.1.8. ∗. Hint. Newton’s law of motion is F = ma. The work done

over a displacement dr is W = F · dr.
1.6.1.9. ∗. Hint. Sketch C and determine the normal vectors from the
sketch. You can use x or y as the integration variable in your integrals.
2 2
1.6.1.10. ∗. Hint. (c) How is x(t) + y(t) related to z(t)?
(d) First, sketch x(t) , y(t) .
R
xρ ds
1.6.1.11. Hint. Remember x̄ = RC , etc. The integrals you evalu-
C
ρ ds
ate should all be straightforward applications of the power rule.
1.7 · Sliding on a Curve

1.7.4 · Exercises
1.7.4.1. Hint. Gravity pulls straight down, while the direction of the
normal force depends on the curve of the wire. There is not enough in-
formation to know the magnitude of the forces, but you can approximate
their directions.
1.7.4.2. Hint. This equation stems from F = ma. In that equation, a
is what kind of derivative?
1.7.4.3. Hint. A thought experiment might help you avoid any calcu-
lations. If the wire were perfectly vertical or perfectly horizontal, what
would W N̂ be?
1.7.4.4. Hint. The skater reaches their highest point when |v| = 0.
1.7.4.5. Hint. The highest vertical height occurs just as the skate-
E
boarder’s speed reduces to 0, at yS = mg .
1.7.4.6. Hint. At the bottom of the culvert, all the skater’s energy is
kinetic, not potential. That is, in the equation E = 12 m|v|2 +mgy, we have
y = 0.
1.7.4.7. Hint. Equation 1.7.2 tells us the normal force exerted by the
track is W N̂, where W = mκ|v|2 + mg k̂ · N̂. Equation 1.3.3 part (c) says
d2 s ds 2

a(θ) = dθ 2 T̂ + κ dθ N̂.
d2 s
1.7.4.8. Hint. When θ = 13π/3, dθ 2 = 0, which is handy for a quicker
calculation.
Important equations: the normal force exerted by the track is W N̂,

d2 s ds 2

where W = mκ|v|2 + mg̂ · N̂ (Equation 1.7.2); a(θ) = dθ 2 T̂ + κ dθ N̂
(Equation 1.3.3, part (c) ).
1.7.4.9. Hint. According to the equation in §1.7.2, the skiier will become
airborne when: r
g
|v| > |̂ · N̂|
κ
q
κ |̂ · N̂| for some point on the
g
So, we need |v| to be greater than 
curve inside the range 1/e ≤ t ≤ e.
Note that g is given in metres per second, while the other quantities
are in kilometres and hours.
1.7.4.10. Hint. There are now three forces acting on the bead: one
parallel to ̂ (exerted by gravity), one parallel to N̂ (exerted by the wire),
and one parallel to T̂ (exerted by the jet pack).
Follow the reasoning in the sliding bead section of the text, focusing on
the tangential forces.
1.7.4.11. Hint. If the snowmachine is moving at a constant speed, the
tangential component of its acceleration is zero. Part (a) is similar to
Question 1.7.4.10.
1.7.4.12. Hint. Follow the discussion in §1.7.3.
It’s fine to leave part (b) pretty messy. Your answer for part (c) involves
the root of a cubic function, but you don’t need a high degree of accuracy
to decide between the three options given.
1.8 · Optional — Polar Coordinates

1.8.1 · Exercises
1.8.1.2. Hint. r is allowed to be negative.
1.8.1.3. Hint. Compute, for each angle θ, the dot product êr (θ) · êθ (θ).

1.8.1.5. Hint. The curve can be parametrized by r(θ) = f (θ) cos θ ı̂ı +
sin θ ̂
1.9 · Optional — Central Forces

1.9.1 · Exercises
1.9.1.1. ∗. Hint. (a) Review §1.9.
(b) Any straight line can be parametrized as r(s) = r0 + T̂ s.
(c) Review §1.10.

1.9.1.2. ∗. Hint. (a) For any central force r(t) × v(t) is independent
of t.
(b) Review Lemma 1.8.2.
2 · Vector Fields
2.1 · Definitions and First Examples
2.1.1 · Exercises
2.1.1.1. Hint. Not all blanks represent a single interval.

2.1.1.2. Hint. Write down all coordinates where v(x, y)·ı̂ı = 0 or v(x, y)·
̂ = 0, and look for a pattern.
2.1.1.3. Hint. If you know the speed and direction of an object, you can
find its velocity.
2.1.1.5. Hint. When the twig is at (x, y) it has velocity v(x, y).
2.1.1.6. Hint. Whenever the twig is on the y-axis, its velocity is parallel
to the y-axis. So it remains on the y-axis for all time.
2.1.1.7. Hint. If you know the speed and direction of an object, you can
find its velocity.
2.1.1.8. Hint. Set your face to be at the origin, (0, 0, 0).
If A is “inversely proportional” to B, then there exists a constant α such
that AB = α. That way when |B| goes up, |A| goes down, and vice-versa.
2.1.1.9. Hint. Start with the regions where v(x, y) · ı̂ı and v(x, y) · ̂ are
positive and negative. As you move up/down/left/right, do the vectors get
longer or shorter? More horizontal or more vertical?
2.1.1.10. Hint. v(x, y) · ı̂ı is the distance from (x, y) to the origin, while
v(x, y) · ̂ is the distance from (x, y) to the point (1, 1).
2.1.1.11. Hint. Factor x2 + xy = x(x + y) and y 2 − xy = y(x − y). Chop
the plane up into eight regions using the two coordinate axes and the lines
y = x, y = −x.
2.1.1.12. Hint. What is the geometric interpretation of each summand?
2.1.1.13. Hint. (a), (c) Intrepret the vector field geometrically.
2.1.1.14. Hint. The constant G is the same for all masses, but M differs.
The net force is the sum of three force vectors.
2.1.1.15. Hint. For part a., make a triangle with P as one of its vertices
that is similar to the triangle made by the pole, the wall, and the ground.
Its hypotenuse has length p; let its base be b and its height be h. Find a
way to translate between (b, h) and (x, y).
For part b., use your answer from part a. Start by describing a point
as itsdistance from the lower end of the pole, p. Then, consider
on a pole
dz dx dy
dt and dt , dt separately. If you’re having a hard time simplifying your
p √
answer, note x + y 2 = 3(1 − z) for any point (x, y, z) on a pole when
2
H = 1.
2.2 · Optional — Field Lines

2.2.2 · Exercises
2.2.2.2. Hint. Express x0 (t) and y 0 (t) purely in terms of x(t) and y(t).
2.2.2.3. ∗. Hint. Review §2.2.
2.3 · Conservative Vector Fields

2.3.1 · Exercises
2.3.1.1. Hint. Carefully consider the context that lead to each of these
equations.
2.3.1.2. Hint. One of the three options will NEVER be true, for any F.
2.3.1.3. Hint. Modify ϕ, the potential for F.
2.3.1.4. Hint. a. If F + G is conservative, what has to be true?
b. What if F and G are quite similar?
c. Find a potential for F + G.

2.3.1.5. ∗. Hint. Note that the domain is D = (x, y) x > 1 .
Compare to Example 2.3.14.
2.3.1.6. Hint. A potential does exist.
d
2.3.1.7. Hint. Recall dx ln |x| = x1 .
2.3.1.8. Hint. Try the screening test, Theorem 2.3.9.
Z
x
2.3.1.9. Hint. dx can be evaluated by inspection, or
x2 + y 2 + z 2
with the substitution u = x2 + y 2 + z 2 .
2.3.1.11. Hint. For what values of the constants A and B does the
vector field F pass the screening test ∇ × F = 0?
2.3.1.12. Hint. Review Example 2.1.2.

2.3.1.13. Hint. Following Example 2.3.3, the particle can never escape
the region
(x, y, z) ϕ(x, y, z) ≥ −E
where E is the energy of the system.
2.3.1.14. Hint. Example 2.3.3 tells us 21 m|v(t)|2 −ϕ x(t), y(t), z(t) = E

is a constant quantity, provided F is conservative with potential ϕ(x, y, z).

2.3.1.15. Hint. Find a potential ϕ. Notice f , g, and h are functions of
one variable each — this simplifies things.
2.3.1.16. Hint. Write the points with curl 0 as multiples of a constant
vector.
2.4 · Line Integrals

2.4.2 · Exercises
2.4.2.1. Hint. The top and bottom of the square can be easily paramer-
ized using x as the parameter. The other two sides can be easily parame-
terized using y as the parameter.
2.4.2.2. Hint. Contrast Theorems 2.4.8 and 2.3.9.
2.4.2.3. Hint. Please don’t do any computation, especially not to find
C!
2.4.2.4. Hint. Review properties of conservative vector fields.
2.4.2.5. ∗. Hint. Use Theorems 2.4.7 and 2.4.8.
2.4.2.6. Hint. Review Theorem 2.4.7.
2.4.2.7. ∗. Hint. Part (d) is a hint.
2.4.2.8. Hint. The last part of the question is a huge hint.
2.4.2.11. ∗. Hint. Parametrize the curve using y as a parameter.

2.4.2.12. Hint. (a) Use Theorem 2.4.8.
(c) You may parametrize the curve using x as the parameter. Exploit
the fact that, for the value of λ found in part (a), F + λG is conservative.
2.4.2.14. ∗. RHint. Parametrize the path using sines and cosines. The
work done is C F · dr
2.4.2.15. ∗. Hint. Is F conservative?
2.4.2.16. ∗. Hint. Is F = xy ̂ conservative? Sketch C.
2.4.2.17. ∗. Hint. That the line integral is to be independent of path
is a huge hint.
2.4.2.18. ∗. Hint. Note that
• y = 0 on the line segment from (1, 0, 0) to (0, 0, 1) and
• x = 0 on the line segment from (0, 0, 1) to (0, 1, 0) and
• z = 0 on the line segment from (0, 1, 0) to (1, 0, 0)

2.4.2.19. ∗. Hint. That F is conservative should be a dead giveaway.
2.4.2.20. ∗. Hint. To calculate the integral, it might be easier to find a
potential for F and use Theorem 2.4.2.
2.4.2.21. ∗. Hint. Your answer from (b) can help you in (c). Also,
cos(1) = cos(−1), because cosine is an even function.
2.4.2.22. ∗. Hint. Review §2.4.1 of the text.
2.4.2.23. ∗. Hint. Relate the integral of part (d) to the integral of part
(c).
R
2.4.2.24. ∗. Hint. Write the integral of part (c) as C
G · dr. What is
the difference between G and F?
2.4.2.25. ∗. Hint. (d) How are G and F related?
2.4.2.26. ∗. Hint. (a) Start with ∂f 2 yz

∂z = y e .
(b) Use the result of part (a) to do part (b).
2.4.2.27. ∗. Hint. The integral in part (b) is path independent. That’s
a big hint.
2.4.2.28. ∗. Hint. Part (a) is a hint for part (b).
2.4.2.29. ∗. Hint. The three parts of this problem are closely related.
2.4.2.30. ∗. Hint. We can rewrite x2 +y 2 +z 2 = 2z as x2 +y 2 +(z−1)2 =
1.
2.4.2.31. Hint. (a) The curve can be easily parametrized by using x as

a parameter.
(b) Don’t evaluate the integral directly.
2.4.2.32. ∗. Hint. Refer to Example 1.4.4 for a parametrization of a

helix.
2.4.2.33. ∗. Hint. (b) Parametrize each side of the square by arc length,
and make use of the plentiful zeroes that arise.
2.4.2.34. ∗. Hint. Force is mass times acceleration, where acceleration
is the second derivative of position, r(t), with respect to time, t. The work
Rb
done by F between time a and time b is a F · dr.
2.4.2.35. ∗. Hint. Note that the curve goes from (2, 2) to (1, 1) — not
the other way around.
For part (b), one possibility is to look for a path consisting of the line
segment from (2, 2) to (2, Y ), followed by the line segment from (2, Y ) to
(1, Y ), followed by the line segment from (1, Y ) to (1, 1), with Y being a
parameter to be determined.
2.4.2.36. ∗. Hint. One possibility is to look for a path consisting of the
line segment from (0, 0) to (0, Y ), followed by the line segment from (0, Y )
to (2, Y ), followed by the line segment from (2, Y ) to (2, 0), with Y being
a parameter to be determined.
2.4.2.37. ∗. Hint. Is F conservative?
2.4.2.38. ∗. Hint. On S, note z = 2 + x2 − 3y 2 . Further,R the vector
field F̃(x, y, z) = z 2 k̂ is conservative (with potential 13 z 3 ), so C1 F̃ · dr =
R
C2
F̃ · dr for any two curves C1 and C2 from P1 to P2 . Compare this to
Questions 2.4.2.24 through 2.4.2.25.
2.4.2.39. ∗. Hint. Simplify the answer in part (a) as much as possible.
3x2
For part (c), start with ∂f
∂y = xe and ∂f 2 2
∂z = x cos(x z).
For part (d), notice the difference between the given vector field and the
conservative vector field of part (c). The resulting integral can be directly
evaluated using methods from integral calculus.
dr
2.4.2.40. ∗. Hint. For (b), remember ds dt = dt \\ Is the vector field of

part (c) conservative?
2.4.2.41.
R ∗. Hint. For part (d), what is the difference between J and
C
F · dr?
For part (e), many parts of the integral are zero: find as many as you
can.
2.4.2.42. ∗. Hint. By Newton’s law of motion, mr00 (t) = F(t).
0 00
Recall κ(t) = |r (t)×r
|r0 (t)|3
(t)|
.
2.4.2.43. ∗. Hint. (a) Remember the arclength of the parametrized

Rb
path r(t) from t = a to t = b is given by a |r0 (t)| dt. In this case, |r0 (t)|
can be simplified considerably.
0 00
(b) Remember κ(t) = |r (t)×r
|r0 (t)|3
(t)|
.
(c) Gravity is conservative. Friction is not conservative.
(d) What are the tangential and normal components of acceleration?
3 · Surface Integrals
3.1 · Parametrized Surfaces
3.1.1 · Exercises
3.1.1.1. Hint. Your answer will have the form r(x, y) = ψ1 (x, y)ı̂ı +
ψ2 (x, y)̂ + ψ3 (x, y)k̂.
3.1.1.3. ∗. Hint. First think about what properties r(u, v) has to have
in order to be a parametrization.
3.1.1.4. ∗. Hint. First think about what properties r has to have in
order to be a parametrization.
3.1.1.5. ∗. Hint. First think about what properties r(u, v) has to have
in order to be a parametrization.
3.2 · Tangent Planes

3.2.1 · Exercises
3.2.1.1. Hint. What are the tangent planes to the two surfaces at
(0, 0, 0)?

3.2.1.2. Hint. Apply the chain rule to G r(t) = 0.
3.2.1.4. Hint. To find a tangent vector to the curve of intersection of
the surfaces F (x, y, z) = 0 and G(x, y, z) = 0 at (x0 , y0 , z0 ), use Q[3.2.1.2]
twice, once for the surface F (x, y, z) = 0 and once for the surface G(x, y, z) =
0.
3.2.1.5. Hint. To find a tangent vector to the curve of intersection of
the surfaces z = f (x, y) and z = g(x, y) at (x0 , y0 , z0 ), use Q[3.2.1.2] twice,
once for the surface z = f (x, y) and once for the surface z = g(x, y).
3.2.1.10. ∗. Hint. Review §3.2.

3.2.1.11. ∗. Hint. Review §3.2.
3.2.1.13. ∗. Hint. Let (x, y, z) be a desired point. Then
• (x, y, z) must be on the surface and
• the normal vector to the surface at (x, y, z) must be parallel to the
plane’s normal vector.
3.2.1.14. ∗. Hint. First find a parametric equation for the normal
line to S at (x0 , y0 , z0 ). Then the requirement that (0, 0, 0) lies on that
normal line gives three equations in the four unknowns x0 , y0 , z0 and t.
The requirement that (x0 , y0 , z0 ) lies on S gives a fourth equation. Solve
this system of four equations.
3.2.1.15. ∗. Hint. Two (nonzero) vectors v and w are parallel if and
only if there is a t such that v = t w. Don’t forget that the point has to
be on the hyperboloid.
3.2.1.16. ∗. Hint. (b) If v is tangent, at a point P , to the curve of

intersection of the surfaces S1 and S2 , then v
• has to be tangent to S1 at P , and so must be perpendicular to the
normal vector to S1 at P and
• has to be tangent to S2 at P , and so must be perpendicular to the

normal vector to S2 at P .
3.2.1.17. ∗. Hint. The angle between the curve and the surface at P
is 90◦ minus the angle between the curve and the normal vector to the
surface at P .
3.2.1.18. Hint. At the highest and lowest points of the surface, the
tangent plane is horizontal.
3.3 · Surface Integrals

3.3.6 · Exercises
3.3.6.1. Hint. S is a very simple geometric object.
x y z
3.3.6.2. Hint. The triangle is part of the plane a + b + c = 1.
3.3.6.3. Hint. Flatten S out.
3.3.6.8. ∗. Hint. The total surface area of (b) (ii) can be determined
without evaluating any integrals.
3.3.6.11. Hint. On S, (x, y) runs over the interior of x2 + y 2 = 2x, or
equivalently, the interior of (x − 1)2 + y 2 = 1.
3.3.6.12. Hint. See Example 3.1.5 for a parametrization of the torus.
3.3.6.13. Hint. Call the part of the sphere in the first octant S. By
definition, the centroid is (x̄, ȳ, z̄) with
RR RR RR
S
x dS S
y dS z dS
x̄ = RR ȳ = RR z̄ = RRS
S
dS S
dS S
dS
The integrals will be easy if you use spherical coordinates. You can reduce
the number of integrals evaluated by using symmetry.
3.3.6.14. Hint. Before parametrizing the cylinder, express x2 +y 2 = 2ay
in cylindrical coordinates.
3.3.6.16. Hint. (a) The integral can be easily evaluated by using that
the sphere has surface area 4πa2 .
(c) Use cylindrcial coordinates for the top part of the cone.
3.3.6.20. ∗. Hint. Beware of signs. Note that 0 ≤ z ≤ 1 on S.
3.3.6.21. ∗. Hint. The z-coordinate of the centre of mass is the weighted
average of the z-coordinate over the cone. Since a density has not been
specified, we assume that it is a constant. We RR
may takeRR
the density to be
1, so the z-coordinate of the centre of mass is S z dS/ S dS.
3.3.6.22. ∗. Hint. Use cylindrcal coordinates. Note that because of
the symmetry of the cone, only the z-component of the centre of mass
requires an integral to be calculated. The z-coordinate of the centre of
massRRis the weighted
RR average of the z-coordinate over the cone. That is
z̄ = S z dS/ S dS.
3.3.6.24. ∗. Hint. Review (3.3.2).
3.3.6.25. ∗. Hint. Don’t pbe afraid to tweak spherical coordinates so
as to fit the condition x ≥ y 2 + z 2 well.
√ To do so, first use a sketch to
y 2 +z 2
develop a geometric interpretation of x .
3.3.6.26. ∗. Hint. The surface S may be parametrized by observing

that, for each fixed y, x2 + z 2 = sin2 y is a circle.
3.3.6.27. ∗. Hint. By symmetry, the centre of mass will lie on the z-
axis. By definition, the z-coordinate of the centre of mass is the weighted
average of z over S, which is
RR
z ρ(x, y, z) dS
z̄ = RRS
S
ρ(x, y, z) dS
3.3.6.37. Hint. You can use the the cylindrical coordinates θ and z to
parametrize the hyperboloid.
3.3.6.38. ∗. Hint. (a) Review §3.2.

(b) Review §3.3.1.
3.3.6.39. ∗. Hint. (a) Review §2.4.1.
(b) Use Lemma 2.3.6 to show that the integrand is identically zero.
4 · Integral Theorems
4.1 · Gradient, Divergence and Curl
4.1.6 · Exercises
4.1.6.2. Hint. Compute ∇ × F for some simple vector fields.
4.1.6.3. Hint. For parts(a) and (b), write out the definitions of the left
and right hand sides and observe that they are equal. Part (c) can be done
easily by using other, simpler, vector identities.
4.1.6.6. ∗. Hint. (c) can be done efficiently by using (a) and (b).
4.1.6.12. Hint. (a) Find the magnitude and direction of the velocity
vector. Then verify that Ω × r has that magnitude and direction.
4.2 · The Divergence Theorem

4.2.6 · Exercises
4.2.6.3. Hint. (b) The integral
RRR can be trivially evaluated by exploiting
oddness and the fact that V
dV = Volume(V ).
4.2.6.4. Hint. For part (a), use spherical coordinates.
4.2.6.5. Hint. (a) The RRRintegral is easier in polar coordinates.
(b) Since x is odd, V
x dV = 0.
4.2.6.6. Hint. (a) The integral is easy in polar coordinates.
(b) The volume of the solid can be easily computed by decomposing
the solid into thin horizontal pancakes. See Section 1.6 in the CLP-2 text.
4.2.6.7. ∗. Hint. The divergence theorem, of course.
4.2.6.8. Hint. It’s easier to use the divergence theorem. But don’t forget
the base of the silo.
4.2.6.9. Hint. The divergence theorem, of course. The integral can be

easily evaluated by using that, for any solid V in R3 ,
ZZZ
dV = Volume(V)
V
and
RRR RRR RRR
V
x dV V
y dV V
z dV
x̄ = ȳ = z̄ =
Volume(V) Volume(V) Volume(V)
where (x̄, ȳ, z̄) is the centroid of V.

4.2.6.10. ∗. Hint. The complexity of F is a hint that the flux should
not be evaluated directly.
4.2.6.11. ∗. Hint. The specified surface is not closed.
4.2.6.12. ∗. Hint. (a), (b), (c) Review warning 4.2.3.
(d) The divergence theorem can be used — with care.
(e) The equation can be made more understandable by completing a
square.
4.2.6.13. ∗. Hint. (a) Use a suitable modification of spherical coordi-
nate. Do not forget to specify the range of the parameters.
4.2.6.14. ∗. Hint. Don’t evaluate the flux directly.
4.2.6.15. ∗. Hint. For practice, try doing this question twice — once
using the divergence theorem and once using direct evaluation.
4.2.6.16. ∗. Hint. The question highlights that the vector field has
divergence 0. Thta’s a big hint.
4.2.6.17. ∗. Hint. As F looks complicated, it is probably wise not to
try and evaluate the flux integral directly.
4.2.6.18. ∗. Hint. As F looks complicated, it is probably wise not to
try and evaluate the flux integral directly.
4.2.6.19. ∗. Hint. The vector field F looks complicated. Try to avoid a
direct evaluation of the flux integral.
4.2.6.20. ∗. Hint. The divergence of F is a lot simpler than F itself.
By default, we want the outward flux.
4.2.6.21. ∗. Hint. The vector field F looks very complicated. That
strongly suggests that we not evaluate the integral directly.
4.2.6.22. ∗. Hint. The divergence of F is a lot simpler than F itself.
4.2.6.23. ∗. Hint. Note that F(x, y, z) is not defined at (x, y, z) =
(0, 0, 0).
4.2.6.25. ∗. Hint. The surface S is not a closed surface.
4.2.6.29. ∗. Hint. The complexity of F is a hint that the flux should
not be evaluated directly.
4.2.6.31. ∗. Hint. The flux can be calculated directly, but it is rather
easier to calculate it using the Divergence Theorem.
4.2.6.32. ∗. Hint. Use that y is odd to easily evaluate some integrals.
4.2.6.34. Hint. (a) Use cylindrical coordinates.
(b) The volume of the V can be easily computed by decomposing V
into thin horizontal washers. See Section 1.6 in the CLP-2 text.
4.2.6.35. Hint. Review the derivation of the heat equation in Section

4.2.1.
4.2.6.37. ∗. Hint. Make a judicious choice of parametrization.
4.2.6.38. ∗. Hint. Do not compute the integral directly.
4.2.6.39. ∗. Hint. Be careful about which normals to use in part (c).
For practice, try to do part (c) in two different ways, with one being direct
evaluation.
4.2.6.40. ∗. Hint. For part (b), do not evaluate the flux directly. In
part (c), the flux can be related to the volume enclosed by the surface, and
the centre of mass of the volume enclosed by the surface.
4.2.6.41. ∗. Hint. (b) We have several different methods for evaluating
flux integrals. Think about what would be involved in applying each of
them before settling on which one to use.
(c) Be sneaky — don’t evaluate this integral directly.
4.2.6.42. ∗. Hint. For parts (b) and (c), write out carefully the integral
that the divergence theorem gives you.
4.2.6.43. ∗. Hint. Note that 22 + 2(12 ) + 3(1)2 = 9 < 16 so that (2, 1, 1)
is inside S, while 32 + 2(22 ) + 3(2)2 = 29 > 16 so that (3, 2, 2) is outside S.
4.2.6.44. ∗. Hint. Review §4.2.2.
4.2.6.46. ∗. Hint. Consider very small a’s.
4.2.6.47. ∗. Hint. Carefully draw a side view of S.
4.2.6.48. ∗. Hint. Both the divergence theorem and a vector identity
in Theorem 4.1.4 are useful.
4.2.6.49. ∗. Hint. x is an odd function.
4.2.6.50. ∗. Hint. You should be RR centre of mass, (x̄, ȳ)
RR able to guess the
of the disk D. RR
Then the integrals RR D x dxdy and D y dxdy can be found
x dxdy y dxdy
by using x̄ = RRD and ȳ = RRD .
dxdy dxdy
D D
4.3 · Green’s Theorem

4.3.1 · Exercises
4.3.1.2. Hint. Let r(s) = x(s)ı̂ı+y(s) ̂ be a counterclockwise parametriza-
tion of C by arc length. Then T̂(s) = r0 (s) = x0 (s)ı̂ı + y 0 (s) ̂ is the
forward pointing unit tangent vector to C at r(s) and n̂(s) = r0 (s) × k̂ =
y 0 (s)ı̂ı − x0 (s) ̂. To see that r0 (s) × k̂ really is n̂(s), note that y 0 (s)ı̂ı − x0 (s) ̂
• has the same length, namely 1, as r0 (s) (recall that r(s) is a parametriza-
tion by arc length),
• lies in the xy-plane and

• is perpendicular to r0 (s). (Check that r0 (s) · y 0 (s)ı̂ı − x0 (s) ̂ = 0.)

• Use the right hand rule to check that r0 (s) × k̂ is n̂ rather than −n̂.
4.3.1.3. Hint. Use direct evaluation!
x −y
4.3.1.4. Hint. The functions x2 +y 2 and x2 +y 2 are not defined, let alone
continuous or differentiable, at x = y = 0.
4.3.1.5. Hint. For practice, evaluate this integral twice — once directly
and once using Green’s theorem.
4.3.1.6. Hint. The sin y 2 and cos y 2 in the integrand look hard to inte-
grate. Try Green’s theorem.
4.3.1.7. ∗. Hint. Don’t do the integral directly.
4.3.1.8. ∗. Hint. Don’t do the integral directly. Sketch the rectangle.
4.3.1.9. ∗. Hint. Do not compute the integral directly.
4.3.1.10. ∗. Hint. Don’t do the integral directly. Sketch the triangle.
4.3.1.11. ∗. Hint. The integrand for direct evaluation looks complicated
— don’t evaluate this integral directly.
4.3.1.12. ∗. Hint. Direct evaluation is not the most efficient method
available.
4.3.1.13. ∗. Hint. Green’s theorem must be applied to a closed curve;
note that the curve C is not closed.
Consider carefully the pointR (0, 0) in your analysis.
dt
You may use the fact that 1+t 2 = arctan(t) + C.
4.3.1.14. ∗. Hint. If we were to try to evaluate this integral directly,

then on the y = x2 − 4x + 3 part of C, the integrand would contain
2
x2 ey = x2 ex −4x+3 . That looks hard to integrate, so try Green’s theorem.
4.3.1.15. ∗. Hint. Beware the point (0, 0).
4.3.1.18. ∗. Hint. It is possible to evaluate this integral by three differ-

ent methods, one of them being direct evaluation (though it requires some
ingenuity). Try to find all three.
H H H
4.3.1.20. ∗. Hint. Write C F · dr − A C G · dr = C (F − AG) · dr.
4.3.1.21. ∗. Hint. Note that F(x, y) is not defined at (x, y) = (0, 0).
4.3.1.22. ∗. Hint. Note that F(x, y) is not defined at (x, y) = (0, 0).
4.3.1.24. ∗. Hint. (a) All points on the curve obey an equation that
contains x’s and y’s, but no z’s.
(b) Exploit conservativeness as much as possible.
4.3.1.25. Hint. Use Green’s theorem to convert the integral over C
into an integral over the region R in the xy-plane whose boundary is C.
Consider the sign of the integrand of the integral over R.
4.4 · Stokes’ Theorem

4.4.3 · Exercises
4.4.3.1. Hint. One approach is to first do
S
∂S
Then imagine slowly deforming the sketch to the get specified S’s
4.4.3.2. Hint. Define the vector field F(x, y, z) = F1 (x, y)ı̂ı + F2 (x, y) ̂.
4.4.3.3. Hint. First verify the vector identity ∇ × [φ∇
∇ψ + ψ∇
∇φ] = 0
4.4.3.4. Hint. To parametrize the curve x2 + y 2 = 1, z = y 2 , first

parametrize the circle x2 + y 2 = 1. That is, find x(t) and y(t) obeying
x(t2 ) + y(t)2 = 1. Then set z(t) = y(t)2 .
4.4.3.5. Hint. Apply Stokes’ theorem. Note that r(t) = x(t)ı̂ı + y(t) ̂ +
z(t) k̂ obeys x(t) + y(t) + z(t) = 3, for every t, and that x(t)ı̂ı + y(t) ̂ =
(1 + cos t)ı̂ı + (1 + sin t) ̂ runs counterclockwise around the circle of radius
1 centered on (1, 1).
4.4.3.6. ∗. Hint. The form of the integral should be quite suggestive.
4.4.3.7. ∗. Hint. The form of the integral should be quite suggestive.
4.4.3.8. ∗. Hint. What’s the title of this section?
RR
4.4.3.9. Hint. We are to evaluate a flux integral of the form S
∇×F·
n̂ dS. Sure looks like one side of Stokes’ theorem.
4.4.3.10. Hint. The vector field F looks too complicated for a direct
evaluation of the line integral. So, try Stokes’ theorem.
4.4.3.16. ∗. Hint. All three vertices of part (a) lie in the plane of part
(b).
4.4.3.17. ∗. Hint. The curve C is the boundary of a surface. To
guess the surface express the z component of r(t) in terms of the x and y
components.
4.4.3.18. ∗. Hint. The fact that the surface is not completely specified
is a big hint.
4.4.3.19. ∗. Hint. We are to evaluate the line integral of a complicated
vector field around a relatively complicated closed curve. (Sketch it!) That
certainly suggests that we should not try to evaluate the integral directly.
4.4.3.20. ∗. Hint. The integral looks messy. Compute the curl of F to
help gauge if Stokes’ theorem would be easier.
4.4.3.21. ∗. Hint. The form of the integrand is sugestive.
4.4.3.26. ∗. Hint. Let D be the disk in the plane x + y + z = 3 whose
boundary is C. Suppose that, as (x, y, z) runs over D, (x, y) runs over the
ellipse Dxy . We are told that the area of D is πR2 , but
RR we are not told the
0
area of DRR . So it is easier to deal with the integral D
dS than with the
integral D0 dxdy.
4.4.3.29. Hint. Given the form of F, direct evaluation looks hard.

The integral evaluations can be greatly simplified by using that the
centroid (x̄, ȳ) of any region R in the xy-plane is
RR RR
x dx dy y dx dy
x̄ = R ȳ = R
Area(R) Area(R)
4.4.3.30. ∗. Hint. Part (a) is a hint for part (b). Sketch the curve in
part (b).
4.4.3.31. ∗. Hint. For practice, evaluate the flux of part (a) twice —
once by direct evaluation and once using Stokes’ theorem.
4.4.3.32. ∗. Hint. By definition, D is connected if any two points in D
can be joined by a curve that lies completely in D.
By definition, D is simply connected if any simple closed curve in D
can be shrunk to a point continuously in D.
4.4.3.33. ∗. Hint. Review §4.1.2.
4.4.3.34. ∗. Hint. Considering that there are ten line segments in C, it
is probably not very efficient to use direct evaluation.
4.4.3.35. ∗. Hint. Direct evaluation looks hard.
H
4.4.3.36. ∗. Hint. Rewrite C E · dr as a surface integral.
4.4.3.37. ∗. Hint. What is x(t)2 +y(t)2 +z(t)2 = 2? How is x(t) relatex

to z(t)?
4.4.3.38. ∗. Hint. The intersection of the plane x + y + z = 1 with the
sphere x2 + y 2 + z 2 = 1 is a circle. Use symmetry to guess the centre of
the circle.
4.4.3.39. ∗. Hint. Sketch S.
4.4.3.40. Hint. You can avoid evaluating any integral by identifying S 0
as a simple geometric figure.
5 · True/False and Other Short Questions

5.2 · Exercises
5.2.2. ∗. Hint. Read (d), (e), (f), (g), (h) very carefully.
5.2.3. ∗. Hint. Beware that in part (f) a surface is defined to be closed if
and only if it is the boundary of a solid region E. Even though that is not the
usual definition, it is be used in this question.
5.2.4. ∗. Hint. (b) In general, for which values of x is the curvature of
y = f (x) zero?
(c) First parametrize x2 + z 2 = 1.
(d) First determine when r(u, v) has z = 0.
(e) What type of curve has curvature zero?
(f) What theorem relates the divergence of a vector field with flux integrals
of the vector field?
(g) What is the screening test for conservativeness in two dimensions?
(h) What is the definition of “parametrized by arclength”?
(i) What theorem relates line integrals to curls?
(j) What theorem relates flux integrals to divergences?
(k) Use Stokes’ theorem.
5.2.5. ∗. Hint. Read all of the statements very carefully. The details are
critical.
(a) Note the word anywhere.
(b) If you have not learned about simply connected domains, skip this part.
If you have, read the statement very carefully.
(d) If you have not learned about Kepler’s three laws, skip this part.
(h) Read the statement very carefully. It does not specify that C is closed.
(i) Review §1.5.
critical.
For part (d), note that the curve need not lie in a plane.
For part (g), note that the domain can have holes in it.
For parts (h) and (i), by definition, D is simply connected if any simply
closed curve in D can be shrunk to a point continuously in D.
critical.
For part (b), note that the curve need not lie in a plane.
For part (d), note that the domain can have holes in it.
For parts (i) and (j), by definition, D is simply connected if any simply
closed curve in D can be shrunk to a point continuously in D.
critical. R R
(a) The integral C f ds = 0 is not of the form C F · dr.
(d) F and G can be any R vector fields.
(e) Think about how C f ds is defined.
∂r
(f) Look at ∂u ∂r
× ∂u very closely.
(g) The integral is completely independent of x(u, v) and y(u, v).
critical.
(b) Read the statement very carefully. “simply connected” plays no role
here. The vector field F is not required to be conservative.
(e) Recall that S is closed when it is the boundary of a solid region V .
(g) Assume that the constant |v| is not zero.
(j) If you have not learned about Kepler’s three laws, skip this part.
5.2.13. ∗. Hint. (g) Be careful. The power in the denominator is important.
(j) Beware the sign.
5.2.20. ∗. Hint. Review Corollary 4.3.5.
RR
5.2.21. ∗. Hint. (b) S ∇ ×F· n̂ dS is a flux integral over the closed surface
S. H H H
(c) Consider C F · dr − C G · dr = C (F − G) · dr.
Appendix C
Answers to Exercises
1 · Curves
1.1.1 · Exercises
p
1.1.1.1. Answer. (a) r(y) = a2 − y 2 ı̂ı + y ̂, 0 ≤ y ≤ a
(b) x(φ), y(φ) = a sin φ, −a cos φ , π2 ≤ φ ≤ π
(c) x(s), y(s) = a cos( π2 − as ), a sin( π2 − as ) , 0 ≤ s ≤ π2 a
√
1.1.1.2. Answer. (1, 25), (−1/ 2, 0), (0, 25).
1.1.1.3. Answer. The curve crosses itself at all points (0, (πn)2 ) where
n is an integer. It passes such a point twice, 2πn time units apart.
1.1.1.4. Answer. (a) (a + aθ, a)
(b)(a + aθ + a sin θ, a + a cos θ)
q
2
1.1.1.5. Answer. z = − 12 1 − y2 − y
4
1.1.1.6. Answer. The particle is moving upwards from t = 1 to t = 2,

and from t = 3 onwards. The particle is moving downwards from t = 0 to
t = 1, and from t = 2 to t = 3.
The particle is moving faster when t = 1 than when t = 3.
1.1.1.7. Answer.
r(t + h)
r(t)
r(0)
The red vector is r(t+h)−r(t). The arclength of the segment indicated

by the blue line is the (scalar) s(t + h) − s(t).
Remark: as h approaches 0, the curve (if it’s differentiable at t) starts
to resemble a straight line, with the length of the vector r(t + h) − r(t)
359
APPENDIX C. ANSWERS TO EXERCISES 360
approaching the scalar s(t + h) − s(t). This step is crucial to understanding

Lemma 1.1.4.
1.1.1.8. Answer. Velocity is a vector-valued quantity, so it has both
a magnitude and a direction. Speed is a scalar — the magnitude of the
velocity. It does not include a direction.
1.1.1.9. ∗. Answer. (c)
1.1.1.10. Answer. See the solution.
1.1.1.11. ∗. Answer. (d)
1.1.1.12. Answer.
√ velocity = −a sin tı̂ı + a cos t ̂ + c k̂
speed = a2 + c2
acceleration = −a cos tı̂ı − a sin t ̂
The path is a helix with radius a and with each turn having height 2πc.
(2,0,1)
1.1.1.13. ∗. Answer. (a) T̂(1) = √
5
(b) 13 53/2 − 8

1.1.1.14. Answer. 2
√
1.1.1.15. Answer. length = a2 + b2 T
1.1.1.16. Answer. 1
1.1.1.17. ∗. Answer. (a) 20 3
π3
(b) x(t) = −2π − 2t, y(t) = −2πt, z(t) = 3 + π2 t
(a) r0 (t) = − 3 sin t, 3 cos t, 4

1.1.1.18. ∗. Answer.
(b) 5
1.1.1.19. Answer. (a) x(θ) = 3 cos θ, y(θ) = 3 sin θ, z(θ) = 6 cos θ +
9 sin θ, 0 ≤ Rθ ≤√2π
2π
(b) s = 0 45 + 45 cos2 θ − 108 sin θ cos θ dθ
1
√
1.1.1.20. ∗. Answer. (a) 27 10 10 − 1
2
√
(b) 27 10 10 − 1
t3 t
1.1.1.21. ∗. Answer. s(t) = 3 + 2
h 3/2 3/2 i
8
1.1.1.22. ∗. Answer. 27 2 + 94 bm − 2 + 94 am
√
1.1.1.23. Answer. (a) r(x) = xı̂ı + x ̂ + 23 x3/2 k̂
(b) 21
(c) 6ı̂ı + 3 ̂ + 6 k̂
(d) −6ı̂ı − 12 ̂ + 12 k̂
1.1.1.24. Answer. |t|
1.1.1.25. ∗. Answer. (a) r(u) = u3 ı̂ı + 3u2 ̂ + 6u k̂

(b) 7
(c) 2
(d) 1
π2 t 3
− t2 ı̂ı + (t − sin t) ̂ + 1 2t

1.1.1.26. ∗. Answer. (a) r(t) = 2 2e − t k̂
(b) t = π

(c) −π 2 ı̂ı + 2 ̂ + e2π − 1 k̂
1.1.1.27. ∗. Answer. (a) 21
(b) 6
(c) 2ı̂ı + 4 ̂ + 4 k̂
(d) − 83 2ı̂ı + ̂ − 2 k̂
x(t)y 0 (t)−y(t)x0 (t)
1.1.1.28. Answer. x2 +y 2
1.1.1.29. Answer. Volume: 540π

Surface area: 360π
50
1.1.1.30. Answer. q ≈ 5.3 cm
1
π 9 + 400π 2
e−αt −1 −αt
1.1.1.31. Answer. r(t) = r0 − α v0 + g 1−αt−e
α2 k̂
1.2.1 · Exercises
1.2.1.1. Answer. t − 1
√
1.2.1.2. Answer. sin(1/2), cos(1/2), 3/2
1.2.1.3. Answer. A
√
1.2.1.4. ∗. Answer. (a) 43 , − 43 , − 21

(b) R(s) = 2 sin3 ( 3s ), 2 cos3 ( 3s ), 3 sin( 3s ) cos( 3s )

√
1.2.1.5. ∗. Answer.
(a) 2
(b) √s2 cos ln √s2 , sin ln √s2 with s > 0
1.2.1.6. Answer. (cos z, z sin z, z) for 0 ≤ z < π/2. The curve is (the
first quarter-turn of) a spiral, with width in the x-direction 2, and increas-
ing width in the y-direction. The parameter z is the height, as well as a
radian measure for the spiral.
√
1.2.1.7. Answer. When s ≤ 13 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (2 2 − 3s)2/3 − 1) , − (2 2 − 3s)2/3 − 1
2 3
√
and when s > 13 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (3s + 2 − 2 2)2/3 − 1 , (3s + 2 − 2 2)2/3 − 1
2 3
1.3 · Curvature
1.3.1 · Exercises
1.3.1.1. Answer.
T̂
N̂
x
1
1
ρ = 3, κ = 3
1.3.1.2. Answer. T̂(t) = (cos t, − sin t), T̂(s) = (cos(s/3), − sin(s/3)),

N̂(t) = (− sin t, − cos t), N̂(s) = (− sin(s/3), − cos(s/3))
1.3.1.3. Answer. lim κ(t) = 0
t→∞
ds
√
1.3.1.4. Answer. dt = e2t + 9 + cos2 t
1.3.1.5. Answer.
dT̂ 1
= √ − sin t − cos t, − sin t + cos t
dt 2
dT̂ 1 √ √
=√ − sin ln s/ 2 − cos ln s/ 2 ),
ds 2s
√ √
− sin ln s/ 2 + cos ln s/ 2
1.3.1.7. Answer.
A v(t) = (et , 2t + 1)
B a(t) = (et , 2)
ds p 2t
C = e + (2t + 1)2
dt
!
et 2t + 1
D T̂(t) = p ,p
e2t + (2t + 1)2 e2t + (2t + 1)2
et |1 − 2t|
E κ(t) =
(e2t + (2t + 1)2 )3/2
1.3.1.8. Answer. κ(t) = √1
2
a b
1.3.1.9. Answer. κmax = b2 , κmin = a2 .
1.3.1.10. ∗. Answer. (a) κ(0) = 2−3/2

(b) (x + 2)2 + (y − 3)2 = 8
1.3.1.11. ∗. Answer. (a)

r(t) = t − sin t , 1 − cos t
y
2π 4π x
(b) κ(t) = √1
23/2 1−cos t
(c) 4
(d) (x − π)2 + (y + 2)2 = 16
1.3.1.12. Answer. κ(s) = πs

√
1.3.1.13. ∗. Answer. The maximum values occur at (x, y) = ± 1/ 4 5 , 13 5−3/4 .

The limits limx→±∞ κ(x) = 0.

1.4.1 · Exercises
1.4.1.1. Answer.
T̂
N̂
B̂ points out of the page (towards the reader).

1.4.1.2. Answer. arclength
1.4.1.3. Answer. a(t) and b(t) have negative torsion, c(t) has zero
torsion.
1.4.1.4. Answer. See solution.
1.4.1.5. ∗. Answer. (a), (b)
z = x2 + y 2
N̂ T̂
B̂
y
x=y
x
(c) The torsion is zero.
1.4.1.6. ∗. Answer. (a) r0 (t) = et + e−t ı̂ı + et − e−t ̂ + 2 k̂, r00 (t) =

et − e−t ı̂ı + et + e−t ̂,

κ(t) = 2+e2t1+e−2t
√ h i
(b) 2 e − 1e
3
1.4.1.7. Answer. 181
ı̂ı+t ̂+t2 k̂
1.4.1.8. Answer. T̂(t) = √
1+t2 +t4
t2 ı̂ı−2t ̂+k̂
B̂(t) = √
1+4t2 +t4
3
) ı̂ı+(1−t4 ) ̂+(2t+t3 )k̂
N̂(t) = −(t+2t √ √
1+t2 +t4 1+4t2 +t4
√
1+4t +t2 4
κ(t) = [1+t 2 +t4 ]3/2
τ (t) = 1+4t22 +t4
1.4.1.9. Answer. When c = 0, the plane is z = 1. When c = 1/5, the

plane is (1/25)x + 3y − (30/e)z = −10.
1.4.1.10. ∗. Answer.
√
(a) 2x + y + 3z = 6
2 1+9t2 +9t4
(b) κ(t) = [1+4t2 +9t4 ]3/2
1.4.1.11.√∗. Answer. (a) 2

(b) − 23 ı̂ı − 12 ̂ + π6 k̂
√
1
(c) B̂ = 2√ 2
ı̂ı − 2√32 ̂ + √12 k̂
1.4.1.12. ∗. Answer. (a) R(t) = (−1, 0, π 2 ) + t(0, −1, 2π)

4t
(b) aT (t) = √1+4t2
√
1.4.1.13. ∗. Answer. (a) 5 t
(b) aT (t) = sin tı̂ı + cos t ̂ + 2 k̂
(c) aN (t) = t cos tı̂ı − t sin t ̂
1
(d) κ(t) = 5t
1.4.1.14. ∗. Answer. (a) r(θ) = [−1+3 cos θ]ı̂ı +3 sin θ ̂ +[10−6 cos θ] k̂,
0 ≤ θ < 2π √
(b) At (2, 0, 4), T̂ = ̂, N̂ = −ı̂ı√+2
5
k̂ ı+k̂
, B̂ = 2ı̂√ 5
, κ(0) = 35
√
t2 ı̂ı+ 2 t ̂+k̂
1.4.1.15. ∗. Answer. (a) T̂(t) = t2 +1
√
2
(b) (t2 +1)2
√
4 ı̂ı−3√ 2 ̂−4k̂
(c) 50
1.4.1.16. ∗. Answer. (a) One possible parametrization is r(θ) =

− cos θ − sin θ) k̂ with 0 ≤ θ ≤ 2π.
cos θ ı̂ı + sin θ ̂ + (1 √
3
(b) κ(θ) = [2−sin(2θ)] 3/2
√ √
(c) maximum curvature = 3 at √ı̂ı2 + √̂2 +(1− 2) k̂ and − √ı̂ı2 − √̂2 +
√
(1 + 2) k̂ and minimum curvature = 31 at − √ı̂ı2 + √̂2 + k̂ and √ı̂ı2 − √̂2 + k̂
2t2 ı̂ı+2t ̂+k̂

1.4.1.17. ∗. Answer. T̂(t) = 2t2 +1
2
N̂(t) = 2t ı̂ı−(2t2t2−1) ̂−2t k̂
ı 
+1
κ(t) = (2t22t +1)2
1
3/2
1.4.1.18. ∗. Answer. (a) 3 2 −1
(b) −ı̂√ı−̂
2

(c) 12

1.4.1.19. ∗. Answer. (a) v(t) = 1 , −1 , t
√
(b) ds
dt (t) = 2 + t2
(c) a(t) = 0 , 0 , 1
√
(d) κ(t) = [2+t22]3/2
(−t , t , 2)
(e) N̂(t) = √ 2 2(2+t )
(f) x + y = 3
(g) (2, 1, 2)
√
t2 ı̂ı+ 2t ̂+k̂
1.4.1.20. ∗. Answer. (a) T̂(t) = t2 +1
√
2
(b) κ(t) = (t2 +1)2
√
(c) κ(0) = 2
(d) N̂(0) = ̂
(e) B̂(0) = −ı̂ı
1.4.1.21. ∗. Answer. (a) x = 1 − 2t, y = −1 + t, z = −1 + 3t
(b) 3x − 3y − z = −1
√
5 π2
1.4.1.22. ∗. Answer. (a) 2
1
(b) κ(t) = 5t
1.4.1.23. ∗. Answer. (a) 8
(b) T̂(1) = √12 (1, 1, 0), N̂(1) = (0, 0, −1)
(c) κ(1) = 18
1.4.1.24. ∗. Answer. (a) 52

√ √
(b) T̂ π6 = 15 − 32 , 3 2 3 , 4 , N̂ π 1
3, 1, 0 , B̂( π6 = 1

6 = 2 5 −
√
2, 2 3, −3)
1.4.1.25. Answer. (a) r(θ) = cos θ ı̂ı + sin θ ̂ + cos(2θ) k̂ 0 ≤ θ < 2π
(b) 51
√ √
(c) z = 2 x − 2 y √ √
(d) radius 1/κ(π/4) = 5 and centre − 2 2 , −2 2 , 0
4
1.4.1.26. ∗. Answer. ı
9 (ı̂ − 4 ̂ + k̂)
√
ı̂ı+t̂
√+ 3tk̂
1.4.1.27. ∗. Answer. (a) T̂(t) = 1+4t2
√
(b) N̂(t) = −4t2√ı̂ı+̂+ 3k̂
1+4t2
(c) \textcircled{3}
√
(d) − 3y + z = 0
−3/2
(e) κ(t) = (1 + 4t2 )
(f) The curvature κ(t) achieves its maximum value at r(0) = (0, 0, 0).
(g) The curvature√never achieves √
a minimum.
(h) ı̂ı = u2 , ̂ = v− 4 3 w , k̂ = 3 v+w
4 , r(t) = t u + t2 v
y
y = x2
v
C
u x

The curve a(t), b(t) = (t, t2 ) is the curve y = x2 . It is “curviest” at
the origin, which is consistent with part (f). It becomes flatter and flatter
as |t| increases, but never achieves “perfect flatness”, which is consistent
with (g).
1.4.1.28. ∗. Answer. See the solution.
√ √
1.4.1.29. ∗. Answer. (a) T̂ = √16 0, 2, − 2 , N̂ = − √139 6, 1, 2 ,
√ √ √
B̂ = √113 − 1, 2, 2 2 , κ = 3√133 = 939
√
dv
(b) (i) = 5√32
dt
√
(ii) v = (0, 2, −1).

1.4.1.30. ∗. Answer. (a) v(t) = − sin t , cos t , c cos t , a(t) = −
cos t , − sin t , −c sin t
√
(b) v(t) = 1 + c2 cos2 t
2
(c) −c
√ sin t cos t
1+c2 cos2 t
(d) The curve lies on the plane z = cy.
√
17
1.4.1.31. ∗. Answer. (a) 4
(b) √417
√
(c) (i) 4 π
(ii) 16π , −4 , −4π
√
(iii) 4 17 π

1.6.1 · Exercises
R
1.6.1.1. Answer. C
ds
1.6.1.2. Answer. (a) See the solution.
(b) 8
4√ 2
1.6.1.3. Answer. 21 3
(27 − 1) + √
5 3
(25 − 1)
1.6.1.4. Answer. π kg
1.6.1.5. Answer. 26
53/2 −1
1.6.1.6. Answer. (a) 12
3/2
8−3
(b) 3/2
1
1.6.1.7. Answer. 2 ln 2

t2
1.6.1.8. ∗. Answer. (a) r(t) = tı̂ı + 1 + 2 ̂ + sin t k̂
2

(b) r(π/2) = π2 ı̂ı + 1 + π8 ̂ + k̂
π2 1
(c) 8 − 2
1.6.1.9. ∗. Answer. 2e3

√ 1

1.6.1.10. ∗. Answer. (a) 1+5π 2
− ı̂ı − π ̂ + 2π k̂
1

(b) 15 (1 + 5π 2 )3/2 − 1
(c) z = x2 + y 2
(d)
z
x
y
412 92 4736

1.6.1.11. Answer. 55 , − 55 , 693

1.7.4 · Exercises
1.7.4.1. Answer.
W N̂
−mg̂
1.7.4.2. Answer. time

1.7.4.3. Answer. positive
E
1.7.4.4. Answer. y = mg — just like a circular culvert (if the culvert is
high enough).
1.7.4.5. Answer. 2940 J

√
1.7.4.6. Answer. at least 5 9.8 m/s
√
1.7.4.7. Answer. − √32 + 2.352, − √52 + 3.92, −2 2 + 3.136
r
9.8 √1
1.7.4.8. Answer. √
6
100 + 2
≈ 20 m/s
1.7.4.9. Answer. |v| > 504 kph
1.7.4.10. Answer. U = mg dy
ds
1.7.4.11. Answer. (a) M = mg̂ · T̂

(b) negative
(c) − 1960
√
3
≈ −1131.6 N
E
1.7.4.12. Answer. (a) yS =
mg 
yA −3
24 (E − mgyA ) 3
(b) = 4mg  q  (or equivalent)
3/2
yA −3 2 yA −3 2

9+7 3
9 + 7 3
(c) The skateboarder makes it up to the ceiling, but falls off rather than
making it all the way around. Ouch.
1.7.4.13. Answer. (a), (b) See the solution.
h 2 2 i1/2
(c) 2 a gb
+b
π

1.8.1 · Exercises
1.8.1.1. Answer. The upper sketch below contains the points, (x1 , y1 ),
(x3 , y3 ), (x5 , y5 ), that are on the axes. The lower sketch below contains
the points, (x2 , y2 ), (x4 , y4 ), that are not on the axes.
y
(0, 1)
π
π 2
(−2, 0) (3, 0)
x
y
(−1, 1) (1, 1)
√
2
3π
4 π
4
x
r1 =√3, θ1 = 0
π
r2 = 2, θ2 = 4
r3 = 1, θ3 = π2
√ 3π
r4 = 2, θ4 = 4
r5 = 2, θ5 = π

1.8.1.2. Answer. (a) r = 2 , θ = nπ, n odd integer or r = −2 , θ =
nπ, n even integer
√ π
√ 5π

(b) r = 2 , θ = 4 + 2nπ or r = − 2 , θ = 4 + 2nπ , with n
integer. √ √
5π π

(c) r = 2 , θ = 4 + 2nπ or r = − 2 , θ = 4 + 2nπ , with n
integer.
1.8.1.3. Answer. (a) Both êr (θ) and êθ (θ) have length 1. The angle
between them is π2 . The cross product is êr (θ) × êθ (θ) = k̂.
(b) Here is a sketch of (xi , yi ), êr (θi ), êθ (θi ) for i = 1, 3, 5 (the points
on the axes)
y
er ( π2 )
eθ ( π2 ) eθ (0)
(0, 1)
er (π) (−2, 0) er (0)

(3, 0) x
eθ (π)
and here is a sketch (to a different scale) of (xi , yi ), êr (θi ), êθ (θi ) for
i = 2, 4 (the points off the axes).
er ( 3π ) y eθ ( π ) er ( π4 )
4 4
(−1, 1) (1, 1)
3π
eθ ( 3π
4
) 4 π
4
x
1.8.1.4. ∗. Answer. (a) ↔ (E)
(b) ↔ (B)
(c) ↔ (F)
(d) ↔ (C)
(e) ↔ (A)
(f) ↔ (D)

f (θ)2 +2f 0 (θ)2 −f (θ)f 00 (θ)
1.8.1.5. Answer. κ(θ) = [f (θ)2 +f 0 (θ)2 ]3/2
1.8.1.6. Answer. κ(θ) = √3 = √ 3

23/2 a 1−cos θ 2 2ar(θ)

1.9.1 · Exercises
1.9.1.1. ∗. Answer. (a) See the solutions.
(b) f (r) = 0 for all r ≥ 0.
(c) Any f (r) which is a positive constant times − r13 works.
1.9.1.2. ∗. Answer. (a) See the solution.

h2
(b) |a(t)| = r(t) 3
2 · Vector Fields
2.1.1 · Exercises
2.1.1.1. Answer.
 
> 0
 when x > 0 

v(x, y) · ı̂ı = 0 when x = 0

 

<0 when x < 0
and  
> 0

 when −2 < x < 2 


v(x, y) · ̂ = 0 when x ∈ {−2, 2}

 

< 0 when x < −2 or x > 2 
at least for (x, y) shown in the sketch.
2.1.1.2. Answer.
 
> 0

 when y > −x 


v(x, y) · ı̂ı = 0 when y = −x

 

< 0 when y < −x 
and  
> 0

 when y < x 


v(x, y) · ̂ = 0 when y = x

 

< 0 when y > x 
at least for (x, y) shown in the sketch.
2.1.1.3. Answer. v(x, y) = √−|y|

2 2
(x, y)
x +y
2.1.1.4. Answer. P > 0

Q>0
∂Q
∂x < 0
∂Q
∂y > 0
2.1.1.5. Answer. (a) (1.01 , 1.01)

(b) (0 , 0)
(c) (0 , 0)
2.1.1.6. Answer. (0 , −10)
2.1.1.7. Answer. v(x, y) = √ −y (x, y)

2 x +y 2
2.1.1.8. Answer. If your face is at the origin, then v(x, y, z) = − x2 +yα2 +z2 (x, y, z)
for some positive constant α.
2.1.1.9. Answer.
2.1.1.10. Answer.
y
2.1.1.11. Answer.
2.1.1.12. Answer.
y
2.1.1.13. Answer. (a)

y
(b)
(c)
y
−5G(x,y) 3G(2−x,3−y) 7G(4−x,−y)

2.1.1.14. Answer. f (x, y) = (x2 +y 2 )3/2
+ ((x−2) 2 +(y−3)2 )3/2 + ((x−4)2 +y 2 )3/2

a. v(p) = 1 − p2 2√ p
1
2.1.1.15. Answer. 3
, − 4
b. V(x, y, z) = − x6 , − y6 , z2 or equivalent


2.2.2 · Exercises
2.2.2.1. Answer.
y
3
x
1 2 3
2.2.2.2. Answer. v(x, y) = (−x − y , x − y)
x2 y2
2.2.2.3. ∗. Answer. (a) 2 = 2 +C
(b)
2.2.2.4. ∗. Answer. x = y 2 , z = ey
2.2.2.5. ∗. Answer. The field lines are y = C 0 x3 with C 0 a nonzero
constant, as well as x = 0 and y = 0.

2.3.1 · Exercises
2.3.1.1. Answer. In general, false.
2.3.1.2. Answer. a. C
b. B
c. C
d. B
2.3.1.3. Answer. Let ϕ be a potential for F. Define φ = ϕ+ax+by+cz.
Then ∇ φ = ∇ ϕ + (a, b, c) = F + (a, b, c).
2.3.1.4. Answer.
a If F+G is conservative for any particular F and G, then by definition,
there exists a potential ϕ with F + G = ∇ ϕ.
Since F is conservative, there also exists a potential ψ with F = ∇ ψ.
But now G = (F + G) − F = ∇ ϕ − ∇ ψ = ∇ (ϕ − ψ). That means
the function (ϕ − ψ) is a potential for G. However, this is impossible:
since G is non-conservative, no function with this property exists.
So it is not possible that F + G is conservative. It must be non-
conservative.
b Counterexample: if F = −G, then F + G = 0 = ∇ c for any constant

c.
c Since both fields are conservative, they both have potentials, say
F = ∇ ϕ and G = ∇ ψ. Then F + G = ∇ ϕ + ∇ ψ = ∇ (ϕ + ψ). That
is, (ϕ + ψ) is a potential for F + G, so F + G is conservative.
2.3.1.5. ∗. Answer. Yes, F is conservative on D. A potential is

ϕ(x, y) = arctan xy .
2.3.1.6. Answer. ϕ = 12 x2 + xy − 12 y 2
x
2.3.1.7. Answer. ϕ = ln |x| − y
2.3.1.8. Answer. None exists: ∂F 1 3

∂z = 3 x , while
2 ∂F3
∂y = 13 x3 + 1, so F
fails the screening test, Theorem 2.3.9.
1
2.3.1.9. Answer. ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 )
2.3.1.10. Answer. (a) F is conservative with potential φ(x, y, z) =
1 2 2 3 2
2 x − y + 2 z + C for any constant C.
(b) F is not conservative.
2.3.1.11. Answer. (a) A = 2, B is arbitrary.
2
(b) ϕ(x, y, z) = xe(z ) + By 2 z 3 + C for any constant C.
2.3.1.12. Answer. v = m xxı̂2ı+y̂ 

+y 2
ϕ = 12 m ln(x2 + y 2 ) + C for any constant C
2.3.1.13. Answer.
√ It can never escape the sphere centred at the origin
with radius 20.
√
2.3.1.14. Answer. 14
2.3.1.15. Answer. ϕ = f 2 (x) + g(y)h(z) is a potential for F, so F is
conservative.
2.3.1.16. Answer. The line through the origin in the direction of the
vector (2, 1, 2).

2.4.2 · Exercises
1
2.4.2.1. Answer. 6
2.4.2.2. Answer. a. A
b. B
c. A
d. B
2.4.2.3. Answer. 0
2.4.2.4. Answer. 5
2.4.2.5. ∗. Answer. a = 1, b = c = 0
2.4.2.6. Answer. (a) Not conservative
(b) Not conservative
(c) Not conservative
(d) Conservative

2.4.2.7. ∗. Answer. (a) The (largest possible) domain is D = (x, y, z) x2 +
y 2 6= 0 .
(b) ∇ × F = 0 on D
R
(c) C F · dr = 4π
(d) F is not conservative.
2.4.2.8. Answer. 9 12 for all paths from (1, 0, −1) to (0, −2, 3)
π2
2.4.2.9. Answer. 2(e − 1) + 2 + 3π
2.4.2.10. Answer. (a) − 14

(b) −1
2.4.2.11. ∗. Answer. − 40
3
2.4.2.12. Answer. (a) λ = −1

(b) φ(x, y, z) = 2x3 yz 2 − xyz + y 2 + K, for any constant K
(c) e2 + 2e − 2
7
2.4.2.13. ∗. Answer. 3
1 1

2.4.2.14. ∗. Answer. 3 1− 23/2
≈ 0.2155
π3 π2
2.4.2.15. ∗. Answer. 8 + 4 −1
2.4.2.16. ∗. Answer. − 32
2.4.2.17. ∗. Answer.
R The line integral is independent of path because
it is of the form C F · dr with F being a conservative field. The value of
the integral is 1 + π2 .
1
2.4.2.18. ∗. Answer. 2
2.4.2.19. ∗. Answer. πeπ

2.4.2.20. ∗. Answer. (a) α = 1, β = γ
(b) ee − β(e + 1)
2.4.2.21. ∗. Answer. (a) 0
(b) Yes. In fact F = ∇ f with f = sin x + 2y − cos y + ez .
(c) −4
R ∗. Answer.
2.4.2.22. (a) ∇ × F = 0. F is conservative.
(b) C F · dr = 2π 2
2.4.2.23. ∗. Answer. (a) a = −1, b = 3
(b) f (x, y, z) = xyex + yz 3 + C works for any constant C
(c) πeπ − 2
(d) πeπ − 32 15
2.4.2.24. ∗. Answer. (a) A = 2, B = 3

(b) ϕ(x, y, z) = xy 2 e3z + x2 y 3 is one allowed scalar potential.
(c) 6 + e − 2[e − 1] = 8 − e ≈ 5.2817
2.4.2.25. ∗. Answer. (a) a = π, b = 3
(b) ϕ(x, y, z) = x2 sin(πy) − xez − 3yez + C for any constant C
(c) −8
(d) − 13
2
2.4.2.26. ∗. Answer. (a) f (x, y, z) = yeyz + y cos2 x + C works for any

constant C
2
(b) 2eπ − e−π − 1
2.4.2.27. ∗. Answer. (a) 0.
(b) F is conservative with potential ϕ(x, y, z) = x2 + y 2 + z 2 . So the
integral is ϕ(a1 , a2 , a3 ) − ϕ(0, 0, 0) = a · a.
2.4.2.28. ∗. Answer. (a) ∇ × F = 0
(b) πe
2 −1
2.4.2.29. ∗. Answer. (a), (b) f (x, y) = y sin(x2 ) + cos(y) + C is a

potential for any constant C. Because F has a potential, it is conservative.
(c) −1 − π2 sin(1)
2.4.2.30. ∗. Answer. (a) p = 2, m = 2, n = 2, but q ∈ R is completely
free
(b) 4q
2
3/2
2.4.2.31. Answer. (a) 3 14 − 1 ≈ 34.26
(b) sin 1 + 32 ≈ 2.3415
1
2.4.2.32. ∗. Answer. 2π + 3
2.4.2.33. ∗. Answer. (a) 18

(b) 3 − e
t2

2.4.2.34. ∗. Answer. (a) r(t) = tı̂ı + 1 + 2 ̂ + sin t k̂
2
(b) r1 = π2 ı̂ı + 1 + π8 ̂ + k̂
2
(c) π8 − 12
2.4.2.35. ∗. Answer. (a) -5

(b) One possibility is the path consisting of the line segment from (2, 2)
to (2, −3), followed by the line segment from (2, −3) to (1, −3), followed
by the line segment from (1, −3) to (1, 1).
Another possibility is the path from (2, 2) to (1, 1) along the parabola
27x2 − 80x + 54.
2.4.2.36. ∗. Answer. One possibility is the path consisting of the line
segment from (0, 0) to (0, 1), followed by the line segment from (0, 1) to
(2, 1), followed by the line segment from (2, 1) to (2, 0).
Another possibility is the path tracing out the half ellipse cos t + 1 , π4 sin t ,

with t running from π to 0.

2.4.2.38. ∗. Answer. a = 4
2.4.2.39. ∗. Answer. (a) ∇ ×F = [−(b+2)x cos(x2 z)+(b+2)x3 z sin(x2 z)] ̂+
2
(6 − a)x2 e3x k̂
(b) a = 6, b = −2
2
(c) f (x, y, z) = xye3x + sin(x2 z) + C for any constant C
(d) 31 e3 + sin 1 − 13
23
2.4.2.40. ∗. Answer. (a) 15 = 1.53̇
(b) 23 143/2 − 1 ≈ 34.26

(c) sin 1 + 32 ≈ 2.3415

2.4.2.41. ∗. Answer. (a) A = −4, B = −2

(b) ϕ(x, y, z) = −x4 y 2 z + yz 3 + C with C being an arbitrary constant.
(c) −2
37
(d) − 24 ≈ −1.5417
1
(e) 2

2.4.2.42. ∗. Answer.
3 (a) v(t) = t2 , t3 , −t2
4 3
(b) r(t) = t3 + 1 , t4 + 2 , − t3 + 3
√
2
(c) κ(t) = t2 (2+t2 )3/2
(d) 2T 4 + T 6
2.4.2.43. ∗. Answer. (a) 8

(b) 81
16 5
(c) −
5 (3 −1) ≈ −774.4
(d) 0, 0, − 98
3 · Surface Integrals
3.1 · Parametrized Surfaces
3.1.1 · Exercises
3.1.1.1. Answer. r(x, y) = xı̂ı + y̂ + (ex+1 + xy)k̂
3.1.1.2. ∗. Answer. parabolic bowl
3.1.1.3. ∗. Answer. (a) No

(b) Yes
(c) Yes
(d) Yes
(e) No
3.1.1.4. ∗. Answer. (a) No.
(b) Yes.
(c) No.
(d) Yes.
(e) Yes.
3.1.1.5. ∗. Answer. (a) No
(b) Yes
(c) Yes
3.1.1.6. ∗. Answer. (a) A, F

(b) B, E
(c) G, J
(d) H, L
3.1.1.7. Answer. (a) (x, y, z) = (2 + √1 cos θ, 2 + √1 cos θ, 4 + sin θ),
2 2
0 ≤ θ ≤ 2π.

3.2.1 · Exercises
3.2.1.1. Answer. Yes. The plane z = 0 is the tangent plane to both

surfaces at (0, 0, 0).
3.2.1.3. Answer.

(x − x0 , y − y0 , z − z0 ) = t − fx (x0 , y0 ) , −fy (x0 , y0 ) , 1 or
x = x0 − tfx (x0 , y0 ) y = y0 − tfy (x0 , y0 ) z = f (x0 , y0 ) + t
3.2.1.4. Answer. The normal plane is n · (x − x0 , y − y0 , z − z0 ) = 0,
where the normal vector n = ∇ F (x0 , y0 , z0 ) × ∇ G(x0 , y0 , z0 ).
3.2.1.5. Answer. Tangent line is

x = x0 + t gy (x0 , y0 ) − fy (x0 , y0 )

y = y0 + t fx (x0 , y0 ) − gx (x0 , y0 )

z = z0 + t fx (x0 , y0 )gy (x0 , y0 ) − fy (x0 , y0 )gx (x0 , y0 )
3.2.1.6. ∗. Answer. 2x + y + 9z = 2
3.2.1.7. ∗. Answer. 2x + y + z = 6
3.2.1.8. ∗. Answer. z = − 43 x − 32 y + 11
4
3.2.1.9. ∗. Answer. (a) 2ax − 2ay + z = −a2

(b) a = 21 .
3.2.1.10. ∗. Answer. x + 3y − 2z = 1
3.2.1.11. ∗. Answer. y = 2x − 2
8 6
3.2.1.12. ∗. Answer. The tangent plane is 25 x − 25 y − z = − 85 .
4 8 6
The normal line is (x, y, z) = (−1, 2, 5 ) + t( 25 , − 25 , −1).
3.2.1.13. ∗. Answer. ±(1, 0, −2)
√1
, −1 , − 12 and √1 , −1 , − 21

3.2.1.14. ∗. Answer. 2
− 2
3.2.1.15. ∗. Answer. ± 21 , −1, −1

3.2.1.16. ∗. Answer. (a) (1, 0, 3)

(b) (3, 3, −1)
(c) r(t) = (1, 1, 3) + t(3, 3, −1)
3.2.1.17. ∗. Answer. 49.11◦ (to two decimal places)
3.2.1.18. Answer. The horizontal tangent planes are z = 0, z = e−1
and z = −e−1 . The largest and smallest values of z are e−1 and −e−1 ,
respectively.

3.3.6 · Exercises
p
3.3.6.1. Answer. ab 1 + tan2 θ = ab sec θ
√
3.3.6.2. Answer. (a) 12 a2 b2 + a2 c2 + b2 c2
(b) See the solution.
πah
3.3.6.3. Answer. 2
116
3.3.6.4. Answer. 15 π
π 3/2
3.3.6.5. ∗. Answer. (1 + 4a2 ) − 1
6
√
3.3.6.6. ∗. Answer. 5 2π
4
√ √
3.3.6.7. ∗. Answer. 15 9 3−8 2+1
p
3.3.6.8. ∗. Answer. (a) F (x, y) = 1 + fx (x, y)2 + fy (x, y)2
R 2π R 1 2r
(b) (i) 0 dθ 0 dr √4−r 2
(b) (ii) 16π

√
3.3.6.9. ∗. Answer. 255 2π ≈ 1132.9
3.3.6.10. ∗. Answer. a2 [π − 2]
√
3.3.6.11. Answer. 2π
3.3.6.12. Answer. (2π)2 Rr
a a a

3.3.6.13. Answer. 2 , 2 , 2
3.3.6.14. Answer. 16a2

π 3
√
3.3.6.15. ∗. Answer. 2a a 2 + b2
3.3.6.16. Answer. (a) 4πa2n+3
(b) 3abc
(c) π3
8
3.3.6.17. ∗. Answer. (a) 3
(b) 16
3
1
3.3.6.18. ∗. Answer. (a) 6
(b) 21
h i
8 13 3/2

3.3.6.19. ∗. Answer. 27 4 −1
3.3.6.20. ∗. Answer. 9π
3.3.6.21. ∗. Answer. (a)
2 2
r(θ, z) = (3 − z) cos θ ı̂ı + (3 − z) sin θ ̂ + z k̂
3 3
0 ≤ θ < 2π, 0 ≤ z ≤ 3
(b) 1
0, 0, a3

3.3.6.22. ∗. Answer.
3.3.6.23. ∗. Answer. 2π
3.3.6.24. ∗. Answer. − 14
3
√
2π
3.3.6.25. ∗. Answer. 4
16
3.3.6.26. ∗. Answer. 3 π
0, 0, 32

3.3.6.27. ∗. Answer.
3.3.6.28. ∗. Answer. 4
81
3.3.6.29. ∗. Answer. 16
3.3.6.30. ∗. Answer. (a) r(Y, θ) = eY sin θ ı̂ı + Y ̂ + eY cos θ k̂ 0≤

Y ≤ 1, 0 ≤ θ ≤ 2π
z
x2 +z 2 = ey
x
h i
2π
(b) 3 (1 + e2 )3/2 − 23/2

(c) π 1 − e2
3.3.6.31. ∗. Answer. 12π
√
5 π
3.3.6.32. ∗. Answer. 8
3.3.6.33. ∗. Answer. −20 π

3.3.6.34. ∗. Answer. 3
3.3.6.35. ∗. Answer. 2π
3.3.6.36. ∗. Answer. 2π
3.3.6.37. Answer. 192π
3.3.6.38. ∗.√Answer. (a) x + y + z = 1 + π/4

(b) 2π

3 2 2−1
3.3.6.39. ∗. Answer. (a) Yes. See the solution for the explanation.
(b) See the solution for the proof.

3.3.6.40. ∗. Answer. (a) (i) r(u, v) = u , v , 13 (16−2u−4v) k̂ u≥
0, v ≥ 0, u + 2v ≤ 8
(a) (ii) r(u, v) = 4 cos u sin v , 4 sin u sin v , 4 cos v 0 ≤ u ≤ 2π, 0 ≤
v ≤ π4
√
(a) (iii) r(u, v) = u , v , 1 + u2 + v 2 u2 + v 2 ≤ 99
√ √
or r(u, v) = u cos v , u sin v , 1 + u2 0 ≤ v ≤ 2π, 0 ≤ u ≤ 99
h i
(b) 32π 1 − √1
2
π
3.3.6.41. ∗. Answer. (a) 2
(b) π4 + 23
3.3.6.42. ∗. Answer. − 65
4.1.6 · Exercises
4.1.6.1. ∗. Answer. (a) A

(b) B
(c) C
(d) A
(e) B
4.1.6.2. Answer. No.
4.1.6.3. Answer. See solution.
4.1.6.4. Answer. (a) ∇ · F = 3, ∇ × F = 0

(b) ∇ · F = y 2 − z 2 + x2 , ∇ × F = 2yz ı̂ı − 2xz ̂ − 2xy k̂
(c) ∇ · F = √ 21 2 , ∇ × F = 0
x +y
k̂
(d) ∇ · F = 0, ∇ × F = √
x2 +y 2
4.1.6.5. ∗. Answer. (a) 2r

(b) xexy − 2x ı̂ı + y 1 − exy ̂ + z k̂
4.1.6.6. ∗. Answer. (a) k = −3
(b) k = 2
(c) k = −2
4.1.6.7. ∗. Answer. (a) 3
(b) 2r
(c) −2a
(d) 2r
4.1.6.8. ∗. Answer. (a) a = −3
(b) a = 4
(c) a = 12
4.1.6.9. Answer. (a) F cannot have a vector potential.
(b) Two solutions are A = 12 (z 2 −y 2 )xı̂ı − 12 yz 2̂ and A = 12 xz 2ı̂ı + 12 (x2 −
2 
z )y̂ .

4.1.6.10. ∗. Answer. (a) D = (x, y, z) x2 + z 2 6= 0
(b) ∇ × F = 0 on D
(c) ∇ · F = 1 on D
(d) F is not conservative on the domain D of part (a).
4.1.6.11. ∗. Answer. (a) α = β = −1
(b) Any function of the form g(x, y, z) = xyz + w(z) will work.
4.1.6.12. Answer. (a) See the solution
(b) ∇ × (Ω
Ω × r) = 2Ω
Ω ∇ · (Ω
Ω × r) = 0
(c) 1095km/hr

4.2.6 · Exercises
8π
4.2.6.3. Answer. (a), (b) 3
4.2.6.4. Answer. (a), (b) 43 πa3
4.2.6.5. Answer. (a) − 81

4 π
(b) 2|V |
(c) 2|V | + 81
4 π
4.2.6.6. Answer. (a), (b) 2π

4.2.6.7. ∗. Answer. (a) z
(b) 0
4.2.6.8. Answer. π
4.2.6.9. Answer. [3 + 3x0 − y0 ] V
4.2.6.10. ∗. Answer. −π
16
4.2.6.11. ∗. Answer. 3 π
4.2.6.12. ∗. Answer. (a) ∇ · F(x, y, z) = 0 except at (x, y, z) = (0, 0, 0),

where F is not defined.
(b) 4π
(c) No.
(d) 4π
(e) 0
4.2.6.13. ∗. Answer. (a)
r(θ, ϕ) = sin ϕ cos θ ı̂ı + 2 sin ϕ sin θ ̂ + 2 cos ϕ k̂

0 ≤ θ < 2π, 0 ≤ ϕ ≤ π
(b) 16π
(c) 16π, again
4.2.6.14. ∗. Answer. 40π
4.2.6.15. ∗. Answer. 24π
π
4.2.6.16. ∗. Answer. 2
3
4.2.6.17. ∗. Answer. 2π
32
4.2.6.18. ∗. Answer. 3 π
4.2.6.19. ∗. Answer. 3π
26
4.2.6.20. ∗. Answer. 3
4.2.6.21. ∗. Answer. 2π
64
4.2.6.22. ∗. Answer. (a) 3 π
(b) 128
3 π
4.2.6.23. ∗. Answer. (a) ∇ · F = 0 if (x, y, z) 6= 0 and is not defined if

(x, y, z) = 0.
(b) 4π
(c) 0 RR RR
(d) The flux integrals S1 F · n̂ dS and S2 F · n̂ dS are different, be-
cause the one point, (0, 0, 0), where ∇ · F fails to be well-defined and zero,
is contained inside S1 but is not contained inside S2 .
4.2.6.24. ∗. Answer. 72
4.2.6.25. ∗. Answer. (a) 3

(b) −14π
4.2.6.26. ∗. Answer. − 45π 35/2

4.2.6.27. ∗. Answer. (a) 0
(b) 625
2 π
4.2.6.28. ∗. Answer. 4π
4.2.6.29. ∗. Answer. π
4.2.6.30. ∗. Answer. 12π
188
4.2.6.31. ∗. Answer. 15 π ≈ 39.37
4.2.6.32. ∗. Answer. 5π
π 1 3
4.2.6.33. Answer. 6 − 3 a
√
4.2.6.34.√ Answer. (a) −8 2π
(b) 8 √2π
(c) 16 2π

4.2.6.37. ∗. Answer. (a), (b) 36π
e
4.2.6.38. ∗. Answer. 4
4.2.6.39. ∗. Answer. (a) √1 (−1, −1, 1)

3
(b) − 27π
2
(c) − 81π
2
4.2.6.40. ∗. Answer. (a) ∇ · F = 2 + 2z

(b) π 23 3 2875
6 5 = 6 π
(c) Let S be an oriented surface that encloses a solid V and has outward
9
pointing normal. If z̄ = − 2|V | − 1, where |V | is the volume of V and z̄ is
the z-component
RR of the centroid (i.e. centre of mass with constant density)
of V , then S F · n̂ dS = −9. One surface which obeys this condition is
11
the unit cube (with outward normal) centred on 0, 0, − 2 .
√
5π
4.2.6.41. ∗. Answer. (a) 8
(b) 20
(c) 18
π π(b+d)
RR ∗. Answer. (a) 4 + 6
4.2.6.42.
(b) σ1 ∪σ3 F · n̂ dS is zero if and only if d = −b.
RR
(c) σ1 ∪σ3 F · n̂ dS is zero for all a, b, c, d.
4.2.6.43. ∗. Answer. (a) 4π

(b) 0.
3
4.2.6.46. ∗. Answer. 4
4.2.6.47. ∗. Answer. 9πa3 + 9πa2


4.2.6.49. ∗. Answer. (a) 0
(b) 15
2 π
4.2.6.50. ∗. Answer. 30 + 24π

4.3.1 · Exercises
4.3.1.3. Answer. (a) 1
(b) 1
(c) 0
4.3.1.5. Answer. −54

4.3.1.6. Answer. 9
32
4.3.1.7. ∗. Answer. 3 π
4.3.1.8. ∗. Answer. −6
4.3.1.9. ∗. Answer. (a)
y y = x2 + 4x + 4
C y = 4 − x2
C
(−2, 0) x
(b) − 83
4.3.1.10. ∗. Answer. − 31
4.3.1.11. ∗. Answer. 54
10
4.3.1.12. ∗. Answer. 3
4.3.1.13. ∗. Answer. (a) − π2

(b) 3π
2
(c) No.
4.3.1.14. ∗. Answer. 54
4.3.1.15. ∗. Answer. (a) I2 = 0
(b) I3 = π
(c) I4 = π
4.3.1.16. ∗. Answer. (a) Qx − Py = 0 except at (0, 0) where it is not
defined.
(b) −2π
(c) No.
(d) 0
(e) −2π
π
4.3.1.17. ∗. Answer. (a) 2
(b) π4 + 23
3π
4.3.1.18. ∗. Answer. 2
3π
4.3.1.19. ∗. Answer. 8
4.3.1.20. ∗. Answer. A = −2
4.3.1.21. ∗. Answer. −π
H H
4.3.1.22. ∗. Answer. C1 F · dr = 0 and C2 F · dr = 2π
4.3.1.23. ∗. Answer. (a) 12 + 12

1
3/2
5 − 1 ≈ 1.3484
(b) 34
4.3.1.24. ∗. Answer. (a) The projection of the curve on the xy-plane
(i.e. the top view of the curve) is a circle. See the solution for more details.
(b) (i) 0
(b) (ii) 0
4.3.1.25. Answer. 6x2 + 3y 2 = 1

4.4.3 · Exercises
4.4.3.1. Answer.
(a) n̂ (b) (c)
∂S
S n̂ ∂S
S
n̂
S
∂S
4.4.3.4. Answer. (a) 2π

(b) 2π
4.4.3.5. Answer. π
4.4.3.6. ∗. Answer. 8π
4.4.3.7. ∗. Answer. 12π
4.4.3.8. ∗. Answer. π
4.4.3.9. Answer. 8
4.4.3.10. Answer. 1
4.4.3.11. ∗. Answer. 4π
4.4.3.12.√ ∗. Answer. (a) 8
(b) 4 3
4.4.3.13. ∗. Answer. (a)
z
(0,0,2)
C2
C3
(0,2,0)
(2,0,0) y
x C1

(b) S = (x, y, z) x2 + y 2 + z 2 = 4, x ≥ 0, y ≥ 0, z ≥ 0 with
r(θ, ϕ) = 2 cos θ sin ϕ ı̂ı + 2 sin θ sin ϕ ̂ + 2 cos ϕ k̂

π π
0≤θ≤ , 0≤ϕ≤
2 2
and
1
n̂ = cos θ sin ϕ ı̂ı + sin θ sin ϕ ̂ + cos ϕ k̂ = r(θ, ϕ)
2
(c) −4π
4.4.3.14. ∗. Answer. (a) −128π,
(b) −126π
4.4.3.15. ∗. Answer. 4π
4.4.3.16.√ ∗. Answer. (a) 8
(b) 4 3
4.4.3.17. ∗. Answer. 5π/4
10
4.4.3.18. ∗. Answer. − √ 3
4.4.3.19. ∗. Answer. −2
4.4.3.20. ∗. Answer. −π
3π
4.4.3.21. ∗. Answer. 4
π
4.4.3.22. ∗. Answer. 3
4.4.3.23. ∗. Answer. (a)

z
(0, 0, 2)
Sy
Sx y
(0, 3, 0)
(2, 0, 0)
x
R
(b) C
F · dr = 10
4.4.3.24. ∗. Answer. −π
4.4.3.25. ∗. Answer. 24π
√
4.4.3.26. ∗. Answer. 2 3πR2
4
4.4.3.27. Answer. 3
4.4.3.28. Answer. −2π
4.4.3.29. Answer. 24π

4.4.3.30. ∗. Answer. (a) ∇ × F = (1 − 2xz) ̂
(b) 20
3
4.4.3.31. ∗. Answer. (a) −18π

(b) −18π

4.4.3.32. ∗. Answer. (a) D = (x, y, z) x > 0, y > 0, z > 0
(b) The domain D isboth connected and simply connected.
(c) ∇ × F = 2x − x1 k̂
(d) 2 ln 2 − 24
(e) No. F is not conservative.
4.4.3.33. ∗. Answer. (a) a = 2, b = −1
(b) π4
4.4.3.34. ∗. Answer. −15
4.4.3.35. ∗. Answer. 12π
H
4.4.3.36. ∗. Answer. Rewrite C
E · dr as a surface integral. For the
details, see the solution.
√
4.4.3.37. ∗. Answer. 2π
2π
4.4.3.38. ∗. Answer. √
3 3
4.4.3.39. ∗. Answer. (a) One possible parametrization is r(r, θ) =

r cos θ ı̂ı + r sin θ ̂ + r k̂ with 0 ≤ r ≤ 1, 0 ≤ θ ≤ π.
(b) π
4.4.3.40. Answer. − √43 π

5.2 · Exercises
5.2.1. ∗. Answer. (a) True
(b) True
(c) True
(d) False
(e) True
(f) That depends. If κ = 0, the curve is part of a straight line. If κ > 0 it
is part of a circle of radius κ1 .
(g) False.
(h) False.
(i) False.
5.2.2. ∗. Answer. (a) False
(b) False
(c) False
(d) False
(e) True
(f) True
(g) False
(h) False
(i) False
(j) True
5.2.3. ∗. Answer. (a) False.
(b) N̂(t), B̂(t)
(c) True.
(d) False.
(e) False.
(f) True.
5.2.4. ∗. Answer. (a) decreasing
(b) f (x) is D
(c) r(θ) = cos θ ı̂ı + sin θ k̂ + sin θ cos θ ̂, 0 ≤ θ < 2π
(d) We want parametrisation (d) with domain |u| ≥ 2, 0 ≤ v ≤ 5.
(e) One possible answer is r(t) = tı̂ı, 0 ≤ t ≤ 1.
(f) C=6
(g) (a, b, c, d) a, b, c, d all real and b = c
(h) 2
(i) (1) True
(2) False
(3) False
(4) False
(5) False
(j) Any vector field whose divergence is 1 everywhere will work. One such
vector field is F = xı̂ı.
(k) negative
5.2.5. ∗. Answer. (a) false
(b) false
(c) true
(d) false
(e) true, assuming that the second derivatives of the vector field exist and
are continuous.
(f) silly, but true
(g) true
(h) false
(i) false
(j) false
(b) False
(c) True
(d) True
(e) True
(f) True
(g) True
(h) False
5.2.7. ∗. Answer. (a) Py < 0
(b) Qx > 0
(c) ∇
R × F is in the direction of +k̂ at A
(d) C1 F · dr > 0
R
(e) C2 F · dr < 0
(f) F is not conservative

(b) True
(c) True
(d) False
(e) True
(f) False
(g) False
(h) False
(i) True
(j) True
(b) False
(c) True
(d) False
(e) False
(f) True
(g) True
(h) False
(i) False
(j) True
(b) False.
(c) True.
(d) False.
(e) True.
(f) False.
(g) False.
(h) True.
(i) False.
(b) False
(c) True
(d) False
(e) True
(f) True
(g) True
(h) False
(i) False
(j) True
5.2.12. ∗. Answer. (b)
(b) False.
(c) False.
(d) True.
(e) False.
(f) True.
(g) False.
(h) False.
(i) True.
(j) False.

(b) False
(c) True, assuming that r(t) is not indentically 0.
(d) False
(e) False
5.2.15. ∗. Answer. (a) 2xy + ey sin x + xexz
(b) y 3 ı̂ı − z ̂
(c) (iii)
(d) False.
(b) True
(c) True
(d) False
(e) True
(f) True
(g) True
(b) False
(b) False
5.2.19. ∗. Answer. (a), (b), (c) See the solution.
(d) Yes
(d) No
5.2.20. ∗. Answer. (a) yes
(b) no
(c) no
(d) yes
(b) True
(c) False
5.2.22. ∗. Answer. (a), (c) See the solution.
(b) 8πb2
Appendix D
Solutions to Exercises
1 · Curves
1.1.1 · Exercises
1.1.1.1. Solution. (a) Since, on the specified part of thep
p circle, x =
a2 − y 2 and y runs from 0 to a, the parametrization is r(y) = a2 − y 2 ı̂ı+
y ̂, 0 ≤ y ≤ a.
(b) Let θ be the angle between
• the radius vector from the origin to the point (a cos θ, a sin θ) on the
circle and
• the positive x-axis.
The tangent line to the circle at (a cos θ, a sin θ) is perpendicular to the
radius vector and so makes angle φ = π2 + θ with the positive x axis. (See
the figure on the left below.) As θ = φ − π2 , the desired parametrization is
x(φ), y(φ) = a cos(φ − π2 ), a sin(φ − π2 ) = a sin φ, −a cos φ ,

π
2 ≤φ≤π
y y (0, a)
φ
s a cos θ , a sin θ
2 2 2
x +y =a
(a cos θ, a sin θ)
θ θ
x x
x2 + y 2 = a2
(c) Let θ be the angle between

• the radius vector from the origin to the point (a cos θ, a sin θ) on the
circle and
• the positive x-axis.
The arc from (0, a)to (a cos θ, a sin θ) subtends an angle π2 − θ and so has
length s = a π2 − θ . (See the figure on the right above.) Thus θ = π2 − as
and the desired parametrization is
π s π s π
x(s), y(s) = a cos − , a sin − , 0≤s≤ a
2 a 2 a 2
392
APPENDIX D. SOLUTIONS TO EXERCISES 393
1.1.1.2. Solution. We can find the time at which the curve hits a given
point by considering the two equations that arise from the two coordinates.
For the y-coordinate
√ to be 0, we must have (t − 5)2 = 0, i.e. t = 5. So, the
point (−1/ 2, 0) happens when t = 5.
Similarly, for the y-coordinate to be 25, we need (t − 5)2 = 25, so
(t − 5) = ±5. When t = 0, the curve hits (1, 25); when t = 10, the curve
hits (0, 25). √
So, in order, the curve passes through the points (1, 25), (−1/ 2, 0),
and (0, 25).
1.1.1.3. Solution. The curve “crosses itself” when the same coordinates
occur for different values of t, say t1 and t2 . So, we want to know when
sin t1 = sin t2 and also t21 = t22 . Since t1 and t2 should be different, the
second equation tells us t2 = −t1 . Then the first equation tells us sin t1 =
sin t2 = sin(−t1 ) = − sin t1 . That is, sin t1 = − sin t1 , so sin t1 = 0. That
happens whenever t1 = πn for an integer n.
So, the points at which the curve crosses itself are those points (0, (πn)2 )
where n is an integer. It passes such a point at times t = πn and t = −πn.
So, the curve hits this point 2πn time units apart.
1.1.1.4. Solution. (a) Pretend that the circle is a spool of thread. As
the circle rolls, it dispenses the thread along the ground. When the circle
rolls θ radians, it dispenses the arc length θa of thread and the circle
advances a distance θa. So the centre of the circle has moved θa units to
the right from its starting point, x = a. The centre of the circle always
has y-coordinate a. So, after rolling θ radians, the centre of the circle is at
position c(θ) = (a + aθ, a).
(b) Now, let’s consider the position of P on the circle, after the circle
has rolled θ radians.
a sin θ P
a cos θ
From the diagram, we see that P is a cos θ units above the centre of
the circle, and a sin θ units to the right of it. So, the position of P is
(a + aθ + a sin θ, a + a cos θ).
Remark: this type of curve is known as a cycloid.
1.1.1.5. Solution. We aren’t concerned with x, so we can eliminate it
by solving for it in one equation, and plugging that into the other. Since
C lies on the plane, x = −y − z, so:

1 1
1 = x2 − y 2 + 3z 2 = (−y − z)2 − y 2 + 3z 2
4 4
3 2 2
= y + 4z + 2yz
4
Completing the square,
1 2 y 2
1= y + 2z +
2 2
y2 y 2
1− = 2z +
2 2
Since y is small, the left hand is close to 1 and the right hand side is close
to (2z)2 . So (2z)2 ≈ 1. Since z is negative, z ≈ − 12 and 2z + y2 < 0. Also,
y2
1− 2 is positive, so it has a real square root.
r
y2 y
− 1− = 2z +
2 2
r
1 y2 y
− 1− − =z
2 2 4
1.1.1.6. Solution. To determine whether the particle is rising or falling,
we only need to consider its z-coordinate: z(t) = (t − 1)2 (t − 3)2 . Its
derivative with respect to time is z 0 (t) = 4(t − 1)(t − 2)(t − 3). This is
positive when 1 < t < 2 and when 3 < t, so the particle is increasing on
(1, 2) ∪ (3, ∞) and decreasing on (0, 1) ∪ (2, 3).
If r(t) is the position of the particle at time t, then its speed is |r0 (t)|.
We differentiate:
1
r0 (t) = −e−t ı̂ı − ̂ + 4(t − 1)(t − 2)(t − 3)k̂
t2
So, r(1) = − 1e ı̂ı − 1 ̂ and r(3) = − e13 ı̂ı − 19 ̂. The absolute value of every
component of r(1) is greater than or equal to that of the corresponding
component of r(3), so |r(1)| > |r(3)|. That is, the particle is moving more
swiftly at t = 1 than at t = 3.
q Note: We could also compute q thesizes of both vectors directly: |r0 (1)| =
1 2 1 2 2
+ (−1)2 , and |r0 (3)| = + − 19 .

e e3
1.1.1.7. Solution.
r(t + h)
r(t)
r(0)
The red vector is r(t+h)−r(t). The arclength of the segment indicated

by the blue line is the (scalar) s(t + h) − s(t).
Remark: as h approaches 0, the curve (if it’s differentiable at t) starts

to resemble a straight line, with the length of the vector r(t + h) − r(t)
approaching the scalar s(t + h) − s(t). This step is crucial to understanding
Lemma 1.1.4.
1.1.1.8. Solution. Velocity is a vector-valued quantity, so it has both
a magnitude and a direction. Speed is a scalar — the magnitude of the
velocity. It does not include a direction.
1.1.1.9. ∗. Solution. By the product rule
d
(r × r0 ) · r00 = (r0 × r0 ) · r00 + (r × r00 ) · r00 + (r × r0 ) · r000

dt
The first term vanishes because r0 × r0 = 0. The second term vanishes
because r × r00 is perpendicular to r00 . So
d
(r × r0 ) · r00 = (r × r0 ) · r000

dt
which is (c).
1.1.1.10. Solution. we are told that r(t) ⊥ r0 (t), so that r(t) · r0 (t) = 0,
for all t. Consequently
d d
|r(t)|2 = r(t) · r(t) = 2r(t) · r0 (t) = 0

dt dt
So |r(t)|2 is a constant,
√ say A, independent of time and r(t) always lies on
the sphere of radius A centred on the origin.
1.1.1.11. ∗. Solution. We have

√
v(t) = r0 (t) = 5 2ı̂ı + 5e5t ̂ + 5e−5t k̂
and hence
√ p
|v(t)| = |r0 (t)| = 5 2ı̂ı + e5t ̂ + e−5t k̂ = 5 2 + e10t + e−10t

2
Since 2 + e10t + e−10t = e5t + e−5t , that’s (d).
1.1.1.12. Solution. We are told that
r(t) = a cos tı̂ı + a sin t ̂ + ct k̂
So, by definition,
velocity = v(t) = r0 (t) = −a sin tı̂ı + a cos t ̂ + c k̂

ds p
speed = (t) = |r0 (t)| = a2 + c2
dt
acceleration = a(t) = r00 (t) = −a cos tı̂ı − a sin t ̂
As t runs over an interval of length 2π, (x, y) traces out a circle of radius
a and z increases by 2πc. The path is a helix with radius a and with each
turn having height 2πc.
1.1.1.13. ∗. Solution. (a) Since r0 (t) = (2t, 0, t2 ), the specified unit
tangent at t = 1 is
(2, 0, 1)
T̂(1) = √
5
(b) We are √ √ 3, −1/3). As
to find the arc length between (0, 3, 0) and (1,
ds
dt = |r0 (t)| = 4t2 + t4 , that arc length is the integral of 4t2 + t4 with
t running between the value of t for which r(t) = (0, 3, 0) and the value of
t for which r(t) = (1, 3, −1/3). To find those two values of t, we observe
that
• the first component of r(t), namely t2 , matches the first component
of (0, 3, 0), namely 0, only when t = 0. So we would guess that the t
corresponding to (0, 3, 0) is t = 0. As r(0) = (02 , 3, 31 03 ) = (0, 3, 0),
this is indeed the case.
• The first component of r(t), namely t2 , matches the first component

of (1, 3, −1/3), namely 1, only when t = ±1. For the third component
of r(t), namely 31 t3 , to match the third component of (1, 3, −1/3),
namely −1/3, we need t < 0. So we would guess that the t corre-
sponding to (1, 3, −1/3) is t = −1. To check that this is indeed the
case, we compute r(−1) = (−1)2 , 3, 31 (−1)3 = (1, 3, −1/3).
So t runs from −1 to 0 and
Z 0 p
arc length = 4t2 + t4 dt
−1
The integrand is even, so

Z 1p Z 1 p
3/2 1
h i
arc length = 4t2 + t4 dt = t 4 + t2 dt = 31 (4 + t2 )
0 0 0
1 3/2

= 3 5 −8
1.1.1.14. Solution. By Lemma 1.1.4 the arclength of r(t) from t = 0 to
R1
t = 1 is 0 dr
dt (t) dt. We’ll calculate this in a few pieces to make the steps

clearer.
r !
3 2 3
r(t) = t, t ,t
2
dr √
(t) = 1, 6t, 3t2
dt q
dr √ p
(t) = 12 + ( 6t)2 + (3t2 )2 = 1 + 6t2 + 9t4
dt
p
= (3t2 + 1)2 = 3t2 + 1
Z 1 Z 1
dr
3t2 + 1 dt = 2

(t) dt =
dt
0 0
1.1.1.15. Solution. Since
x0 (t) = a cos2 t − sin2 t = a cos 2t

y 0 (t) = 2a sin t cos t = a sin 2t

z 0 (t) = b
we have
ds p p
(t) = x0 (t)2 + y 0 (t)2 + z 0 (t)2 = a2 + b2
dt
ds ds
√
As the speed dt (t) is constant, the length is just dt T = a2 + b2 T .
1.1.1.16. Solution. Since r(t) is the position of the particle, its acceler-
ation is r00 (t).
r(t) = (t + sin t, cos t)

r0 (t) = (1 + cos t, − sin t)
r00 (t) = (− sin t, − cos t)
p
|r00 (t)| = sin2 t + cos2 t = 1
The magnitude of acceleration is constant, but its direction is changing,

since r00 (t) is a vector with changing direction.
1.1.1.17. ∗. Solution. (a) The speed is
ds
(t) = r0 (t) = 2 cos t − 2t sin t , 2 sin t + 2t cos t , t2

dt q 2 2
= 2 cos t − 2t sin t + 2 sin t + 2t cos t + t4
p
= 4 + 4t2 + t4
= 2 + t2
so the length of the curve is

2 2 2
t3
Z Z
ds 20
length = dt = (2 + t2 ) dt = 2t + =
0 dt 0 3 0 3
π3

(b) A tangent vector to the curve at r(π) = − 2π , 0 , 3 is
r0 (π) = 2 cos π − 2π sin π , 2 sin π + 2π cos π , π 2 = (−2 , −2π , π 2 )

So parametric equations for the tangent line at r(π) are
x(t) = −2π − 2t
y(t) = −2πt
π3
z(t) = + π2 t
3

1.1.1.18. ∗. Solution. (a) As r(t) = 3 cos t, 3 sin t, 4t , the velocity of
the particle is
r0 (t) = − 3 sin t, 3 cos t, 4

ds
(b) As dt , the rate of change of arc length per unit time, is
ds
(t) = |r0 (t)| = − 3 sin t, 3 cos t, 4 = 5

dt
the arclength of its path between t = 1 and t = 2 is
Z 2 Z 2
ds
dt (t) = dt 5 = 5
1 dt 1
1.1.1.19. Solution. (a) We can parametrize the circle x2 + y 2 = 9 as

x(θ) = 3 cos θ, y(θ) = 3 sin θ with θ running from 0 to 2π. As z = 2x + 3y,
the ellipse can be parametrized by
x(θ) = 3 cos θ, y(θ) = 3 sin θ, z(θ) = 2x(θ) + 3y(θ) = 6 cos θ + 9 sin θ

0 ≤ θ ≤ 2π
(b) As
ds p 0 2
= x (θ) + y 0 (θ)2 + z 0 (θ)2
dθ p
= 9 sin2 θ + 9 cos2 θ + 36 sin2 θ + 81 cos2 θ − 108 sin θ cos θ
p
= 45 + 45 cos2 θ − 108 sin θ cos θ
the circumference is
Z 2π p
s= 45 + 45 cos2 θ − 108 sin θ cos θ dθ
0
1.1.1.20. ∗. Solution. (a) As
r0 (t) = − sin t cos2 tı̂ı + sin2 t cos tı̂ı + 3 sin2 t cos t k̂

= sin t cos t − cos tı̂ı + sin t ̂ + 3 sin t k̂
ds p p
(t) = | sin t cos t| cos2 t + sin2 t + 9 sin2 t = | sin t cos t| 1 + 9 sin2 t
dt
π
the arclength from t = 0 to t = 2 is
Z π/2 Z π/2
ds p
(t) dt = sin t cos t 1 + 9 sin2 t dt
0 dt 0
Z 10
1 √
= u du with u = 1 + 9 sin2 t, du = 18 sin t cos t dt
18 1
1 h 2 3/2 i10
= u
18 3 1
1 √
= 10 10 − 1
27
(b) The arclength from t = 0 to t = π is
Z π Z π
ds p
(t) dt = | sin t cos t| 1 + 9 sin2 t dt
0 dt 0
Don’t forget the absolute value signs!
Z π/2 p
=2 | sin t cos t| 1 + 9 sin2 t dt
0
Z π/2 p
=2 sin t cos t 1 + 9 sin2 t dt
0
since the integrand is invariant under t → π−t. So the arc length from
√ t = 0
2
to t = π is just twice the arc length from part (a), namely 27 10 10 − 1 .
1.1.1.21. ∗. Solution. Since
t3 t2 t
r(t) = ı̂ı + ̂ + k̂
3 2 2
0 2 1
r (t) = t ı̂ı + t ̂ + k̂
2
r r
ds 1 1 2 1
(t) = |r0 (t)| = t4 + t2 + = t2 + = t2 +
dt 4 2 2
the length of the curve is
Z t Z t
ds 1 t3 t
s(t) = (u) du = u2 + du = +
0 dt 0 2 3 2
r(t) = tm ı̂ı + tm ̂ + t3m/2 k̂

3m 3m/2−1
r0 (t) = mtm−1 ı̂ı + mtm−1 ̂ + t k̂
r 2
r
ds 0 9m2 3m−2 m−1 9
= |r (t)| = 2m t 2 2m−2 + t = mt 2 + tm
dt 4 4
the arc length is
Z b Z b
r
ds m−1 9
(t) dt = mt 2 + tm dt
a dt a 4
Z 2+ 94 bm √
4 9 9m m−1
= u du with u = 2 + tm , du = t
9 2+ 49 am 4 4
m
4 h 2 3/2 i2+ 49 b
u=
9 3 2+ 94 am
8 h 9 3/2 9 3/2 i
= 2 + bm − 2 + am
27 4 4
√
1.1.1.23. Solution. (a) Since y = x and z = 23 xy = 23 x3/2 ,
√ 2
r(x) = xı̂ı + x ̂ + x3/2 k̂
3
For the remaining parts of this problem we will also need
1 √
r0 (x) = ı̂ı + √ ̂ + x k̂
2 x
1 1
r00 (x) = − 3/2 ̂ + √ k̂
4x 2 x
s
√ 2
r
√

ds 0 1 1 1
= |r (x)| = 1 + +x= √ + x = √ + x
dx 4x 2 x 2 x
ds 3
(1) =
dx 2
(b)
9 9 9
√ √
Z Z Z
ds 1 2
ds = dx = √ + x dx = x + x3/2
C 0 dx 0 2 x 3 0
= 3 + 18 = 21
(c) Denote by
• r(x) the position of the particle when its first coordinate is x,
• R(t) the position of the particle at time t,
• x(t) the x--coordinate of the particle at time t, and
• s(x) the arc length of the curve from the origin to r(x).
We are told that |R0 (t)| = 9 for all t. So
dx
=⇒ R0 (t) = r0 x(t)

R(t) = r x(t) (t)
dt !
0 ds dx 1 p dx
=⇒ 9 = |R (t)| = x(t) (t) = p + x(t) (t)
dx dt 2 x(t) dt
In particular, if the particle is at (1, 1, 23 ) at time 0, then x(0) = 1 and
√

1 dx dx
9= √ + 1 (0) =⇒ (0) = 6
2 1 dt dt
so that

0 dx
0 1
R (0) = r (1) (0) = ı̂ı + ̂ + k̂ 6 = 6ı̂ı + 3 ̂ + 6 k̂
dt 2
(d) By the product and chain rules,

2
d2 x

dx dx
R0 (t) = r0 x(t) (t) =⇒ R00 (t) = r00 x(t) (t) + r0 x(t) (t)
dt dt dt2
p
We saw in part (c) that 9 = |R0 (t)| = √1 + x(t) dx dt (t) so that
2 x(t)
!−1
dx 1 p
(t) = 9 p + x(t)
dt 2 x(t)
Differentiating that gives

!−2 !
d2 x 1 p 1 1 dx
(t) = −9 + x(t) − + p (t)
dt2
p
2 x(t) 4x(t)3/2 2 x(t) dt
dx
In particular, when t = 0, x(0) = 1 and dt (0) =6
−2
d2 x 3 1
(0) = −9 6 = −6
dt2 2 4
so

00 00
2 0
1 1 1
R (0) = r (1) 6 + r (1) − 6 = 36 − ̂ + k̂ − 6 ı̂ı + ̂ + k̂
4 2 2
= −6ı̂ı − 12 ̂ + 12 k̂
1.1.1.24. Solution. Given the position of the particle, we can find its
velocity:
v(t) = r0 (t) = (cos t, − sin t, 1)
Applying the given formula,
L(t) = r × v = (sin t, cos t, t) × (cos t, − sin t, 1).
• [Solution 1:] We can first compute the cross product, then differ-
entiate:
L(t) = (cos t + t sin t)ı̂ı + (t cos t − sin t)̂ − k̂
L0 (t) = t cos tı̂ı − t sin t ̂

q √
|L0 (t)| = t2 (sin2 t + cos2 t) = t2 = |t|
• [Solution 2:] Using the product rule:
L0 (t) = r0 (t) × v(t) + r(t) × v0 (t)

= r0 (t) × r0 (t) +r(t) × v0 (t)
| {z }
0
= (sin t, cos t, t) × (− sin t, − cos t, 0)
= t cos tı̂ı − t sin t ̂
p
|L0 (t)| = t2 cos2 t + t2 sin t2 = |t|
z2 yz
1.1.1.25. ∗. Solution. (a) Since z = 6u, y = 12 = 3u2 and x = 18 = u3 ,
r(u) = u3 ı̂ı + 3u2 ̂ + 6u k̂
(b)
r0 (u) = 3u2 ı̂ı + 6u ̂ + 6 k̂

r00 (u) = 6uı̂ı + 6 ̂
ds p
(u) = |r0 (u)| = 9u4 + 36u2 + 36 = 3 u2 + 2

du
Z Z 1 Z 1
ds 1
3 u2 + 2 du = u3 + 6u 0 = 7

ds = du =
C 0 du 0
(c) Denote by R(t) the position of the particle at time t. Then

du
R(t) = r u(t) =⇒ R0 (t) = r0 u(t)

dt
In particular, if the particle is at (1, 3, 6) at time t1 , then u(t1 ) = 1 and
du du
6ı̂ı + 12 ̂ + 12 k̂ = R0 (t1 ) = r0 (1) (t1 ) = 3ı̂ı + 6 ̂ + 6 k̂ (t1 )
dt dt
which implies that du
dt (t1 ) = 2.
du du 2 d2 u
R0 (t) = r0 u(t) =⇒ R00 (t) = r00 u(t) + r0 u(t)
dt dt dt2
In particular,
du 2 d2 u
27ı̂ı + 30 ̂ + 6 k̂ = R00 (t1 ) = r00 (1) (t1 ) + r0 1 (t1 )
dt dt2
d2 u
= 6ı̂ı + 6 ̂ 22 + 3ı̂ı + 6 ̂ + 6 k̂

(t1 )
dt2
Simplifying
d2 u d2 u
3ı̂ı + 6 ̂ + 6 k̂ = 3ı̂ı + 6 ̂ + 6 k̂ 2
(t1 ) =⇒ (t1 ) = 1
dt dt2
1.1.1.26. ∗. Solution. (a) According to Newton,
mr00 (t) = F(t) so that r00 (t) = −3tı̂ı + sin t ̂ + 2e2t k̂
Integrating once gives
t2
r0 (t) = −3 ı̂ı − cos t ̂ + e2t k̂ + c
2
π2
for some constant vector c. We are told that r0 (0) = v0 = 2 ı̂ı. This forces
2
c = π2 ı̂ı + ̂ − k̂ so that
π2 3t2

r0 (t) = ı̂ı + (1 − cos t) ̂ + e2t − 1 k̂

−
2 2
Integrating a second time gives

2
π t t3

1 2t
r(t) = − ı̂ı + (t − sin t) ̂ + e − t k̂ + c
2 2 2
1
for some (other) constant vector c. We are told that r(0) = r0 = 2 k̂. This
forces c = 0 so that
2
π t t3

1 2t
r(t) = − ı̂ı + (t − sin t) ̂ + e − t k̂
2 2 2
(b) The particle is in the plane x = 0 when

2
π t t3

t
0= − = (π 2 − t2 ) ⇐⇒ t = 0, ±π
2 2 2
So the desired time is t = π.

(c) At time t = π, the velocity is
2
3π 2

π
r0 (π) = ı̂ı + (1 − cos π) ̂ + e2π − 1 k̂

−
2 2
= −π ı̂ı + 2 ̂ + e2π − 1 k̂
2

1.1.1.27. ∗. Solution. (a) Parametrize C by x. Since y = x2 and

z = 32 x3 ,
2
r(x) = xı̂ı + x2 ̂ + x3 k̂
3
0
r (x) = ı̂ı + 2x ̂ + 2x2 k̂
r00 (x) = 2 ̂ + 4x k̂
ds p
= |r0 (x)| = 1 + 4x2 + 4x4 = 1 + 2x2
dx
and
3 3
2 i3
Z Z Z
ds h
1 + 2x2 dx = x + x3 = 21

ds = dx =
C 0 dx 0 3 0
7
(b) The particle travelled a distance of 21 units in 2 time units. This
21
corresponds to a speed of 7/2 = 6.
(c) Denote by R(t) the position of the particle at time t. Then

dx
=⇒ R0 (t) = r0 x(t)

R(t) = r x(t)
dt
By parts (a) and (b) and the chain rule
ds ds dx dx dx 6
6= = = (1 + 2x2 ) =⇒ =
dt dx dt dt dt 1 + 2x2
In particular, the particle is at 1, 1, 32 at x = 1. At this time dx

dt =
6
1+2×1 = 2 and
dx
R 0 = r0 1

= ı̂ı + 2 ̂ + 2 k̂ 2 = 2ı̂ı + 4 ̂ + 4 k̂
dt
dx dx 2 d2 x
R0 (t) = r0 x(t) =⇒ R00 (t) = r00 x(t) + r0 x(t)
dt dt dt2
d
to 6 = 1 + 2x(t)2 dx

Applying dt dt (t) gives
dx 2 d2 x
0 = 4x + (1 + 2x2 )
dt dt2
2 2
In particular, when x = 1 and dx
dt = 2, 0 = 4 × 1 2 + (3) ddt2x gives
d2 x 16
dt2 = − 3 and
2 16 8
R00 = 2 ̂ + 4 k̂ 2 − ı̂ı + 2 ̂ + 2 k̂

= − 2ı̂ı + ̂ − 2 k̂
3 3
1.1.1.28. Solution. The question is already set up as an xy-plane, with
the camera at the origin, so the vector in the direction the camera is point-
ing is (x(t), y(t)). Let θ be the angle the camera makes with the positive
x-axis (due east). The camera, the object, and the due-east direction (pos-
itive x-axis) make a right triangle.
y
y(t) object
θ
x
camera
x(t)
y
tan θ =
x
Differentiating implicitly with respect to t:
dθ xy 0 − yx0
sec2 θ =
dt x2
0 !2
xy − yx0 xy 0 − yx0

dθ x
= cos2 θ =
x2 x2
p
dt x2 + y 2
xy 0 − yx0
=
x2 + y 2
1.1.1.29. Solution. Using the Theorem of Pappus, we can calculate the
surface area and volume of a pipe with the same length and radius as this
pipe. So, we need to find the length of the pipe, L.
dr √
= ( 2t, t, 1)
dt p
dr
= 2t + t2 + 1 = |t + 1|
dt
Z 10
L= (t + 1)dt = 60
0
A pipe with radius 3 and length 60 has surface area 60(2π · 3) = 360π
and volume 60(π · 32 ) = 540π.
1.1.1.30. Solution. In general a helix can be parametrized by
r(θ) = a cos θ ı̂ı + a sin θ ̂ + bθ k̂
Our first task is to determine a and b. The radius of the helix is 3 cm,
so a = 3 cm. After 10 turns (i.e. θ = 20π) the height, bθ, is 1 cm. So
1 1
b(20π) = 1 and b = 20π cm/rad. Thus r(θ) = 3 cos θ ı̂ı + 3 sin θ ̂ + 20π θk̂.
With each full turn of the helix (i.e. each increase of θ by 2π) the height
1
of the helix increases by 2πb = 10 cm. So if we can determine the length of
wire in one full turn of the helix, we can easily determine how many turns
the helix goes through in total, and from that we can determine the total
height of the helix.
q
As r0 (θ) = −3 sin θ ı̂ı+3 cos θ ̂+ 1 k̂ we have ds = r0 (θ) = 9 + 1 2 .
20π dθ 400π
So the length of one full turn of the helix is
Z 2π r r
1 1
9+ 2
dθ = 2π 9 + 2
0 400π 400π
and 1000cm of wire generates

1000 500
q = q
1 1
2π 9 + 400π 2 π 9 + 400π 2
1
turns. Each turn adds 10 cm to the height, so the total height is
500 1 50
q · = q ≈ 5.3 cm
π 9+ 1 10 1
π 9 + 400π
400π 2 2
Remark. We can check that this answer is reasonable by taking advan-

tage of the fact that each coil adds only a very small height (relative to
the radius). So we expect the length of one coil to be about the same as
the circumference of a circle of the same radius, namely 6π. If we were
making actual circles of the wire, there would be 1000
6π of them. Stacking
1000
up at 10 per centimetre, this would make a pile of height 6π·10 cm. Since
this number is also approximately 5.3cm, we feel our result is reasonable.
1.1.1.31. Solution. Define u(t) = eαt dr

dt (t). Then
du dr d2 r
(t) = αeαt (t) + eαt 2 (t)
dt dt dt
αt dr dr
= αe (t) − ge k̂ − αeαt (t)
αt
dt dt
= −geαt k̂
Integrating both sides of this equation from t = 0 to t = T gives
eαT − 1
u(T ) − u(0) = −g k̂
α
so that
eαT − 1 dr eαT − 1 eαT − 1
u(T ) = u(0) − g k̂ = (0) − g k̂ = v0 − g k̂
α dt α α
−αT
Substituting in u(T ) = eαt dr
dt (T ) and multiplying through by e gives
dr 1 − e−αT
(T ) = e−αT v0 − g k̂
dt α
Integrating both sides of this equation from T = 0 to T = t gives
e−αt − 1 t e−αt − 1
r(t) − r(0) = v0 − g k̂ + g k̂
−α α −α2
so that
e−αt − 1 1 − αt − e−αt
r(t) = r0 − v0 + g k̂
α α2
1.2.1 · Exercises
1.2.1.1. Solution. By Lemma 1.1.4.c, |r0 (s)| = 1 under arclength parametriza-
Rt Rt
tion. So 1 |r0 (s)|ds = 1 ds = t − 1.
1.2.1.2. Solution. The arclength from P to P will √ be 0, so P is the
point where s = 0. That is, r(0), or sin(1/2), cos(1/2), 3/2 .
1.2.1.3. Solution 1. We consider the situation geometrically. If we plot
R in space (of the relevant dimension), regardless of its parametrization,
the derivative at a point will give a vector tangent to R, in the direction
the curve moves when the parameter is increasing. Since a(t0 ) and b(s0 )
describe the same spot on the curve, a0 (t0 ) and b0 (s0 ) will be parallel1
— they’re both tangent to the same piece of curve. Furthermore, as t
increases, so does s, so the direction of increasing t is the same as the
direction of increasing s. Therefore, A. holds.
tangent direction
a(t0 ) = b(s0 )
R
Now we consider the magnitudes of the vectors, to rule out E. Recall

|a0 (t)| is the speed at which the curve changes relative to t; this could be
any (nonnegative) number. By the same token, |b0 (s)| = 1. So, b0 (s0 ) is a
unit vector, while a0 (t0 ) may or may not be. Then the two vectors are not
necessarily equal (although they could be).
So, the best answer is A.
Solution 2. The chain rule gives us a relationship between b0 (s) and
a0 (t).
db d da dt
= [a(t(s))] =
ds ds dt ds
So, the vectors db da dt
ds and dt differ only by the scalar function ds . So, at any
point along the curve, these vectors are parallel.
Furthermore, we know that t and s are positively correlated: as t in-
dt
creases, so does s, because we’re covering more arclength. So, ds is nonneg-
dt
ative. Furthermore, since the derivatives are nonzero, ds is nonzero. So,
b0 (s0 ) and a0 (t0 ) are positive scalar multiples of each other. That is, they
dt
are parallel, and pointing in the same direction. However, unless ds =1
(that is, t(s) = s + C for some constant C), the vectors do not have the
same magnitude, and hence are not equal.
So, A is the best solution.
1.2.1.4. ∗. Solution. (a) The velocity vector is
r0 (t) = (6 sin2 (t) cos t , −6 sin t cos2 (t) , 3 cos2 t − 3 sin2 t)

= 3 sin t sin(2t) , − cos t sin(2t) , cos(2t)
√
3
In particular, since sin(π/3) = sin(2π/3) = 2 and cos(π/3) = − cos(2π/3) =
1
2,
√
0 3 3 1
r (π/3) = 3 , − ,−
4 4 2
and the specified unit tangent vector is
3
√
3
√
, − 12

4 ,− 4 3 3 1
T̂ = √ = ,− ,−
3
, − 43 1
, −2 4 4 2
4
(b) The speed is
ds
q
0
= |r (t)| = 3 sin2 t sin2 (2t) + cos2 t sin2 (2t) + cos2 (2t)
dt q
= 3 sin2 (2t) + cos2 (2t)
=3
So s = 3t and the reparametrized form is

s s s s
R(s) = 2 sin3 ( ), 2 cos3 ( ), 3 sin( ) cos( )
3 3 3 3
1.2.1.5. ∗. Solution. (a) We have |r(t)| = et ≤ 1 for t ≤ 0. So the
part of the spiral contained in the unit circle is the part of the spiral with
−∞ < t ≤ 0. As
r0 (t) = et (cos t, sin t) + et (− sin t, cos t) = et cos t − sin t , sin t + cos t

the speed
ds 0 p √
= r (t) = et (cos t − sin t)2 + (sin t + cos t)2 = 2et
dt
and the arclength from t = −∞ to r(t) is
Z t
ds
Z t √ √
s(t) = (t̃) dt̃ = 2et̃ dt̃ = 2et
−∞ dt −∞
In particular
√ the length of the part of the spiral contained in the unit circle
is s(0) = 2.
√
(b) The inverse function of s(t) = 2et is t(s) = ln √s2 with s > 0.
(As t → −∞, the arc length s → 0 and as t → +∞, the arc length
s → +∞.) So the reparametrization is
s
t
s s
R(s) = e (cos t, sin t) =√ cos ln √ , sin ln √

t=ln √s 2 2 2
2
with s > 0
1.2.1.6. Solution. Using arctan t = z, and so t = tan z:

1 arctan t
r(t) = √ ,√ , arctan t
1 + t2 1 + t−2
!
1 z
= p ,p ,z
1 + tan2 z 1 + cot2 z

1 z
= , ,z
| sec z| | csc z|
= (| cos z|, z| sin z|, z)
Since 0 ≤ t, and arctan t < π/2 we have 0 ≤ z < π/2, so cos z and sin z are
both nonnegative.
= (cos z, z sin z, z)
If we didn’t have the restricted domain, this would make a spiral going up:
z is both the height of the spiral and a radian measure. The ı̂ı-component
of the spiral stays between −1 and 1, while the ̂-component increases. So,
our spiral gets increasingly “wide,” while staying the same “thickness.”
Due to the restricted domain, our actual curve is only one-quarter of a

“turn” of this spiral, indicated in red above.
The parameter z is a measure of height, and it is also a radian measure
as the spiral turns.
1.2.1.7. Solution.
r(t) = ( 21 t2 , 13 t3 )
r0 (t) = (t, t2 )
p p
|r0 (t)| = t2 + t4 = |t| 1 + t2
Z t p
s(t) = |x| 1 + x2 dx
−1
(R t √
−x 1 + x2 dx when t ≤ 0
= R−1 0 √ Rt √
−1
−x 1 + x2 dx + 0 x 1 + x2 dx when t > 0
Let u = 1 + x2 , 12 du = xdx
( R 1+t2 √
1
− 2 udu when t ≤ 0
= R 1 1 √2 R 1+t2 1 √
− 2 2 udu + 1 2 udu when t > 0
2
(
− 13 u3/2 |1+t
2 when t ≤ 0
= 1 3/2 1 1 3/2 1+t2
− 3 u |2 + 3 u |1 when t > 0
( 3/2
2 1 2 3/2
= 3 − 3 (1 + t ) when t ≤ 0
2 23/2 1 2 3/2
− 3 + 3 + 3 (1 + t ) when t > 0
Solving for t in terms of s:

( √
2 (2 2 − 3s)2/3 when t ≤ 0
1+t = √ 2/3
(3s + 2 − 2 2) when t > 0
( √
(2 2 − 3s)2/3 − 1 when t ≤ 0
t2 = √
(3s + 2 − 2 2)2/3 − 1 when t > 0
√
Remembering that t2 = |t|:
 q √
− (2 2 − 3s)2/3 − 1 when t ≤ 0
t= q
 (3s + 2 − 2√2)2/3 − 1 when t > 0
√
Noting that t = 0 when s = 13 (2 2 − 1), we find our reparametrization of
√
( 12 t2 , 13 t3 ). When s ≤ 31 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (2 2 − 3s)2/3 − 1) , − (2 2 − 3s)2/3 − 1
2 3
√
and when s > 13 (2 2 − 1),
1 h √ i 1h √ i3/2
R(s) = (3s + 2 − 2 2)2/3 − 1 , (3s + 2 − 2 2)2/3 − 1
2 3
Remark: after a computation with this much detail, it’s nice to find a
few points to check, to verify that our answer is reasonable. For instance,
when s = 0, t should be√−1, and vice-versa. Also, we found that t = 0
corresponds to s = 31 (2 2 − 1). So, we should be able to verify that
√
r(0) = R 13 (2 2 − 1) and r(−1) = R(0).

1.3 · Curvature
1.3.1 · Exercises
1.3.1.1. Solution. The curve is a circle of radius 3, centred at the
origin. So, the “circle of best fit” is just the curve itself. T̂ is the unit
vector tangent to the circle in direction of increasing t, and N̂ is the unit
vector pointing towards the origin.
y
T̂
N̂
x
1
1
The radius of the (osculating) circle is 3, so ρ = 3 and κ = ρ = 13 .
1.3.1.2. Solution. The arclength of r(t) traced out by an interval of t

of length θ is 3θ. That is, s = 3t. Our reparametrization of the circle in
terms of arclength is R(s) = (3 sin(s/3), 3 cos(s/3)).
We can calculate the vectors tangent to the circle, then normalize them
(i.e. make them length one) to find T̂.
v(t) = r0 (t) = (3 cos t, −3 sin t)

r0 (t) (3 cos t, −3 sin t)
T̂(t) = 0
= = (cos t, − sin t)
|r (t)| 3
T̂(s) = R0 (s) = (cos(s/3), − sin(s/3))
Note R0 (s), because it’s parametrized in terms of arclength, has derivative

vectors of length one. So, we don’t need to normalize them (although if we
did, it wouldn’t change anything).
Note also that we can check out answers using Question 1.3.1.1. In that
question, we found T̂ was ı̂ı when t = s = 0; this fits with the vectors we
just found.
As in Question 1.3.1.1, κ = 13 . So, using Theorem 1.3.3 Part (b):
dT̂
(s) = κ(s) N̂(s)
ds
1 1 1
− sin(s/3), − cos(s/3) = N̂(s)
3 3 3
(− sin(s/3), − cos(s/3)) = N̂(s)
Remember s = 3t. Using Theorem 1.3.3 Part (c):
dT̂ ds
= κ N̂(t)
dt dt
1
(− sin t, − cos t) = (3)N̂(t)
3
(− sin t, − cos t) = N̂(t)
1.3.1.3. Solution. As t increases, the arms of the spiral “flatten out,”
looking like a circle of bigger and bigger radius. So, we would expect the
curvature to decrease: lim κ(t) = 0.

t→∞
√
1.3.1.4. Solution. ds
dt = |v(t)| = |r0 (t)| = |(et , 3, cos t)| = e2t + 9 + cos2 t
1.3.1.5. Solution.
v(t) r0 (t)
T̂(t) = = 0
|v(t)| |r (t)|

et (cos t − sin t), et (cos t + sin t)
=p
e2t (cos t − sin t)2 + e2t (cos t + sin t)2
1
= √ cos t − sin t, cos t + sin t
2
dT̂ 1
= √ − sin t − cos t, − sin t + cos t
dt 2
Since R(s) is parametrized with respect to arclength, |R0 (s)| = 1.
T̂(s) = R0 (s)
√
Making ample use of the chain rule, and setting U (s) = ln s/ 2 , we
have U 0 (s) = 1s :
1
T̂(s) = √ (cos U (s) − sin U (s), cos U (s) + sin U (s))
2
dT̂ 1
= √ (− sin U (s) − cos U (s), − sin U (s) + cos U (s))
ds 2s
1.3.1.6. Solution. The circle of radius r centred at (0, r) is x2 +(y−r)2 =
r2 . The bottom half of this circle is
p
y = g(x) = r − r2 − x2
So
x
g 0 (x) = √ g 0 (0) = 0
r2 − x2
1 x2 1
g 00 (x) = √ + 3/2
g 00 (0) =
r2 − x2 [r2 − x2 ] r
As f (x) and g(x) have the same second order Taylor approximation at
x = 0, f 00 (0) = g 00 (0) = 1r .
We may parametrize the curve by r(x) = xı̂ı + f (x) ̂. So
r0 (x) = ı̂ı + f 0 (x) ̂ r0 (0) = ı̂ı + f 0 (0) ̂ = ı̂ı

r00 (x) = f 00 (x) ̂ r00 (0) = f 00 (0) ̂
and
|r0 (0) × r00 (0)| |f 00 (0)ı̂ı × ̂|
κ(0) = = = f 00 (0)
|r0 (0)|3 |ı̂ı|3
So κ(0) = f 00 (0) = 1
r and r is indeed the radius of curvature of y = f (x)
at x = 0.
1.3.1.7. Solution.
A v(t) = r0 (t) = (et , 2t + 1)
B a(t) = r00 (t) = (et , 2)

ds p
C = |v(t)| = e2t + (2t + 1)2
dt
D
v(t) (et , 2t + 1)
T̂(t) = =p
|v(t)| e2t + (2t + 1)2
!
et 2t + 1
= p ,p
e2t + (2t + 1)2 e2t + (2t + 1)2
|v(t) × a(t)| |(et , 2t + 1) × (et , 2)| et |1 − 2t|

E κ(t) = 3 = p 3 = 2t
ds (e + (2t + 1)2 )3/2

dt e2t + (2t + 1)2
1.3.1.8. Solution 1. Note that (cos t + sin t)2 + (sin t − cos t)2 = 2 for
2 2
all t. So, the
√ points (x, y) of our curve lie on x + y = 2, which is a circle
of radius 2. Indeed
√
x(t) = cos t + sin t = 2 cos t cos π4 + sin t sin π4

√
= 2 cos t − π4

√
y(t) = sin t − cos t = 2 sin t cos π4 − cos t sin π4

√
= 2 sin t − π4

√
So, r(t) circumnavigates a circle of radius 2 and consequently has curva-
ture κ = √12 .
|v(t) × a(t)|
Solution 2. We use the formula κ = ds 3 , remembering that

dt
v(t) = r0 (t), a(t) = r00 (t), and ds
dt = |r0 (t)|.
v(t) = r0 (t) = (− sin t + cos t, cos t + sin t)

a(t) = r00 (t) = (− cos t − sin t, − sin t + cos t)
v(t) × a(t) = (− sin t + cos t)2 + (cos t + sin t)2 k̂ = 2k̂

√

ds dv p
= = (− sin t + cos t)2 + (cos t + sin t)2 = 2
dt dt

v(t) × a(t) 2k̂ 1
κ= = √ 3 = √

ds 3 2
2
dt
1.3.1.9. Solution. For the given ellipse
r(t) = a cos t ı̂ı + b sin t ̂

v(t) = −a sin t ı̂ı + b cos t ̂
p
|v(t)| = a2 sin2 t + b2 cos2 t
a(t) = −a cos t ı̂ı − b sin t ̂
 
ı̂ı ̂ k̂
v(t) × a(t) = det  −a sin t b cos t 0  = ab k̂
−a cos t −b sin t 0
|v(t) × a(t)| ab
κ(t) = =
|v(t)|3 2
[a sin t + b2 cos2 t]
2 3/2
So the maximum (minimum) curvature is achieved when the denominator

is a minimum (maximum) which is the case when sin t = 0 (cos t = 0). So
κmax = ba2 and κmin = ab2 .
1.3.1.10. ∗. Solution. Parametrize the curve by r(t) = tı̂ı + et ̂. Then
v(t) = ı̂ı + et ̂ v(0) = ı̂ı + ̂

ds p ds √
= |v(t)| = 1 + e2t (0) = 2
dt dt
v(t) ı̂ı + et̂ v(0) ı̂ı + ̂
T̂(t) = =√ T̂(0) = = √
|v(t)| 1 + e2t |v(0)| 2
t
a(t) = e ̂ a(0) = ̂
(a) We’re given y in terms of x, so let’s use Part (e) of Theorem 1.3.3:
d2 y
2
dx ex
κ= 3/2 = 2 3/2
dy 2

1 + dx 1 + ex
1
κ(0) = = 2−3/2
[1 + 1]3/2
(b)
• The radius of the circle we want is ρ = κ1 = 23/2 . If its centre is at
(a, b), then the circle will have equation (x − a)2 + (y − b)2 = 23 . So,
we will find its centre.
• The unit vector N̂ points from our point (0, 1) towards the centre
of the circle. Since the radius of the circle is 23/2 , the centre of the
circle will be at (0, 1) + 23/2 N̂. So, we’ll find N̂.
ı̂ı + ̂
• Since N̂ is a unit vector perpendicular to T̂ = √ , we know N̂ will
2
ı̂ı − ̂ −ı̂ı + ̂
be either √ or √ .
2 2
• Using Part (d) of the proof of Theorem 1.3.3:
3
ds
v(t) × a(t) = κ T̂ × N̂
dt
√ 3 ı̂ı + ̂
(ı̂ı + ̂) × (̂) = 2−3/2 2 √ × N̂
2
1
k̂ = √ (ı̂ı + ̂) × N̂
2
−ı̂ı + ̂
N̂ = √
2
So, the centre of our circle is at point (0, 1)+ρN̂ = (0, 1)+23/2 −ı̂
ı+̂
21/2
=
2 2
(−2, 3). Then the equation of the circle is (x + 2) + (y − 3) = 8.
1.3.1.11. ∗. Solution. (a) Think of
r(t) = (t, 1) − (sin t, cos t)
The (t, 1) part gives the position of the centre of the wheel at time t. The
other part gives the position of the thumbtack with respect to the centre
of the wheel. In particular,
• at time t = 0, r(0) = (0, 0). The thumbtack is on the ground (i.e. at
y = 0).
• At time t = π, r(π) = (π, 2). The thumbtack is at its highest point
(i.e. at y = 2) and is above the centre of the wheel at x = π.
• At time t = 2π, r(2π) = (2π, 0). The thumbtack is back on the
ground (i.e. at y = 0) and is below the centre of the wheel at x = 2π.
• At time t = 3π, r(3π) = (3π, 2). The thumbtack is again at its
highest point (i.e. at y = 2) and is above the centre of the wheel at
x = 3π.
• At time t = 4π, r(4π) = (4π, 0). The thumbtack is back on the
ground (i.e. at y = 0) and is below the centre of the wheel at x = 4π.
Here is a sketch of the curve.

r(t) = t − sin t , 1 − cos t
y
2π 4π x
(b) Since

r(t) = t − sin t , 1 − cos t
v(t) = r0 (t) = 1 − cos t , sin t

ds √
(t) = |v(t)| = 2 − 2 cos t
dt
a(t) = v0 (t) = sin t , cos t

 
ı̂ı ̂ k̂
v(t) × a(t) = det 1 − cos t sin t 0  = cos t − 1 k̂
sin t cos t 0
the curvature
|v(t) × a(t)| | cos t − 1| 1
κ(t) = 3
= 3/2
= 3/2 √
|v(t)| (2 − 2 cos t) 2 1 − cos t
(c) The radius of curvature at time t = π is
1 1
ρ(π) = = √ =4
κ(π) 1/23/2 2
(d) At time π, the tack is at r(π) = (π, 2), which is at the top of its
trajectory. Looking at the sketch in part (a), we see that, at that time
N̂(π) = −̂. So the osculating circle at time t = π has center
r(π) + ρ(π)N̂(π) = (π, 2) + 4(0, −1) = (π, −2)
and radius ρ(π) = 4. So the equation of the osculating circle at time π is
(x − π)2 + (y + 2)2 = 16
1.3.1.12. Solution. The velocity vector is
v(θ) = x0 (θ)ı̂ı + y 0 (θ) ̂ = cos 1 2 1 2

2 πθ ı̂ı + sin 2 πθ ̂
Consequently the speed

ds
(θ) = |v(θ)| = 1 =⇒ s(θ) = θ + s(0)
dθ
Since s(θ) is zero when θ = 0, we have s(θ) = θ and hence
T̂(s) = v(s) = cos 21 πs2 ı̂ı + sin 12 πs2 ̂

so that

dT̂
2 2
1
1

κ(s) = (s) = − πs sin 2 πs ı̂ı + πs cos 2 πs ̂ = πs

ds
x3
1.3.1.13. ∗. Solution. The curve is y = y(x) = 3 . Since y 0 (x) = x2
and y 00 (x) = 2x, the curvature is
d2 y
2 (x) 2x
dx
κ(x) = h =
2 3/2
i 3/2
dy
1 + dx (x) 1 + x4
We’d like to find the critical points of κ(x), but differentiating it looks
messy. Since κ(x) has only nonnegative values, its maxima correspond the
the maxima of the function κ2 (x). So, we find the critical points of κ2 (x)
instead, to save ourselves some computational toil.
d d 4x2 8x 16x5
0= κ(x)2 = 3 = 3 −3 4
dx dx (1 + x4 ) (1 + x4 ) (1 + x4 )
8x(1 + x4 ) − 3 × 16x5 8x(1 − 5x4 )
= 4 = 4
(1 + x4 ) (1 + x4 )
Note that κ(0)√= 0 and κ(x) → 0 as x → ±∞. So the maximum occurs

when x = ±1/ 4 5.

1.4.1 · Exercises
1.4.1.1. Solution. T̂ is tangent to the curve, while N̂ is perpendicular
to it.
T̂
N̂
Using the right-hand rule and B̂ = T̂ × N̂, B̂ points out of the page
(towards the reader).
To see this, point the fingers of your right hand in the direction of T̂,
and curl them inwards until they are in the direction of N̂. To do this,
your thumb must be pointing towards you, not away from you. Your thumb
shows the direction of T̂ × N̂.
1.4.1.2. Solution. In this equation, s stands for arclength.
When we take a very small interval from t to t + h, the change in
arclength s(t + h) − s(t) is approximately |r(t + h) − r(t)|, because our
curve is approximated by a straight line. So, s(t+h)−s(t)
h ≈ |r(t+h)−r(t)|
h ,
ds dr
leading to dt = | dt | = |v(t)|.
The magnitude of velocity is speed; in this text we generally call this v.
That is, v = |v(t)|. This leads to the potentially confusing (but standard)
convention that s stands for arclength, while v stands for speed.
1.4.1.3. Solution 1. Curves a and b are the same curve, just parametrized
differently (replace t with −t to convince yourself if the picture isn’t enough).
So, they ought to have the same torsion.
As in Example 1.4.4, we imagine that the curve is the thread on a bolt.
Take a look at your right hand. If your thumb is pointing up (corresponding
to the +z direction), and you’re looking at the tip of your thumb, your
fingers curl anticlockwise. Imagine a screw has threads matching the curves
a and b, and we turn it anticlockwise. The screw would move down — not
in the same direction as our thumb. So these curves are not right-handed
helices, so they have negative torsion.
The curve c sits entirely in a plane (the plane x = 0) so its torsion is
zero everywhere.
Solution 2. Here is the conventional computation for both a(t) and b(t).
(The upper sign is for a and the lower sign is for b.)

r(t) = cos t , ∓2 sin t , ±t/2

v(t) = − sin t , ∓2 cos t , ±1/2

a(t) = − cos t , ±2 sin t , 0

v(t) × a(t) = − sin t , ∓ cos t/2 , ∓2
da
(t) = sin t , ±2 cos t , 0
dt
da
v(t) × a(t) · (t) = −1
dt
v(t) × a(t) · da
dt (t) 1
τ (t) = 2
=− 2 1 <0
|v(t) × a(t)| sin t + 4 cos2 t + 4
T̂
1.4.1.4. Solution. (a) If κ(s) ≡ 0, then dds = κ(s)N̂(s) ≡ 0 so that T̂ is
dr
a constant. As a result ds (s) = T̂ and r(s) = sT̂ + r(0) so that the curve
is the straight line with direction vector T̂ that passes through r(0).
B̂
(b) If τ (s) ≡ 0, then dds = −τ (s)N̂(s) ≡ 0 so that B̂ is a constant. As
T̂(s) ⊥ B̂,
d
(r(s) − r(0)) · B̂ = T̂(s) · B̂ = 0
ds
and (r(s) − r(0)) · B̂ must be a constant. The constant must be zero (set
s = 0), so (r(s) − r(0)) · B̂ = 0 and r(s) always lies in the plane through
r(0) with normal vector B̂.
(c) Parametrize the curve by arc length. Define the “centre of curva-
ture” at s by
1
rc (s) = r(s) + N̂(s)
κ(s)
Since κ(s) = κ0 is a constant and τ (s) ≡ 0,
d 1 1
rc (s) = T̂(s) + τ (s)B̂ − κ(s)T̂ = T̂(s) + 0B̂ − κ0 T̂ = 0
ds κ0 κ0

Thus rc (s) = rc is a constant and r(s) − rc = κ10 lies on the sphere of
radius κ10 centred on rc . Since τ (s) ≡ 0, the curve also lies on a plane, so
it is a circle.
1.4.1.5. ∗. Solution. (a), (b): T̂ points in the direction of the curve;
N̂ is perpendicular to it, in the same plane, pointing towards the centre of
curvature. Using the right-hand rule in the picture, we see B̂ is pointing
to the left.
z
z = x2 + y 2
N̂ T̂
B̂
y
x=y
x
(c) The torsion is zero, since the curve lies in a plane (the plane x = y).
1.4.1.6. ∗. Solution. (a) As
r0 (t) = et + e−t ı̂ı + et − e−t ̂ + 2 k̂

ds p √
(t) = |r0 (t)| = 4 + 2e2t + 2e−2t = 2 et + e−t

dt
r00 (t) = et − e−t ı̂ı + et + e−t ̂

r0 (t) × r00 (t) = −2 et + e−t ı̂ı + 2 et − e−t ̂ + 4 k̂

the curvature
√
|v(t) × a(t)| 2 4 + 2e2t + 2e−2t 1
κ(t) = = =
ds 3 [4 + 2e2t + 2e−2t ]3/2 2 + e2t + e−2t

dt
(b) The length of C between r(0) and r(1) is

Z 1
ds √ Z 1 √ h t i1 √ h 1i
(t) dt = 2 (et + e−t ) dt = 2 e − e−t = 2 e −
0 dt 0 0 e
1.4.1.7. Solution. The point (2, 4, 8) occurs when t = 2.
v(t) = (1, 2t, 3t2 ) v(2) = (1, 4, 12)

a(t) = (0, 2, 6t) a(2) = (0, 2, 12)
da da
(t) = (0, 0, 6) (2) = (0, 0, 6)
dt dt
v(2) × a(2) = (24, −12, 2)
√
|v(2) × a(2)| = 2 181
Now, we use a formula for torsion:
(v(t) × a(t)) · da
dt (t)
τ (t) =
|v(t) × a(t)|2
(24, −12, 2) · (0, 0, 6) 3
τ (2) = √ =
(2 181) 2 181
1.4.1.8. Solution. For the specified curve
t2 t3
r(t) = tı̂ı + ̂ + k̂
2 3
v(t) = r0 (t) = ı̂ı + t ̂ + t2 k̂
a(t) = r00 (t) = ̂ + 2t k̂
 
ı̂ı ̂ k̂
v(t) × a(t) = det 1 t  t2 
0 1 2t
= t2 ı̂ı − 2t ̂ + k̂
a0 (t) = 2 k̂
From this, we read off
v(t) ı̂ı + t ̂ + t2 k̂
T̂(t) = =√
|v(t)| 1 + t2 + t4
√
|v(t) × a(t)| 1 + 4t2 + t4
κ(t) = 3
=
|v(t)| [1 + t2 + t4 ]3/2
v(t) × a(t) t2 ı̂ı − 2t ̂ + k̂
B̂(t) = =√
|v(t) × a(t)| 1 + 4t2 + t4
N̂(t) = B̂(t) × T̂(t)
 
ı̂ı ̂ k̂
1
=√ √ det t2 −2t 1
1 + t2 + t4 1 + 4t2 + t4
1 t t2
−(t + 2t3 )ı̂ı + (1 − t4 ) ̂ + (2t + t3 )k̂

= √ √
1 + t2 + t4 1 + 4t2 + t4
0
v(t) × a(t) · a (t) 2
τ (t) = 2
=
|v(t) × a(t)| 1 + 4t2 + t4
1.4.1.9. Solution. First, some preliminaries:
r(t) = (t3 , t, ect ) r(5) = (53 , 5, e5c )

v(t) = (3t2 , 1, cect ) v(5) = (3 · 52 , 1, ce5c )
a(t) = (6t, 0, c2 ect ) a(5) = (6 · 5, 0, c2 e5c )
da da
(t) = (6, 0, c3 ect ) (5) = (6, 0, c3 e5c )
dt dt
v(5) × a(5) = (c2 e5c , 15ce5c (2 − 5c), −30)
Second, we figure out what value of c makes τ (5) = 0.
(v(5) × a(5)) · da
dt (5)
0 = τ (5) =
|v(5) × a(5)|2
da
0 = (v(5) × a(5)) · (5)
dt
= (c2 e5c , 15ce5c (2 − 5c), −30) · (6, 0, c3 e5c )
= 6c2 e5c (1 − 5c)
1
c = 0 or c =
5
If c = 0, then r(t) = (t3 , t, 1), and so the entire curve is contained inside
the plane z = 1. (Its torsion is zero everywhere — not just at t = 5.)
Consider the case c = 15 . When t = 5, our curve (and its osculating
circle) passes through the point r(5) = (53 , 5, e). The normal vector to
v(5)×a(5)
the plane of the osculating curve is the binormal vector B̂(5) = |v(5)×a(5)| .
Since we don’t need the normal vector to the plane to be a unit vector,
we can take as the normal vector to the plane simply v(5) × a(5), or
(e/25, 3e, −30). Then, an equation of the plane containing the osculating
circle is (e/25)x + (3e)y − 30z = −10e. An equivalent equation for this
plane is (1/25)x + 3y − (30/e)z = −10.
1.4.1.10. ∗. Solution. (a) Since r0 (t) = (2t, 1, 3t2 ), we have r0 (1) =
(2, 1, 3). So the normal plane must pass through r(1) = (1, 1, 1) and be
perpendicular to (2, 1, 3). The equation of the normal plane is then
2(x − 1) + (y − 1) + 3(z − 1) = 0 or 2x + y + 3z = 6
(b) As
ds p
v(t) = r0 (t) = 2t, 1, 3t2

= 1 + 4t2 + 9t4
dt
a(t) = v0 (t) = 2, 0, 6t v(t) × a(t) = 6t, −6t2 , −2

the curvature
√
|v(t) × a(t)| 2 1 + 9t2 + 9t4
κ(t) = =
ds 3 [1 + 4t2 + 9t4 ]3/2

dt
1.4.1.11. ∗. Solution. First some preliminaries.
v(t) = r0 (t) = − sin tı̂ı + cos t ̂ + k̂

a(t) = r00 (t) = − cos tı̂ı − sin t ̂
(a), (b) From v(t) we read off
ds √
= |v(t)| = 2
dt
2 2 d2 s
From a(t) = ddt2s (t) T̂(t) + κ(t) ds
dt (t) N̂(t), and the fact that dt2 = 0, we
read off that
ds −2 1 a
κ(t) = (t) |a| = N̂(t) = = − cos tı̂ı − sin t ̂
dt 2 |a|
So the radius of curvature is κ1 = 2 and the centre of curvature is

1 h i
r(t)+ N̂(t) = cos tı̂ı +sin t ̂ +tk̂ + 2 − cos tı̂ı −sin t ̂
κ(t) t= π t= π
6
6
h i
= − cos tı̂ı − sin t ̂ + t k̂
t= π
6
√
3 1 π
=− ı̂ı − ̂ + k̂
2 2 6
(c) From
 
ı̂ı ̂ k̂
v(t) × a(t) = det  − sin t cos t 1  = sin tı̂ı − cos t ̂ + k̂
− cos t − sin t 0
|v(t) × a(t)|2 = 2
we read off
v(t) × a(t) 1 1 1
B̂(t) = = √ sin tı̂ı − √ cos t ̂ + √ k̂
|v(t) × a(t)| 2 2 2
so that
√
π 1 3 1
B̂ = √ ı̂ − √ ̂ + √ k̂
ı
6 2 2 2 2 2
1.4.1.12. ∗. Solution. (a) The velocity vector is
0
r (t) = (− sin(t), cos(t), 2t)
So a tangent vector at t = π is T = (0, −1, 2π) and a parametric form for
the tangent line is
R(t) = r(π) + tT = (−1, 0, π 2 ) + t(0, −1, 2π)
(b) The speed is
ds p
= |r0 (t)| = 1 + 4t2
dt
By Theorem 1.3.3, the tangential component of acceleration is
d2 s d p 4t
aT (t) = 2 = 1 + 4t2 = √
dt dt 1 + 4t2
1.4.1.13. ∗. Solution. (a) The velocity vector of the particle at time t

is
r0 (t) = (cos t − cos t + t sin t)ı̂ı + (− sin t + sin t + t cos t) ̂ + 2t k̂

= t sin tı̂ı + t cos t ̂ + 2t k̂
so its speed at time 1 ≤ t < ∞ is

ds p √
= |r0 (t)| = t2 sin2 t + t2 cos2 t + 4t2 = 5 t
dt
(b) The unit tangent at time t is
r0 (t) 1
T̂(T ) = = √ sin tı̂ı + cos t ̂ + 2 k̂
|r0 (t)| 5
So the tangential component of acceleration at time t is
d2 s
aT (t) = (t) T̂(t) = sin tı̂ı + cos t ̂ + 2 k̂
dt2
(c) The (full) acceleration is
d 0
r00 (t) =

r (t) = sin t + t cos t ı̂ı + cos t − t sin t ̂ + 2 k̂
dt
So the normal component of acceleration at time t is
aN (t) = a(t) − aT (t) = t cos tı̂ı − t sin t ̂
ds
2
(d) Another formula for the normal component of acceleration is κ(t) dt (t) N̂(t).
ds
2
So the magnitude of the normal component of acceleration is κ(t) dt (t)
and, by part (c),
2
ds
κ(t) (t) = t cos tı̂ı − t sin t ̂ = t
dt
Consequently, by part (a),

t 1
κ(t) = 2 =
ds 5t
dt (t)
1.4.1.14. ∗. Solution. (a) If the point (x, y, z) is on the curve, it obeys

both z = x2 + y 2 and z = 8 − 2x and hence is also obeys
x2 + y 2 = 8 − 2x or (x + 1)2 + y 2 = 9
So the curve C is also the intersection of
(x + 1)2 + y 2 = 9 and z = 8 − 2x
(x + 1)2 + y 2 = 9 is the circle of radius 3 centred on (−1, 0) and can be

parametrized by x(θ) = −1 + 3 cos θ, y(θ) = 3 sin θ, 0 ≤ θ ≤ 2π. So C can
be parametrized by
x(θ) = −1 + 3 cos θ
y(θ) = 3 sin θ
z(θ) = 8 − 2x(θ) = 10 − 6 cos θ
or r(θ) = [−1 + 3 cos θ]ı̂ı + 3 sin θ ̂ + [10 − 6 cos θ] k̂
with 0 ≤ θ < 2π. √

Remark: if we tried to parametrize the equation as (x, y, z) = (x, 8 − 2x − x2 , 8−
2x), then we would miss the negative y-values.
(b) Note that r(θ) is (2, 0, 4) when θ = 0. As
v(θ) = r0 (θ) = −3 sin θ ı̂ı + 3 cos θ ̂ + 6 sin θ k̂ v(0) = 3̂

0
a(θ) = v (θ) = −3 cos θ ı̂ı − 3 sin θ ̂ + 6 cos θ k̂ a(0) = −3ı̂ı + 6k̂
the unit tangent vector at (2, 0, 4) is
v(0)
T̂(0) = = ̂
|v(0)|
and, since v(0) × a(0) = 9k̂ + 18ı̂ı, the unit binormal vector and curvature
at (2, 0, 4) are
√ √
v(0) × a(0) 2ı̂ı + k̂ |v(0) × a(0)| 9 5 5
B̂(0) = = √ κ(0) = 3
= 3 =
|v(0) × a(0)| 5 |v(0)| 3 3
and the unit normal vector N̂ at (2, 0, 4)

1 1
N̂(0) = B̂(0) × T̂(0) = √ (2ı̂ı + k̂) × ̂ = √ (2k̂ − ı̂ı)
5 5
1.4.1.15. ∗. Solution. We have
√ p
v(t) = r0 (t) = t2 ı̂ı + 2 t ̂ + k̂ |v(t)| = t4 + 2t2 + 1 = t2 + 1
√
a(t) = v0 (t) = 2tı̂ı + 2 ̂
(a) The unit tangent vector is

√
v(t) t2 ı̂ı + 2 t ̂ + k̂
T̂(t) = =
|v(t)| t2 + 1
(b) Since
 
ı̂ı √̂ k̂ √ √
v(t) × a(t) = det t2  = − 2ı̂ı + 2t ̂ − 2 t2 k̂
√2 t 1
2t 2 0
The curvature is
√ √
|v(t) × a(t)| 2 + 4t2 + 2t4 2
κ(t) = = 3 = 2
|v(t)|3 2
(t + 1) 2
(t + 1)
√
(c) Note that r(2) is 83 , 2 2, 2 .
• Solution 1: Since
√ √
0 2tı̂ı + 2 ̂ t2 ı̂ı + 2 t ̂ + k̂
T̂ (t) = − 2t 2
t2 + 1 (t2 + 1)
√ √
4ı̂ı + 2 ̂ 4ı̂ı + 2 2 ̂ + k̂
T̂0 (2) = −4
5 25
√
4ı̂ı − 3 2 ̂ − 4k̂
=
25
√
0 5 2
|T (2)| =
25
8
√
the principal normal vector N̂ at 3, 2 2, 2 is
√
T̂0 (2) 4ı̂ı − 3 2 ̂ − 4k̂
N̂(2) = = √
|T̂0 (2)| 5 2
• Solution 2: Perhaps we’d rather not differentiate T̂(t).
v(t) × a(t)
B̂ = and N̂ = B̂ × T̂
|v(t) × a(t)|
Using our previous work:

√ √
v(2) × a(2) − 2ı̂ı + 4̂ − 4 2k̂
B̂(2) = = √
|v(2) × a(2)| 2 + 16 + 32
1 √
= −ı̂ı + 2 2̂ − 4k̂
5
1 √
T̂(2) = 4ı̂ı + 2 2̂ + k̂
5
1 √ 1 √
N̂(2) = B̂(2) × T̂(2) = −ı̂ı + 2 2̂ − 4k̂ × 4ı̂ı + 2 2̂ + k̂
5 5
√ √
2 2 3 2 2
= ı̂ı + − ̂ + − k̂
5 5 5
1.4.1.16. ∗. Solution. (a) The curve x2 + y 2 = 1 is a circle of radius 1.
So we can parametrize it by x(θ) = cos θ, y(θ) = sin θ, 0 ≤ θ < 2π. The
z-coordinate of any point on the intersection is determined by z = 1−x−y.
So we can use
r(θ) = cos θ ı̂ı + sin θ ̂ + (1 − cos θ − sin θ) k̂ 0 ≤ θ < 2π
(b) As
v(θ) = r0 (θ) = − sin θ, cos θ, sin θ − cos θ

a(θ) = v0 (θ) = − cos θ, − sin θ, cos θ + sin θ

we have
ds
q
= |r0 (θ)| = sin2 θ + cos2 θ + (sin θ − cos θ)2
dθ √
= 2 − 2 sin θ cos θ
p
= 2 − sin(2θ)

v(θ) × a(θ) = 1, 1, 1
and the curvature

√
|v(θ) × a(θ)| 3
κ(θ) = =
ds 3 [2 − sin(2θ)]3/2

dθ
(c) The curvature is

• a maximum (minimum) when 2 − sin(2θ) is a minimum (maximum),
• which is the case when sin(2θ) = 1 (sin(2θ) = −1),
π 5π 3π 7π
• which in turn is the case when θ = 4, 4 (θ = 4 , 4 ).
So
√
3 √ ı̂ı ̂ √
maximum curvature = = 3 at √ + √ + (1 − 2) k̂
[2 − 1]3/2 2 2
ı̂ı ̂ √
and − √ − √ + (1 + 2) k̂
2 2
√
3 1 ı̂ı ̂
minimum curvature = = at − √ + √ + k̂
[2 − (−1)]3/2 3 2 2
ı̂ı ̂
and √ − √ + k̂
2 2
1.4.1.17. ∗. Solution. For r(t) to be well-defined, we need t > 0
(because of the ln t.)
r
0 1 ds 1 1
v(t) = r (t) = 2tı̂ı + 2 ̂ + k̂ = 4t2 + 4 + 2 = 2t +
t dt t t
The unit tangent vector is
1
r0 (t) 2tı̂ı + 2 ̂ + t k̂ 2t2 ı̂ı + 2t ̂ + k̂
T̂(t) = = =
0
|r (t)| 2t + 1t 2t2 + 1
so, from §1.5,
ds 4tı̂ı + 2 ̂ 2t2 ı̂ı + 2t ̂ + k̂

(t) κ(t) N̂(t) = T̂0 (t) = − 4t 2
dt 2t2 + 1 (2t2 + 1)
4tı̂ı + (−4t2 + 2) ̂ − 4t k̂
= 2
(2t2 + 1)
2tı̂ı − (2t2 − 1) ̂ − 2t k̂
=2 2
(2t2 + 1)
Since the length of 2tı̂ı − (2t2 − 1) ̂ − 2t k̂ is

p p p
4t2 + (2t2 − 1)2 + 4t2 = 8t2 + 4t4 − 4t2 + 1 = 4t4 + 4t2 + 1
p
= (2t2 + 1)2 = 2t2 + 1
we have
2tı̂ı − (2t2 − 1) ̂ − 2t k̂

N̂(t) =
2t2 + 1
and
2
|T̂0 (t)| 2 2t
κ(t) = ds
= 2t +11 = 2
dt (t)
2t + t 2
(2t + 1)
1.4.1.18. ∗. Solution. (a) Since
t2 t3
r(t) = ı̂ı + ̂ + k̂
2 3
0 2
r (t) = t ̂ + t k̂
ds p p
(t) = |r0 (t)| = t2 + t4 = t 1 + t2
dt
the length of the curve is
Z 1 Z 1 p
ds 1 3/2 1 1
t 1 + t2 dt = 1 + t2 = 23/2 − 1

(t) dt =
0 dt 0 3 0 3
(b) For the specified curve
r(t) = cos(t)ı̂ı + sin(t) ̂ + t k̂

r0 (t) = − sin(t)ı̂ı + cos(t) ̂ + 1 k̂
− sin(t)ı̂ı + cos(t) ̂ + 1 k̂
T̂(t) = √
2
− cos(t)ı̂ı − sin(t) ̂
T̂0 (t) = √
2
π −ı̂
ı − 
̂
T̂0

=
4 2
0
T̂ π4

π −ı̂ı − ̂
N̂ = π
= √
4 0
|T̂ 4 | 2
(c) Recalling, from §1.5, that
ds
T̂0 (t) = κ(t) (t) N̂(t)
dt
we have, by part (d),
√
π |T̂0 π

4 | 1/ 2 1
κ = 0 π = √ =
4 |r 4 | 2 2
1.4.1.19. ∗. Solution. (a), (b), (c) We have
r(t) = t + 2 , 1 − t , t2 /2

v(t) = r0 (t) = 1 , −1 , t

ds p
(t) = |v(t)| = 2 + t2
dt
a(t) = v0 (t) = 0 , 0 , 1

(d) By §1.5, the curvature

√
|v(t) × a(t)| |(−1, −1, 0)| 2
κ(t) = = =
( ds
dt (t)) 3 2
[2 + t ]
3/2
[2 + t2 ]
3/2
√ 2
(e) Since ds
dt (t) = 2 + t2 , we have ddt2s (t) = √2+t
t
2
and
d2 s ds 2
0 , 0 , 1 = a(t) = 2 T̂(t) + κ(t) N̂(t)
dt dt √
t (1 , −1 , t) 2 p 2
=√ √ + 3/2
2 + t2 N̂(t)
2+t 2 2+t 2
[2 + t2 ]
or
√
2 (t , −t , t2 ) (−t , t , 2)
√ N̂(t) = 0 , 0 , 1 − 2
=
2+t2 2 + t 2 + t2
which implies
(−t , t , 2)
N̂(t) = p
2(2 + t2 )
(f) At t = 0
r(0) = (2, 1, 0)
(1, −1, 0)
T̂(0) = √
2
N̂(0) = (0, 0, 1)
1 1
B̂(0) = T̂(0) × N̂(0) = √ (1, −1, 0) × (0, 0, 1) = √ (−1, −1, 0)
2 2
The osculating plane is the plane through r(0) which is perpendicular to
B̂(0), which is
1
√ (−1, −1, 0) · (x, y, z) − (2, 1, 0) = 0 or x+y =3
2
(g) The osculating circle has centre
1 1
r(0) + N̂(0) = (2, 1, 0) + (0, 0, 1) = (2, 1, 2)
κ(0) 1/2
1.4.1.20. ∗. Solution. First some preliminary computations.
3
t t2
r(t) = ı̂ı + √ ̂ + t k̂
3 2
0
√
r (t) = t2 ı̂ı + 2t ̂ + k̂
p
|r0 (t)| = t4 + 2t2 + 1 = t2 + 1
√
r00 (t) = 2tı̂ı + 2 ̂
 
ı̂ı √̂ k̂ √ √
r0 (t) × r00 (t) = det t2 2t 1  = − 2ı̂ı + 2t ̂ − 2t2 k̂
√
2t 2 0
(a) The unit tangent vector is
√
r0 (t) t2 ı̂ı + 2t ̂ + k̂
T̂(t) = 0 =
|r (t)| t2 + 1
(b) The curvature is (see §1.5)
√ √ √
|r0 (t) × r00 (t)| | − 2ı̂ı + 2t ̂ − 2t2 k̂| 2 + 4t2 + 2t4
κ(t) = 0 3
= 3 = 3
|r (t)| 2
(t + 1) (t2 + 1)
√
2
= 2
2
(t + 1)
(c) At t = 0 √
κ(0) = 2
For ease of computation, we’ll find B̂ first, then use it to find N̂.
(e) At t = 0, the binormal vector is (see §1.5)
√
r0 (0) × r00 (0) − 2ı̂ı
B̂(0) = 0 = √ = −ı̂ı
|r (0) × r00 (0)| 2
(d) At t = 0, the principal normal vector is (see §1.5)
N̂(0) = B̂(0) × T̂(0) = −ı̂ı × k̂ = ̂

1.4.1.21. ∗. Solution. The curve has
r(t) = (t2 , t , t3 )
v(t) = r0 (t) = (2t , 1 , 3t2 )
a(t) = r00 (t) = (2 , 0 , 6t)
(a) In particular, a (non unit) tangent vector at r(−1) = (1, −1, −1) is
r0 (−1) = (−2, 1, 3). So the tangent line to the curve at (1, −1, −1) is
(x, y, z) − (1, −1, −1) = t(−2, 1, 3)
or
x = 1 − 2t
y = −1 + t
z = −1 + 3t
(b) At r(1) = (1, 1, 1),
v(1) = r0 (1) = (2, 1, 3)

a(1) = r00 (1) = (2, 0, 6)
v(1) × a(1) = (6, −6, −2)
So the unit binormal vector is

v(1) × a(1) (3, −3, −1) 1
B̂(1) = = = √ (3, −3, −1)
|v(1) × a(1)| |(3, −3, −1)| 19
An equation for the osculating plane is
(3, −3, −1) · (x − 1 , y − 1 , z − 1) = 0 or 3x − 3y − z = −1

1.4.1.22. ∗. Solution. (a) For this curve
r0 (t) = t sin tı̂ı + t cos t ̂ + 2t k̂

ds √
(t) = |r0 (t)| = 5 t
dt
so the length of the curve from t = 0 to t = π is
√
Z π
ds √ Z π
5 π2
(t) dt = 5 t dt =
0 dt 0 2
(b) The unit tangent vector is
r0 (t) 1
T̂ (t) = = √ sin tı̂ı + cos t ̂ + 2 k̂
|r0 (t)| 5
so that
ds dT̂ 1
κ(t) (t) N̂(t) = (t) = √ cos tı̂ı − sin t ̂
dt dt 5
which implies that

ds
dt (t)
√
z}|{
1 1 1
κ(t) 5 t = √ cos tı̂ı − sin t ̂ = √ =⇒ κ(t) =
5 5 5t
1.4.1.23. ∗. Solution. (a) For the specified curve
√ √
4 2 3/2 4 2 3/2
r(t) = t , t , t(2 − t)
3 3
√ 1/2 √ 1/2
v(t) = 2 2t , 2 2t , 2 − 2t
p p
|v| = 8t + 8t + 4 − 8t + 4t2 = 4(1 + 2t + t2 ) = 2(1 + t)
The rocket is at z = 0 when t = 0 and when t = 2. So the distance travelled

is
Z 2 Z 2 2
t2

|v(t)| dt = 2(1 + t) dt = 2 t + =8
0 0 2 0
(b) The rocket is at its maximum height when dz dt = 2 − 2t = 0. That

√ √
is, when t = 1. Its velocity then is (2 2, 2 2, 0). A unit vector in this
direction is T̂(1) = √12 (1, 1, 0). That is the unit tangent vector.
At general t, the unit tangent is
√ 1/2 √ 1/2
v(t) 2t , 2t , 1 − t
T̂(t) = =
|v(t)| 1+t
So
√ √ √ 1/2 √ 1/2
2t−1/2 /2, 2t−1/2 /2, −1

0 2t , 2t , 1 − t
T̂ (t) = −
1+t (1 + t)2
√ √ √ √
2/2, 2/2, −1 2, 2, 0
T̂0 (1) = −
2 4
1
= 0, 0, −
2
So the principal unit normal vector is
T̂0 (1)
N̂(1) = = (0, 0, −1)
|T̂0 (1)|
(c) As
dT̂ 1 ds
(1) = 0, 0, − (1) = |v(1)| = 4
dt 2 dt
the curvature
|T̂0 (1)| 1
κ(1) = =
|v(1)| 8
r(t) = cos3 t, sin3 t, 2 sin2 t

v(t) = − 3 cos2 t sin t, 3 sin2 t cos t, 4 sin t cos t

= sin t cos t(−3 cos t, 3 sin t, 4)

p
|v(t)| = sin t cos t 9 cos2 t + 9 sin2 t + 16 = 5 sin t cos t
So the distance travelled is

Z π/2 Z π/2 π/2
5 5
|v(t)| dt = 5 sin t cos t dt = sin2 t =

0 0 2 0 2
(b) Since
v(t) = sin t cos t(−3 cos t, 3 sin t, 4) |v(t)| = 5 sin t cos t
we have
π √
v(t) 1 1 3 3 3
T̂(t) = = (−3 cos t, 3 sin t, 4) T̂ = − , ,4
|v(t)| 5 6 5 2 2
π √
1 13 3 3
T̂0 (t) = (3 sin t, 3 cos t, 0) T̂0 = , ,0
5 6 5 2 2
3 √
= 3, 1, 0
10
T̂0 π
1 √
π
6
π π π
N̂ = π
= 3, 1, 0 B̂( = T̂ × N̂
6 |T̂0 6 |
2 6 6 6
1 √
= − 4, 4 3, −6)
10
1 √
= − 2, 2 3, −3)
5
1.4.1.25. Solution. (a) The curve x2 + y 2 = 1 is a circle of radius 1.
So we can parametrize it by x(θ) = cos θ, y(θ) = sin θ, 0 ≤ θ < 2π. The
z-coordinate of any point on the intersection is determined by z = x2 − y 2 .
So we can use the parametrization
r(θ) = cos θ ı̂ı + sin θ ̂ + [cos2 θ − sin2 θ] k̂
= cos θ ı̂ı + sin θ ̂ + cos(2θ) k̂ 0 ≤ θ < 2π
√ √
(b) Note that r(θ) = 1/ 2 , 1/ 2 , 0 when θ = π4 . For general θ, the

velocity and acceleration are

v(θ) = r0 (θ) = − sin θ ı̂ı + cos θ ̂ − 2 sin(2θ) k̂
a(θ) = v0 (θ) = − cos θ ı̂ı − sin θ ̂ − 4 cos(2θ) k̂
In particular,
1 1
v(π/4) = − √ ı̂ı + √ ̂ − 2 k̂
2 2
1 1
a(π/4) = − √ ı̂ı − √ ̂
2 2
ds √
(π/4) = |v(π/4)| = 5
dθ √ √
v(π/4) × a(π/4) = − 2ı̂ı + 2 ̂ + k̂
√
|v(π/4) × a(π/4)| = 5
So the curvature
|v(π/4) × a(π/4)| 1
κ(π/4) = =
|v(π/4)|3 5
√ √
(c) The binormal to C at 1/ 2 , 1/ 2 , 0 is
√ √
v(π/4) × a(π/4) − 2ı̂ı + 2 ̂ + k̂
B̂(π/4) = = √
|v(π/4) × a(π/4)| 5
√ √
So the osculating plane to C at 1/ 2 , 1/ 2 , 0 is
√ √ √ √
− 2 , 2 , 1 · x − 1/ 2 , y − 1/ 2 , z − 0 = 0 or
√ √
z = 2x − 2y
(d) From the computations in parts (b) and (c), we have

√ √
v(π/4) −1/ 2ı̂ı + 1/ 2 ̂ − 2 k̂
T̂(π/4) = = √
|v(π/4)| 5
√ √
v(π/4) × a(π/4) − 2ı̂ı + 2 ̂ + k̂
B̂(π/4) = = √
|v(π/4) × a(π/4)| 5
−ı̂ı − ̂
N̂(π/4) = B̂(π/4) × T̂(π/4) = √
2
So the osculating circle has radius 1/κ(π/4) = 5 and centre
N̂(π/4) √ √ √ √
rc (π/4) = r(π/4) + = 1/ 2 , 1/ 2 , 0 − 5 1/ 2 , 1/ 2 , 0
κ(π/4)
√ √
= − 2 2 , −2 2 , 0
1.4.1.26. ∗. Solution. We’ll solve this problem twice, using two differ-
ent strategies. (The second strategy will be much more efficient than the
first one.) Both strategies use that F = ma. Since we are told that m = 2,
we just have to find the acceleration a at (1, 1, 1).
Strategy 1: In the first strategy, we’ll find the position r(t), as a function
of time t and then differentiate twice to get the acceleration a(t).
• First we’ll find any old parametrization. We are told that, on the
path, z = x and z = y 2 . So let’s use y as the parameter. Then
x = z = y 2 . So the parametrization is R(y) = y 2 ı̂ı + y ̂ + y 2 k̂.
(We’ll save the notation “r(t)” for the parametrization with respect
to time.)
• Next we’ll reparametrize to get the time t as the parameter. Since
dR
= 2y ı̂ı + ̂ + 2y k̂
dy
ds p
=⇒ = 2y ı̂ı + ̂ + 2y k̂ = 1 + 8y 2
dy
We are told that the speed dsdt = 3 for all t. So, choosing our zero
point for time to coincide with our zero point for s, we have s = 3t,
or t = s/3 so that
dt 1p
= 1 + 8y 2
dy 3
We could now integrate to get t as a function of y. But that looks
quite messy. Fortunately we only need the acceleration at one point,
namely (1, 1, 1). We’ll now see that that saves quite a bit of work.
Pretend that we have integrated to get t as a function of y and call

the answer t(y). Call the inverse function, which gives y as a function
of t, y(t).

• We now have r(t) = R y(t) . So, by the chain rule,
r0 (t) = R0 y(t) y 0 (t)

r00 (t) = R0 y(t) y 00 (t) + R00 y(t) y 0 (t)2

We’re only interest in the time, call it t0 , at which y(t0 ) = 1. The

acceleration at time t0 is
r00 (t0 ) = R0 y(t0 ) y 00 (t0 ) + R00 y(t0 ) y 0 (t0 )2

= R0 (1) y 00 (t0 ) + R00 (1) y 0 (t0 )2

= 2ı̂ı + ̂ + 2 k̂ y 00 (t0 ) + 2ı̂ı + 2 k̂ y 0 (t0 )2

so we just have to find y 0 (t0 ) and y 00 (t0 ).

dt
p
• We know that dy = 13 1 + 8y 2 . So by the inverse function theorem
dy 3
(t) = p
dt 1 + 8y(t)2
1 3 16y(t)y 0 (t)

d2 y
(t) = −
dt2 2 [1 + 8y(t)2 ]3/2
In particular
3 3
y 0 (t0 ) = p =√ =1
1 + 8y(t0 )2 1+8
24 y(t0 )y 0 (t0 ) 24 × 1 × 1 8
y 00 (t0 ) = − 3/2
=− 3/2
=−
[1 + 8y(t0 )2 ] (1 + 8) 9
• Finally, the force is
2r00 (t0 ) = 2 2ı̂ı + ̂ + 2 k̂ y 00 (t0 ) + 2 2ı̂ı + 2 k̂ y 0 (t0 )2

16
=− 2ı̂ı + ̂ + 2 k̂ + 2 2ı̂ı + 2 k̂
9
4 16 4
= ı̂ı − ̂ + k̂
9 9 9
Strategy 2: The second strategy will be based on (see §1.5)

d2 s ds 2
a= T̂ + κ N̂
dt2 dt
ds d2 s
In this problem, we are told that dt = 3 for all t, so that dt2 = 0 and
a = 9κN̂
So we just have to find the curvature, κ, and unit normal, N̂, at (1, 1, 1). We
have already found one parametrization of the path in strategy 1, namely
R(y) = y 2 ı̂ı + y ̂ + y 2 k̂
Note that R(1) = (1, 1, 1). Since
R0 (y) = 2y ı̂ı + ̂ + 2y k̂

R0 (y) 2y ı̂ı + ̂ + 2y k̂

T̂(y) = 0
= p
|R (y)| 1 + 8y 2
2ı̂ı + 2 k̂ 16y 2y ı̂ı + ̂ + 2y k̂
T̂0 (y) = p −
1 + 8y 2 2 [1 + 8y 2 ]3/2
2ı̂ı + 2 k̂ 2ı̂ı + ̂ + 2 k̂ 2ı̂ı − 8 ̂ + 2 k̂
T̂0 (1) = −8 =
3 27 27
we have (again see §1.5)
|T̂0 (1)|
κ(1) =
|R0 (1)|
T̂0 (1)
N̂(1) =
|T̂0 (1)|
T̂0 (1) 2ı̂ı − 8 ̂ + 2 k̂
F = ma = 2 × 9κ(1)N̂(1) = 18 = 18 √
|R0 (1)| 27 1 + 8 × 12
4
= (ı̂ı − 4 ̂ + k̂)
9
1.4.1.27. ∗. Solution. (a) As
√
r(t) = 2tı̂ı + t2̂ + 3t2 k̂
√
r0 (t) = 2ı̂ı + 2t̂ + 2 3tk̂
the unit tangent vector is
√ √
ı̂ı + t̂ + 3tk̂ ı̂ı + t̂ + 3tk̂
T̂(t) = √ = √
|ı̂ı + t̂ + 3tk̂| 1 + 4t2
(b) Since
√ √ √
dT̂ ̂ + 3k̂ ı̂ı + t̂ + 3tk̂ −4tı̂ı + ̂ + 3k̂
(t) = √ − 4t 3/2
= 3/2
dt 1 + 4t2 (1 + 4t2 ) (1 + 4t2 )
the unit normal is
√ √
−4tı̂ı + ̂ + 3k̂ −4tı̂ı + ̂ + 3k̂
N̂(t) = √ = √
| − 4tı̂ı + ̂ + 3k̂| 2 1 + 4t2
(c) The unit binormal is
B̂(t) = T̂(t) × N̂(t)
 
ı̂ı ̂ √k̂
1
= det  1 t √3t
2(1 + 4t2 )
−4t 1 3
√ 2  2
− 3(1 + 4t )̂ + (1 + 4t )k̂
=
2(1 + 4t2 )
√
3 1
=− ̂ + k̂
2 2
which is (3).
(d) The
√
plane contains the point r(0) = 0 and is perpendicular to the
vector − 23 ̂ + 12 k̂ and so is
√
− 3y + z = 0
(e) The curvature is

dT̂ ds | − 4tı̂ı + ̂ + √3k̂| 1
κ(t) = (t)/ = √

dt dt 3/2
(1 + 4t2 ) |2ı̂ı + 2t̂ + 2 3tk̂|
√
4 + 16t2 1 1
= 3/2
√ = 3/2
2
(1 + 4t2 ) 2 1 + 4t (1 + 4t2 )
3/2
(f), (g) The denominator (1 + 4t2 ) of κ(t) is a minimum at t = 0 and
grows without bound as |t| increases. So the denominator never achieves
a maximum. Consquently, the curvature κ(t) achieves its maximum value
when t = 0 and so at r(0) = (0, 0, 0). The curvature never achieves a
minimum. √ √
(h) Since 3 v + w = 4 k̂ and v − 3 w = 4 ̂,
√ √
u v − 3w 3v + w
ıı̂ = ̂ = k̂ =
2 4 4
√
Since u = 2ı̂ı and v = ̂ + 3 k̂,
r(t) = t u + t2 v = a(t)u + b(t)v + c(t)w

with a(t) = t, b(t) = t2 , c(t) = 0

The curve a(t), b(t) = (t, t2 ) is the curve y = x2 . It is “curviest” at the
origin, which is consistent with part (f). It becomes flatter and flatter as
|t| increases, but never achieves “perfect flatness”, which is consistent with
(g).
y
y = x2
v
C
u x
1.4.1.28. ∗. Solution. The three unit vectors T̂, N̂ and B̂ are mutually
perpendicular and form a right handed triple.
B̂
N̂
T̂
So
N̂ = B̂ × T̂ N̂ × T̂ = −B̂ B̂ × N̂ = −T̂
and
dN̂ dB̂ dT̂

= × T̂ + B̂ × = −τ N̂ × T̂ + B̂ × κN̂ = τ B̂ − κT̂
ds ds ds
1.4.1.29. ∗. Solution. (a) Parametrizing the curve by θ gives

r(θ) = sin(2θ), 1 − cos(2θ), 2 cos θ ,
v = r0 (θ) = 2 cos(2θ), 2 sin(2θ), −2 sin θ ,

a = r00 (θ) = − 4 sin(2θ), 4 cos(2θ), −2 cos θ .

At the point P , we have θ = π/4, giving instantaneous values

√ √ √ √
r = (1, 1, 2), v = 0, 2, − 2 , v = |v| = 6, a = − 4, 0, − 2 .
v
√
Hence T̂ = |v| = √16 0, 2, − 2 .
v×a
√ √ √
Now B̂ = |v×a| = √126 − 2, 2 2, 4 = √113 − 1, 2, 2 2 , since

ı̂ı ̂ k̂
√ √ √ √ √

v × a = 0 2 − 2 = − 2 2, 4 2, 8 ,
|v × a| = 104 = 2 26.
−4 0 − 2 √
This leads to

ı̂ı ̂ k̂
1 √ 1 √ √
N̂ = B̂ × T̂ = √ −1 2 2 √2 = − √ 6 2, 2, 2
78 78
0 2 − 2
1 √
= −√ 6, 1, 2 .
39
Finally, √ √ √ √ √
|v × a| 2 26 2 2 13 13 39
κ= = √ 3 = √ √ = √ = .
v3 ( 6) 6 2 3 3 3 9
(b) Now parametrize the curve by time, t, and write v = r0 (t), v =
|r (t)| and a = r00 (t). Note that in part (a) we used v, v and a with
0
different meanings. We use the dot product to extract the tangential and
normal components of a = dv 2
dt T̂ + v κN̂:

dv
a · T̂ = T̂ + v 2 κN̂ · T̂
dt
dv
= T̂ · T̂ + (v 2 κ)N̂ · T̂
dt
Since T̂ is a unit vector, T̂·T̂ = kT̂k2 = 1; since T̂ and N̂ are perpendicular,
T̂ · N̂ = 0.
dv
=
dt
dv
This gives us a nice way to compute dt , the rate of change of speed.
dv √ 1 √
= a · T̂ = (−2, 3, −2 2) · √ 0, 2, − 2
dt 6
1 10 5√
= √ [0 + 6 + 4] = √ = 6.
6 6 3
Similarly, a · N̂ = v 2 κ, so
1 9 −1 √ √ 9 × 13
v2 = a · N̂ = √ √ (−2, 3, −2 2) · 6, 1, 2 = = 3.
κ 39 39 39
√ √
Hence |v| = 3; since v = |v|, v = 3. Then v = |v|T̂ = v T̂ =
√ √ √
√3 0, 2, − 2 = (0, 2, −1).
6
1.4.1.30. ∗. Solution. (a) The position, velocity and acceleration are

r(t) = cos t , sin t , c sin t
v(t) = r0 (t) = − sin t , cos t , c cos t

a(t) = r00 (t) = − cos t , − sin t , −c sin t

(b) The speed is

p
v(t) = |v(t)| = 1 + c2 cos2 t
(c) By Theorem 1.3.3.c, the tangential component of the acceleration

is
d2 s d p −c2 sin t cos t
2 cos2 t = √
= 1 + c
dt2 dt 1 + c2 cos2 t
(d) y(t) = sin t and z(t) = c sin t obey z(t) = cy(t) for all t. So the
curve lies on the plane z = cy.
1.4.1.31. ∗. Solution. (a) For the specified curve r(π) = − 4, 0, 41

and
1
r(θ) = 4 cos θ , 2 sin θ , cos(2θ)
4
0
1
v(θ) = r (θ) = − 4 sin θ , 2 cos θ , − sin(2θ)
2
00
a(θ) = r (θ) = − 4 cos θ , −2 sin θ , − cos(2θ)

v(π) = 0 , −2 , 0

a(π) = 4 , 0 , −1

v(π) × a(π) = 2 , 0 , 8
So the curvature at θ = π is
√
|v(π) × a(π)| |(2, 0, 8)| 17
κ(π) = 3
= =
|v(π)| |(0, −2, 0)|3 4
(b) The radius is
1 4
=√
κ(π) 17
(c) Set R(t) = r(t2 ). Then
R0 (t) = 2t r0 (t2 )
R00 (t) = 2 r0 (t2 ) + 4t2 r00 (t2 )
In particular,
√ 1
R( π) = − 4 , 0 ,
√ √ 4 √
R0 ( π) = 2 π v(π) = 0 , −4 π , 0

√ √
speed = R0 ( π) = 4 π

√
acceleration = R00 ( π) = 2 v(π) + 4π a(π) = 16π , −4 , −4π

The normal component of the acceleration has magnitude

√
ds 2 17 √ 2 √
κ = 4 π = 4 17 π
dt 4

1.6.1 · Exercises
1.6.1.1. Solution. We want to add up all the tinyR pieces of arclength
ds along a curve C. So, the integral would simply be C ds.
To see this another way, if we define r = (x(t), y(t), z(t)) for a ≤ t ≤ b
to be the equation of C, we could calculate the arclength as:
Z b Z b p
|r0 (t)|dt = x0 (t)2 + y 0 (t)2 + z 0 (t)2 dt
a a
This fits the formR of Definition 1.6.1 with f (x, y,

R z) = 1, so we write it as
a line integral as C 1ds, which is equivalent to C ds.
1.6.1.2. Solution. (a) The curve is r(θ) = x(θ)ı̂ı + y(θ) ̂ with x(θ) =
r(θ) cos θ, y(θ) = r(θ) sin θ and θ1 ≤ θ ≤ θ2 . On this curve
dr
v(θ) = (θ) = x0 (θ)ı̂ı + y 0 (θ)̂
dθ
= r0 (θ) cos θ − r(θ) sin θ ı̂ı + r0 (θ) sin θ + r(θ) cos θ ̂

ds
q 2 2
=⇒ (θ) = r0 (θ) cos θ − r(θ) sin θ + r0 (θ) sin θ + r(θ) cos θ
dθ p
= r0 (θ)2 + r(θ)2
Hence
Z Z θ2 ds
f (x, y) ds = f x(θ), y(θ) dθ
C θ1 dθ
s 2
Z θ2
dr
= f r(θ) cos θ, r(θ) sin θ r(θ)2 + (θ) dθ
θ1 dθ
(b) In this case f (x, y) = 1, r(θ) = 1 + cos θ, θ1 = 0 and θ2 = 2π,

Z Z 2π p Z 2π p
ds = [1 + cos θ]2 + [− sin θ]2 dθ = 2(1 + cos θ) dθ
C 0 0
Z 2π
r Z 2π Z π
θ cos θ dθ = 4
θ
= 4 cos2 dθ = 2 cos dθ
0 2 0
2 0 2
π
θ
= 8 sin = 8
2 0
1.6.1.3. Solution. Following Definition 1.6.1:

Z 2 2 3 √ 2!q √
3t · 3t
Z
xy
ds = (2t2 )2 + (2 3t)2 + (3)2 dt
C z 1 3t
Z 2
2 4
= √ t (2t2 + 3) dt
1 3 3
4 2
= √ (27 − 1) + √ (25 − 1)
21 3 5 3
1.6.1.4. Solution. We parametrize the unit circle as (cos t, sin t), 0 ≤
t ≤ 2π.
A tiny slice of the hoop with length ds has mass (x2 kg/m)(dsm) =
x2 dskg. So, the entire hoop has mass:

Z Z 2π p Z 2π
x2 ds = cos2 t (− sin t)2 + (cos t)2 dt = cos2 t dt
C 0 0
Z 2π 2π
1 + cos(2t) t sin(2t)
= dt = + = πkg
0 2 2 4 0
R 2π
For an efficient, sneaky, way to evaluate 0
cos2 t dt, see Example 2.4.4.
1.6.1.5. Solution. To parametrize C, we note the vector between the
two points is (2 − 1, 4 − 2, 5 − 3) = (1, 2, 2). So, the line is (1, 2, 3) + t(1, 2, 2)
for 0 ≤ t ≤ 1. That is, x(t) = 1 + t, y(t) = 2 + 2t, and z(t) = 3 + 2t.
Z Z 1 p
(xy + z)ds = ((1 + t)(2 + 2t) + (3 + 2t)) 11 + 22 + 22 dt
C 0
Z 1
3 5 + 6t + 2t2 dt = 26

=
0
dr
1.6.1.6. Solution. (a) In this case r(t) = tı̂ı + t2̂, so that v(t) = dt (t) =
√
ı̂ı + 2t̂ and ds
dt = 1 + 4t2 . Hence
Z Z 1 Z 1
ds p
f (x, y, z) ds = x(t) cos z(t) dt = t(cos 0) 1 + 4t2 dt
C 0 dt 0
2 3/2 1 3/2

1 (1 + 4t ) 5 −1
= =
8 3/2 0 12
(b) In this case r(t) = t, 23 t3/2 , t , so that v(t) = dr

1/2

√ dt (t) = 1, t ,1
and ds
dt = 2 + t. Hence
2 2
t + 32 t3/2 √
Z Z Z
x(t) + y(t) ds
f (x, y, z) ds = dt = 2 3/2 2 + t dt
C 1 y(t) + z(t) dt 1 3t +t
2
(2 + t)3/2 8 − 33/2
= =
3/2 1 3/2
1.6.1.7. Solution. In the figure below, we construct a triangle with
θ = arcsec t; the hypotenuse has length t, while the side adjacent to θ has
length 1. By the Pythagorean Theorem, the remaining side has length
√ √
2
t2 − 1, so sin θ = sin(arcsec t) = t t−1 .
t √
t2 − 1
θ
1
d 1 d √1
Remember dt {ln t}
= and t dt {arcsec t} = |t| t2 −1
. In our range,
√
1 ≤ t ≤ 2, we have |t| = t.
√ s 2 2
Z Z 2
1 1
sin x ds = sin (arcsec t) √ + dt
C 1 t t2 − 1 t
√
2
√ s
t2 − 1
Z
1 1
= + dt
1 t t2 (t2 − 1) t2
√
Z 2
1 1
= dt = ln 2
1 t 2
1.6.1.8. ∗. Solution. (a) Since the particle has mass m = 1, Newton’s
law of motion ma = F simplifies to
r00 (t) = ̂ − sin t k̂
Integrating once gives
r0 (t) = t ̂ + cos t k̂ + C
for some constant vector C. To satisfy the initial condition that r0 (0) =
v0 = ı̂ı + k̂, we need
ı̂ı + k̂ = r0 (0) = k̂ + C =⇒ C = ı̂ı
So
r0 (t) = ı̂ı + t ̂ + cos t k̂
Integrating a second time, and imposing the initial condition that r(0) = ̂,
gives
t2 t2
r(t) = tı̂ı + ̂ + sin t k̂ + ̂ = tı̂ı + 1 + ̂ + sin t k̂
2 2
(b) The particle has x(t) = π/2 when t = π/2. So
π π2
r(π/2) = ı̂ı + 1 + ̂ + k̂
2 8
(c) The work done is
Z π/2
Work = F(t) · r0 (t) dt
0
Z π/2
= ̂ − sin t k̂ · ı̂ı + t ̂ + cos t k̂ dt
0
Z π/2
= t − sin t cos t dt
0
h t2
1 iπ/2
= cos2 t +
2 2 0
π2 1
= −
8 2
1.6.1.9. ∗. Solution. Here is a sketch of the rectangle R.

y n̂
(0, 1) L3 (3, 1)
L4 n̂
x
R L2
n̂
(0, −1) L1 (3, −1)

n̂
Its boundary consists of four line segments.

• L1 from (0, −1) to (3, −1), with n̂ = −̂
• L2 from (3, −1) to (3, 1), with n̂ = ı̂ı

• L3 from (3, 1) to (0, 1), with n̂ = ̂
• L4 from (0, 1) to (0, −1), with n̂ = −ı̂ı
So
Z Z Z Z Z
F · n̂ ds = F · (−̂) ds+ F · ı̂ı ds+ F · (̂) ds+ F · (−ı̂ı) ds
C L1 L2 L3 L4
y x y ds
Z 3 z }| { Z 1 z}|{ Z 0 z}|{ z }| {
= − (−1) ex dx + (3) y 2 dy + (1) ex (−dx)
0 −1 3
x ds
z}|{ z }| { Z −1
(0) y 2 (−dy)
+
1
= e3 − 1 + 13 − (−1)3 + e3 − 1 + 0

= 2e3
The trickiest part of this computation is getting ds correct on L3 and L4

(remembering that ds is the arc length traveled and so is positive, while
dxR< 0 on L3 and dy < 0 on L4 ). To make a more detailed computation
of L3 F · (̂) ds, parametrize L3 by

r(t) = (3, 1) + t (0, 1) − (3, 1) = 3 − 3t, 1 0≤t≤1
so that r(0) = (3, 1) is the initial point of L3 and r(1) = (0, 1) is the final
point of L3 . Then
ds
r0 (t) = (−3, 0) (t) = |r0 (t)| = 3
dt
and
y(t)ex(t) ds
dt (t)
Z Z 1 ds
Z 1 z }| { z}|{
3−3t
F · ̂ ds = F r(t) · ̂ (t) dt = e 3 dt
L3 0 dt 0
1
= −e3−3t = e3 − 1

0
1.6.1.10. ∗. Solution. (a) Since r(t) = t cos tı̂ı + t sin t ̂ + t2 k̂

r0 (t) = cos t − t sin t ı̂ı + sin t + t cos t ̂ + 2t k̂

ds
q 2 2
= |r0 (t)| = cos t − t sin t + sin t + t cos t + (2t)2
dt p
= 1 + 5t2
r0 (π) = −ı̂ı − π ̂ + 2π k̂
r0 (t) 1
T̂(π) = 0
=√ − ı̂ı − π ̂ + 2π k̂
|r (t)| 1 + 5π 2
(b)
Z p Z π Z π p
2 2
p
2 2
ds
x + y ds = x (t) + y (t) dt = t 1 + 5t2 dt
C 0 dt 0
π
1 1
(1 + 5t2 )3/2 (1 + 5π 2 )3/2 − 1

= =
15 0 15
(c) For every t, the coordinates x(t) = t cos t, y(t) = t sin t, z(t) = t2
obey x(t)2 + y(t)2 = t2 = z(t) and so the 2
curve lies on z = x + y .
2
(d) First concentrate

on x(t) , y(t) . As t runs from 0 to π, the curve
r cos t , r sin t sweeps out half of a circle of radius r. Our x(t) , y(t)
does something similar, but the radius r = t increases from 0 to π. Thus
our x(t) , y(t) sweeps out the beginning of a spiral. At the same time
z(t) increases from 0 to π 2 . So the curve C looks like
z
x
y
R
xρ ds
1.6.1.11. Solution. We use the centre of mass formulae x̄ = RC ,
C
ρ ds
etc. To make the working clearer, we’ll break these calculations into several
steps.
1
x(t) = t + t2 x0 (t) = 1 + t
2
1
y(t) = t − t2 y 0 (t) = 1 − t
2
4 3/2 √
z(t) = t z 0 (t) = 2 t
3
p p
x0 (t)2 + y 0 (t)2 + z 0 (t)2 =
1 + 2t + t2 + 1 − 2t + t2 + 4t
p √
= 2(t2 + 2t + 1) = 2(t + 1)
x(t) + y(t) (t + t2 /2) + (t − t2 /2)
ρ(x(t), y(t), z(t)) = = =t
2 2
√
Z Z 4 √ 23 · 11 2
ρ ds = t 2(t + 1) dt =
C 0 3
2
Z 4 √ √ Z 2 t4
Z
1 3
xρ ds = t + t2 t 2(t + 1) dt = 2 + t3 + t2 dt
C 0 2 0 2 2
√
√
9 6
5
2 2 2 · 103 2
= 2 + 3(25 ) + =
5 3 15
2
Z 4 √ √ Z 2 t4 t3
Z
1
yρ ds = t − t2 t 2(t + 1) dt = 2 − + + t2 dt
C 0 2 0 2 2
√
√
9 6
5
2 2 2 · 23 2
= 2 − + 25 + =−
5 3 15
√ Z 2
4 3/2 √
Z 4
4 2 2 7/2
Z
zρ ds = t t 2(t + 1) dt = t + t5/2 dt
C 0 3 3 0
√ 10 8
10
√
4 2 2 2 2 · 37 2
= + =
3 9 7 7 · 33
√
25 ·103 2
R
xρ ds 15 412
x= R = 3 √ = ≈ 7.5
ρ ds 2 ·11 2 55
3
5
√
− 2 ·23 2
R
yρ ds 15 92
y= R = 3 √ =− ≈ −1.7
ρ ds 2 ·11 2 55
3
√
210 ·37 2
R
zρ ds 3 4736
z= R = 37·3 √ = ≈ 6.8
ρ ds 2 ·11 2 693
3
After these long calculations, it’s nice to do a sanity check. Using

0 ≤ t ≤ 4, we see our wire takes up space in the following intervals:
0 ≤ x ≤ 12, −4 ≤ y ≤ 1/2, and 0 ≤ z ≤ 32/3. The coordinates of
our centre of mass all fall in these intervals, which doesn’t guarantee our
answer is correct, but it is a nice sign. If, say x had been negative, or z
were greater than 11, we would have known there was something wrong.

1.7.4 · Exercises
1.7.4.1. Solution. We don’t have enough information to gauge the size
of the vectors, but we can figure out their direction. Gravity pulls straight
down, so the vector −mg̂ points straight down. The normal force will be
normal to the curve.
W N̂
−mg̂
1.7.4.2. Solution. This equation stems from F = ma. In that equation,

a is acceleration — the second derivative of position with respect to time.
So, v is the derivative of position with respect to time.
We previously used v as the derivative of position with respect to the
parameter we use to define our position — which was often called t, but
was not the necessarily time. So this is a good point to keep straight.
1.7.4.3. Solution 1. For large, negative values of x, the wire is closer
and closer to a vertical line. If the bead were sliding down a vertical
wire, it could do so without even touching the wire, so the force exerted
on the bead would be zero. As x approaches 0 from the left, the wire
approximates a horizontal line. If the bead were sitting on a horizontal
line, the wire would be pushing up to counter gravity. So, we imagine the
magnitude of the force exerted by the wire might increase as x increases.
That is, dW
dx > 0.
Solution 2. The net force exerted on the bead is
F = ma = W N̂ − mg̂
We dot both sides with N̂.
W N̂ · N̂ − mg̂ · N̂ = ma · N̂
d2 s ds 2

Using the equation a(t) = dt2 T̂ +κ dt N̂,
2
ds
W − mg̂ · N̂ = mκ
dt
2
ds
W = mg n̂ · N̂ + mκ
dt
2
ds
= mg cos θ + mκ
dt
where θ is the angle between ̂ and N̂.

As x moves from a highly negative number to zero, θ moves from nearly
π/2 to nearly 0. Therefore cos θ increases from nearly zero to nearly one.
Then mg cos θ is increasing.
Furthermore, as x increases, we see from the picture that the curvature
κ increases, and speed ds
dt increases as well (kinetic energy is increasing as
potential energy decreases).
So, dW
dx > 0.
1.7.4.4. Solution. Equation 1.7.1 defines E = 12 m|v|2 + mgy. The

E
skater reaches their highest point when |v| = 0, so when y = mg . This is
the same equation as a sufficiently large circular culvert: it’s the height
where all the kinetic energy has been converted into potential energy.
That’s why we never even used the equation y = x2 !
1.7.4.5. Solution. The skateboarder starts going back down at yS =

E E kg·m2
mg , so we solve 3 m = 100kg·9.8 sm2 to find E = 2940 s2 = 2940J
Remark: we needed the diameter to be greater than 3m for the skate-
boarder to not be going all the way around the culvert, but choosing r = 5
leads to an answer no different from, say, r = 50.
1.7.4.6. Solution. From the text, the skateboarder will make it all the
E
way around when 52 (5) ≤ mg . Energy E is given by E = 12 m|v|2 + mgy,
the sum of the kinetic and potential energy of the system. At y = 0, all
the energy is kinetic, so E = 12 m|v|2 , where |v| is the skater’s velocity at
the bottom of the culvert.
So, we solve:
1
25 E m|v|2
≤ = 2
2 mg m · 9.8
√
|v| ≥ 5 9.8
√
So, a speed of 5 9.8 m/s or higher is needed. (That’s about 56 kph.)
1.7.4.7. Solution. Equation 1.7.2 tells us the normal force exerted by
the track is W N̂, where W = mκ|v|2 + mg k̂ · N̂. (Note in our problem,
the vertical direction is k̂, not ̂ as in §1.7.1.) So, we ought to find κ and
N.
r(θ) = (3 cos θ, 5 sin θ, 4 + 4 cos θ)
v(θ) = (−3 sin θ, 5 cos θ, −4 sin θ)
ds p
|v(θ)| = = 9 sin2 θ + 25 cos2 θ + 16 sin2 θ = 5
dθ
a(θ) = (−3 cos θ, −5 sin θ, −4 cos θ)
v × a = 5(−4, 0, 3)
|v × a| 25 1
κ(θ) = = 3 =
ds 3 5 5
dθ
2
d s
Since dθ 2 = 0, we use the following theorem to find N̂:
2
d2 s ds
a(θ) = 2 T̂ + κ N̂
dθ dθ
25
(−3 cos θ, −5 sin θ, −4 cos θ) = 0 + N̂
5
3 4
N̂(θ) = − cos θ, − sin θ, − cos θ
5 5
Using the given quantity |v(t)| = 5 at the specified point,

W = mκ|v|2 + mg k̂ · N̂

θ=π/4

1 2 4 39.2
= (1) 5 + 1(9.8) − cos(π/4) = 5 − √
5 5 5 2

39.2 3 4
W N̂ = 5− √ − cos(π/4), − sin(π/4), − cos(π/4)

θ=π/4 5 2 5 5

39.2 3 1 4
= 5− √ − √ , −√ , − √
5 2 5 2 2 5 2
√

3 5
= − √ + 2.352, − √ + 3.92, −2 2 + 3.136
2 2
1.7.4.8. Solution. Equation 1.7.2 tells us the normal force exerted by
the track is W N̂, where W = mκ|v|2 + mg̂ · N̂. So, we need to find κ and
̂ · N̂ at the point θ = 13π3 .
the track, but it is not
Note that θ is the parameter used to describe
time. So |v(θ)| = dr is not the same as |v| = dr , the speed of the bead.
dθ dt
r(θ) = (sin θ, sin θ − θ)

v(θ) = (cos θ, cos θ − 1)
ds p
|v(θ)| = = 2 cos2 θ − 2 cos θ + 1
dθ
a(θ) = (− sin θ, − sin θ)
|v × a| = | sin θ|
|v × a| | sin θ|
κ(θ) = =
ds 3 (2 cos2 θ − 2 cos θ + 1)3/2
dθ
d2 s sin θ(1 − 2 cos θ)
2
=√
dθ 2 cos2 θ − 2 cos θ + 1
d2 s ds 2

Equation 1.3.3 part (c) gives us the relation a(θ) = dθ 2 T̂ + κ dθ N̂. We
use this to find ̂ · N̂ at θ = 13π/3 without differentiating (actually, without
even finding) T̂.
√ √
a(13π/3) = − 3/2, − 3/2
d2 s
(13π/3) = 0
dθ2 √
κ(13π/3) = 6
ds √
(13π/3) = 1/ 2
dθ
2 !
d2 s ds
a(θ) · ̂ = T̂ + κ N̂ · ̂
dθ2 dθ
√
3 √
− = 0 + 6(1/2)N̂ · ̂
2
1
N̂ · ̂ = − √
2
Now we can find the speed |v| of the bead when |W | = 100 and it
breaks off the track.
W = mκ|v|2 + mg̂ · N̂
1 √

9.8 1
±100 = 6|v|2 + −√
9.8 9.8 2
s
9.8 1
|v| = √ 100 + √ ≈ 20m/s ≈ 72kph
6 2
(Because |v| > 0, the equation above has no solution for W = −100.)
Quite fast! 100 N is a lot of force for such a light object.
1.7.4.9. Solution. According to the equation in §1.7.2, the skier will
become airborne when: r
g
|v| > |̂ · N̂|
κ
We’ll use the equation of the curve to find κ and N̂.
Note that g is given in metres per second, while the other quanti-
ties are in kilometres and hours. Converting, 9.8 m/s2 is the same as
9.8 m
1 km 3600 s 2
1 s2 1000 m 1 hr = 98 · 64 km
h2
= 25 · 34 · 72 km
h2
.
r(t) = (ln t, 1 − t)
ds p
r0 (t) = v(t) = (t−1 , −1) = |v(t)| = 1 + t−2
dt
r00 (t) = a(t) = (−t−2 , 0)
|v × a| t−2 |t| t
κ(t) = 3 =√ 3 = 2 )3/2
=
ds (1 + t (1 + t2 )3/2

1+t −2
dt
Note t is positive in the interval in question.

v(t) 1 −1 1 −t
T̂(t) = =√ (t , −1) = √ ,√
|v(t)| 1 + t−2 1 + t2 1 + t2

−t −1 1
T̂0 (t) = 2 3/2
, 2 3/2
|T̂0 (t)| = 2
(1 + t ) (1 + t ) t +1
T̂0 (t)

−t −1
N̂(t) = = √ ,√
|T̂0 (t)| 1 + t2 1 + t2
1
|N̂ · ̂| = √
1 + t2
Now, we have all the pieces we need to find the “escape velocity” of the
ground.
r s r
g g · (1 + t2 )3/2 g(1 + t2 )
|v| = |N̂ · ̂| = =
κ t(1 + t2 )1/2 t
Since the skier can take off anywhere on the

q hill, we just need their velocity
g(1+t ) 2
to be larger than the smallest value of t when 1/e ≤ t ≤ e. To
find that minimum, we find the location of the minimum of the simpler
2
function g(t) = 1+t t . Using first-semester calculus, we find it to occur
q
g(1+t2 )
when t = 1. So, the minimum value of t (that is, smallest speed
to achieve lift-off) occurs at t = 1. We therefore need a minimum speed
greater than:
s
g(1 + t2 ) p √
= 2g = 26 · 34 · 72 = 23 · 32 · 7 = 504 kph
|t|

t=1
(It seems unlikely that one could reach this speed on skis. The skier is
probably earth-bound until they find a curvier hill.)
1.7.4.10. Solution. We now have three forces acting on the bead, rather
than the two in §1.7.1. The wire still exerts a normal force W N̂ on the
bead to keep it on the wire; gravity still exerts a force −mg̂ straight down.
Now our jet-pack force also exerts a force parallel to the direction of the
bead’s motion, i.e. parallel to T̂. This force is U T̂.
W N̂
U T̂
−mg̂
The net force acting on the bead is the sum of these three forces:
F = ma = U T̂ + W N̂ − mg̂
To focus on the force

in the direction of T̂, we dot both sides of the equa-
dy
tion with T̂(s) = dx ,
ds ds . (Recall r(s) was parametrized with respect
dr
to arclength, so T̂(s) = ds everywhere.) Since the speed of the bead is
constant, the tangential component of its acceleration, a · T̂, is 0 (see The-
orem 1.3.3.c).
0 = (U T̂ + W N̂ − mg̂) · T̂
= (U T̂ · T̂) + (W N̂ · T̂) − mg̂ · T̂
dy
= U + 0 − mg
ds
dy
U = mg
ds
1.7.4.11. Solution. (a) There are three forces acting on the snowma-
chine. If it’s not accelerating, then F = ma = 0: that is, the forces all
cancel out.
W N̂
M T̂
−mg̂
So, we have the equation
ma = W N̂ + M T̂ − mg̂
To isolate M , we dot both sides of the equation with T̂. Remember T̂ is a

unit vector, and it is perpendicular to N̂.
ma · T̂ = W N̂ · T̂ + M T̂ · T̂ − mg̂ · T̂
= 0 + M − mg̂ · T̂
Since the speed of the snowmachine is constant, the tangential component

of its acceleration, a · T̂, is 0 (see Theorem 1.3.3.c).
0 = M − mg̂ · T̂
M = mg̂ · T̂
(b) We would expect, from looking at the situation, that the engine
would have to provide a “backwards” force to slow the acceleration due to
gravity. So, we would expect M < 0. Indeed, if T̂ points downhill, then
the y-component of T̂ is negative, so M = mg̂ · T̂ is negative.
(This is the purpose of driving downhill in a low gear: the friction inside
the motor provides a force opposing the direction of motion, slowing the
vehicle.)
(c) To use the equation M = mg̂ · T̂, we’ll need to find ̂ · T̂.
r(x) = (x, 1 + cos x) r0 (x) = (1, − sin x)

p 1
|r0 (x)| = 1 + sin2 x T̂(x) = p (1, − sin x)
1 + sin2 x
r !
2 1
T̂(3π/4) = , −√
3 3
So,
21 1960
M = (200kg)(9.8m/s ) − √ m = − √ N ≈ −1131.6N
3 3
1.7.4.12. Solution. We begin with the usual computations.
r(θ) = (4 cos θ, 3(1 + sin θ))

v(θ) = r0 (θ) = (−4 sin θ, 3 cos θ)
ds p p
|v(θ)| = = 16 sin2 θ + 9 cos2 θ = 9 + 7 sin2 θ
dθ
a(θ) = (−4 cos θ, −3 sin θ)
|v(θ) × a(θ)| = 12
|v(θ) × a(θ)| 12
κ(θ) = =
ds 3 (9 + 7 sin2 θ)3/2

dθ
(−4 sin θ, 3 cos θ)
T̂(θ) = p
9 + 7 sin2 θ
(36 cos θ, 48 sin θ)
T̂0 (θ) =
−(9 cos2 θ + 16 sin2 θ)3/2
12
|T̂0 (θ)| =
9 cos2 θ + 16 sin2 θ
(3 cos θ, 4 sin θ)
N̂(θ) = p
− 9 cos2 θ + 16 sin2 θ
We want to find the height yS where |v| = 0, and the height yA where
W = 0. Remember that v in these equations is the derivative of position
with respect to time, and is not the same as v(θ).
1
Equation 1.7.1 : E= m|v|2 + mgy
2
E
If |v| = 0 : E = mgyS =⇒ yS =
mg
This answers part a.
Equation 1.7.2 : W = 2κ(E − mgy) + mg̂ · N̂

If W = 0 : 0 = 2κ(E − mgyA ) + mg̂ · N̂
!
24 (E − mgyA ) sin θ
= 2 3/2
− mg 4 p
(9 + 7 sin θ) 9 + 7 sin2 θ
Using y = 3 + 3 sin θ:
 
yA −3
24 (E − mgyA ) 3
= − 4mg  q 
3/2
yA −3 2 yA −3 2

9+7 3
9+7 3
So, for part b., we can write (say)

 
yA −3
24 (E − mgyA ) 3
= 4mg  q 
3/2
yA −3 2 yA −3 2

9+7 3
9+7 3
Now, suppose the skater’s speed at the bottom of the culvert (y = 0)

is 11 m/s. Then their energy is E = 12 m(112 ) + 0, or 121m 2 joules, where
E 121
m is their mass. Then yS = mg = 2·9.8 ≈ 6.2. Since the half-way height
of the culvert is at height y = 3, the skater makes it onto the ceiling of the
culvert. Now the question is: did they make it all around, or fall off the
ceiling?
For this, we need to find yA . If they go airborne on the ceiling, they
fall; but if yA > 6, then they never lose contact with the culvert, and they
go all the way around.
 
yA −3
24 (E − mgyA ) 3
= 4mg  q 
3/2
yA −3 2 yA −3 2

9+7 3
9 + 7 3

E
6 mg − yA yA −3
3
⇔ =q
2 3/2 2
9 + 7 yA3−3 9 + 7 yA3−3
2
11
6 2·9.8 − yA yA −3
3
⇔ = q
3/2
yA −3 2
2
9+7 3
9 + 7 yA3−3
3/2
yA −3 2
To simplify to a more standard form, we multiply both sides by 9 + 7 3 :
2 !
112

yA − 3 yA − 3
6 − yA = 9+7
2 · 9.8 3 3
Now, we simplify to
7 3 2 7797
0= y − 7yA + 48yA −
9 A 49
Now, solving for yA involves solving a cubic function, which is no small
task. We could ask a computer, but we can also get an idea of its root(s)
by plugging in numbers and using the intermediate value theorem. In
particular, we need to know whether yA is greater than 6 (the skater makes
it!) or between 3 and 6 (they fall off the ceiling).
Let f (y) = 79 y 3 − 7y 2 + 48y − 7797 12941
49 . Note f (4) = − 441 , which is
1367
negative, and f (5) = 441 , which is positive. So, by the intermediate value
theorem, there is a root of f (y) between y = 4 and y = 5. That is, yA
is between 4 and 5, so the skater falls off the ceiling somewhere between
these heights, rather than making it all the way around.
1.7.4.13. Solution. (a) By Newton’s law of motion

d 1
E 0 (t) = m|v(t)|2 + mgr(t) · k̂ = mv(t) · v0 (t) + mgv(t) · k̂
dt 2

= v(t) · N r(t) − mg k̂ + mgv(t) · k̂
=0

since v(t) · N r(t) = 0. So E(t) is a constant, independent of t.
(b) By part (a),
1 2
E(t) = E(0) =⇒ 2 m|v(t)|
+ mgbθ(t) = mg(2πb)
2

=⇒ |v(t)| = 2gb 2π − θ(t)
(c) We wish to determine the time it takes to go from θ = 2π to θ = 0.

We’ll first determine dθ
dt .
dr dr dθ dθ
v= = = − a sin θ, a cos θ, b
dt dθ dt dt
2
dθ
=⇒ |v|2 = a2 + b2 ]

dt
2
1/2 1/2
dθ |v| 2gb(2π − θ)
=⇒ =− 2 =−
dt a + b2 a2 + b2
We have chosen the negative sign because θ must decrease from 2π to 0.

The time required to do so is

0 2 1/2 Z 0
a + b2
Z Z
dt 1
dt = dθ = − 1/2
dθ
2π dθ 2gb 2π (2π − θ)
2 1/2 Z 2π
a + b2 1
= dθ
2gb 0 (2π − θ)1/2
2 1/2 h 1/2
a + b2
2
1/2
i2π a + b2
= −2(2π − θ) =2 π
2gb 0 gb

1.8.1 · Exercises
1.8.1.1. Solution. The upper sketch below contains the points, (x1 , y1 ),
(x3 , y3 ), (x5 , y5 ), that are on the axes. The lower sketch below contains
the points, (x2 , y2 ), (x4 , y4 ), that are not on the axes.
y
(0, 1)
π
π 2
(−2, 0) (3, 0)
x
y
(−1, 1) (1, 1)
√
2
3π
4 π
4
x
Recall that the polar coordinates r, θ are related
p to the cartesian coor-
dinates x, y, by x = r cos θ, y = r sin θ. So r = x2 + y 2 and tan θ = xy
(assuming that x 6= 0) and
(x1 , y1 ) = (3, 0) =⇒ r1 = 3, tan θ1 = 0

=⇒ θ1 = 0 as (x1 , y1 ) is on the positive x-axis
√
(x2 , y2 ) = (1, 1) =⇒ r2 = 2, tan θ2 = 1
π
=⇒ θ2 = as (x2 , y2 ) is in the first octant
4
(x3 , y3 ) = (0, 1) =⇒ r3 = 1, cos θ3 = 0
π
=⇒ θ3 = as (x3 , y3 ) is on the positive y-axis
2
√
(x4 , y4 ) = (−1, 1) =⇒ r4 = 2, tan θ4 = −1
3π
=⇒ θ4 = as (x4 , y4 ) is in the third octant
4
(x5 , y5 ) = (−2, 0) =⇒ r5 = 2, tan θ5 = 0
=⇒ θ5 = π as (x5 , y5 ) is on the negative x-axis

1.8.1.2. Solution. Note that the distance from the point r cos θ , r sin θ
to the origin is p √
r2 cos2 θ + r2 sin2 θ = r2 = |r|
Thus r can be either the distance to the origin or minus the distance to
the origin.
(a) The distance from (−2, 0) to the origin is 2. So either r = 2 or

r = −2.
• If r = 2, then θ must obey

(−2, 0) = 2 cos θ , 2 sin θ ⇐⇒ sin θ = 0, cos θ = −1
⇐⇒ θ = nπ, n integer , cos θ = −1
⇐⇒ θ = nπ, n odd integer
• If r = −2, then θ must obey

(−2, 0) = − 2 cos θ , −2 sin θ ⇐⇒ sin θ = 0, cos θ = 1
⇐⇒ θ = nπ, n integer , cos θ = 1
⇐⇒ θ = nπ, n even integer
In the figure on the left below, the blue half-line is the set of all points with
polar coordinates θ = π, r > 0 and the pink half-line is the set of all points
with polar coordinates θ = π, r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = 0, r > 0
and the pink half-line is the set of all points with polar coordinates θ = 0,
r < 0.
y y
(−2, 0) π (−2, 0)
x x
√ √
(b)√The distance from (1, 1) to the origin is 2. So either r = 2 or
r = − 2.
√
√ √ 1
(1, 1) = 2 cos θ , 2 sin θ ⇐⇒ sin θ = cos θ = √
2
π
⇐⇒ θ = + 2nπ, n integer
4
√
• If r = − 2, then θ must obey
√ √ 1
(1, 1) = − 2 cos θ , − 2 sin θ ⇐⇒ sin θ = cos θ = − √
2
5π
4
polar coordinates θ = π4 , r > 0 and the pink half-line is the set of all points
with polar coordinates θ = π4 , r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = 5π 4 , r > 0
and the pink half-line is the set of all points with polar coordinates θ = 5π 4 ,
r < 0.
y y
√ (1, 1) √ (1, 1)
2 2
5π
π 4
4
x x
√ √
√ distance from (−1, −1) to the origin is
(c) The 2. So either r = 2
or r = − 2.
√
√ √ 1
(−1, −1) = 2 cos θ , 2 sin θ ⇐⇒ sin θ = cos θ = − √
2
5π
4
√
• If r = − 2, then θ must obey
√ √ 1
(−1, −1) = − 2 cos θ , − 2 sin θ ⇐⇒ sin θ = cos θ = √
2
π
4
polar coordinates θ = 5π 4 , r > 0 and the pink half-line is the set of all points
with polar coordinates θ = 5π 4 , r < 0. In the figure on the right below, the
blue half-line is the set of all points with polar coordinates θ = π4 , r > 0
and the pink half-line is the set of all points with polar coordinates θ = π4 ,
r < 0.
y y
5π
4 π
4
√ x √ x
2 2
(−1, −1) (−1, −1)
1.8.1.3. Solution. (a) The lengths are

p
|êr (θ)| = cos2 θ + sin2 θ = 1
p
|êθ (θ)| = (− sin θ)2 + cos2 θ = 1
As
êr (θ) · êθ (θ) = (cos θ)(− sin θ) + (sin θ)(cos θ) = 0
π
the two vectors are perpendicular and the angle between them is 2. The
cross product is
 
ı̂ı ̂ k̂
êr (θ) × êθ (θ) = det  cos θ sin θ 0  = k̂
− sin θ cos θ 0
(b) Note that for θ determined by x = r cos θ, y = r sin θ,

• the vector êr (θ) is a unit vector in the same direction as the vector
from (0, 0) to (x, y) and
• the vector êθ (θ) is a unit vector that is perpendicular to êr (θ).
• The y-component of êθ (θ) has the same sign as the x-component of
êr (θ). The x-component of êθ (θ) has opposite sign to that of the
y-component of êr (θ).
Here is a sketch of (xi , yi ), êr (θi ), êθ (θi ) for i = 1, 3, 5 (the points on the
axes)
y
er ( π2 )
eθ ( π2 ) eθ (0)
(0, 1)
er (π) (−2, 0) er (0)

(3, 0) x
eθ (π)
and here is a sketch (to a different scale) of (xi , yi ), êr (θi ), êθ (θi ) for
i = 2, 4 (the points off the axes).
er ( 3π ) y eθ ( π ) er ( π4 )
4 4
(−1, 1) (1, 1)
3π
eθ ( 3π
4
) 4 π
4
x
1.8.1.4. ∗. Solution. (a) Since −1 ≤ sin(4θ) ≤ 1, the coordinate
r = 2 + sin(4θ) oscillates between r = 1 and r = 3 as θ runs from 0 to
2π. The maximum value r = 3 is achieved when sin(4θ) = 1, i.e when
4θ = π2 + 2nπ, i.e. when θ = π8 + nπ 2 . That matches figure (E).
(b) Since −1 ≤ sin(4θ) ≤ 1, the coordinate r = 1 + 2 sin(4θ) takes its
maximum value r = 3 when sin(4θ) = 1, i.e. when θ = π8 + nπ 2 , just as the
case with (a). But now r can also take the value 0. That matches figure
(B).
(c) r = 1 is completely indepedent of θ. All points on the curve r = 1
are a distance 1 from the origin. That is, r = 1 is the circle of radius 1
centred on the origin. That’s figure (F).
(d) In this case, θ is subject to the restriction − π2 ≤ θ ≤ π2 , like figure
(C). Figure (C) looks like a circle. We can verify that r = 2 cos(θ) is indeed
a circle by converting to Cartesian coordinates. We can convert the right
hand side to exactly 2x = 2r cos(θ) by multiplying the whole equation by
r.
r = 2 cos(θ) ⇐⇒ r2 = 2r cos(θ) ⇐⇒ x2 + y 2 = 2x
⇐⇒ (x − 1)2 + y 2 = 1
So r = 2 cos(θ) is the circle of radius 1 centred on x = 1, y = 0, which

indeed matches figure (C).
(e) When θ = 0, r = eθ/10 + e−θ/10 = 2. As
d θ/10 1 θ/10
+ e−θ/10 = − e−θ/10 > 0

e e for all θ > 0
dθ 10
r = eθ/10 + e−θ/10 increases as θ increases for all θ ≥ 0. Furthermore the
rate of increase gets bigger and bigger as θ gets bigger and bigger. So r
starts at r = 2 when θ = 0 and increases faster and faster as θ increases.
That matches figure (A).
(f) When θ = 0, r = θ = 0. As
d
θ=1 for all θ
dθ
r = θ increases as θ increases for all θ ≥ 0. Furthermore the rate of increase
is independent of θ. So r starts at r = 0 when θ = 0 and increases at a
constant rate as θ increases. That matches figure (D).
1.8.1.5. Solution. Think of θ as a time parameter and recall that κ(θ) =

|v(θ)×a(θ)|
|v(θ)|3 . The given curve has
x(θ) = f (θ) cos θ

y(θ) = f (θ) sin θ

r(θ) = f (θ) cos θ ı̂ı + sin θ ̂
v(θ) = r0 (θ) = f 0 (θ) cos θ ı̂ı + sin θ ̂ + f (θ) − sin θ ı̂ı + cos θ ̂

a(θ) = r00 (θ) = f 00 (θ)−f (θ) cos θ ı̂ı +sin θ ̂ +2f 0 (θ) − sin θ ı̂ı +cos θ ̂

The efficient way to compute |v(θ)| and the cross product v(θ) × a(θ) is to
observe that
v(θ) = f 0 (θ) êr (θ) + f (θ) êθ (θ)
a(θ) = f 00 (θ) − f (θ) êr (θ) + 2f 0 (θ) êθ (θ)

where êr (θ) and êθ (θ) are the vectors of Q[1.8.1.3]. As êr (θ) and êθ (θ)
are mutually perpendicular unit vectors obeying êr (θ) × êθ (θ) = k̂ and
êr (θ) × êr (θ) = êθ (θ) × êθ (θ) = 0,
|v(θ)|2 = v(θ)·v(θ) = f 0 (θ) êr (θ)+f (θ) êθ (θ) · f 0 (θ) êr (θ)+f (θ) êθ (θ)

= f 0 (θ)2 êr (θ) · êr (θ)+f (θ)2 êθ (θ) · êθ (θ)+2f 0 (θ) f (θ) êr (θ) · êθ (θ)
= f 0 (θ)2 + f (θ)2
p
|v(θ)| = f 0 (θ)2 + f (θ)2
v(θ)×a(θ) = f 0 (θ) êr (θ)+f (θ) êθ (θ) × f 00 (θ)−f (θ) êr (θ)+2f 0 (θ) êθ (θ)

= 2f 0 (θ)2 êr (θ) × êθ (θ) + f (θ)[f 00 (θ) − f (θ)] êθ (θ) × êr (θ)
= 2f 0 (θ)2 − f (θ)[f 00 (θ) − f (θ)] k̂

So

|v(θ) × a(θ)| f (θ)2 + 2f 0 (θ)2 − f (θ)f 00 (θ)
κ(θ) = =
|v(θ)|3 [f (θ)2 + f 0 (θ)2 ]
3/2
1.8.1.6. Solution. By the Q[1.8.1.5] with
f (θ) = a(1 − cos θ) f 0 (θ) = a sin θ f 00 (θ) = a cos θ
we have

f (θ)2 + 2f 0 (θ)2 − f (θ)f 00 (θ)
κ(θ) =
[f (θ)2 + f 0 (θ)2 ]3/2
2
a − 2a2 cos θ + a2 cos2 θ + 2a2 sin2 θ − a2 cos θ + a2 cos2 θ
=
[a2 − 2a2 cos θ + a2 cos2 θ + a2 sin2 θ]3/2
3a − 3a2 cos θ
2
3 3
= 2 2 3/2
= 3/2 √ = p
[2a − 2a cos θ] 2 a 1 − cos θ 2 2ar(θ)

1.9.1 · Exercises
1.9.1.1. ∗. Solution. (a) Both (i) and (ii) were proven in §1.9. Here
are those arguments.
Define
Ω (t) = r(t) × v(t)
By the product rule,
Ω
dΩ d
(t) = r(t) × v(t) = v(t) × v(t) + r(t) × a(t)
dt dt
= mr(t) × f r(t))r(t) = 0
So Ω (t) is in fact independent of t. It is a constant vector that we’ll just

denote Ω .
As r(t) × v(t) = Ω , we have that r(t) is always perpendicular to Ω and
r(t) · Ω = 0
• If Ω 6= 0, this is exactly the statement that r(t) always lies in the

plane through the origin with normal vector Ω .
• If Ω = 0, then r(t) is always parallel to v(t) and there is some function
α(t) such that
dr
(t) = v(t) = α(t) r(t)
dt
This is a first order, linear, ordinary differential equation that we can
solve by using an integrating factor. Set
Z t
β(t) = α(t) dt
0
Then
dr dr
(t) = α(t) r(t) ⇐⇒ e−β(t) (t) − α(t)e−β(t) r(t) = 0
dt dt
d −β(t)
⇐⇒ e r(t) = 0
dt
⇐⇒ e−β(t) r(t) = r(0)
⇐⇒ r(t) = eβ(t) r(0)
so that r(t) lies on a line through the origin. This makes sense — the
particle is always moving parallel to its radius vector.
This completes the verification that r(t) lies in a plane.

Now we show that the radius vector r(t) sweeps out equal areas in equal
times. In other words, we now verify that the rate at which r(t) sweeps
out area is independent of time. To do so we rewrite the statement that
|r(t) × v(t) is constant in polar coordinates. Writing r(t) = r(t)r̂ θ(t)
and then applying Lemma 1.8.2.b gives that
dr dθ dθ
θ̂θ = r2

constant = r × v = rr̂ × r̂ + r
dt dt dt
since |r̂ × r̂| = 0, |r̂ × θ̂θ | = 1
is constant. It now suffices to observe that r(t)2 dθ

dt (t) is exactly twice the
rate at which r(t) sweeps out area. To see this, just look at the figure
below. The shaded area is essentially a wedge of a circular disk of radius
r. (If r(t) were independent of t, it would be exactly a wedge of a circular
dθ
disk.) Its area is the fraction 2π of the area of the full disk, which is
dθ 1 rpt ` dtq
πr2 = r2 dθ
2π 2 dθ
rptq
(b) If f (r) is identically zero, then r00 (t) = 0, so that r0 (t) is a constant,
say v0 , and r(t) = r0 + v0 t, for some constant r0 . That’s a straight line.
We’ll now show that if the motion of the particle always lies on a straight
line, then f (r) must be identically zero. Suppose that r(t) is a straight line.
Then there are constant vectors r0 and T̂ such that r(t) = r0 + g(t)T̂ for
some scalar valued function g(t). Then
r00 (t) = f r(t) r(t)

becomes
g 00 (t)T̂ = f r(t) r0 + g(t)T̂ = f r(t) r0 + f r(t) g(t)T̂

We may always choose the initial conditions so that, for example, r0 = ı̂ı
and T̂ = ̂. So
g 00 (t)̂ = f r(t) ı̂ı + f r(t) g(t)̂

Taking the dot product of both sides with ı̂ı gives f r(t) = 0, as desired.
(c) We saw in §1.10 that the gravitational force − GM m
r 3 r can produce
elliptical orbits. So any f (r) which is a positive constant times − r13 does
the job.
1.9.1.2. ∗. Solution. (a) Our object is subject to a central force. So
the acceleration a(t) is parallel to r(t) and
d
r(t) × v(t) = v(t) × v(t) + r(t) × a(t) = 0 + 0 = 0
dt
By Lemma 1.8.2, v(t) = dr r̂ θ(t) + r(t) dθ

dt (t)
θ
dt (t) θ̂ θ(t) . Because r(t) =

r(t)r̂ θ(t) is parallel to r̂ θ(t) and is perpendicular to θ̂θ θ(t) ,
r(t) × v(t) = r(t)2 θ̇(t) r̂ θ(t) × θ̂θ θ(t)

and, in particular,
|r(t) × v(t)| = r(t)2 |θ̇(t)|
is a constant. As θ̇(t) is continuous, r(t)2 θ̇(t) is also constant.
(b) By Lemma 1.8.2, the acceleration
d2 r dθ 2
a(t) = 2
(t) − r(t) (t) r̂ θ(t)
dt dt
d2 θ dr dθ
+ r(t) 2 (t) + 2 (t) (t) θ̂θ θ(t)
dt dt dt
Because our object is subject to a central force, the acceleration a(t) is

parallel to r̂ θ(t) . So the θ̂θ θ(t) component of the acceleration is zero
and d2 r dθ 2
a(t) = 2
(t) − r(t) (t) r̂ θ(t)
dt dt
so that d2 r dθ 2
|a(t)| = 2 (t) − r(t) (t)

dt dt
1
Since r(t) = θ(t)+α
1
ṙ(t) = − θ̇(t) = −r(t)2 θ̇(t) = −h
[θ(t) + α]2
d2 r
So dt2 (t) = 0 and
r(t)4 θ̇(t)2 h2
|a(t)| = r(t)θ̇(t)2 = 3
=
r(t) r(t)3
2 · Vector Fields
2.1.1 · Exercises
2.1.1.1. Solution. The vectors are pointing to the right when x > 0, to
the left when x < 0, and are vertical when x = 0. So, at least for (x, y)
shown in the sketch,

> 0
 when x > 0
v(x, y) · ı̂ı = 0 when x = 0


<0 when x < 0
The behaviour of the y-values is more complicated. Vectors in one

vertical line seem to be all pointing up, or all pointing down. So, the sign
of v·̂ depends only on x, not on y (although the magnitude of v·̂ depends
on both). Roughly, the vectors are pointing
• Down when x < −2;
• horizontally when x = −2 (remember the vector is positioned with

the base of v(x, y) at (x, y);
• up when −2 < x < 2;
• horizontally when x = 2;
• up when 2 < x.
Since we’re assuming there’s nothing surprising happening between the

samples pictured, at least for (x, y) shown in the sketch,

> 0 when −2 < x < 2


v(x, y) · ̂ =0 when x ∈ {−2, 2}


< 0 when x < −2 or x > 2
2.1.1.2. Solution. To start out, we find the places where v(x, y) · ı̂ı = 0
(vertical vectors) or v(x, y) · ̂ = 0 (horizontal vectors). Remember the
vector v(x, y) has its tail at (x, y).
We see the vertical vectors (those with v(x, y) · ı̂ı = 0) occur at every
point along the line y = −x, while horizontal vectors (those with v(x, y)·̂ =
0) occur at every point along the line y = x.
Indeed, below the line y = −x, vectors point to the left, while above the
line y = −x they point to the right. Similarly, vectors point down when
they’re above the line y = x, and the point up when they’re below the line
y = x.
y
RIGHT
LEFT
y
DOWN
UP
So, at least for (x, y) shown in the sketch,

 
> 0 when y > −x > 0 when y < x

 

v(x, y) · ı̂ı =0 when y = −x and v(x, y) · ̂ = 0 when y = x

 

< 0 when y < −x < 0 when y > x
2.1.1.3. Solution. Since all conveyors point towards the origin, the
(−x,−y)
direction of motion of an object at location (x, y) is √ 2 2
. Its magnitude
x +y
is |y|, so v(x, y) = √−|y|
2 2
(x, y).
x +y
2.1.1.4. Solution. The arrows near the point A are pointing to the right,
indicating that P > 0, and upward, indicating that Q > 0. Moving from
left to right near A, the vertical component of the arrows is decreasing,
indicating that ∂Q
∂x < 0. Moving vertically upwards near A, the vertical
component of the arrows is increasing, indicating that ∂Q
∂y > 0.
2.1.1.5. Solution. (a) At time 0 the velocity of the twig is v(1, 1) = ı̂ı +̂.
So at time t = 0.01, the position of the twig is approximately
(1, 1) + 0.01(1, 1) = (1.01 , 1.01)
(b) At time 0 the velocity of the twig is v(0, 0) = 0. So at time t = 0.01,

the position of the twig is
(0, 0) + 0.01(0, 0) = (0 , 0)
(c) At time 0 the velocity of the twig is v(0, 0) = 0. So it is stationary

and its velocity remains zero for all time. The position of the twig at time
10, and in fact at all times, is (0 , 0).
2.1.1.6. Solution. The velocity of the fluid at all points of the y-axis
is −̂. So the twig will remain on the y-axis and will consequently have
velocity −̂ for all time. The position of the twig at time 10 will be
(0, 0) + 10(0, −1) = (0 , −10)
2.1.1.7. Solution. Since all conveyors point towards the origin, the
(−x,−y)
direction of motion of an object at location (x, y) is √ 2 2
. Its magnitude
x +y
is y, so v(x, y) = √ −y
2 2
(x, y).
x +y
2.1.1.8. Solution. Set your face to be at the origin of our p coordinate

system, (0, 0, 0). A bee at position (x, y, z) is a distance of x2 + y 2 + z 2
from your face, heading in the direction (−x, −y, −z). So, the unit vector
indicating the direction of one friendly bee is √ 2 −12 2 (x, y, z). Now all
x +y +z
we need to find is the length of this vector, i.e. the speed of the friendly
bee. p
The speed of the friendly bee is inversely proportional to x2 + y 2 + z 2 ,
its distance from your face. (Bees that are farther away are buzzing towards
you more excitedly.) So, speed is given by √ 2 α 2 2 for some constant α.
x +y +z
−1
The bee velocity has the direction of the unit vector √ (x, y, z)
x2 +y 2 +z 2
α
with length √ for some positive constant α. That is,
x2 +y 2 +z 2
α
v(x, y, z) = − (x, y, z)
x2 + y2 + z2
2.1.1.9. Solution. Beginning as in Example 2.1.4, we note
(
2 > 0 x 6= 0
v(x, y) · ı̂ı = x
=0 x=0
and 
> 0
 y>0
v(x, y) · ̂ = y =0 y=0.


<0 y<0
That leads to the following picture:
y
This gives us a general idea to start with. Refining, we notice that when
x2 > |y|, then the vector v(x, y) will be more horizontal than vertical. As
we move away from the y-axis in a horizontal line, the difference between
x2 and |y| grows, so the vectors get more and more horizontal. However,
for a fixed value of x, vectors farther from the axis will be more vertical
than vectors closer to it.
2.1.1.10. Solution. Although ultimately we’ll sketch only unit-length

vectors, we can still find the direction of v(x, y) by finding its x- and y
components.
Note v(x, y)·ı̂ı is the distance from (x, y) to the origin, while v(x, y)·̂ is
the distance from (x, y) to the point (1, 1). Both these numbers are always
nonnegative. This leads to the following sketch:
y
When (x, y) is far from the origin, its distance from (0, 0) is almost the
same as its distance from (1, 0). So, we expect v(x, y) to be approximately
a scalar multiple of (1, 1).
At (0, 0), v(0, 0)·ı̂ı = 0, so our vector is horizontal; similarly, v(1, 1)·̂ = 0
so this vector is horizontal. Vectors very near to (0, 0) are nearly horizontal,
while vectors near to (1, 1) are nearly vertical.
For the direction field, we normalize our vectors to have unit length.
y
2.1.1.11. Solution. The sign of v(x, y) · ı̂ı = x(x + y) depends on the

signs of x and x + y. When they have the same signs, v(x, y) ·ı̂ı is positive,
so v(x, y) points to the right; when they have different signs, v(x, y) points
to the left.
y = −x
Similarly, the sign of v(x, y) · ̂ = y(y − x) depends on the signs of y
and y − x.
y=x
All together:
y
y=x
↑ ↑
← →
↑
→ →
↓
x
↑
→ →
↓
↑ ↑
→ ←
y = −x
Refining, we notice that as we move straight up or down, |v(x, y) ·ı̂ı| has
its minimum along the lines y = −x and x = 0. So, the vectors become
more strongly vertical as we approach y = −x and x = 0 from above or
below.
Similarly, |v(x, y) · ̂| has its minima along the lines y = x and y = 0,
so the vectors become more strongly horizontal as we approach y = x
horizontally.
2.1.1.12. Solution. The field v(x, y) is the sum, scaled by 1/3, of the
unit vector pointing away from the origin and the unit vector pointing away
from (1, 0). This tells us about a few regions:
• Along the x axis between (0, 0) and (1, 0), the vectors away from
these points are pointing in opposite directions (and have the same
length), so they cancel each other out. That is, v(x, 0) = 0 for all
x ∈ (0, 1).
• v(0, 0) and v(1, 0) are not defined.

• Along the x-axis outside of [0, 1], the vector pointing away from the
point (0, 0) is the same as the vector pointing away from the point
(1, 0). So, v(x, 0) = (−2/3, 0) for x < 0 and v(x, 0) = (2/3, 0) for
x > 1.
y
• As the distance from (x, y) to the origin grows, the vector pointing
away from (0, 0) looks more and more like the vector pointing away
from (1, 0). So, our vectors far away from the origin look like vectors
of length about 2/3, pointing away from the origin.
2.1.1.13. Solution. (a) The vector field v(x, y) = xı̂ı + y ̂ is the same
as the radius vector. It points radially outward and has length growing
linearly with the distance from the origin.
y
(b) The vertical component of v(x, y) = 2xı̂ı − ̂ is always −1. Its
horizontal component is 2x, so that
• v(x, y) is rightward pointing when x > 0 and leftward pointing when
x < 0, and
• the magnitude of the horizontal component grows linearly with the
distance from the y-axis.
It is sketched in the figure on the left below.
y y
x x
(c) For every (x, y) the vector v(x, y) = √y ı̂ı−x

2
̂
2
x +y
• is of length 1 and
• is perpendicular to the radius vector xı̂ı + y ̂.

• v(x, y) is rightward pointing when y > 0 and leftward pointing when
y < 0, and
• v(x, y) is downward pointing when x > 0 and upward pointing when
x < 0.
It is sketched in the figure on the right above.

2.1.1.14. Solution.
p A particle of unit mass at position (x, y) has dis-
tance D1 = x2 + y 2 from the 5kg mass, so that mass exerts a force of
magnitude xG(5)
2 +y 2 on the particle. This force has direction (−x, −y). So,
−5G
the force exerted by the 5kg mass is f1 (x, y) = (x2 +y 2 )3/2 (x, y).
3G
Similarly, the 3 kg mass at (2, 3) exerts a force of f2 (x, y) = ((x−2)2 +(y−3)2 )3/2
(2−
7G
x, 3−y); and the 7 kg mass at (4, 0) exerts a force of f3 (x, y) = ((x−4)2 +y 2 )3/2
(4−
x, −y).
The net force on a unit mass is therefore
f (x, y) = f1 (x, y) + f2 (x, y) + f3 (x, y)

−5G(x, y) 3G(2 − x, 3 − y) 7G(4 − x, −y)
= 2 2 3/2
+ 2 2 3/2
+
(x + y ) ((x − 2) + (y − 3) ) ((x − 4)2 + y 2 )3/2
2.1.1.15. Solution.
a. Consider a point P on the pole that is a distance p away from the
bottom end. Use this point to make a smaller right triangle, as in
the picture below.
y
H P p
h
x
b
√
4 − H2
Using similar triangles:

p pp
h= H b= 4 − H2
2 2
If P is at position (x, y), then:
p p p p
y=h= H x=4 − H2 − b = 1 − 4 − H2
2 2
dy p dH p dx p −H dH
= =− = 1− √
dt 2 dt 4 dt 2 4 − H 2 dt
p H
= 1− √
2 2 4 − H2
When H = 1:

dy p dx p 1
=− = 1 − √
dt H=1
4 dt H=1 2 2 3
Therefore,

dx dy p 1 p
v(p) = , = 1− √ ,−
dt dt H=1 2 2 3 4
For our model, we set the domain of this function to be [0, 2].
b. Let’s start by seeing what we can salvage from our work on part a.
As in part a., consider a point P on one of the poles, p metres from
the bottom end.
(0, 0, H)
2
p
P = (x, y, z)
p
x2 + y 2
√
4 − H2
Let P have position (x, y, z). Noting that dH dt is now positive, not
negative, if we stick to this two-dimensional slice,

p −1 p
V(p) = 1− √ ,
2 2 3 4
where the second coordinate is z and the first coordinate refers to

the (horizontal) line in the direction of the vector (x, y, 0).
(0, 0, H)
(x, y, z)
(0, 0, 0)
p
(x, y, 0)
c · (x, y, 0)

dz p dx dy
So, we know = , and we know , = (x, y)c
dt H=1 4 dt dt H=1
for some negative constant c with |(x, y)c| = 1 − p2 2√ 1
3
. Since we
have the direction and the magnitude of the vector, we can find the
vector:
1 − p2

dx dy
, = (x, y)c = − √ p (x, y)
dt dt H=1 2 3 x2 + y 2
We want our equation to be in terms of√ x, y, and

√ z, so we need to get
p 4−H 2 − x2 +y 2
rid of p. Using similar triangles, 2 = √
4−H 2
. When H = 1,
√ 2 2
x +y
then 1 − p2 = √3 . So:

dx dy 1
, = − (x, y)
dt dt H=1
6
Finally:

dx dy dz 1 1 1
V(x, y, z) = , , = − x, − y, z
dt dt dt H=1 6 6 2
Not all values of (x, y, z) are on the frame. But, for those values of
(x, y, z) that are on the frame, this equation holds.

2.2.2 · Exercises
2.2.2.1. Solution. (a) At every point of the positive y-axis, the velocity
vector v(0, y) points straight down. So a rubber ducky placed in the water
at (0, 2) just floats straight down the positive y-axis towards the origin.
y
3
x
1 2 3
(b) At every point of the positive x-axis, the velocity vector v(x, 0)
points straight to the right. So a rubber ducky placed in the water at (1, 0)
just floats rightward along the positive x-axis.
(c) At every point of the first quadrant away from the axes, the velocity
vector v(x, y) points downwards and towards the right. So a rubber ducky
placed in the water at (1, 2) always floats down and to the right. The closer
the ducky gets to the x--axis the more rightward its motion becomes.
2.2.2.2. Solution. The derivatives
x0 (t) = −e−t cos t − e−t sin t = −x(t) − y(t)

y 0 (t) = −e−t sin t + e−t cos t = −y(t) + x(t)

So x(t), y(t) is a solution of the system of differential equations
dx
= v1 (x, y) = −x − y
dt
dy
= v2 (x, y) = x − y
dt

So the vector field is v(x, y) = v1 (x, y) , v2 (x, y) = (−x − y , x − y).
2.2.2.3. ∗. Solution. (a) The field lines of F(x, y) = ∇ f = y ı̂ı + x ̂

obey
dx dy x2 y2
= ⇐⇒ x dx = y dy ⇐⇒ = +C
y x 2 2
for any constant C.
(b) The sign data
 
> 0
 if y > 0 

ı̂ı · F(x, y) = y = 0 if y = 0

 

<0 if y < 0
 
> 0
 if x > 0 

̂ · F(x, y) = x = 0 if x = 0

 

<0 if x < 0
is visually displayed in the figure on the left below. The arrows in the figure
2 2
on the left gives us the direction of motion along the field lines x2 = y2 + C
(in red) in the figure on the right below. Some equipotential curves xy = C
are also sketched (in blue) in the figure on the right below.
y
2.2.2.4. ∗. Solution. The field lines obey

dx dy dz
= 2
= y if x, y 6= 0
2y x/y e
In particular
dx y 2 dy 1 1
= =⇒ x dx = 2y 3 dy =⇒ x2 = y 4 + C
2y x 2 2
Since y = 1 when x = 1, C = 0. So x = y 2 and

dy dz
= y =⇒ ey dy = dz =⇒ z = ey + D
x/y 2 e
Since z = e when y = 1, D = 0. So the field line is
x = y2 z = ey
2.2.2.5. ∗. Solution. The field lines obey
dx dy
= if x, y 6= 0
x 3y
=⇒ 3 ln |x| = ln |y| + C
=⇒ |x|3 = eC |y|
=⇒ y = ±e−C x3
=⇒ y = C 0 x3
with C 0 a nonzero constant. x = 0 and y = 0 are also field lines, since on

the y-axis F k ̂ and on the x-axis F k ı̂ı.

2.3.1 · Exercises
2.3.1.1. Solution. False, in general.
In the context of Equation 1.7.1, the only forces acting on the particle
are gravity, −mg̂, and the normal force, W N̂.
We make no such constraints on the force in Example 2.3.3. Certainly
F could arise from gravity and the normal force of a track, but there’s
nothing saying it has to. For example, suppose ϕ is an equation that does
not depend on m and/or g. Alternately, suppose the y-coordinate of our
three-dimensional system is not “up.”
2.3.1.2. Solution. Remember that the screening test can only rule out
conservativity — it can never, by itself, guarantee conservativity. So, A is
never the case.
a.
F = xı̂ı + z̂ + y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (0 − 0)k̂ = 0
This field passes the screening test. That means the screening test doesn’t
rule out the possibility of F being conservative. So, we have option C.
b.
F = y 2 zı̂ı + x2 z̂ + x2 y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (x2 − x2 )ı̂ı + (y 2 − 2xy)̂ + (2xz − 2yz)k̂ 6= 0
So, F fails the screening test — it’s not conservative. That’s option B.
c.

1
F = (yexy + 1)ı̂ı + (xexy + z)̂ + + y k̂
z
∂F ∂F ∂F ∂F ∂F ∂F1
3 2 1 3 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (exy (xy + 1) − exy (xy + 1))k̂ = 0
F passes the screening test, so it may or may not be conservative. That is

Option C.
d.
F = y cos(xy)ı̂ı + x sin(xy)̂
∂F2
= xy cos(xy) + sin(xy)
∂x
∂F1
= −xy sin(xy) + cos(xy)
∂y
∂F2 ∂F1
6=
∂x ∂y
F fails the screening test, so it is not conservative. That is Option B.
2.3.1.3. Solution. Let ϕ be a potential for F. Define φ = ϕ+ax+by+cz.

Then ∇φ = ∇ϕ+(a, b, c) = F+(a, b, c). So, F+(a, b, c) is also conservative.
2.3.1.4. Solution.
a If F+G is conservative for any particular F and G, then by definition,
there exists a potential ϕ with F + G = ∇ ϕ.
Since F is conservative, there also exists a potential ψ with F = ∇ ψ.
But now G = (F + G) − F = ∇ ϕ − ∇ ψ = ∇ (ϕ − ψ). That means
the function (ϕ − ψ) is a potential for G. However, this is impossible:
since G is non-conservative, no function with this property exists.
So it is not possible that F + G is conservative. It must be non-
conservative.
b Counterexample: if F = −G, then F + G = 0 = ∇ c for any constant
c.
c Since both fields are conservative, they both have potentials, say
F = ∇ ϕ and G = ∇ ψ. Then F + G = ∇ ϕ + ∇ ψ = ∇ (ϕ + ψ). That
is, (ϕ + ψ) is a potential for F + G, so F + G is conservative.
2.3.1.5. ∗. Solution. Set ϕ(x, y) = arctan xy (using the standard arctan

that takes values between − π2 and π2 ). Note that ϕ(x, y) is well-defined,
with all partial derivatives continuous, on D since x > 1 there. Then
∂ϕ − xy2 y
(x, y) = =− 2
y 2
∂x 1+ x x + y2
1
∂ϕ x x
(x, y) = 2 =
∂y 1 + xy x2 + y 2
so that F = ∇ ϕ.
2.3.1.6. Solution. If ϕ is a potential for F, then:
∂ϕ
• ∂x = x + y, so ϕ = 12 x2 + xy + ψ1 (y)
∂ϕ
• ∂y = x − y, so ϕ = xy − 12 y 2 + ψ2 (x)
So, for instance, ϕ = 12 x2 + xy − 12 y 2 is a potential for F.

2.3.1.7. Solution. If ϕ is a potential for F, then:
∂ϕ 1 1 x
• ∂x = x − y, so ϕ = ln |x| − y + ψ1 (y)
∂ϕ x
• ∂y = y2 , so ϕ = − xy + ψ2 (x)
x
So, for instance, ϕ = ln |x| − y is a potential for F.
2.3.1.8. Solution. None exists: ∂F 1 3

∂z = 3 x , while
2 ∂F3
∂y = 13 x3 + 1, so F
fails the screening test, Theorem 2.3.9.
2.3.1.9. Solution. If ϕ(x, y, z) is a potential for F(x, y, z), then:
∂ϕ x 1
• ∂x (x, y, z) = x2 +y 2 +z 2 , so ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 ) + ψ1 (y, z)
∂ϕ y 1
• ∂y (x, y, z) = x2 +y 2 +z 2 , so ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 ) + ψ2 (x, z)
∂ϕ z 1
• ∂z (x, y, z) = x2 +y 2 +z 2 , so ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 ) + ψ2 (x, y)
1
So, for instance, ϕ(x, y, z) = 2 ln(x2 + y 2 + z 2 ) is a potential for F(x, y, z).
2.3.1.10. Solution. (a) We shall show that F(x, y, z) is conservative by
finding a potential for it. ϕ(x, y, z) is a potential for this F if and only if
∂ϕ
(x, y, z) = x
∂x
∂ϕ
(x, y, z) = −2y
∂y
∂ϕ
(x, y, z) = 3z
∂z
Integrating the first of these equations gives
x2
ϕ(x, y, z) = + f (y, z)
2
Substituting this into the second equation gives
∂f
(y, z) = −2y
∂y
which integrates to
f (y, z) = −y 2 + g(z)
x2
Finally, substituting ϕ(x, y, z) = 2 − y 2 + g(z) into the last equation gives
g 0 (z) = 3z
which integrates to
3 2
g(z) = z +C
2
with C being an arbitrary constant. So, F(x, y, z) is conservative and
ϕ(x, y, z) = 12 x2 − y 2 + 23 z 2 is one allowed potential.
(b) The field F = F1 ı̂ı + F2 ̂ can be conservative only if it passes the
screening test
∂F1 ∂F2
=
∂y ∂x
In this case
∂F1 ∂ x 2xy
= =− 2
∂y ∂y x2 + y 2 (x2 + y 2 )
is different from
∂F2 ∂ −y 2xy
= = 2
∂x ∂x x2 + y 2 (x + y 2 )
2
for all (x, y) with x and y both nonzero. So F is not conservative.

2.3.1.11. Solution. By Theorem 2.4.8, the field F = F1 ı̂ı + F2 ̂ + F3 k̂
is conservative only if it passes the screening test ∇ × F = 0. That is, if
and only if
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
or,
∂ (z2 ) ∂
2Byz 3

e = ⇐⇒ 0=0
∂y ∂x
∂ (z2 ) ∂ 2 2 2
Axze(z ) + 3By 2 z 2 2ze(z )
= Aze(z )

e = ⇐⇒
∂z ∂x
∂ ∂ 2 2 2
2Byz 3 = Axze(z ) + 3By 2 z 2 6Bye(z )
= 6Bye(z )

⇐⇒
∂z ∂y
Hence only A = 2 works. We shall see in part (b) that any B works.
(b) When A = 2, and B is any real number.
2 2
F = e(z ) ı̂ı + 2Byz 3 ̂ + 2xze(z )
+ 3By 2 z 2 k̂

ϕ(x, y, z) is a potential for this F if and only if
∂ϕ 2
(x, y, z) = e(z )
∂x
∂ϕ
(x, y, z) = 2Byz 3
∂y
∂ϕ 2
(x, y, z) = 2xze(z ) + 3By 2 z 2
∂z
2
ϕ(x, y, z) = xe(z )
+ f (y, z)

∂f
(y, z) = 2Byz 3
∂y
which integrates to
f (y, z) = By 2 z 3 + g(z)
2
Finally, substituting ϕ(x, y, z) = xe(z ) +By 2 z 3 +g(z) into the last equation
gives
2 2
2xze(z )
+ 3By 2 z 2 + g 0 (z) = 2xze(z )
+ 3By 2 z 2 or g 0 (z) = 0
which integrates to
g(z) = C
with C being an arbitrary constant. So, for each real number B, ϕ(x, y, z) =
2
xe(z ) + By 2 z 3 is one allowed potential.
2.3.1.12. Solution. In each second 2πm cm2 of fluid crosses each circle
of radius r (and hence circumference 2πr) centred on the origin. So the
speed of flow at radius r is m
r . As the direction of flow is radially outward
xı̂ı + y̂
v=m
x2 + y 2
ϕ(x, y) is a potential for this F if and only if
∂ϕ x
(x, y) = m 2
∂x x + y2
∂ϕ y
(x, y) = m 2
∂y x + y2
ϕ(x, y) = 12 m ln(x2 + y 2 ) + f (y)

y y
m + f 0 (y) = m 2 or f 0 (y) = 0
x2 + y 2 x + y2
which integrates to
f (y) = C
with C an arbitrary constant. So one possible potential is
ϕ = 21 m ln(x2 + y 2 )
2.3.1.13. Solution. Following Example 2.3.3, the particle can never
escape the region {(x, y, z) : ϕ(x, y, z) ≥ −E}. So, we should find E, then
figure out the region.
The kinetic energy of the particle is 12 m|v|2 , so the total energy of
the system (also the kinetic energy when the potential energy is 0) is
1 2
2 (10)(2 ) = 20 J.
Therefore, a region it can never escape is

(x, y, z) ϕ(x, y, z) ≥ −20
that is,
(x, y, z) x2 + y 2 + z 2 ≤ 20

√
So, it can never escape the sphere centred at the origin with radius 20.
1 2

2.3.1.14. Solution. Example 2.3.3 tells us 2 m|v(t)| −ϕ
x(t), y(t), z(t) =
E is a constant quantity, provided F is conservative with potential ϕ. So,
it would be nice if F were conservative.
If F = ∇ ϕ, then
∂ϕ
• ∂x = 0, so ϕ = ψ1 (y, z)
∂ϕ
• ∂y = 1, so ϕ = y + ψ2 (x, z)
∂ϕ
• ∂z = 3z 1/3 , so ϕ = 94 z 4/3 + ψ3 (x, y)
We can choose ϕ(x, y, z) = y+ 94 z 4/3 . So, 12 m|v(t)|2 −ϕ x(t), y(t), z(t) =

E is a constant quantity, as desired. Using the information that the particle

has mass 1/2, and speed 1 when it is at the origin:
1 1 2 1
E= · |1| − ϕ 0, 0, 0 =
2 2 4
When the particle is at (1, 1, 1):
|v|2

1 1 1 2 9
= · |v| − ϕ(1, 1, 1) = − 1+
4 2 2 4 4
√
|v| = 14
√
So, at the point (1, 1, 1), the particle has speed 14.
2.3.1.15. Solution. We can start with the screening test, Theorem 2.3.9.
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y

= g 0 (y)h0 (z) − g 0 (y)h0 (z) ı̂ı + 0 − 0 ̂ + 0 − 0 k̂ = 0
So, it’s possible that the field is conservative. Remember, this test alone
isn’t enough to tell us it’s conservative. (Had the test come out differently,
though, we’d be done.)
Suppose F = ∇ ϕ(x, y, z). Then:
∂ϕ
• ∂x = 2f (x)f 0 (x). By inspection, we 2
R see ϕ 0= f (x) + ψ1 (y, z). (We
could also find this by evaluating 2f (x)f (x)dx with the substitu-
tion u = f (x).)
∂ϕ
• ∂y = g 0 (y)h(z), so ϕ = g(y)h(z) + ψ2 (x, z).
∂ϕ
• ∂z = g(y)h0 (z), so ϕ = g(y)h(z) + ψ2 (x, y).
All together, we can choose ϕ(x, y, z) = f 2 (x) + g(y)h(z).

2.3.1.16. Solution. Following Definition 2.3.8, The curl of a vector field
is defined by
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= −
ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y

When F = xy, xz, y 2 + z ,
∇ × F = (2y − x)ı̂ı + (0 − 0)̂ + (z − x)k̂
When the curl is 0ı̂ı +0̂ +0k̂, we have x = 2y and x = z. That is, our points
are of the form (2c, c, 2c) for any constant c. So, the region in question is
the line through the origin in the direction of the vector (2, 1, 2).

2.4.2 · Exercises
2.4.2.1. Solution. The square has four sides, each of which is a line
segment.
• On the first side, y = 0 and dy = 0. That is, we may parametrize
the first side by r(x) = xı̂ı with 0 ≤ x ≤ 1.
• On the second side, x = 1 and dx = 0. We may parametrize the
second side by r(y) = ı̂ı + y ̂ with 0 ≤ y ≤ 1.
• On the third side, y = 1 and dy = 0. We may parametrize the third

side by r(x) = xı̂ı + ̂ with x running from 1 to 0.
• On the final side, x = 0 and dx = 0. We may parametrize the fourth
side by r(y) = y ̂ with y running from 1 to 0.
dy=0
(0, 1) y=1 (1, 1)
dx=0 dx=0
x=0 x=1
(0, 0) dy=0 (1, 0)

y=0
So
Z Z 1 Z 1 Z 0
x2 y 2 dx + x3 y dy = x2 × 02 dx + 13 × y dy + x2 × 12 dx
C 0 0 1
Z 0
+ 03 × y dy
1
1 1 1
= − =
2 3 6
2.4.2.2. Solution. Every F in this problem is defined and has continuous
first-order partial derivatives on all of R2 or R3 . The characterization in
Theorem 2.4.8 tells us that our fields will be conservative if and only if
they pass the screening test, i.e. have curl 0.
a.
F = xı̂ı + z̂ + y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + (0 − 0)k̂ = 0
This field passes the screening test. Since F is defined and has continuous
first-order partial derivatives on all of R3 , it is conservative. So, we have
option A.
b.
F = y 2 zı̂ı + x2 z̂ + x2 y k̂
∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (x2 − x2 )ı̂ı + (y 2 − 2xy)̂ + (2xz − 2yz)k̂ 6= 0
So, F fails the screening test. So, it’s not conservative. That’s option B.
c.
F = (yexy + 1)ı̂ı + (xexy + z)̂ + (z + y) k̂

∂F ∂F2 ∂F ∂F3 ∂F ∂F1
3 1 2
∇×F= − ı̂ı + − ̂ + − k̂
∂y ∂z ∂z ∂x ∂x ∂y
= (1 − 1)ı̂ı + (0 − 0)̂ + {exy (xy + 1) − exy (xy + 1)}k̂ = 0
F passes the screening test. Since F is defined and has continuous first-
order partial derivatives on all of R3 , it is conservative. So, we have option
A.
d.
F = y cos(xy)ı̂ı + x sin(xy)̂
∂F2
= xy cos(xy) + sin(xy)
∂x
∂F1
= −xy sin(xy) + cos(xy)
∂y
∂F2 ∂F1
6=
∂x ∂y
F fails the screening test, so it is not conservative. That is Option B.
R
2.4.2.3. Solution. Since F is conservative, C F · dr = 0 over any closed
curve C. The given curve is closed, so the integral is simply zero.
R and A and
2.4.2.4. Solution. Since F is conservative, R B start and end
at the same points, by path-independence B F · dr = A F · dr = 5.
3
2.4.2.5. ∗. Solution. Note F R is defined and continuous on all of R .
By Theorem 2.4.7, the integral C F · dr = 0 for all closed paths C if and
only if F is conservative. Furthermore, F has continuous first-order partial
derivatives on all of R3 . Using Theorem 2.4.8, F is conservative if and only
if it passes the scsreening test ∇ × F = 0:
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
0 = ∇ × F = det  ∂x ∂y ∂z 
ex sin y aex cos y + bz cx
= (0 − b)ı̂ı − (c − 0)̂ + (aex cos y − ex cos y)k̂
This is the case if and only if a = 1, b = c = 0.

2.4.2.6. Solution. (a) Consider the circle C in the figure (a) on the left
below, oriented clockwise. The vector field F is in the same direction as dr
dt
dr
at every point of the
H curve. So F · dt > 0 at every point of C and C is a
closed curve with C F · dr > 0. As a consequence F is not conservative.
(a) y (b) y
L1
C
x
L2 L4
L3 x
(b) Consider the square in the figure (b) on the right above, oriented
counterclockwise. It consists of the fourline segments L1 , L2 , L3 and L4 .
On all of L1 , L2 , L3 we have that F r(t) ·r0 (t) = 0 because the vector field
is perpendicular to the line segment. On L4 we have F r(t) · r0 (t) > 0. So
I Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr
C L1 L2 L3 L4
Z
=0+0+0+ F · dr > 0
L4
H
So C is a closed curve with C F · dr > 0 and F is not conservative.
(c) Consider the square in the figure (c) on the left below, oriented
counterclockwise. It consists of the four line segments L1 , L2 , L3 and L4 .
On L1 and L3 we have that the dot product F r(t) · r0 (t) = 0 because the

vector field is perpendicular to the line segment. On L2 we have F r(t) ·
r0 (t) < 0 while on L4 we have F r(t) 0
0· r (t) > 0. The vector field F is
longer on L4 than on L2 . So F r(t) · r (t) has a larger magnitude on L4
than L2 and
I Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr
C L1 L2 L3 L4
Z Z
=0+ F · dr + 0 + F · dr > 0
L2 L4
H
So C is a closed curve with C
F · dr > 0 and F is not conservative.
(c) y (d) y
L1
L2 L4
L3 x x
(d) We are told that one of the four vector fields is conservative. Only
the vector field in (d) is left, so it is conservative.
Remark: We can verify that vector field (d) is indeed conservative by
observing (look at the figure (d) on the right above) that the ı̂ı component
of the vector field is exactly zero and that the ̂ component depends only
on y. So the vector field is of the form
F(x, y) = a(y) ̂
for some function a(y). If A(y) is any antiderivative of a(y), we have

F = ∇ A, so that F is conservative with potential A(y).

2.4.2.7. ∗. Solution. (a) The (largest possible) domain is D = (x, y, z) x2 +
y 2 6= 0 . That is, all of R3 except the points lying along the z-axis.
(b) As preliminary computations, let’s find
−2x2 + 2y 2 − 2xy

∂ x − 2y −2 2y(x − 2y)
2 2
= 2 2
− 2 = 2
∂y x + y x +y (x2 + y 2 ) (x2 + y 2 )
−2x2 + 2y 2 − 2xy

∂ 2x + y 2 2x(2x + y)
= − 2 = 2
∂x x2 + y 2 x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
So the curl of F is
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det 

∂x ∂y ∂z 
x−2y 2x+y
x2 +y 2 x2 +y 2 z
!
−2x2 + 2y 2 − 2xy −2x2 + 2y 2 − 2xy
= 2 − 2 k̂
(x2 + y 2 ) (x2 + y 2 )
=0
on the domain of F.
(c) Parametrize the circle by
r(t) = 2 cos tı̂ı + 2 sin t ̂ + 3 k̂ r0 (t) = −2 sin tı̂ı + 2 cos t ̂
with 0 ≤ θ ≤ 2π. So the integral is

x−2y 2x+y
x2 +y 2 x2 +y 2
z }| { z }| { z
2π z}|{
2 cos t − 4 sin t
Z Z
4 cos t + 2 sin t
F · dr = ı̂ı + ̂ + 3 k̂ ·
C 0 4 4
r0 (t)
nz }| {o
−2 sin tı̂ı + 2 cos t ̂ dt
2π
−4 sin t cos t + 8 sin2 t + 8 cos2 t + 4 sin t cos t
Z
= dt
0 4
Z 2π
=2 dt = 4π
0
(d) As the integral of F around the simple closed curve C is not zero, F
cannot be conservative. See Theorem 2.4.7 and Examples 2.3.14 and 4.3.8.
2.4.2.8. Solution. The point here is that F is conservative, as F = ∇φ
with
x2 z2
+ yx − yz +
φ=
2 2
So, for all paths from r(t0 ) = (1, 0, −1) to r(t1 ) = (0, −2, 3),
Z

F · dr = φ r(t1 ) − φ r(t0 ) = φ(0, −2, 3) − φ(1, 0, −1)
C

9 1 1
= 0+0+6+ − +0−0+
2 2 2
1
=9
2
2.4.2.9. Solution. Note that:

• Along the line segment from (0, 0) to (1, 0), x increases from 0 to 1,
while y is held fixed at y = 0. So we may parametrize this segment
by r(x) = xı̂ı, 0 ≤ x ≤ 1.
• Along the line segment from (1, 0) to (1, π), y increases from 0 to π,
while x is held fixed at x = 1. So we may parametrize this segment
by r(x) = ı̂ı + y ̂, 0 ≤ y ≤ π.
• Along the line segment from (1, π) to (0, π), x decreases from 1 to 0,
while y is held fixed at y = π. So we may parametrize this segment
by r(x) = xı̂ı + π ̂ with x running from 1 to 0.
Hence
Z Z 1 Z π Z 0
V · dr = V(x, 0) · ı̂ı dx + V(1, y) · ̂ dy + V(x, π) · ı̂ı dx
C 0 0 1
Z 1 Z π Z 0
x
= 2
(e + x ) dx + (y + 3) dy + (−ex + x2 ) dx
0 0 1
Z 1 Z π
x
=2 e dx + (y + 3) dy
0 0
π2
= 2(e − 1) + + 3π
2
2
2.4.2.10. Solution. (a) We may parametrize the curve by r(t)
dr
=3tı̂ı +t2 ̂
with 0 ≤ t ≤ 1. Then v(t) = dt (t) = ı̂ı + 2t ̂ and F x(t), y(t) = t ı̂ı − t ̂
so
Z Z 1 Z 1
dr 3
t ı̂ı − t2 ̂ · ı̂ı + 2t ̂ dt

F · dr = F x(t), y(t) · (t) dt =
C 0 dt 0
Z 1
− t3 dt

=
0
1
=−
4
(b) The path is the union of three line segments.
• On the first segment of the path y = z = 0 so F simplifies to xı̂ı − x k̂
and dr = ı̂ı dx (i.e. we can parametrize the first segment of the path
by r(x) = xı̂ı with 0 ≤ x ≤ 1), so F · dr = x dx.
• On the second segment of the path x = 1, z = 0 so F simplifies to
ı̂ı + y̂ − (1 + y)k̂ and dr = ̂ dy (parametrize the second segment of
the path by r(y) = ı̂ı + y ̂ with 0 ≤ y ≤ 1), so F · dr = y dy.
• On the final segment of the path x = y = 1 so F simplifies to (1 −
z)ı̂ı + (1 − z)̂ − 2k̂ and dr = k̂ dz (parametrize the third segment of
the path by r(z) = ı̂ı + ̂ + z k̂ with 0 ≤ z ≤ 1), so F · dr = −2 dz.
So
Z Z 1 Z 1 Z 1
1 1
F · dr = x dx + y dy + (−2) dz = + − 2 = −1
C 0 0 0 2 2
2.4.2.11. ∗. Solution. Parametrize the curve using y as a parameter.
8
Then y = t, x = 2y = 2t and z = xy = 2t82 so that:
4
r(t) = 2tı̂ı + t ̂ + k̂, 1≤t≤2
t2
8
r0 (t) = 2ı̂ı + ̂ −
k̂
t3
F(r(t)) = 4t2 ı̂ı + 4t3 k̂
F(r(t)) · r0 (t) = 8t2 − 32
Then
Z Z 2 Z 2 2
8 3
F(r(t)) · r0 (t) dt = 8t2 − 32 dt =

F · dr = t − 32t
C 1 1 3 1
40
=−
3
2.4.2.12. Solution. (a), (b) The curls of F and G are
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
6x2 yz 2 2x3 z 2 + 2y − xz 4x3 yz
= (4x3 z − 4x3 z + x)ı̂ı − (12x2 yz − 12x2 yz) ̂ + (6x2 z 2 − z − 6x2 z 2 ) k̂

= xı̂ı − z k̂
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × G = det  ∂x ∂y ∂z 
yz 0 xy
= xı̂ı − z k̂
Hence the screening test for

∇ × F + λG = (x + λx)ı̂ı − (z + λz) k̂
passes for λ = −1. Furthermore
F − G = (6x2 yz 2 − yz)ı̂ı + (2x3 z 2 + 2y − xz) ̂ + (4x3 yz − xy) k̂
= ∇ 2x3 yz 2 − xyz + y 2

The potential was found by guessing. Alternatively, we can find it by using

that φ(x, y, z) is a potential for F − G if and only if
∂φ
(x, y, z) = 6x2 yz 2 − yz
∂x
∂φ
(x, y, z) = 2x3 z 2 + 2y − xz
∂y
∂φ
(x, y, z) = 4x3 yz − xy
∂z
φ(x, y, z) = 2x3 yz 2 − xyz + f (y, z)
∂f ∂f
2x3 z 2 − xz + (y, z) = 2x3 z 2 + 2y − xz or (y, z) = 2y
∂y ∂y
which integrates to
f (y, z) = y 2 + g(z)
Finally, substituting φ(x, y, z) = 2x3 yz 2 − xyz + y 2 + g(z) into the last
equation gives
4x3 yz − xy + g 0 (z) = 4x3 yz − xy or g 0 (z) = 0
which integrates to
g(z) = K
with K being an arbitrary constant. Choosing K = 0 gives the potential
φ(x, y, z) = 2x3 yz 2 − xyz + y 2 as in the guess above.
2
(c) Any point (x, y, z) on the curve must have z = x and y = exz = ex .
2
So we may parametrize the curve by r(x) = xı̂ı + ex ̂ + x k̂, 0 ≤ x ≤ 1.
Hence
Z Z Z
F · dr = (F − G) · dr + G · dr
C C C
dr
G(r(x)) dx
h i(1,e,1) z Z 1 }| { z }| {
2 2
3 2 2 x x x2
= 2x yz − xyz + y + [xe ı̂ + xe k̂] · [ı̂ + 2xe ̂ + k̂] dx
ı ı 
2.4.2.13. ∗. Solution. Parametrize
(0,1,0) 0 the line segment by
Z 1 h 2 i1
2
e2 − 1,x0)dx = e0,+1)e2 − + t,ex1 − t= e2 +0 2e

= e=
r(t) + (0, 1 ++ t (2,
0, 1) 2xe − (0, = 1 2t, ≤ t−≤2 1
0 0
so that r(0) = (0, 0, 1) is the initial point of the line segment and r(1) =
(2, 1, 0) is the final point of the segment. Then
r0 (t) = (2, 1, −1)
and the work is

Z Z 1
F r(t) · r0 (t) dt

F · dr =
0
Z 1
2t − t2 , t − (1 − t)2 , (1 − t) − 4t2 · (2, 1, −1) dt

=
0
Z 1
4t − 2t2 + t − 1 + 2t − t2 − 1 + t + 4t2 dt

=
0
Z 1
t2 + 8t − 2 dt

=
0
1 7
= +4−2=
3 3
2.4.2.14. ∗. Solution. On P , z = ln x1 = − ln(x). So parametrize the
curve P by
π
r(θ) = cos θ ı̂ı + sin θ ̂ − ln(cos θ) k̂ 0≤θ≤
4
Then
r 0 (θ) = − sin θ ı̂ı + cos θ ̂ + tan θ k̂

F r(θ) = cos θ ı̂ı + sin θ ̂ + cos3 θ k̂

F r(θ) · r 0 (θ) = sin θ cos2 θ

so that
Z Z π/4 Z π/4
F r(θ) · r 0 (θ) dθ = sin θ cos2 θ dθ

Work = F · dr =
P 0 0
1 π/4
= − cos3 θ

3 0
1 1
= 1 − 3/2 ≈ 0.2155
3 2
2.4.2.15. ∗. Solution. Hmmm. F looks suspiciously complicated. Let’s
guess that F is conservative and look for a potential for it. φ(x, y, z) is a
potential for this F if and only if
∂ϕ
(x, y, z) = yz cos x
∂x
∂ϕ
(x, y, z) = z sin x + 2yz
∂y
∂ϕ
(x, y, z) = y sin x + y 2 − sin z
∂z
ϕ(x, y, z) = yz sin x + f (y, z)
∂f ∂f
z sin x + (y, z) = z sin x + 2yz or (y, z) = 2yz
∂y ∂y
which integrates to
f (y, z) = y 2 z + g(z)
Finally, substituting ϕ(x, y, z) = yz sin x + y 2 z + g(z) into the last equation
gives
y sin x + y 2 + g 0 (z) = y sin x + y 2 − sin z or g 0 (z) = − sin z
which integrates to
g(z) = cos z + C
with C being an arbitrary constant. So φ(x, y, z) = yz sin x + y 2 z + cos z
is one allowed scalar potential and the specified integral is
π3 π2
Z r(π/2)
F · dr = ϕ(r) = ϕ(π/2 , π/2 , π/2) − ϕ(0, 0, 0) = + −1

C r(0) 8 4
R2.4.2.16. ∗. Solution 1. We are being asked to evaluate the line integral

C
F·dr with C being the specified semi-circle and F = xy ̂. As ∇ ×F 6= 0,
the vector field F is not conservative. So we’ll evaluate the integral directly.
First, using the figure,
y 1 − cos θ , sin θ
(x − 1)2 + y 2 = 1
θ
x
1, 0
we parametrize C by

r(θ) = x(θ) , y(θ) = (1 − cos θ)ı̂ı + sin θ ̂ 0≤θ≤π
So the integral is
Z Z π Z π
xy dy = x(θ) y(θ) y 0 (θ) dθ = (1 − cos θ) sin θ cos θ dθ
C 0 0
Making the substitution u = cos θ, du = − sin θ dθ, u(0) = 1, u(π) = −1,

Z Z −1 Z 1
xy dy = (1 − u) u (−du) = (u − u2 ) du
C 1 −1
1
13
Z
2
= −2 u2 du = −2 = −
0 3 3
Solution 2. We can write x in terms of y over C in two pieces:

p
• Let C1 be the quarter-circle x = 1 − 1 − y 2 as y goes from 0 to 1,
and
p
• Let C2 be the quarter-circle x = 1 + 1 − y 2 as y goes from 1 to 0.
Then:
Z Z Z
xydy = xydy + xydy
C C1 c2
Z 1 p Z 0 p
= 1− 1− y2 y dy + 1+ 1 − y 2 y dy
0 1
Z 1 Z 1 p Z 0 Z 0 p
= y dy − y 1− y2 dy + y dy + y 1 − y 2 dy
0 0 1 1
Z 1 p
= −2 y 1 − y 2 dy
0
Using the substitution u = 1 − y 2 , du = −2y dy:

Z 0
2
= u1/2 du = −
1 3
2.4.2.17. ∗. Solution. The line integral is C F · dr with F = (yex +
R
sin y)ı̂ı + (ex + sin y + x cos y) ̂. We are to show that it is independent of
path. That is the case if and only if F is conservative. So let’s look for a
potential ϕ for F. That is, let’s look for a function ϕ that obeys
∂ϕ
(x, y) = yex + sin y
∂x
∂ϕ
(x, y) = ex + sin y + x cos y
∂y
ϕ(x, y) = yex + x sin y + f (y)
ex + x cos y + f 0 (y) = ex + sin y + x cos y or f 0 (y) = sin y
which integrates to
f (y) = − cos y + C
So F is indeed conservative with one potential being ϕ(x, y) = yex +x sin y−
cos y and the line integral is
Z Z (0,π/2)
x x
(ye + sin y) dx + (e + sin y + x cos y) dy = F · dr = ϕ(x, y)

C C (1,0)
h i(0,π/2)
= yex + x sin y − cos y
(1,0)
π
=1+
2
2.4.2.18. ∗. Solution. Here is a sketch of C.
z
(0, 0, 1)
C1 C2
(0, 1, 0)
y
(1, 0, 0) C3
x
Note that
• y = 0 on the
R line segment from (1, 0, 0) to (0, 0, 1) so that the integral
reduces to zx dz on that line segment and
• x = 0 on the line segment from (0, 0, 1) to (0, 1, 0) so that the integral
R
reduces to yz dy on that line segment and
• z = 0 on the
R line segment from (0, 1, 0) to (1, 0, 0) so that the integral
reduces to xy dx on that line segment.
So it looks feasible to evaluate the integral directly. Label the sides of the
triangle C1 , C2 and C3 as in the sketch above.
• We parametrize C1 by r(t) = (1, 0, 0) + t[(0, 0, 1) − (1, 0, 0)] = (1 −
t , 0 , t), 0 ≤ t ≤ 1. So
x z z 0 (t)
Z Z Z 1 z }| { z}|{ z}|{
xy dx + yz dy + zx dz = zx dz = (1 − t) (t) (1) dt
C1 C1 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
• We parametrize C2 by r(t) = (0, 0, 1)+t[(0, 1, 0)−(0, 0, 1)] = (0 , t , 1−

t), 0 ≤ t ≤ 1. So
y z y 0 (t)
Z Z Z 1 z}|{ z }| { z}|{
xy dx + yz dy + zx dz = yz dy = (t) (1 − t) (1) dt
C2 C2 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
• We parametrize C3 by r(t) = (0, 1, 0) + t[(1, 0, 0) − (0, 1, 0)] = (t , 1 −

t , 0), 0 ≤ t ≤ 1. So
x y x0 (t)
Z Z Z 1 z}|{ z }| { z}|{
xy dx + yz dy + zx dz = xy dx = (t) (1 − t) (1) dt
C3 C3 0
Z 1
= (t − t2 ) dt
0
1 1 1
= − =
2 3 6
All together
Z 3 Z
X 1 1
xy dx + yz dy + zx dz = xy dx + yz dy + zx dz = 3 × =
C C` 6 2
`=1
2.4.2.19. ∗. Solution. We are told that F is conservative. Let’s find a

potential ϕ obeying ∇ ϕ = F. That is,
∂ϕ
= y + zex
∂x
∂ϕ
= x + ey sin z
∂y
∂ϕ
= z + ex + ey cos z
∂z
The first equation forces ϕ(x, y, z) = xy + zex + ψ(y, z). Substituting this
into the second equation gives x + ∂ψ y ∂ψ
∂y (y, z) = x + e sin z or ∂y (y, z) =
y y
e sin z which forces ψ(y, z) = e sin z + ζ(z). So far, we have ϕ(x, y, z) =
xy + zex + ey sin z + ζ(z). Substituting this into the third equation gives
2
ex +ey cos z+ζ 0 (z) = z+ex +ey cos z or ζ 0 (z) = z which forces ζ(z) = z2 +C,
for some constant C, which we take to be zero. So our potential is
z2
ϕ(x, y, z) = xy + zex + ey sin z +
2
So the line integral
Z
F · dr = ϕ r(π) − ϕ r(0) = ϕ π, eπ , 0 − ϕ 0, 1, 0 = πeπ

C
2.4.2.20. ∗. Solution. (a) Note F is defined and continuous on all of

R3 . Furthermore, F has continuous first-order partial derivatives on all of
R3 . Using Theorem 2.4.8, F is conservative if and only if it has zero curl:
0 = ∇ × F = ∇ × αey ı̂ı + (xey + β cos z) ̂ − γy sin z k̂

= (−γ sin z + β sin z)ı̂ı + (ey − αey )k̂
which is the case if and only if α = 1, β = γ.

(b) We use Theorem 2.4.2: if ϕ is a potential for F, then
Z
F · dr = ϕ(P1 ) − ϕ(P0 )
C
where C runs from P0 to P1 . So, we find ϕ.

Assume that α = 1, β = γ. We find a potential ϕ for F by antidiffer-
entiating.
∂ϕ
(x, y, z) = ey =⇒ ϕ(x, y, z) = xey + ψ1 (y, z)
∂x
∂ϕ
(x, y, z) = xey + β cos z =⇒ ϕ(x, y, z) = xey + βy cos z + ψ2 (x, z)
∂y
∂ϕ
(x, y, z) = −βy sin z =⇒ ϕ(x, y, z) = βy cos z + ψ3 (x, y)
∂z
for some functions ψ1 (y, z), ψ2 (x, z) and ψ3 (x, y) to be determined.
We’d like a single function ϕ(x, y, z) that simultaneously obeys all three
of these equations, for some ψj ’s. An initial guess is simply the sum of all of
the distinct terms, other that the ψj ’s, that appear in the three equations
above. The term xey appears in the ψ1 and ψ2 equations and the term
βy cos z appears in the ψ2 and ψ3 equations. So we guess
?
ϕ(x, y, z) = xey + βy cos z
If we let ψ1 (y, z) = βy cos z, ψ2 (x, z) = 0, and ψ3 (x, y) = xey , then we

see this function ϕ(x, y, z) does indeed obey all three equations and so is a
potential for F.
The curve C runs from P0 = (02 , e0 , π · 0) = (0, 1, 0) to P1 = (12 , e1 , π ·
1) = (1, e, π). Using Theorem 2.4.2:
Z
F · dr = ϕ(1, e, π) − ϕ(0, 1, 0) = ee − βe − β = ee − β(e + 1)

C
2.4.2.21. ∗. Solution. (a) The curl of F is

 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  =0
z
cos x 2 + sin y e
Because F1 is a function only of x, F2 is a function only of y, and F3 is a

function only of z, that all partial derivatives used in computing the curl
are 0.
(b) The vector field F passes the screening test on all of R3 and so is
conservative by Theorem 2.4.8 in the text. Alternatively, we can see that
F = ∇ sin x + 2y − cos y + ez

by inspection. Alternatively, f can be found by antidifferentiating its par-

tial derivatives:
∂f
(x, y, z) = cos x =⇒ f (x, y, z) = sin x + ψ1 (y, z)
∂x
∂f
(x, y, z) = 2 + sin y =⇒ f (x, y, z) = 2y − cos y + ψ2 (x, z)
∂y
∂f
(x, y, z) = ez =⇒ f (x, y, z) = ez + ψ3 (x, y)
∂z
We’d like a single function f (x, y, z) that simultaneously obeys all three
of these equations, for some ψj ’s. An initial guess is simply the sum of
all of the distinct terms, other than the ψj ’s, that appear in the three
equations. The term sin x appears in the ψ1 equation, the terms 2y and
− cos y appears in the ψ2 equation, and the term ez appears in the ψ3
equation. So we guess
?
f (x, y, z) = sin x + 2y − cos y + ez
If we let ψ1 (y, z) = 2y − cos y + ez , ψ2 (x, z) = sin x + ez , and ψ3 (x, y) =

sin x + 2y − cos y, then we see this function f (x, y, z) is indeed a potential
for F.
(c) Since F = ∇ f ,
Z

F · dr = f r(3π) − f r(0)
C
= f (3π, −1, 0) − f (0, 1, 0)

= 0 − 2 − cos(−1) + 1 − 0 + 2 − cos 1 + 1
Since cosine is an even function, cos(−1) = cos 1.
= −4
2.4.2.22. ∗. Solution. (a) The curl is
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x =0

∂y ∂z
z + ey xey − e sin y
z
1 + x + ez cos y
so F passes the screening test. Since its first-order partial derivatives are
continuous on all of R3 , it is conservative by Theorem 2.4.8 in the text.
By inspection, the potential is ϕ(x, y, z) = xz + xey + ez cos y + z —
this is another way to verify that F is conservative. Alternatively, ϕ can

be found by antidifferentiating its partial derivatives.
∂ϕ
(x, y, z) = z + ey
∂x
=⇒ ϕ(x, y, z) = zx + xey + ψ1 (y, z)
∂ϕ
(x, y, z) = xey − ez sin y
∂y
=⇒ ϕ(x, y, z) = xey + ez cos y + ψ2 (x, z)
∂ϕ
(x, y, z) = 1 + x + ez cos y
∂z
=⇒ ϕ(x, y, z) = z + zx + ez cos y + ψ3 (x, y)
We’d like a single function ϕ(x, y, z) that simultaneously obeys all three of
these equations. An initial guess is simply the sum of the distinct terms
(without the ψj ’s) that appear in the equations above:
?
ϕ(x, y, z) = zx + xey + ez cos y + z
If we let ψ1 (y, z) = ez cos y + z, ψ2 (x, z) = zx + z, and ψ3 (x, y) = xey ,

then we see this function ϕ(x, y, z) is indeed a potential for F.
(b) Since F = ∇ ϕ, with ϕ = xz + xey + ez cos y + z,
Z ir(π)
h
F · dr = ϕ r(π) − ϕ r(0) = xz + xey + ez cos y + z

C r(0)
h i(π2 ,0,1)
= xz + xey + ez cos y + z
(0,0,1)
= π 2 + π 2 + e + 1 − (0 + 0 + e + 1) = 2π 2

2.4.2.23. ∗. Solution. (a) For F to be conservative, it must pass the

screening test
 
ı̂ı ̂ k̂
∂ ∂ ∂
0 = ∇ × F = det 
 
∂x ∂y ∂z 
(x − a)yex x
xe + z 3 byz 2
= bz 2 − 3z 2 )ı̂ı − 0 − 0 ̂ + ex + xex − (x − a)ex k̂

This is the case if and only if b = 3 and a = −1

(b) Set a = −1 and b = 3. For f to be a potential for F, it must obey
∂f
(x, y, z) = (x + 1)yex
∂x
∂f
(x, y, z) = xex + z 3
∂y
∂f
(x, y, z) = 3yz 2
∂z
Integrating the second of these equations gives
f (x, y, z) = xyex + yz 3 + g(x, z)
Substituting this into the last equation gives

∂g ∂g
3yz 2 + (x, z) = 3yz 2 or (x, z) = 0
∂z ∂z
which forces
g(x, z) = h(x)
Finally, substituting f (x, y, z) = xyex + yz 3 + h(x) into the first equation
gives
xyex + yex + h0 (x) = (x + 1)yex or h0 (x) = 0
So h(x) = C and hence f (x, y, z) = xyex + yz 3 + C works for any constant
C.
(c) Since F = ∇ f ,
Z Z

F · dr = ∇ f · dr = f r(π)) − f ( r(0) = f (π, 1, −1) − f (0, 1, 1)
C
C π
= πe − 1 − 1 = πeπ − 2

(d) Since
Z Z
F · dr = (x + 1)yex dx + (xex + z 3 ) dy + 3yz 2 dz
C C
we have
Z Z
I= F · dr + yz 2 dz
C C
y z2 dz
Z π z }| { z }| { z }| {
= πeπ − 2 + 2
(cos 2t) cos t (− sin t) dt
Z0 π
= πeπ − 2 + (2 cos2 t − 1) cos2 t(− sin t) dt
0
Z −1
= πeπ − 2 + (2u2 − 1)u2 du with u = cos t, du = − sin t dt
1
u3 i−1
h 2u5
= πeπ − 2 + −
5 3 1
h 4 2i
π
= πe − 2 + − +
5 3
π 32
= πe −
15
2.4.2.24. ∗. Solution. (a) The vector field F is conservative if and only

if it passes the screening test ∇ × F = 0. That is, if and only if,

ı̂ı ̂ k̂

∂ ∂ ∂
0 = ∇ × F = det

2 3z ∂x ∂y ∂z
y e + Axy 3 2xye3z + 3x2 y 2 Bxy 2 e3z

= 2Bxye3z − 6xye3z ı̂ı − By 2 e3z − 3y 2 e3z ̂

+ 2ye3z + 6xy 2 − 2ye3z − 3Axy 2 k̂

So F is conservative if and only if A = 2 and B = 3.

(b) Let A = 2 and B = 3. We find a potential ϕ for F by antidifferen-
tiating its partial derivatives.
∂ϕ
(x, y, z) = y 2 e3z + 2xy 3
∂x
=⇒ ϕ(x, y, z) = xy 2 e3z + x2 y 3 + ψ1 (y, z)
∂ϕ
(x, y, z) = 2xye3z + 3x2 y 2
∂y
=⇒ ϕ(x, y, z) = xy 2 e3z + x2 y 3 + ψ2 (x, z)
∂ϕ
(x, y, z) = 3xy 2 e3z
∂z
=⇒ ϕ(x, y, z) = xy 2 e3z + ψ3 (x, y)
Let’s guess that

ϕ(x, y, z) = xy 2 e3z + x2 y 3
(This was obtained by summing the distinct terms in the above three
equations, without the ψi ’s.) If we set ψ1 (y, z) = ψ2 (x, z) = 0 and
ψ3 (x, y) = x2 y 3 , we see our choice of ϕ is indeed a potential
R for F.
(c) Set A = 2 and B = 3. We are asked the evaluate C G · dr with
G = (y 2 e3z + xy 3 )ı̂ı + (2xye3z + 3x2 y 2 ) ̂ + 3xy 2 e3z k̂ = F − xy 3 ı̂ı
So
Z
(y 2 e3z + xy 3 ) dx + (2xye3z + 3x2 y 2 ) dy + 3xy 2 e3z dz
C
Z Z
= F · dr − xy 3 dr
C C
3
xy ı̂ı r0 (t)
Z 1z }| { z }| {
2t −t 3
2t −t 1
= ϕ r(1) − ϕ r(0) − e e ı̂ı · 2e ı̂ı − e ̂ + k̂ dt
0 1+t
Z 1
2 1
2et dt

= ϕ e , , ln 2 − ϕ 1, 1, 0 −
e 0
1 2 3 ln 2 1 3
= e2 + e4

e − 1 + 1 − 2(e − 1)
e e
= 23 + e − 2 − 2e + 2
=8−e
2.4.2.25. ∗. Solution. (a) The field is conservative only if
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
That is,
∂ ∂
(2x sin(πy) − ez ) = ax2 cos(πy) − 3ez

∂y ∂x
⇐⇒ 2πx cos(πy) = 2ax cos(πy)
∂ ∂
(2x sin(πy) − ez ) = − (x + by) ez
∂z ∂x
⇐⇒ −ez = −ez
∂ ∂
ax2 cos(πy) − 3ez = − (x + by) ez

∂z ∂y
⇐⇒ −3ez = −bez
Hence only a = π, b = 3 works.

(b) When a = π, b = 3
F = (2x sin(πy) − ez )ı̂ı + πx2 cos(πy) − 3ez ̂ − (x + 3y) ez k̂

= ∇ x2 sin(πy) − xez − 3yez + C

so ϕ(x, y, z) = x2 sin(πy) − xez − 3yez + C for any constant C. Here ϕ was

guessed. Alternatively, it can be found by antidifferentiating the partial
derivatives of F.
∂ϕ
(x, y, z) = 2x sin(πy) − ez
∂x
=⇒ ϕ(x, y, z) = x2 sin(πy) − xez + ψ1 (y, z)
∂ϕ
(x, y, z) = πx2 cos(πy) − 3ez
∂y
=⇒ ϕ(x, y, z) = x2 sin(πy) − 3yez + ψ2 (x, z)
∂ϕ
(x, y, z) = −(x + 3y)ez
∂z
=⇒ ϕ(x, y, z) = −xez − 3yez + ψ3 (x, y)
Summing the distinct terms on the right hand sides of the three equations
above, we guess
ϕ(x, y, z) = x2 sin(πy) − xez − 3yez
is a potential for F. Setting ψ1 (y, z) = −3yez , ψ2 (x, z) = −xez , and

ψ3 (x, y) = x2 sin(πy) convinces us that our guess is indeed a valid potential.
(c) By part (b),
Z
F · dr = ϕ(1, 1, ln 2) − ϕ(0, 0, 0)
C
= sin π − eln 2 − 3eln 2 − (sin(0) − 0 − 0)

= −8
(d) Observe that G = F + 3yez k̂, with F evaluated with a = π, b = 3.

Hence
Z Z Z Z
z
G · dr = F · dr + 3ye k̂ · dr = −8 + 3yez k̂ · dr
C C C C
To evaluate the remaining integral, parametrize the curve by r(t) = tı̂ı +t̂ +
ln(1 + t)k̂ with 0 ≤ t ≤ 1. Then r0 (t) = ı̂ı +̂ + 1+t
1
k̂ and 3yez k̂ = 3t(1 + t)k̂
z
so that 3ye k̂ · dr = 3t dt. Subbing in
Z Z 1
3 13
G · dr = −8 + 3t dt = −8 + = −
C 0 2 2
2.4.2.26. ∗. Solution. (a) The potential f must obey
∂f
(x, y, z) = −2y cos x sin x
∂x
∂f
(x, y, z) = cos2 x + (1 + yz)eyz
∂y
∂f
(x, y, z) = y 2 eyz
∂z
Integrating the last of these equations with respect to z gives
f (x, y, z) = yeyz + g(x, y)


∂g ∂g
eyz + yzeyz + (x, y) = cos2 x + (1 + yz)eyz or (x, y) = cos2 x
∂y ∂y
which forces
g(x, y) = y cos2 x + h(x)
Finally, substituting f (x, y, z) = yeyz +y cos2 x+h(x) into the first equation
gives
−2y sin x cos x + h0 (x) = −2y cos x sin x or h0 (x) = 0
So h(x) = C and hence f (x, y, z) = yeyz + y cos2 x + C works for any
constant C.
(b) By part (a)
Z Z
∇ f · dr = f π, eπ , 0 − f 0, 1, −π 2

F · dr =
C C
h i(π,eπ ,0)
= yeyz + y cos2 x
(0,1,−π 2 )
π −π 2

= 2e − e +1
2.4.2.27. ∗. Solution. (a) The curl of F is zero because F1 is a function
only of x, F2 is a function only of y, and F3 is a function only of z. That
is:
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = (0 − 0)ı̂ + (0 − 0)̂ + (0 − 0)k̂ = 0
ı 
2x 2y 2z
(b) All first-order partial derivative of F are continuous on all of R3 .

By part (a), F passes the screening test and is conservative by Theorem
2.4.8 in the text. By inspection, a potential is ϕ = x2 + y 2 + z 2 . Since
F = ∇ ϕ,
Z h i(a1 ,a2 ,a3 )
F · dr = x2 + y 2 + z 2 = a21 + a22 + a23 = a · a
C (0,0,0)

 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
yz yz y yz y
e xze + ze xye + e
= (xe + xyzeyz + ey ) − (xeyz + xyzeyz + ey ) ı̂ı − yeyz − yeyz ̂
yz
+ zeyz − zeyz k̂

=0
(b) F is defined on all of R3 and passes the conservative field screening

test ∇ × F = 0. So F is conservative. We find a potential ϕ for F by
antidifferentiating its partial derivatives.
∂ϕ
(x, y, z) = eyz =⇒ ϕ(x, y, z) = xeyz + ψ1 (y, z)
∂x
∂ϕ
(x, y, z) = xzeyz + zey =⇒ ϕ(x, y, z) = xeyz + zey + ψ2 (x, z)
∂y
∂ϕ
(x, y, z) = xyeyz + ey =⇒ ϕ(x, y, z) = xeyz + zey + ψ3 (x, y)
∂z
All together, ϕ(x, y, z) = xeyz + zey + C works for any constant C. So the
specified work integral is
Z
πe
F · dr = ϕ r(π/2) − ϕ r(0) = ϕ 0, 1, π/2 − ϕ 1, 0, 0 = −1
C 2
2.4.2.29. ∗. Solution. (a), (b) The function f (x, y) is a potential for
F(x, y) if and only if it obeys
∂f
(x, y) = 2xy cos(x2 )
∂x
∂f
(x, y) = sin(x2 ) − sin(y)
∂y
f (x, y) = y sin(x2 ) + g(y)
sin(x2 ) + g 0 (y) = sin(x2 ) − sin(y) or g 0 (y) = − sin(y)
which integrates to
g(y) = cos(y) + C
with C an arbitrary constant. Hence f (x, y) = y sin(x2 ) + cos(y) + C is a
potential for any constant C. Because F has a potential, it is conservative.
(c) We may parametrize C by
π
r(t) = sin(t)ı̂ı + t ̂ ≤t≤π
2
As f (x, y) = y sin(x2 ) + cos(y) is a potential for F
Z
π π
F · dr = f r(π) − f r( ) = f 0, π − f 1,
C 2 2
π
= −1 − sin(1)
2
π
= −1 − sin(1)
2
2.4.2.30. ∗. Solution. (a) The stated integral property is characteristic
of conservative fields (Theorem 2.4.7). Since all partial derivatives of F are
defined on all of R3 , an equivalent property is

ı̂ı ̂ k̂

∂ ∂ ∂
0=∇×F=

∂x ∂y ∂z
(mxyz + z 2 − ny 2 ) (x2 z − 4xy) (x2 y + pxz + qz 3 )

= ı̂ı x2 − x2 − ̂ (2xy + pz − mxy − 2z) + k̂ (2xz − 4y − mxz + 2ny).

This requires p = 2, m = 2, and n = 2, but leaves q ∈ R completely free.

(b) Solution 1: The choices from (a) give
F = (2xyz + z 2 − 2y 2 )ı̂ı + (x2 z − 4xy) ̂ + (x2 y + 2xz + qz 3 ) k̂.

We find a potential ϕ for F by antidifferentiating its partial derivatives.

∂ϕ
(x, y, z) = 2xyz + z 2 − 2y 2
∂x
=⇒ ϕ(x, y, z) = x2 yz + xz 2 − 2xy 2 + ψ1 (y, z)
∂ϕ
(x, y, z) = x2 z − 4xy
∂y
=⇒ ϕ(x, y, z) = x2 yz − 2xy 2 + ψ2 (x, z)
∂ϕ
(x, y, z)ϕ(x, y, z) = x2 y + 2xz + qz 3
∂z
q
=⇒ ϕ(x, y, z) = x2 yz + xz 2 + z 4 + ψ3 (x, y)
4
All together, F = ∇ϕ for
1
ϕ(x, y, z) = x2 yz + xz 2 − 2xy 2 + qz 4 + C
4
where C is any constant.
Rearranging the sphere’s equation to x2 + y 2 + (z − 1)2 = 1 reveals that
its bottom is at r0 = (0, 0, 0), and its top is at r1 = (0, 0, 2). Hence the
work done is
Z Z
W = F · dr = ∇ϕ · dr = ϕ(0, 0, 2) − ϕ(0, 0, 0) = 4q
C C
Solution 2: Since the integral is path-independent, all paths from r0

to r1 produce the same result. A simple choice is
C: r = (0, 0, t), 0≤t≤2
Here r0 (t) = (0, 0, 1), so direct calculation gives

Z Z 2 Z 2 h1 i2
F r(t) · r0 (t) dt = qt3 dt = qt4

F · dr = = 4q
C t=0 t=0 4 t=0
2.4.2.31. Solution. (a) Parametrize C by x. When the first component

of a point on the curve is x, then the second component, y, must be x2 and
the third component, z, must be x3 . So
r(x) = xı̂ı + x2 ̂ + x3 k̂ 0≤x≤1

0 2
r (x) = ı̂ı + 2x ̂ + 3x k̂
ds p
(x) = 1 + 4x2 + 9x4
dx
and
ds p
ρ(x) (x) = 8x + 36x3 1 + 4x2 + 9x4
dx
Z Z 1
p
ρ ds = 8x + 36x3 1 + 4x2 + 9x4 dx
C 0

Substituting u = 1 + 4x + 9x4 , du = 8x + 36x3 dx, u(0) = 1, u(1) = 14,
2
14 14
√
Z Z
2 3/2
ρ ds = u du = u
C 1 3 1
2 3/2
= 14 − 1 ≈ 34.26
3
(b) Since F(x, y, z) = ∇f (x, y, z) with f (x, y, z) = x sin y + yz + 12 z 2 ,
Z
3
F · dr = f (1, 1, 1) − f (0, 0, 0) = sin 1 + ≈ 2.3415
C 2
The potential f was just guessed. Alternatively, it can be found by solving

∂f
(x, y, z) = sin y
∂x
∂f
(x, y, z) = x cos y + z
∂y
∂f
(x, y, z) = y + z
∂z
f (x, y, z) = x sin y + g(y, z)

∂g ∂g
x cos y + (y, z) = x cos y + z or (x, z) = z
∂y ∂y
which forces
g(y, z) = yz + h(z)
Finally, substituting f (x, y, z) = x sin y + yz + h(z) into the last equation
gives
y + h0 (z) = y + z or h0 (z) = z
2
z2
So h(x) = z2 + C and hence f (x, y, z) = x sin y + yz + 2 + C for any
constant C.
2.4.2.32. ∗. Solution. First, we’ll parametrize (x, y), which wraps once,
counterclockwise, aroung the circle x2 +y 2 = 1. So x(t) = cos t, y(t) = sin t,
0 ≤ t ≤ 2π works. As (x, y) wraps around the circle, z has to start at 0
t
(when t = 0) and end at 1 (when t = 2π). So z(t) = 2π works and our
parametrization is
t
r(t) = cos tı̂ı + sin t ̂ + k̂
2π
(Compare to Example 1.4.4.) With this parametrization
1
r0 (t) = − sin tı̂ı + cos t ̂ +
k̂
2π
t2
F x(t), y(t), z(t) = − sin tı̂ı + cos t ̂ + 2 k̂
4π
2
0 t
F x(t), y(t), z(t) · r (t) = 1 + 3
8π
and
2π 2π
t2
Z Z Z
F x(t), y(t), z(t) · r0 (t) dt =

F · dr = 1 + 3 dt
C 0 0 8π
1
= 2π +
3
2.4.2.33. ∗. Solution. (a) Let’s evaluate the integral directly using the
parametrization
r(x) = xı̂ı + (9 − x2 ) ̂
with −3 ≤ x ≤ 3.
Since r0 (x) = ı̂ı − 2x ̂,
dy
y dx
Z Z 3 z }| { z }| {
(x2 + y) dx + x dy = x2 + 9 − x2 +x (−2x) dx
C −3
Z 3
9 − 2x2 dx

=
−3
3
33
Z
9 − 2x2 dx = 2 27 − 2

=2
0 3
= 18
(b) In this solution, we’ll evaluate the integral directly. Label the four
sides of the square L1 , L2 , L3 and L4 as in the figure
y
(0, 1) (1, 1)
L3
L4 L2
L1
x
(1, 0)
The parametrization of L1 by arc length is r(s) = sı̂ı, 0 ≤ s ≤ 1. As

the outward pointing normal to L1 is −̂,
Z Z 1 Z 1
F · n̂ ds = F(s, 0) · (−̂) ds = (−0) ds = 0
L1 0 0
The parametrization of L2 by arc length is r(s) = ı̂ı + s ̂, 0 ≤ s ≤ 1. As

the outward pointing normal to L2 is ı̂ı,
Z Z 1 Z 1
F · n̂ ds = F(1, s) · ı̂ ds =
ı 2 ds = 2
L2 0 0
The parametrization of L3 by arc length (starting at (1, 1)) is r(s) = (1 −

s)ı̂ı + ̂, 0 ≤ s ≤ 1. As the outward pointing normal to L3 is ̂,
Z Z 1 Z 1 h i1
F · n̂ ds = F(1 − s, 1) · ̂ ds = e1−s ds = − e1−s = e − 1
L3 0 0 0
The parametrization of L4 by arc length (starting at (0, 1)) is r(s) = (1 −

s) ̂, 0 ≤ s ≤ 1. As the outward pointing normal to L4 is −ı̂ı,
Z Z 1 Z 1
F · n̂ ds = F(0, 1 − s) · (−ı̂ı) ds = (−0) ds = 0
L2 0 0
All together
Z Z Z Z Z
F · n̂ ds = F · n̂ ds + F · n̂ ds + F · n̂ ds + F · n̂ ds
C L1 L2 L3 L4
= 0 + 2 + (e − 1) + 0 = e + 1
2.4.2.34. ∗. Solution. (a) Since m = 1, Newton’s law of motion gives
a(t) = v0 (t) = F(t) = ̂ − sin t k̂
Integating gives
v(t) = t ̂ + cos t k̂ + c
for some constant vector c. Since v(0) = ı̂ı + k̂, we have c = ı̂ı so that
r0 (t) = v(t) = ı̂ı + t ̂ + cos t k̂
Integating again gives
t2
r(t) = tı̂ı + ̂ + sin t k̂ + c
2
for some (new) constant vector c. Since r(0) = ̂, we have c = ̂ so that
t2

r(t) = tı̂ı + 1 + ̂ + sin t k̂
2
(b) The particle has x = π/2 when t = π/2 and then
π2

π
r1 = r(π/2) = ı̂ı + 1 + ̂ + k̂
2 8
(c) The work done between time t = 0 and time t = π/2 is

Z π/2 Z π/2
dr
F(t) · dr = F(t) · (t) dt
0 0 dt
Z π/2
= [̂ − sin t k̂] · [ı̂ı + t ̂ + cos t k̂] dt
0
π/2 h t2 π2
Z iπ/2
1 1
= [t − sin t cos t] dt = + cos2 t = −
0 2 2 0 8 2
2.4.2.35. ∗. Solution. (a) We can parametrize L by

r(t) = x(t), y(t) = (t, t),
with t running from 2 to 1. Using this parametrization,

Z Z 1 Z 1
F x(t), y(t) · (x0 (t), y 0 (t)) dt =

F · dr = 3t , t − 1 · 1, 1 dt
L 2 2
Z 1
= (4t − 1) dt = −5
2
(b) First, we note

R that such a choice of path is even possible: if F were
conservative, then c F · dr would be −5 for every path starting at (2, 2)
and ending at (1, 1), because it would be path independent. Since ∂F ∂y = 3
1
∂x = 1 6= ∂y , by Theorem 2.4.7, F is not path-independent.

and ∂F 2 ∂F1
Solution 1:
(2, 2)
L
(1, 1)
L1
L3
x
(1, Y ) (2, Y )
L2
Let’s try a family of polygonal paths CY that consist of
• the line segment L1 from (2, 2) to (2, Y ) followed by
• the line segment L2 from (2, Y ) to (1, Y ) followed by
• the line segment L3 from (1, Y ) to (1, 1).
This is a way of characterizing a family of alternate paths withR only one
parameter, Y . We are hoping that the value of the integral CY F · dr
depends on Y and that we R can choose a specific value of Y so as to make
the value of the integral CY F · dr exactly 4.
Note that
• On L1 , x = 2 is a constant (so that dx = 0) and y runs from 2 to Y .
• On L2 , y = Y is a constant (so that dy = 0) and x runs from 2 to 1.
• On L3 , x = 1 is a constant (so that dx = 0) and y runs from Y to 1
So,
Z Z Z

F · dr = (3y dx + (x − 1)dy + (3y dx + (x − 1)dy
CY L1 L2
Z

+ (3y dx + (x − 1)dy
L3
Z Y Z 1 Z 1
= dy + 3Y dx + 0 dy
2 2 Y
= (Y − 2) + 3Y (1 − 2) = −2Y − 2
Since we want our integral to be 4, we set 4 = −2Y − 2, and find

Y = −3. That is, the path D consisting
R of line segments from (2, 2) to
(2, −3) to (1, −3) to (1, 1) gives us D F · dr = 4.
Solution 2: Choosing three straight line segments was a convenient
way to solve this, but not the only way. To emphasize this point, we show
that we also could have considered (for example) the family of parabolas
that pass through (2, 2) and (1, 1).
That is, we consider the family of functions y = ax2 + bx + c with
2 = 4a + 2b + c and 1 = a + b + c. Subtracting the equation a + b + c = 1
from the equation 4a + 2b + c = 2 (in order to eliminate c) gives
(4a + 2b + c) − (a + b + c) = (2) − (1)

=⇒ 3a + b = 1
=⇒ b = 1 − 3a
Using b = 1 − 3a,
a+b+c=1
=⇒ a + (1 − 3a) + c = 1
=⇒ c = 2a
So, the class of functions described by y = ax2 + (1 − 3a)x + 2a for some

constant a are parabolas that pass through (1, 1) and (2, 2).
y
(2, 2)
L
(1, 1)
y = ax2 + (1 − 3a)x + 2a
So, we consider paths of the form:
r(x) = x, ax2 + (1 − 3a)x + 2a

F(r(x)) = 3ax2 + 3(1 − 3a)x + 6a, x − 1

r0 (x) = 1, 2ax + 1 − 3a

F(r(x)) · r0 (x) = 3ax2 + 3(1 − 3a)x + 6a

+ 2ax2 + (1 − 3a)x − 2ax + (3a − 1)

= 5ax2 + (4 − 14a)x + (9a − 1)
So, if C is a portion of this parabola from (2, 2) to (1, 1), then

Z Z 1
5ax2 + (4 − 14a)x + (9a − 1) dx

F · dr =
C 2
1
5a 3
= x + (2 − 7a)x2 + (9a − 1)x
3 2
a
= −5
3
Since we want our integral to have value 4, we set 4 = a3 − 5, which yields
a = 27.
If we choose C to be
R the path from (2, 2) to (1, 1) along the parabola
27x2 − 80x + 54, then C F · dr = 4, as desired.
2.4.2.36. ∗. Solution 1. Let’s try a family of polygonal paths CY
(sketched below) that consist of
• the line segment L1 from (0, 0) to (0, Y ) followed by
• the line segment L2 from (0, Y ) to (2, Y ) followed by
• the line segment L3 from (2, Y ) to (2, 0).
Here Y is a parameter.
y
(0, Y ) L2 (2, Y )
L1 L3
(0, 0) (2, 0) x
R
We are hoping that the value of the integral CY F · dr depends on Y
and thatRwe can choose a specific value of Y so as to make the value of the
integral CY F · dr exactly 8. Note that
• on L1 , x = 0 is a constant (so that dx = 0) and y runs from 0 to Y

and
• on L2 , y = Y is a constant (so that dy = 0) and x runs from 0 to 2
and
• on L3 , x = 2 is a constant (so that dx = 0) and y runs from Y to 0
Since F · dr = (2y + 2) dx,
Z Z Z Z
F · dr = (2y + 2) dx + (2y + 2) dx + (2y + 2) dx
CY L1 L2 L3
Z 2
=0+ (2Y + 2) dx + 0
0
= 2(2Y + 2)
So Y = 1 does the job.

Solution 2. There’s nothing magical about the form of the path from
Solution 1. It’s just a path that’s relatively easy to describe using one con-
stant Y . To emphasize this point, we provide a solution with an alternate
path based on an ellipse.
A partial ellipse running from (0, 0) to (2, 2) can be described by r(t) =
(cos t + 1 , A sin t) for a constant A, with t running from π to 0. (To find
this: we centre a circle of radius 1 at the point (1, 0), then multiply its
y-coordinate by A.)
y
x
(2, 0)
In this case, F(r(t)) = (2A sin t + 2, 0) and r0 (t) = (− sin t, A cos t), so
F(r(t)) · r0 (t) = − sin t(2A sin 2 + 2) = −A(2 sin2 t) − 2 sin t

= −A(1 − cos 2t) − 2 sin t
Z Z 0

F · dr = A(cos 2t − 1) − 2 sin t dt
π
0
1
= A sin(2t) − t + 2 cos t
2 π
= Aπ + 4
Setting Aπ+4 = 8, we find A = π4 . So, the half-ellipse r(t) 4

R = cos t + 1 , π sin t ,
with t running from π to 0, is another path that gives C F · dr = 8.
2.4.2.37. ∗. Solution. The vector field F is conservative, with
Z y
F = ∇ϕ ϕ(x, y) = x + ỹg(ỹ) dỹ
0
Consquently, for P = (x0 , 0) and Q = (x1 , 0),

Z Z 0 Z 0
F · dr = ϕ(Q) − ϕ(P ) = x1 + ỹg(ỹ) dỹ − x0 − ỹg(ỹ) dỹ
C 0 0
= x1 − x0
Thus
Z

F · dr = |x(Q) − x(P )| = distance between P and Q

C
2.4.2.38. ∗. Solution.
2
• First notice that the vector
1 3
R R y, z) = z k̂ is conservative
field F̃(x,
(with potential 3 z ), so C1 F̃ · dr = C2 F̃ · dr for any two curves C1
and C2 from P1 to P2 (whether R or not they R are on the surface S).
Consequently, the statement “ C1 F · dr = C2 F · dr” is true if and
R R
only if the statement “ C1 (F − F̃) · dr = C2 (F − F̃) · dr” is true. So
we may replace the vector field F with the vector field
G(x, y, z) = F(x, y, z) − F̃(x, y, z) = (xz + axy 2 )ı̂ı + yz̂
• We are to consider only curves on the surface S. For any such curve
C, say parametrized by r(t) with a ≤ t ≤ b, the integral
Z Z b dr
G · dr = G r(t) · (t) dt
C a dt
depends only on the values of G on the surface S. In particular, if

another vector field H obeys H(x, y, z) = G(x, y, z), for all points
(x, y, z) on S, we have
Z Z b dr
Z b dr
G · dr = G r(t) · (t) dt = H r(t) · (t) dt
C dt dt
Za a
= H · dr
C
So we may replace G with
H(x, y, z) = G(x, y, 2 + x2 − 3y 2 )
= x(2 + x2 − 3y 2 ) + axy 2 ı̂ı + y(2 + x2 − 3y 2 )̂

= (2x + x3 − 3xy 2 + axy 2 )ı̂ı + (2y + yx2 − 3y 3 )̂
Note that H(x, y, z) is defined on all of R3 . It just happens to not

depend on z.
• The curl of H is
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × H = det 

∂x ∂y ∂z 
2x + x3 − 3xy 2 + axy 2 2y + yx2 − 3y 3 0

= 2xy − [−6xy + 2axy] k̂ = (8 − 2a)xy k̂
This is zero if a = 4. As H has continuous first order partial deriva-

tives on all of R3 , Theorem
R 2.4.8 tells
R us that, when a = 4, H is
conservative and that C1 H · dr = C2 H · dr for any two curves C1
and C2 from P1 to P2
So a = 4 does the job.
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
2 2
(1 + ax2 )ye3x − bxz cos(x2 z) xe3x x2 cos(x2 z)
= 0ı̂ı + [−bx cos(x2 z) + bx3 z sin(x2 z) − 2x cos(x2 z) + 2x3 z sin(x2 z)] ̂
2 2 2
+ [e3x + 6x2 e3x − (1 + ax2 )e3x ] k̂
2
= [−(b + 2)x cos(x2 z) + (b + 2)x3 z sin(x2 z)] ̂ + (6 − a)x2 e3x k̂
(b) For F to be conservative it is necessary that ∇ × F = 0. This is the

case when b = −2 and a = 6.
(c) For f to be a potential, when b = −2 and a = 6, we need
∂f 2
(x, y, z) = (1 + 6x2 )ye3x + 2xz cos(x2 z)
∂x
∂f 2
(x, y, z) = xe3x
∂y
∂f
(x, y, z) = x2 cos(x2 z)
∂z
Integrating the second of these equations gives
2
f (x, y, z) = xye3x + g(x, z)
Substituting this into the last equation gives

∂g
(x, z) = x2 cos(x2 z)
∂z
which integrates to
g(x, z) = sin(x2 z) + h(x)
2
Finally, substituting f (x, y, z) = xye3x + sin(x2 z) + h(x) into the first
equation gives
2 2
(1 + 6x2 )ye3x + 2xz cos(x2 z) + h0 (x) = (1 + 6x2 )ye3x + 2xz cos(x2 z)
or h0 (x) = 0
2
So h(x) = C and hence f (x, y, z) = xye3x + sin(x2 z) + C works for any
constant C.
R 2
(d) Note that the integral is C
Fa=6,b=−2 − 6x2 ye3x ı̂ı · dr. So
Z
2 2
ye3x + 2xz cos(x2 z) dx + xe3x dy + x2 cos(x2 z)dz
C
Z Z
2
= ∇ f · dr − 6 x2 ye3x dx
C C
Z 1
2
= f (1, 1, 1) − f (0, 0, 0) − 6 t3 e3t dt
0
Z 3
1
= e3 + sin 1 − ueu du with u = 3t2 , du = 6t dt
3 0
1h i3
= e3 + sin 1 − ueu − eu integration by parts
3 0
1 3 1
= e + sin 1 −
3 3
2.4.2.40. ∗. Solution. (a) Parametrize C by x. Then
r(x) = xı̂ı + x2 ̂ + x3 k̂ 0≤x≤1
0 2
r (x) = ı̂ı + 2x ̂ + 3x k̂
F r(x) · r0 (x) = (x4 − x2 )ı̂ı + (x + x3 ) ̂ + x2 k̂ · ı̂ı + 2x ̂ + 3x2 k̂

= x4 − x2 + 2x2 + 2x4 + 3x4 = x2 + 6x4

Z 1 h x3 6x5 i1
Z
2 23
x + 6x4 dx =

F · dr = + = = 1.53̇
C 0 3 5 0 15
(b) Parametrize C by x as in part (a). Then
p
ds dr
= = 1 + 4x2 + 9x4
dx dx
ds p
ρ(x, x2 , x3 ) = 8x + 36x3 1 + 4x2 + 9x4
dx
Z Z 1
p
ρ ds = 8x + 36x3 1 + 4x2 + 9x4 dx
C 0
Using the substitution u = 1 + 4x2 + 9x4 , du = (8x + 36x3 ) dx:

2 3/2 1
= 1 + 4x2 + 9x4
3

0
2 3/2
= 14 − 1 ≈ 34.26
3
(c) Since F = ∇ f with f = x sin y + yz + 21 z 2 ,
Z
3
F · dr = f (1, 1, 1) − f (0, 0, 0) = sin 1 + ≈ 2.3415
C 2
The potential f was just guessed. Alternatively, it can be found by antid-
ifferentiating:
∂f
(x, y, z) = sin y =⇒ f (x, y, z) = x sin y + ψ1 (y, z)
∂x
∂f
(x, y, z) = x cos y + z =⇒ f (x, y, z) = x sin y + yz + ψ2 (x, z)
∂y
∂f 1
(x, y, z) = y + z =⇒ f (x, y, z) = yz + z 2 + ψ3 (x, y)
∂z 2
z2
All together, f (x, y, z) = x sin y + yz + 2 + C works for any constant C.
2.4.2.41. ∗. Solution. (a) This field is conservative if and only if it

passes the screening test ∇ × F = 0. That is, if and only if,
∂F1 ∂F2 ∂F1 ∂F3 ∂F2 ∂F3
= = =
∂y ∂x ∂z ∂x ∂z ∂y
That is,
∂ ∂ 3
(Ax3 y 2 z) = z + Bx4 yz 2Ax3 yz = 4Bx3 yz

⇐⇒
∂y ∂x
∂ ∂
(Ax3 y 2 z) = 3yz 2 − x4 y 2 ⇐⇒ Ax3 y 2 = −4x3 y 2

∂z ∂x
∂ 3 ∂
z + Bx4 yz = 3yz 2 − x4 y 2 ⇐⇒ 3z 2 + Bx4 y = 3z 2 − 2x4 y

∂z ∂y
Hence only A = −4, B = −2 works.
(b) When A = −4, B = −2
F = −4x3 y 2 z ı̂ı + z 3 − 2x4 yz ̂ + 3yz 2 − x4 y 2 k̂

We find a potential function ϕ(x, y, z) for this F by antidifferentiating.

∂ϕ
(x, y, z) = −4x3 y 2 z =⇒ ϕ(x, y, z) = −x4 y 2 z + ψ1 (y, z)
∂x
∂ϕ
(x, y, z) = z 3 − 2x4 yz =⇒ ϕ(x, y, z) = yz 3 − x4 y 2 z + ψ2 (x, z)
∂y
∂ϕ
(x, y, z) = 3yz 2 − x4 y 2 =⇒ ϕ(x, y, z) = yz 3 − x4 y 2 z + ψ3 (x, y)
∂z
All together, ϕ(x, y, z) = −x4 y 2 z + yz 3 + C with C being an arbitrary
constant.
(c) I = ϕ(1, −1, 1) −
R ϕ(0, 0, 0) = −2.
(d) Note that J = C G · dr with
G = (z − 4x3 y 2 z)ı̂ı + (z 3 − x4 yz)̂ + (3yz 2 − x4 y 2 )k̂

= F + z ı̂ı + x4 yz ̂
so that
Z Z
4
J= (zı̂ı + x yz ̂ + F) · dr = −2 + (zı̂ı + x4 yz ̂) · dr
C C
dr
Parametrize C by r(x) = xı̂ı −x ̂ +x2 k̂ with 0 ≤ x ≤ 1. As dx = ı̂ı −̂ +2x k̂
Z Z 1
(zı̂ı + x4 yz ̂) · dr = (x2ı̂ı − x7 ̂) · ı̂ı − ̂ + 2x k̂ dx

C 0
Z 1
= (x2 + x7 ) dx
0
1 1
= +
3 8
11
=
24
11 37
=⇒ J = −2 + =− ≈ −1.5417
24 24
R
(e) T is a closed path and F is conservative, so T F · dr = 0. Let T1
be the line segment from (1, 0, 0) to (0, 1, 0), T2 be the line segment from
(0, 1, 0) to (0, 0, 1) and T3 be the line segment from (0, 0, 1) to (1, 0, 0).
z
(0,0,1)
T2
T3
(0,1,0)
y
(1,0,0)
T1
x
R
On T1 , z = 0, so T1 zı̂ı · dr = 0. On T2 , x = 0, so ı̂ı · dr = dx = 0 and
R
T2
zı̂ı · dr = 0. Parametrize T3 by r(t) = tı̂ı + (1 − t)k̂, 0 ≤ t ≤ 1. Then
dr
dt = ı̂ − k̂ and the z-coordinate of the path is parametrized by 1 − t. So,
ı
zı̂ı dr
Z Z Z 1 z }| { z }| {
(zı̂ı + F) · dr = zı̂ı · dr = (1 − t)ı̂ı · (ı̂ı − k̂) dt
T T3 0
Z 1
1
= (1 − t) dt =
0 2
2.4.2.42. ∗. Solution. (a) By Newton’s law of motion
ma = F =⇒ 2v0 (t) = 4t , 6t2 , −4t =⇒ v0 (t) = 2t , 3t2 , −2t

So
Z t Z t
0
2u , 3u2 , −2u du

v(t) = v(0) + v (u) du = (0, 0, 0) +
0 0
= t2 , t3 , −t2

(b) From part (a), r0 (t) = v(t) = t2 , t3 , −t2 . So

Z t Z t
0
u2 , u3 , −u2 du

r(t) = r(0) + r (u) du = (1, 2, 3) +
0 0
3
t4 t3

3 4 3
t
= (1, 2, 3) + t /3 , t /4 , −t /3 = + 1, + 2, − + 3
3 4 3
(c) From parts (a) and (b)
p
|r0 (t)| = t2 (1, t, −1) = t2 2 + t2

and
 
ı̂ı ̂ k̂
r0 (t) × r00 (t) = det t2 t3 −t2 
2t 3t2 −2t
= − 2t4 + 3t4 ı̂ı − − 2t3 + 2t3 ̂ + 3t4 − 2t4 k̂

= t4 ı̂ı + t4 k̂
√
=⇒ r (t) × r00 (t) = 2 t4
0
The curvature is (see \S1.5)

√ 4
|r0 (t) × r00 (t)| 2t
κ(t) = = √ 3
|r0 (t)|3 t 2 + t2
2
√
2
= 3/2
t2 (2 + t2 )
R
(d) W = F · dr:
Z t=T Z T
dr
F(t) · dr = F(t) · (t) dt
t=0 0 dt
Z T
4t , 6t2 , −4t · t2 , t3 , −t2 dt

=
0
Z T
8t3 + 6t5 dt = 2T 4 + T 6

=
0

√
4 2 √
3/2 4 2 3/2

r(t) = t , t , t(2 − t)
√3 √3
v(t) = r0 (t) = 2 2t1/2 , 2 2t1/2 , 2 − 2t

p p
|v| = 8t + 8t + 4 − 8t + 4t2 = 4(1 + 2t + t2 )
= 2|1 + t| = 2(1 + t)
So the distance travelled is

Z 2 2
t2 i2
Z h
|v(t)| dt = 2(1 + t) dt = 2 t + =8
0 0 2 0
(b) As
√ √ √
v(t) = r0 (t) = 2 2t1/2 , 2 2t1/2 , 2 − 2t

v(1) = 2 2 1, 1, 0
√ −1/2 √ −1/2 √ √
a(t) = v0 (t) =

2t , 2t , −2 a(1) = 2 1, 1, − 2
√ √
v(1) × a(1) = 4 − 2, 2, 0 |v(1)| = 4
the curvature
|v(1) × a(1)| 8 1
κ(1) = 3
= 3 =
|v(1)| 4 8
(c) G = ∇ϕ with ϕ(x, y, z) = −M gz, so that gravity is conservative.

The work done is
16 16
ϕ r(2) − ϕ r(0) = ϕ , , 0 − ϕ(0, 0, 0) = 0
3 3
Friction is not conservative, so we have to compute the work long hand.
Z 2 Z 2 Z 2
dr
F · dr = F(t) · (t) dt = − |v(t)|2 v(t) · v(t) dt
0 0 dt 0
Z 2
=− |v(t)|4 dt
0
Z 2 2
4 16
= −2 (1 + t)4 dt = − (1 + t)5

0 5 0
16
= − (35 − 1) ≈ −774.4
5
(d) Solution 1:
We know, from Theorem 1.3.3.c in the text, that

d2 s ds 2
a(t) = 2 T̂ + κ N̂
dt dt
ds
We have also been told that, at the apex, N̂ = −k̂ and that dt (t) = 3 for
2
all t. So ddt2s = 0. As κ = 18 at the apex
1 2 9
a(1) = 0T̂ + 3 (−k̂) = − k̂
8 8
Solution 2:
The bird follows the parametrized path
√ √
4 2 3/2 4 2 3/2
r(u) = u , u , u(2 − u)
3 3
This is the same path as the plane, but the parameter u is not time. Let’s
denote by R(t) the position of the bird at time t. At time t the bird is at
some point on the parametrized path, so there is some u(t) with

R(t) = r u(t)
dr
We saw in part (a) that du = 2(1 + u). Since the bird always has
speed 3,
dR dr du du
3= (t) = (u(t)) = 2 1 + u(t)

dt du dt dt
du 3
=⇒ =
dt 2(1 + u(t))
2
d u 3 du 9
=⇒ =− =−
dt2 2(1 + u(t))2 dt 4(1 + u(t))3
du 3 d2 u 9
At the apex u = 1 so that dt = 4 and dt2 = − 32 . The bird’s acceleration
is
d2 R d dR d dr du
(t) = (t) = (u(t)) (t)
dt2 dt dt dt du dt
d2 r du 2 dr d2 u
= +
du2 dt du dt2
From part (a)
dr √ √
= 2 2u1/2 , 2 2u1/2 , 2 − 2u

du
d2 r √ −1/2 √ −1/2
2
= 2u , 2u , −2
du
At the apex, when u = 1,
dr √ √
= 2 2, 2 2, 0
du
d2 r √ √
2
= 2, 2, −2
du
and the acceleration is
d2 R d2 r du 2 dr d2 u
= +
dt2 du2 dt du dt2
3 · Surface Integrals √ √ 3 2 √ √ 9
= 2, 2, −2 + 2 2, 2 2, 0 −
3.1 · Parametrized 9
Surfaces
4 32
3.1.1 · Exercises = 0, 0, −
8
3.1.1.1. Solution. This parametrization is almost trivial. We know it

will have the form r(x, y) = ψ1 (x, y)ı̂ı + ψ2 (x, y)̂ + ψ3 (x, y)k̂ where ψ1 gives
the x-component (i.e. x), ψ2 gives the y-component (i.e. y), and ψ3 gives
the z-component (i.e. ex+1 + xy). So, r(x, y) = xı̂ı + y̂ + (ex+1 + xy)k̂
3.1.1.2. ∗. Solution. Our parametrization is
x(u, v) = u + v
y(u, v) = u2 + v 2
z(u, v) = u − v
• Adding x(u, v) and z(u, v) gives x(u, v) + z(u, v) = 2u.

• Subtracting z(u, v) from x(u, v) gives x(u, v) − z(u, v) = 2v.
So u = 21 x(u, v) + z(u, v) and v = 12 x(u, v) − z(u, v) . So on our surface

1 2 1 2
y(u, v) = u2 + v 2 = x(u, v) + z(u, v) + x(u, v) − z(u, v)
4 4
1 1
= x(u, v)2 + z(u, v)2
2 2
All points of our surface lie on 2y = x2 + z 2 . This is a parabolic bowl:
• no points have y < 0 and
• the y = Y (with Y > 0) cross-section is the circle x2 + z 2 = 2Y ,
y=Y
• the x = 0 cross-section is the parabola 2y = z 2 , x = 0
• the z = 0 cross-section is the parabola 2y = x2 , z = 0
z
3.1.1.3. ∗. Solution. Note that, since x2 + y 2 = 1 + 2z 2 on S, the

condition z ≥ 1 is equivalent to x2 + y 2 ≥ 3, z ≥ 0. So the hyperboloid is

(x, y, z) x2 + y 2 = 1 + 2z 2 , 3 ≤ x2 + y 2 ≤ 9, z ≥ 0 .
(a) No. Under this parametrization, the condition 3 ≤ x2 + y 2 ≤ 9 is
3 ≤ u2 + v 2 ≤ 9, not 2 ≤ u2 + v 2 ≤ 9.
q Yes. Under this parametrization, x = u sin v, y = −u cos v and
(b)
u2
z= 2 − 12 . So

u2 1
• x2 + y 2 − 2z 2 = u2 − 2 2 − 2 = 1, as desired.
• The condition x2 + y 2 ≤ 9 is equivalent to u ≤ 3, since u ≥ 0.

√
• The condition x2 + y 2 ≥ 3 is equivalent to u ≥ 3, since u ≥ 0.
r
u2 1
• z= − ≥0
2 2
√ √
(c) Yes. Under this parametrization, x = 1 + 2v 2 cos u, y = 1 + 2v 2 sin u
and z = v. So
• x2 + y 2 − 2z 2 = 1 + 2v 2 − 2v 2 = 1, as desired.
• The condition x2 + y 2 ≤ 9 is equivalent to 1 + 2v 2 ≤ 9, which is
equivalent to v ≤ 2, since v ≥ 0.
• The condition x2 + y 2 ≥ 3 is equivalent to 1 + 2v 2 ≥ 3, which is
equivalent to v ≥ 1, since v ≥ 0.
• z=v≥0
√ √
p Under this parametrization, x =
(d) Yes. 1 + u sin v, y = 1 + u cos v
and z = u/2. So
• x2 + y 2 − 2z 2 = 1 + u − 2(u/2) = 1, as desired.
• The condition x2 + y 2 ≤ 9 is equivalent to 1 + u ≤ 9, which is
equivalent to u ≤ 8.
• The condition x2 + y 2 ≥ 3 is equivalent to 1 + u ≥ 3, which is
equivalent to u ≥ 2.
p
• z = u/2 ≥ 0
√ √
p No. Under this parametrization, x = u cos v, y = − u sin v and
(e)
z = (u + 1)/2. So
• x2 + y 2 − 2z 2 = u − 2(u + 1)/2 = −1, not +1

3.1.1.4. ∗. Solution. (a) No. z = sin φ sin θ is negative when 0 < φ ≤ π4 ,
π < θ < 2π. 2 p 2
(b) Yes. Note that x2 + − y + 2 − x2 − y 2 = 2 and that, for
p
x2 + y 2 ≤ 1, we have both x2 + (−y)2 ≤ 1 and 2 − x2 − y 2 ≥ √ 0.
2 2 2
(c) No. (u sin√ θ) + (u cos θ) = u > 1 for 1 < u ≤ 2. Also 2 − u2 is
not defined for 2 < u ≤ 2.
(d) Yes. Note that
√ 2 √ 2 √ 2
• 2 sin φ cos θ + 2 sin φ sin θ + 2 cos φ = 2
√
• For 0 ≤ φ ≤ π4 , we have z = 2 cos φ > 0.
√
• As φ runs from 0 to π4 , r(φ) = 2 sin φ runs from 0 to 1, so that
x = r(φ) cos θ , y = r(φ) sin θ covers all of x2 + y 2 ≤ 1 as φ runs
from 0 to π4 and θ runs from 0 to 2π.
(e) Yes. Note that

p 2 p 2 2
• − 2 − z 2 sin φ + 2 − z 2 cos φ + z = 2
√
• For 1 ≤ z ≤ 2, we have obviously have z > 0.
√ √
• As z runs from 1 to 2, r(z) = 2 − z 2 runs from 1 to 0, so that
x = −r(z) sin φ , y = r(z) cos φ covers all of x2 + y 2 ≤ 1 as z runs
√
from 1 to 2 and φ runs from 0 to 2π.
3.1.1.5. ∗. Solution. (a) No. When u = v = 0, z = 4 is not between 0

and 1. √ √
(b) Yes. Note that when x = 4 − u cos v, y = 4 − u sin v and z = u
with 0 ≤ u ≤ 1, 0 ≤ v ≤ 2π,
• z + x2 + y 2 = 4
• 0≤z=u≤1
• For each fixed z = u between 0 and 1, (x, y) runs once around the
circle x2 + y 2 = 4 − z = 4 − u as v runs from 0 to 2π.
2
√ (c) Yes. Note that when x = u cos v, y = u sin v and z = 4 − u , with
3 ≤ u ≤ 2, 0 ≤ v ≤ 2π
• z + x2 + y 2 = 4
• 0 ≤ z = 4 − u2 ≤ 1
• For each fixed z = 4 − u2 between 0 and 1, (x, y) runs once around
the circle x2 + y 2 = 4 − z = u2 as v runs from 0 to 2π.
3.1.1.6. ∗. Solution. First note that,

• for A, B and C, r(θ, φ) = x(θ, φ)ı̂ı + y(θ, φ) ̂ + z(θ, φ) k̂ obeys
x(θ, φ)2 + y(θ, φ)2 + z(θ, φ)2 = 4
and so lies on S1
• for D, E and F, r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ obeys
x(θ, z)2 + y(θ, z)2 = 4 − z(θ, z)2
and so lies on S1
• for G, H and I, r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ obeys
x(θ, z)2 + y(θ, z)2 = z(θ, z)2
and so lies on S3
• for J, K and L, r(x, y) = x(x, y)ı̂ı + y(x, y) ̂ + z(x, y) k̂ obeys
x(x, y)2 + y(x, y)2 = z(x, y)2
and so lies on S3
(a) To get a part of S1 , we need to use one of the parametrizations A, B,
C, D, E, F. In the cases of A, B, C, for r(θ, φ) = x(θ, φ)ı̂ı +y(θ, φ) ̂ +z(θ, φ) k̂
to lie inside S2 we need (recalling that all points of S1 have z(θ, φ) ≥ 0 and
hence 0 ≤ φ ≤ π2 )
1
x(θ, φ)2 + y(θ, φ)2 ≤ 1 ⇐⇒ 4 sin2 φ ≤ 1 ⇐⇒ sin φ ≤
2
π
⇐⇒ 0 ≤ φ ≤
6
In the cases of D, E, F, for r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ to lie
inside S2 we need (recalling that all points of S1 have z(θ, z) ≥ 0 and hence
z ≥ 0)
√
x(θ, z)2 + y(θ, z)2 ≤ 1 ⇐⇒ 4 − z 2 ≤ 1 ⇐⇒ z ≥ 3
So parametrizations A and F work.

(b) To get a part of S1 , we need to use one of the parametrizations A, B,
C, D, E, F. In the cases of A, B, C, for r(θ, φ) = x(θ, φ)ı̂ı +y(θ, φ) ̂ +z(θ, φ) k̂
to lie inside S3 we need (recalling that all points of S1 have z(θ, φ) ≥ 0 and
hence 0 ≤ φ ≤ π2 )
x(θ, φ)2 + y(θ, φ)2 ≤ z(θ, φ)2 ⇐⇒ 4 sin2 φ ≤ 4 cos2 φ

⇐⇒ tan φ ≤ 1
π
⇐⇒ 0 ≤ φ ≤
4
In the cases of D, E, F, for r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ to lie
inside S3 we need (recalling that all points of S1 have z(θ, z) ≥ 0 and hence
z ≥ 0)
√
x(θ, z)2 + y(θ, z)2 ≤ z(θ, z)2 ⇐⇒ 4 − z 2 ≤ z 2 ⇐⇒ z ≥ 2
So parametrizations B and E work.

(c) To get a part of S3 , we need to use one of the parametrizations G, H,
I, J, K, L. In the cases of G, H, I, for r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂
to lie inside S2 we need (recalling that all points of S3 have z ≥ 0)
x(θ, z)2 + y(θ, z)2 ≤ 1 ⇐⇒ z 2 ≤ 1 ⇐⇒ 0 ≤ z ≤ 1
In the cases of J, K, L, for r(x, y) = x(x, y)ı̂ı + y(x, y) ̂ + z(x, y) k̂ to lie
inside S3 we need
x(x, y)2 + y(x, y)2 ≤ 1 ⇐⇒ x2 + y 2 ≤ 1
So parametrizations G and J work.

(d) To get a part of S3 , we need to use one of the parametrizations G, H,
I, J, K, L. In the cases of G, H, I, for r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂
to lie inside S1 we need (recalling that all points of S3 have z ≥ 0)
√
x(θ, z)2 + y(θ, z)2 + z(θ, z)2 ≤ 4 ⇐⇒ 2z 2 ≤ 4 ⇐⇒ 0 ≤ z ≤ 2
In the cases of J, K, L, for r(x, y) = x(x, y)ı̂ı + y(x, y) ̂ + z(x, y) k̂ to lie
inside S3 we need
x(x, y)2 + y(x, y)2 + z(x, y)2 ≤ 4 ⇐⇒ 2x2 + 2y 2 ≤ 4
So parametrizations H and L work.

3.1.1.7. Solution. (a) In the sketch below, the point (x, y, z) deviates
q centre (2, 2, 4) by sin θ units in the k̂ direction, and by cos θ units
from the
in the √12 (ı̂ı + ̂) direction. So, (x, y, z) = (2 + √12 cos θ, 2 + √12 cos θ, 4 +
sin θ).
(x, y, z)
θ √1 (ı̂ı + ̂)
4 2
(2, 2, 4)
y
2
2
x
So, we can parametrize the circle as (x, y, z) = (2+ √12
cos θ, 2+ √12 cos θ, 4+
sin θ), with 0 ≤ θ ≤ 2π.
Remark: it’s easy to check that this equation satisfies the two prop-
erties we desire. Since the x- and y coordinates match, it’s in the plane
x = y. To check that it’s a circle centred at (2, 2, 4), we note the distance
from (x, y, z) to (2, 2, 4) is:
p
d = (x − 2)2 + (y − 2)2 + (z − 4)2
s 2 2
1 1
= √ cos θ + √ cos θ + (sin θ)2
2 2
r
1 1 p
= cos2 θ + cos2 θ + sin2 θ = cos2 θ + sin2 θ
2 2
=1
So, our points all have distance one from the same point — that is, they
lie on a circle of radius 1.
(b) Consider a point (x, y, z) = (2 + √12 cos θ, 2 + √12 cos θ, 4 + sin θ),
rotating φ radians about the line x = y = 4.
r(t) = (4, 4, t)
R
(x, y, z)
φ
(2, 2, 4)
The new position of the point has the same height, z = 4 + sin θ. Its
distance from the line x = y = 4 is also preserved:
p
R = (x − 4)2 + (y − 4)2 + (z − z)2
s
1 1
= ( √ cos θ − 2)2 + ( √ cos θ − 2)2 + 0)
2 2
√
= cos θ − 2 2
The circle traced out by a point (x, y, z) = (2+ √12 cos θ, 2+ √12 cos θ, 4+
√
sin θ) on the circle√is centred
√ at (4, 4, z) with radius√ 2(4 − √ x), so it has
equation x = 4 + 2(2 − 2 cos θ) cos φ, y = 4 + 2(2 − 2 cos θ) sin φ,
z = 4 sin θ.

3.2.1 · Exercises
3.2.1.1. Solution. Write F (x, y, z) = x2 +y 2 +(z−1)2 −1 and G(x, y, z) =
x2 + y 2 + (z + 1)2 − 1. Let S1 denote the surface F (x, y, z) = 0 and S2
denote the surface G(x, y, z) = 0. First note that F (0, 0, 0) = G(0, 0, 0) = 0
so that the point (0, 0, 0) lies on both S1 and S2 . The gradients of F and
G are

∂F ∂F ∂F
∇ F (x, y, z) = (x, y, z) , (x, y, z) , (x, y, z)
∂x ∂y ∂z
= (2x , 2y , 2(z − 1))

∂G ∂G ∂G
∇ G(x, y, z) = (x, y, z) , (x, y, z) , (x, y, z)
∂x ∂y ∂z
= (2x , 2y , 2(z + 1))
In particular,
∇ F (0, 0, 0) = (0, 0, −2) ∇ G(0, 0, 0) = (0, 0, 2)
so that the vector k̂ = − 12 ∇ F (0, 0, 0) = 12 ∇ G(0, 0, 0) is normal to both

surfaces at (0, 0, 0). So the tangent plane to both S1 and S2 at (0, 0, 0) is
k̂ · (x − 0, y − 0, z − 0) = 0 or z=0
Denote by P the plane z = 0. Thus S1 is tangent to P at (0, 0, 0) and P

is tangent to S2 at (0, 0, 0). So it is reasonable to say that S1 and S2 are
tangent at (0, 0, 0).
3.2.1.2. Solution. Denote by S the surface G(x, y, z) = 0 and by C the
parametrized curve r(t) = x(t), y(t), z(t) . To start, we’ll find the tangent
plane to S at r0 and the tangent line to C at r0 .
• The tangent vector to C at r0 is (x0 (t0 ) , y 0 (t0 ) , z 0 (t0 )), so the para-
metric equations for the tangent line to C at r0 are
x − x0 = tx0 (t0 ) y − y0 = tx0 (t0 ) z − z0 = tx0 (t0 ) (E1)

∂G ∂G ∂G

• The gradient ∂x x0 , y 0 , z 0 , ∂y x0 , y 0 , z 0 , ∂z x0 , y 0 , z 0 is
a normal vector to the surface S at (x0 , y0 , z0 ). So the tangent plane
to the surface S at (x0 , y0 , z0 ) is

∂G ∂G ∂G
x0 , y 0 , z 0 , x0 , y 0 , z 0 , x0 , y 0 , z 0 ·
∂x ∂y ∂z
(x − x0 , y − y0 , z − z0 ) = 0
or
∂G ∂G
x0 , y0 , z0 (x − x0 ) + x0 , y0 , z0 (y − y0 )
∂x ∂y
∂G
+ x0 , y0 , z0 (z − z0 ) = 0
∂z
(E2)
the tangent vector (x0 (t0 ) , y 0 (t0 ) , z 0 (t0 )) to C at r

Next, we’ll show that 0
and the normal vector ∂G
∂G ∂G
∂x x 0 , y 0 , z 0 , ∂y x 0 , y 0 , z 0 , ∂z x 0 , y 0 , z 0
to S at r0 are perpendicular to each other. To do so, we observe that, for
every t, the point x(t), y(t), z(t) lies on the surface G(x, y, z) = 0 and so
obeys

G x(t), y(t), z(t) = 0
Differentiating this equation with respect to t gives, by the chain rule,

d
0= G x(t), y(t), z(t)
dt
∂G ∂G
x(t) , y(t) , z(t) x0 (t) + x(t) , y(t) , z(t) y 0 (t)

=
∂x ∂y
∂G
x(t) , y(t) , z(t) z 0 (t)

+
∂z
Then setting t = t0 gives
∂G ∂G
x0 , y0 , z0 x0 (t0 ) + x0 , y0 , z0 y 0 (t0 )

∂x ∂y
∂G
x0 , y0 , z0 z 0 (t0 ) = 0

+ (E3)
∂z
Finally, we are in a position to show that if (x, y, z) is any point on the
tangent line to C at r0 , then (x, y, z) is also on the tangent plane to S at
r0 . As (x, y, z) is on the tangent line to C at r0 then there is a t such that,
by (E1),
∂G ∂G
x0 , y0 , z0 {x − x0 } + x0 , y0 , z0 {y − y0 }
∂x ∂y
∂G
+ x0 , y0 , z0 {z − z0 }
∂z
∂G 0 ∂G
x0 , y0 , z0 t y 0 (t0 )

= x0 , y0 , z0 t x (t0 ) +
∂x ∂y
∂G
x0 , y0 , z0 t z 0 (t0 )

+
∂z
h ∂G ∂G
x0 , y0 , z0 x0 (t0 ) + x0 , y0 , z0 y 0 (t0 )

=t
∂x ∂y
∂G i
x0 , y0 , z0 z 0 (t0 )

+
∂z
=0
by (E3). That is, (x, y, z) obeys the equation, (E2), of the tangent plane
to S at r0 and so is on that tangent plane. So the tangent line to C at r0
is contained in the tangent plane to S at r0 .
3.2.1.3. Solution. By part (b) of Theorem 3.2.1,
n = −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂
is normal to the surface at (x0 , y0 , z0 ). So the parametric equations of the

normal line are

(x − x0 , y − y0 , z − z0 ) = t − fx (x0 , y0 ) , −fy (x0 , y0 ) , 1 or
x = x0 − tfx (x0 , y0 ) y = y0 − tfy (x0 , y0 ) z = f (x0 , y0 ) + t
3.2.1.4. Solution. Use S1 to denote the surface F (x, y, z) = 0, S2 to
denote the surface G(x, y, z) = 0 and C to denote the curve of intersection
of S1 and S2 .
• Since C is contained in S1 , the tangent line to C at (x0 , y0 , z0 ) is
contained in the tangent plane to S1 at (x0 , y0 , z0 ), by Q[3.2.1.2]. In
particular, any tangent vector, t, to C at (x0 , y0 , z0 ) must be perpen-
dicular to ∇ F (x0 , y0 , z0 ), the normal vector to S1 at (x0 , y0 , z0 ).

contained in the tangent plane to S2 at (x0 , y0 , z0 ), by Q[3.2.1.2]. In
particular, any tangent vector, t, to C at (x0 , y0 , z0 ) must be perpen-
dicular to ∇ G(x0 , y0 , z0 ), the normal vector to S2 at (x0 , y0 , z0 ).
So any tangent vector to C at (x0 , y0 , z0 ) must be perpendiular to both
∇ F (x0 , y0 , z0 ) and ∇ G(x0 , y0 , z0 ). One such tangent vector is
t = ∇ F (x0 , y0 , z0 ) × ∇ G(x0 , y0 , z0 )
(Because the vectors ∇ F (x0 , y0 , z0 ) and ∇ G(x0 , y0 , z0 ) are nonzero and

not parallel, t is nonzero.) So the normal plane in question passes through
(x0 , y0 , z0 ) and has normal vector n = t. Consquently, the normal plane is
n · (x − x0 , y − y0 , z − z0 ) = 0
where n = t = ∇ F (x0 , y0 , z0 ) × ∇ G(x0 , y0 , z0 )
3.2.1.5. Solution. Use S1 to denote the surface z = f (x, y), S2 to denote
the surface z = g(x, y) and C to denote the curve of intersection of S1 and
S2 .
contained in the tangent plane to S1 at (x0 , y0 , z0 ), by Q[3.2.1.2].
In particular, any tangent vector, t, to C at (x0 , y0 , z0 ) must be
perpendicular to −fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂, the normal vector to
S1 at (x0 , y0 , z0 ). (See part (b) of Theorem 3.2.1.)

contained in the tangent plane to S2 at (x0 , y0 , z0 ), by Q[3.2.1.2].
In particular, any tangent vector, t, to C at (x0 , y0 , z0 ) must be
perpendicular to −gx (x0 , y0 )ı̂ı − gy (x0 , y0 ) ̂ + k̂, the normal vector to
S2 at (x0 , y0 , z0 ).
So any tangent vector to C at (x0 , y0 , z0 ) must be perpendicular to both of
the vectors −fx (x0 , y0 )ı̂ı −fy (x0 , y0 ) ̂ + k̂ and −gx (x0 , y0 )ı̂ı −gy (x0 , y0 ) ̂ + k̂.
One such tangent vector is

t = − fx (x0 , y0 )ı̂ı − fy (x0 , y0 ) ̂ + k̂ × − gx (x0 , y0 )ı̂ı − gy (x0 , y0 ) ̂ + k̂
 
ı̂ı ̂ k̂
= det −fx (x0 , y0 ) −fy (x0 , y0 ) 1 
−gx (x0 , y0 ) −gy (x0 , y0 ) 1

= gy (x0 , y0 ) − fy (x0 , y0 ) , fx (x0 , y0 ) − gx (x0 , y0 ) ,

fx (x0 , y0 )gy (x0 , y0 ) − fy (x0 , y0 )gx (x0 , y0 )
So the tangent line in question passes through (x0 , y0 , z0 ) and has direction
vector d = t. Consquently, the tangent line is
(x − x0 , y − y0 , z − z0 ) = t d
or

x = x0 + t gy (x0 , y0 ) − fy (x0 , y0 )

y = y0 + t fx (x0 , y0 ) − gx (x0 , y0 )

z = z0 + t fx (x0 , y0 )gy (x0 , y0 ) − fy (x0 , y0 )gx (x0 , y0 )
3.2.1.6. ∗. Solution. We are going to use part (b) of Theorem 3.2.1.

To do so, we need the first order derivatives of f (x, y) at (x, y) = (−1, 1).
So we find them first.
2xy x2 y(4x3 ) 2 4 2
fx (x, y) = − 2 fx (−1, 1) = − + 2 = −
x4 + 2y 2 (x4 + 2y 2 ) 3 3 9
x2 x2 y(4y) 1 4 1
fy (x, y) = 4 − 2 fy (−1, 1) = − 2 =−
x + 2y 2 (x4 + 2y 2 ) 3 3 9
So (2/9 , 1/9 , 1) is a normal vector to the surface at (−1, 1, 1/3) and the
tangent plane is
2 1 1
(x + 1) + (y − 1) + z − =0
9 9 3
2 1 2 1 1 2
x+ y+z =− + + =
9 9 9 9 3 9
or 2x + y + 9z = 2.
3.2.1.7. ∗. Solution. The equation of the given surface is of the form
G(x, y, z) = 9 with G(x, y, z) = √ 2 272 2 . So, by part (c) of Theorem
x +y +z +3
3.2.1, a normal vector to the surface at (2, 1, 1) is

1 27
∇ G(2, 1, 1) = − 2x , 2y , 2z
2 (x2 + y 2 + z 2 + 3)3/2
(x,y,z)=(2,1,1)
= −(2 , 1 , 1)
and the equation of the tangent plane is
−(2 , 1 , 1) · (x − 2 , y − 1 , z − 1) = 0 or 2x + y + z = 6
3.2.1.8. ∗. Solution. We may use G(x, y, z) = xyz 2 + y 2 z 3 − 3 − x2 = 0
as an equation for the surface. Note that (−1, 1, 2) really is on the surface
since
G(−1, 1, 2) = (−1)(1)(2)2 + (1)2 (2)3 − 3 − (−1)2 = −4 + 8 − 3 − 1 = 0

By part (c) of Theorem 3.2.1, since
Gx (x, y, z) = yz 2 − 2x Gx (−1, 1, 2) = 6
2 3
Gy (x, y, z) = xz + 2yz Gy (−1, 1, 2) = 12
2 2
Gz (x, y, z) = 2xyz + 3y z Gz (−1, 1, 2) = 8
one normal vector to the surface at (−1, 1, 2) is ∇ G(−1, 1, 2) = (6 , 12 , 8)

and an equation of the tangent plane to the surface at (−1, 1, 2) is
(6 , 12 , 8) · (x + 1 , y − 1 , z − 2) = 0 or 6x + 12y + 8z = 22
or
3 3 11
z =− x− y+
4 2 4
3.2.1.9. ∗. Solution. (a) The surface is G(x, y, z) = z−x2 +2xy−y 2 = 0.
When x = a and y = 2a and (x, y, z) is on the surface, we have z =
a2 − 2(a)(2a) + (2a)2 = a2 . So, by part (c) of Theorem 3.2.1, a normal
vector to this surface at (a, 2a, a2 ) is

∇ G(a, 2a, a2 ) = (−2x + 2y , 2x − 2y , 1) = (2a , −2a , 1)

(x,y,z)=(a,2a,a2 )
and the equation of the tangent plane is
(2a , −2a , 1) · (x − a , y − 2a , z − a2 ) = 0 or 2ax − 2ay + z = −a2
(b) The two planes are parallel when their two normal vectors, namely
(2a , −2a , 1) and (1 , −1 , 1), are parallel. This is the case if and only if
a = 21 .
3.2.1.10. ∗. Solution. A plane is determined by one point on the plane
and one vector perpendicular to the plane. We are told that (8, 1, 5) is
on the plane, so it suffices to find a normal vector. The given surface is
parametrized by
r(u, v) = 2u2 ı̂ı + v 2 ̂ + (u2 + v 3 ) k̂
so the vectors
∂r
(u, v) = 4u , 0 , 2u
∂u
∂r
(u, v) = 0 , 2v , 3v 2

∂v
are tangent to S at r(u, v). Note that r(2, 1) = (8, 1, 5). So
∂r
(2, 1) = 8 , 0 , 4
∂u
∂r
(2, 1) = 0 , 2 , 3
∂v
are tangent to S at r(2, 1) = (8, 1, 5) and
∂r ∂r
(2, 1) × (2, 1) = 8 , 0 , 4 × 0 , 2 , 3
∂u ∂v
 
ı̂ı ̂ k̂
= det 8 0 4 
0 2 3
= − 8 , −24 , 16)
1
or −8 − 8 , −24 , 16) = 1 , 3 , −2) is normal to S at (8, 1, 5). So the
tangent plane is

(1, 3, −2) · (x, y, z) − (8, 1, 5) = 0 or x + 3y − 2z = 1
3.2.1.11. ∗. Solution. To find the tangent plane we have to find a
normal vector to the surface at (2, 2, 0). Since
∂r
= 1 , 2u , 1
∂u
∂r
= 1 , 2v , −1
∂v
a normal vector to the surface at r(u, v) is
 
ı̂ı ̂ k̂
∂r ∂r
× = det 1 2u 1 
∂u ∂v
1 2v −1

= − 2u − 2v , 2 , 2v − 2u
As r(u, v) = (2, 2, 0) when (the x-coordinate) u + v = 2 and (the z-

coordinate) u − v = 0, i.e when u = v = 1, a normal vector to the surface
at (2, 2, 0) = r(1, 1) is
(−4, 2, 0) or (−2, 1, 0)
and the equation of the specified tangent plane is
−2(x − 2) + (y − 2) + 0z = 0 or y = 2x − 2
3.2.1.12. ∗. Solution. The first order partial derivatives of f are
4xy 8
fx (x, y) = − 2 fx (−1, 2) =
(x2
+ y2 ) 25
2 4y 2 2 16 6
fy (x, y) = 2 − 2 fy (−1, 2) = − =−
x + y2 (x2 + y 2 ) 5 25 25
So, by part (b) of Theorem 3.2.1, a normal vector to the surface at (x, y) =
8 6
(−1, 2) is ( 25 , − 25 , −1). As f (−1, 2) = 54 , the tangent plane is
8 6 4
, − , −1 · x + 1 , y − 2 , z − =0
25 25 5
8 6 8
or x− y−z =−
25 25 5
and the normal line is
4 8 6
(x, y, z) = − 1, 2, +t , − , −1
5 25 25
3.2.1.13. ∗. Solution. A normal vector to the surface x2 + 9y 2 + 4z 2 =
17 at the point (x, y, z) is (2x , 18y , 8z). A normal vector to the plane
x − 8z = 0 is (1 , 0 , −8). So we want (2x , 18y , 8z) to be parallel to
(1 , 0 , −8), i.e. to be a nonzero constant times (1 , 0 , −8). This is the
case whenever y = 0 and z = −2x with x 6= 0. In addition, we want
(x, y, z) to lie on the surface x2 + 9y 2 + 4z 2 = 17. So we want y = 0,

z = −2x and
17 = x2 + 9y 2 + 4z 2 = x2 + 4(−2x)2 = 17x2 =⇒ x = ±1
So the allowed points are ±(1, 0, −2).

3.2.1.14. ∗. Solution. The equation of S is of the form G(x, y, z) =
x2 + 2y 2 + 2y − z = 1. So one normal vector to S at the point (x0 , y0 , z0 ) is
∇ G(x0 , y0 , z0 ) = 2x0 ı̂ı + (4y0 + 2) ̂ − k̂
and the normal line to S at (x0 , y0 , z0 ) is
(x, y, z) = (x0 , y0 , z0 ) + t(2x0 , 4y0 + 2 , −1)
For this normal line to pass through the origin, there must be a t with
(0, 0, 0) = (x0 , y0 , z0 ) + t(2x0 , 4y0 + 2 , −1)
or
x0 + 2x0 t = 0 (E1)
y0 + (4y0 + 2)t = 0 (E2)
z0 − t = 0 (E3)
Equation (E3) forces t = z0 . Substituting this into equations (E1) and

(E2) gives
x0 (1 + 2z0 ) = 0 (E1)
y0 + (4y0 + 2)z0 = 0 (E2)
The question specifies that x0 6= 0, so (E1) forces z0 = − 21 . Substituting

z0 = − 12 into (E2) gives
−y0 − 1 = 0 =⇒ y0 = −1
Finally x0 is determined by the requirement that (x0 , y0 , z0 ) must lie on S

and so must obey
1
z0 = x20 + 2y02 + 2y0 − 1 =⇒ − = x20 + 2(−1)2 + 2(−1) − 1
2
1
=⇒ x20 =
2
So the allowed points P are √12 , −1 , − 21 and − √12 , −1 , − 21 .

3.2.1.15. ∗. Solution. Let (x0 , y0 , z0 ) be a point on the hyperboloid

z 2 = 4x2 +y 2 −1 where the tangent plane is parallel to the plane 2x−y+z =
0. A normal vector to the plane 2x−y +z = 0 is (2, −1, 1). Because the hy-
perboloid is G(x, y, z) = 4x2 +y 2 −z 2 −1 and ∇ G(x, y, z) = (8x, 2y, −2z), a
normal vector to the hyperboloid at (x0 , y0 , z0 ) is ∇ G(x0 , y0 , z0 ) = (8x0 , 2y0 , −2z0 ).
So (x0 , y0 , z0 ) satisfies the required conditions if and only if there is a
nonzero t obeying
(8x0 , 2y0 , −2z0 ) = t(2, −1, 1) and z02 = 4x20 + y02 − 1

t t
⇐⇒ x0 = , y0 = z0 = − and z02 = 4x20 + y02 − 1
4 2
t2 t2 t2 t t
⇐⇒ = + − 1 and x0 = , y0 = z0 = −
4 4 4 4 2
1

⇐⇒ t = ±2 (x0 , y0 , z0 ) = ± 2 , −1, −1
3.2.1.16. ∗. Solution. (a) A vector perpendicular to x2 + z 2 = 10 at

(1, 1, 3) is
1
∇ (x2 + z 2 )(1,1,3) = (2xı̂ı + 2z k̂)(1,1,3) = 2ı̂ı + 6k̂ or (2, 0, 6) = (1, 0, 3)

2
(b) A vector perpendicular to y 2 + z 2 = 10 at (1, 1, 3) is
1
∇ (y 2 + z 2 )(1,1,3) = (2y̂ + 2z k̂)(1,1,3) = 2̂ + 6k̂ or (0, 2, 6) = (0, 1, 3)

2
A vector is tangent to the specified curve at the specified point if and only
if it perpendicular to both (1, 0, 3) and (0, 1, 3). One such vector is
 
ı̂ı ̂ k̂
(0, 1, 3) × (1, 0, 3) = det 0
 1 3  = (3, 3, −1)
1 0 3
(c) The specified tangent line passes through (1, 1, 3) and has direction
vector (1, 1, 3) and so has vector parametric equation
r(t) = (1, 1, 3) + t(3, 3, −1)

3.2.1.17. ∗. Solution. r(t) = (x(t) , y(t) , z(t)) intersects z 3 +xyz −2 =
0 when
3
z(t)3 + x(t) y(t) z(t) − 2 = 0 ⇐⇒ t2 + t3 )(t) t2 − 2 = 0

⇐⇒ 2t6 = 2 ⇐⇒ t = 1
since t is required to be positive. The direction vector for the curve at

t = 1 is
r0 (1) = 3ı̂ı + ̂ + 2 k̂
A normal vector for the surface at r(1) = (1, 1, 1) is
∇ (z 3 + xyz)(1,1,1) = [yzı̂ı + xz̂ + (3z 2 + xy)k̂](1,1,1) = ı̂ı + ̂ + 4k̂

The angle θ between the curve and the normal vector to the surface is
determined by
√ √
(3, 1, 2) (1, 1, 4) cos θ = (3, 1, 2) · (1, 1, 4) ⇐⇒ 14 18 cos θ = 12
√
⇐⇒ 7 × 36 cos θ = 12
2
⇐⇒ cos θ = √
7
◦
⇐⇒ θ = 40.89
The angle between the curve and the surface is 90 − 40.89 = 49.11◦ (to two
decimal places).
3.2.1.18. Solution. Let (x0 , y0 , z0 ) be any point on the surface. A

vector normal to the surface at (x0 , y0 , z0 ) is

−(x2 +y 2 )/2
∇ xye − z
(x0 ,y0 ,z0 )

−(x20 +y02 )/2 2 2 2 2 2 2
= y0 e − x20 y0 e−(x0 +y0 )/2 , x0 e−(x0 +y0 )/2 − x0 y02 e−(x0 +y0 )/2 , −1
The tangent plane to the surface at (x0 , y0 , z0 ) is horizontal if and only

if this vector is vertical, which is the case if and only if its x- and y-
components are zero, which in turn is the case if and only if
y0 (1 − x20 ) = 0 and x0 (1 − y02 ) = 0

⇐⇒ y0 = 0 or x0 = 1 or x0 = −1 and x0 = 0 or y0 = 1 or y0 = −1
⇐⇒ (x0 , y0 ) = (0, 0) or (1, 1) or (1, −1) or (−1, 1) or (−1, −1)
The values of z0 at these points are 0, e−1 , −e−1 , −e−1 and e−1 , respec-
tively. So the horizontal tangent planes are z = 0, z = e−1 and z = −e−1 .
At the highest and lowest points of the surface, the tangent plane is horizon-
tal. So the largest and smallest values of z are e−1 and −e−1 , respectively.

3.3.6 · Exercises
3.3.6.1. Solution. (a) S is the part of the plane z = y tan θ that
lies above the rectangle in the xy-plane with vertices (0, 0), (a, 0), (0, b),
(a, b). So S is the rectangle with vertices (0, 0, 0), (a, 0, 0), (0, b, b tan θ),
(a, b, b tan θ). So it has side lengths
|(a, 0, 0) − (0, 0, 0)| = a

p
|(0, b, b tan θ) − (0, 0, 0)| = b2 + b2 tan2 θ
p
and hence area ab 1 + tan2 θ = ab sec θ.
(b) S is the part of the surface z = f (x, y) with f (x, y) = y tan θ and
with (x, y) running over

D = (x, y) 0 ≤ x ≤ a, 0 ≤ y ≤ b
Hence by (3.3.2)
q p
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy = 1 + 02 + tan2 θ dx dy
and
ZZ ZZ p
Area(S) = dS = 1 + tan2 θ dx dy
S D
Z a pZ b
= dx dy 1 + tan2 θ
0 0
p
= ab 1 + tan2 θ = ab sec θ
3.3.6.2. Solution. Note that all three vertices (a, 0, 0), (0, b, 0) and
(0, 0, c) lie on the plane xa + yb + zc = 1. So the triangle is part of that
plane.
Method 1. S is the part of the surface z = f (x, y) with f (x, y) =
c 1 − xa − yb and with (x, y) running over the triangle Txy in the xy-plane

with vertices (0, 0, 0) (a, 0, 0) and (0, b, 0). Hence by the first part of (3.3.2),
ZZ q
Area(S) = 1 + fx (x, y)2 + fy (x, y)2 dx dy
Txy
r
c2 c2
ZZ
= 1+ 2
+ 2 dx dy
Txy a b
r
c2 c2
= 1+ 2
+ 2 A(Txy )
a b
where A(Txy ) is the area of Txy . Since the triangle Txy has base a and
height b (see the figure below), it has area 12 ab. So
r
1 c2 c2 1p 2 2
Area(S) = 1 + 2 + 2 ab = a b + a2 c2 + b2 c2
2 a b 2
z
p0, 0, cq Txy
p0, b, 0q
y
pa, 0, 0q
x
Method 2. S is the part of the surface x = g(y, z) with g(y, z) =
a 1 − yb − zc and with (y, z) running over the triangle Tyz in the yz-plane
with vertices (0, 0, 0) (0, b, 0) and (0, 0, c). Hence by the second part of
(3.3.2),
ZZ q
Area(S) = 1 + gy (y, z)2 + gz (y, z)2 dy dz
Tyz
r
a2 a2
ZZ
= 1+ 2
+ 2 dy dz
Tyz b c
r
a2 a2
= 1+ 2
+ 2 A(Tyz )
b c
where A(Tyz ) is the area of Tyz . Since Tyz has base b and height c, it has
area 21 bc. So
r
1 a2 a2 1p 2 2
Area(S) = 1 + 2 + 2 bc = a b + a2 c2 + b2 c2
2 b c 2
Method 3. S is the part of the surface y = h(x, z) with h(x, z) =
b 1 − xa − zc and with (x, z) running over the triangle Txz in the xz-plane
with vertices (0, 0, 0) (a, 0, 0) and (0, 0, c). Hence by the third part of
(3.3.2),
ZZ p
Area(S) = 1 + hx (x, z)2 + hz (x, z)2 dx dz
Txz
r
b2 b2
ZZ
= 1 + 2 + 2 dx dz
Txz a c
r
b2 b2
= 1+ 2
+ 2 A(Txz )
a c
where A(Txz ) is the area of Txz . Since Txz has base a and height c, it has
area 12 ac. So
r
1 b2 b2 1p 2 2
Area(S) = 1+ 2
+ 2 bc = a b + a2 c2 + b2 c2
2 a c 2
(b) We have already seen in the solution to part (a) that
ab ac bc
Area(Txy ) = Area(Txz ) = Area(Tyz ) =
2 2 2
Hence
r
a2 b2 a2 c2 b2 c2
Area(S) = + +
4 4 4
q
= Area(Txy )2 + Area(Txz )2 + Area(Tyz )2
3.3.6.3. Solution. (a) Think of the cylinder as being a piece of paper

that has been partially rolled up. If you flatten the piece of paper out, you
get a rectangle with the length of one side being h and the length of the
other side being one quarter of the circumference of a circle of radius a, i.e.
1 πa πah
4 (2πa) = 2 . So the area of S is 2 .
(b) S is parametrized by
x(θ, y) = a cos θ y(θ, y) = y z(θ, z) = a sin θ

π
with (θ, y) running over 0 ≤ θ ≤ 2, 0 ≤ y ≤ h. Then, by (3.3.1),
∂x ∂y ∂z
, , = (−a sin θ, 0, a cos θ)
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (0, 1, 0)
∂y ∂y ∂y

∂x ∂y ∂z ∂x ∂y ∂z
dS = , , × , , dθ dy
∂θ ∂θ ∂θ ∂y ∂y ∂y

= (−a cos θ, 0, −a sin θ) dθ dy
= a dθ dy
So
ZZ Z π/2 Z h π
Area(S) = dS = dθ dy a = a h
S 0 0 2
3.3.6.4. Solution. The surface is z = f (x, y) with f (x, y) = xy. So, by

(3.3.2), q p
dS = 1 + fx2 + fy2 dx dy = 1 + x2 + y 2 dx dy
and
ZZ ZZ p
I= (x2 + y 2 ) dS = (x2 + y 2 ) 1 + x2 + y 2 dx dy
S x2 +y 2 ≤3
√
Z 2π Z 3 p
= dθ dr r r2 1 + r2
0 0
We switched to polar coordinates in the last step. Making the change of

variables u = 1 + r2 , du = 2r dr
4 4
√
Z
2 2
I=π du (u − 1) u = π u5/2 − u3/2
1 5 3 1

64 16 2 2
=π − − +
5 3 5 3
116
= π
15
3.3.6.5. ∗. Solution. First observe that any point (x, y, z) on the
paraboliod lies above the xy-plane if and only if
0 ≤ z = a2 − x2 − y 2 ⇐⇒ x2 + y 2 ≤ a2
That is, if and only if (x, y) lies in the circular disk of radius a centred on
the origin. The equation of the paraboloid is of the form z = f (x, y) with
f (x, y) = a2 − x2 − y 2 . So, by (3.3.2),
ZZ q
Surface area = 1 + fx (x, y)2 + fy (x, y)2 dx dy
x2 +y 2 ≤a2
ZZ p
= 1 + 4x2 + 4y 2 dx dy
x2 +y 2 ≤a2
Switching to polar coordinates,

Z a Z 2π p
Surface area = dr dθ r 1 + 4r2
0
Z a 0 p
= 2π dr r 1 + 4r2
0
1+4a2
ds √
Z
= 2π s with s = 1 + 4r2 , ds = 8r dr
1 8
s=1+4a2
π 2 3/2
= s
4 3 s=1
π 3/2
(1 + 4a2 ) − 1

=
6
3.3.6.6. ∗. Solution. First observe that any point (x, y, z) on the cone
lies between the planes z = 2 and z = 3 if and only if 4 ≤ x2 + y 2 ≤ 9.
The equation
p of the cone can be rewritten in the form z = f (x, y) with
f (x, y) = x + y 2 . Note that
2
x y
fx (x, y) = p fy (x, y) = p
x2 + y2 x + y2
2
So, by (3.3.2),
ZZ q
4≤x2 +y 2 ≤9
s
x2 y2
ZZ
= 1+ + 2 dx dy
4≤x2 +y 2 ≤9 x2 +y 2 x + y2
√ ZZ
= 2 dx dy
4≤x2 +y 2 ≤9
Now the domain of integration is a circular washer with outside radius 3

and√inside radius 2 and hence of area π(32 − 22 ) = 5π. So the surface area
is 5 2π.
3.3.6.7. ∗. Solution. The equation of the surface is of the form z =
f (x, y) with f (x, y) = 32 x3/2 + y 3/2 . Note that
√ √
fx (x, y) = x fy (x, y) = y
So, by (3.3.2),
Z 1 Z 1 q
Surface area = dx dy 1 + fx (x, y)2 + fy (x, y)2
0 0
Z 1 Z 1 p
= dx dy 1+x+y
0 0
Z 1 h2 iy=1
= dx (1 + x + y)3/2
0 3 y=0
Z 1
2
dx (2 + x)3/2 − (1 + x)3/2

=
3 0
2 2h ix=1
= (2 + x)5/2 − (1 + x)5/2
3 5 x=0
4 5/2 5/2 5/2 5/2

= 3 −2 −2 +1
15
4 √ √
= 9 3−8 2+1
15
p
3.3.6.8. ∗. Solution. (a) By (3.3.2), F (x, y) = 1 + fx (x, y)2 + fy (x, y)2 .
(b) (i) The “dimple” to be painted is part of the upper sphere x2 + y 2 +
√ 2
z − 2 3 = 4. It is on the bottom half of the sphere and so has equation
√ p
z = f (x, y) = 2 3 − 4 − x2 − y 2 . Note that
x y
fx (x, y) = p fy (x, y) = p
4 − x2 − y 2 4 − x2 − y 2
√
The point on the dimple with the largest value of x is (1, 0, 3). (It is
marked by a dot in the figure above.) The dimple is invariant under rota-
tions around the z--axis and so has (x, y) running over x2 + y 2 ≤ 1. So, by
(3.3.2),
ZZ q
x2 +y 2 ≤1
s
x2 y2
ZZ
= 1+ 2 2
+ dx dy
x2 +y 2 ≤1 4−x −y 4 − x2 − y 2
ZZ
2
= p dx dy
x2 +y 2 ≤1 4 − x2 − y 2
Switching to polar coordinates,

Z 2π Z 1
2r
Surface area = dθ dr √
0 0 4 − r2
(b)
√ (ii) Observe that if we flip the dimple up by reflecting it in the plane
z = 3, as in the figure below, the “Death Star” becomes a perfect ball of
radius 2.
z
?
2 3
? ?
z“ 3 p1, 0, 3q
π
6
2
x
The area of the pink dimple in the figure above is identical to the area
of the blue cap in that figure. So the total surface area of the Death Star
is exactly the surface area of a sphere of radius 2 and so is 4π22 = 16π.
3.3.6.9. ∗. Solution. On the upper half of the cone
p x y
z = f (x, y) = x2 + y 2 fx (x, y) = p fy (x, y) = p
x2 + y 2 x2 + y 2
so that
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy
s
x2 y2
= 1+ 2 + dx dy
x + y2 x2 + y 2
√
= 2 dx dy
and
ZZ √
Area = 2 dx dy
1≤x2 +y 2 ≤162
√ h i
2 area of (x, y) x2 + y 2 ≤ 162 − area of (x, y) x2 + y 2 ≤ 1

=
√ √
= 2 π162 − π12 = 255 2π ≈ 1132.9

3.3.6.10. ∗. Solution. We are to find the surface area of part of a

hemisphere. On the hemisphere
p
z = f (x, y) = a2 − x2 − y 2
x
fx (x, y) = − p
a − x2 − y 2
2
y
fy (x, y) = − p
a − x2 − y 2
2
so that
q
dS = 1 + fx (x, y)2 + fy (x, y)2 dx dy
s
x2 y2
= 1+ + dx dy
a2 − x2 − y 2 a2 − x2 − y 2
s
a2
= dx dy
a2 − x2 − y 2
In polar coordinates, this is dS = √a2a−r2 r dr dθ. We are to find the surface

area of the part of the hemisphere that is inside the cylinder, x2 −ax+y 2 =
0, which in polar coordinates becomes r2 − ar cos θ = 0 or r = a cos θ. The
top half of the domain of integration is sketched below.
y
r “ a cos θ
x
pa{2, 0q
So the
Z π/2 Z a cos θ
a
Surface Area = 2 dθ dr r √
0 0 a2 − r2
Z π/2 h p ia cos θ
= 2a dθ − a2 − r2
0 0
Z π/2
= 2a dθ a − a sin θ
0
h iπ/2
2
= 2a θ + cos θ = a2 [π − 2]
0
3.3.6.11. Solution. The upper half cone obeys f (x, y, z) = x2 +y 2 −z 2 =

0. So, by (3.3.3),
p
∇f 2xı̂ı + 2y̂ − 2z k̂ x2 + y 2 + z 2
dS = dx dy = dx dy = dx dy

∇ f · k̂ −2z z
But on the cone x2 + y 2 = z 2 , and z > 0, so that

√
√
p
x2 + y 2 + z 2 2z 2
dS = dx dy = dx dy = 2 dx dy
z z
and
x4 − y 4 + y 2 z 2 − z 2 x2 + 1 = x4 − y 4 + y 2 x2 + y 2 − x2 + y 2 )x2 + 1 = 1

We have to integrate (x, y) over the interior of x2 +y 2 = 2x, or equivalently,

the interior of (x − 1)2 + y 2 = 1, which is the disk
D = (x, y) (x − 1)2 + y 2 ≤ 1

So
=1
ZZ z
4 4
}|
2 2 2 2
{ √ ZZ √ √
(x − y + y z − z x + 1) dS = 2 dx dy = 2 Area(D) = 2 π
S D
3.3.6.12. Solution. As we saw in Example 3.1.5, the torus may be

parametrized by
r(θ, ψ) = (R + r cos θ) cos ψ ı̂ı + (R + r cos θ) sin ψ ̂ + r sin θ k̂

0 ≤ θ, ψ ≤ 2π
Then
∂r
= (R + r cos θ) − sin ψı̂ı + cos ψ̂
∂ψ
∂r
= r − sin θ cos ψ ı̂ı − sin θ sin ψ ̂ + cos θ k̂
∂θ
and
∂r ∂r
× = r(R + r cos θ) − sin ψı̂ı + cos ψ̂ ×
∂ψ ∂θ

− sin θ cos ψ ı̂ı − sin θ sin ψ ̂ + cos θ k̂
 
ı̂ı ̂ k̂
= r(R + r cos θ) det  − sin ψ cos ψ 0 
− sin θ cos ψ − sin θ sin ψ
cos θ

= r(R + r cos θ) cos ψ cos θ ı̂ı + sin ψ cos θ ̂ + sin θ k̂

As cos ψ cos θ ı̂ı + sin ψ cos θ ̂ + sin θ k̂ is a unit vector, (we could have
shortened this computation by observing that − sin ψ ı̂ı+cos ψ ̂ and − sin θ cos ψ ı̂ı−
sin θ sin ψ ̂ + cos θ k̂ are mutually perpendicular unit vectors, so that their
cross product is automatically a unit vector) and

∂r ∂r
∂ψ × ∂θ = r(R + r cos θ) =⇒ dS = r(R + r cos θ) dψ dθ

The total surface area of the torus is

Z 2π Z 2π Z 2π
r dθ dψ (R + r cos θ) = 2πr dθ (R + r cos θ) = (2π)2 Rr
0 0 0
3.3.6.13. Solution. By symmetry, the centroid (x̄, ȳ, z̄) obeys x̄ = ȳ =

z̄. Parametrize the sphere using spherical coordinates.
r(θ, ϕ) = a sin ϕ cos θ ı̂ı + a sin ϕ sin θ ̂ + a cos ϕ k̂
Then
∂r
= −a sin ϕ sin θ ı̂ı + a sin ϕ cos θ ̂
∂θ
∂r
= a cos ϕ cos θ ı̂ı + a cos ϕ sin θ ̂ − a sin ϕ k̂
∂ϕ
so that
 
ı̂ı ̂ k̂
∂r ∂r
× = det −a sin ϕ sin θ
 a sin ϕ cos θ 0 
∂θ ∂ϕ
a cos ϕ cos θ a cos ϕ sin θ −a sin ϕ
= −a2 sin2 ϕ cos θ ı̂ı − a2 sin2 ϕ sin θ ̂ − a2 sin ϕ cos ϕ k̂

∂r ∂r
=⇒ dS =
× dθ dϕ = a2 sin ϕ dθ dϕ
∂θ ∂ϕ
As the surface area of the part of the sphere in the first octant is 18 4πa2 =
πa2
2
RR Z π/2 Z π/2 z
z dS 1 2
x̄ = ȳ = z̄ = RRS
z }| {
= dθ dϕ (a sin ϕ)( a cos ϕ)
S
dS πa2 /2 0 0
π/2
2a π π/2
Z
1 a
= dϕ sin ϕ cos ϕ = a sin2 ϕ =
π 2 0 2 0 2
3.3.6.14. Solution. In cylindrical coordinates
x = r cos θ y = r sin θ z=z
In these coordinates the equation, x2 + y 2 = 2ay, of the cylinder becomes
r2 = 2ar sin θ or r = 2a sin θ
That is, r = f (θ) with f (θ) = 2a sin θ. Parametrize the cylinder by

r(θ, z) = x(θ, z)ı̂ı + y(θ, z) ̂ + z(θ, z) k̂ with
x(θ, z) = f (θ) cos θ = 2a sin θ cos θ = a sin 2θ

y(θ, z) = f (θ) sin θ = 2a sin θ sin θ = a(1 − cos 2θ)
z(θ, z) = z
Under this parametrization,

∂r ∂r
= 2a cos 2θ ı̂ı + 2a sin 2θ ̂ = k̂
∂θ ∂z
∂r ∂r
=⇒ × = −2a cos 2θ ̂ + 2a sin 2θ ı̂ı
∂θ ∂z
∂r ∂r
=⇒ dS = × dθ dz = 2a dθ dz
∂θ ∂z
We still have to determine the limits of integration. The figure on the left
below provides a top view of the cylinder.
y z z=r
x2 + y 2 = 2ax
(0, a) r
r
θ
x z = −r
r=2a sin θ
From it we see that 0 ≤ θ ≤ π. The cone z 2 = x2 + y 2 = r2 (i.e.
z = ±r) and the cylinder r = 2a sin θ intersect at z 2 = r2 = 4a2 sin2 θ. So,
for each fixed θ, z runs from −2a sin θ to z = +2a sin θ. (See the figure on
the right above. It shows a constant θ cross-section.) Finally,
Z Z π Z 2a sin θ Z π
Area = 2a dθ dz = 2a dθ dz = 8a2 dθ sin θ
|z|≤2a sin θ 0 −2a sin θ 0
h iπ
= 8a2 − cos θ
0
= 16a2
3.3.6.15. ∗. Solution. (a) This right circular cone symmetric about the
z-axis projects down onto a disk D in the plane z = 0. Setting z = b gives
D = (x, y, z) x2 + y 2 ≤ a2 , z = 0

Since G(x, y, z) = b2 (x2 + y 2 ) − a2 z 2 is constant on S, the area elements

dS on S are related to area elements dxdy on D as follows:
|∇G(x, y, z)| 2|(b2 x, b2 y, −a2 z)|

dS = dxdy = dxdy
|∇G(x, y, z) · k̂| 2|a2 z|
p
b4 (x2 + y 2 ) + a4 z 2
= dxdy
a2 z
b
p
by (3.3.3). The defining equation for S gives z = a x2 + y 2 , so
p
b4 (x2 + y 2 ) + a2 b2 (x2 + y 2 ) 1p 2
dS = p dxdy = a + b2 dxdy.
a b2 (x2 + y 2 ) a
√
2 2 RR
Hence I = a a+b D
(x2 + y 2 ) dxdy.
(b) Or, parametrize the surface S using θ and t as follows:
bp 2 b
x = t cos θ, y = t sin θ, z = x + y 2 = t, (∗)
a a
0 ≤ θ ≤ 2π, 0 ≤ t ≤ a.
Then we have, by (3.3.1),

 
ı̂ı ̂ k̂
∂r ∂r
× = det  cos θ sin θ b/a
∂t ∂θ
−t sin θ t cos θ 0
b b
= − t cos θ, − t sin θ, t ,
a a
so
∂r ∂r p
dS =
× dt dθ = t 1 + b2 /a2 dt dθ.
∂t ∂θ
It follows that for the rectangular region R of the tθ-plane described in (∗),
ZZ
p
I= t2 t 1 + b2 /a2 dt dθ.
R
(c) Using polar coordinates in (a) would give

√
a2 + b2 2π a 2
Z Z
π p
I= r r dr dθ = a3 a2 + b2 .
a θ=0 r=0 2
Direct integration in (b) gives the same thing, because

√
a2 + b2 2π a 3
ZZ Z Z
2
p
I= 2 2
t t 1 + b /a dt dθ = t dt dθ.
D a θ=0 t=0
3.3.6.16. Solution. (a) The surface is g(x, y, z) = x2 + y 2 + z 2 − a2 = 0.

So, on the surface of the sphere,
∇g xı̂ı + y̂ + z k̂
n̂ = =p
∇
|∇ g| x2 + y 2 + z 2
n++1−1/2 n+1/2
=⇒ F · n̂ = x2 + y 2 + z 2 = a2 = a2n+1
ZZ ZZ
=⇒ F · n̂ dS = a2n+1 dS = a2n+1 Area(S) = 4πa2n+3
S S
since the surface area of a sphere of radius a is 4πa2 .

(b) The box has six faces.

S1 = (x, y, z) 0 ≤ x ≤ a, 0 ≤ y ≤ b, z = c with outward normal n̂ = k̂

S2 = (x, y, z) 0 ≤ x ≤ a, 0 ≤ y ≤ b, z = 0 with outward normal n̂ = −k̂

S3 = (x, y, z) 0 ≤ x ≤ a, 0 ≤ z ≤ c, y = b with outward normal n̂ = ̂

S4 = (x, y, z) 0 ≤ x ≤ a, 0 ≤ z ≤ c, y = 0 with outward normal n̂ = −̂

S5 = (x, y, z) 0 ≤ y ≤ b, 0 ≤ z ≤ c, x = a with outward normal n̂ = ı̂ı

S6 = (x, y, z) 0 ≤ y ≤ b, 0 ≤ z ≤ c, x = 0 with outward normal n̂ = −ı̂ı
z
n̂ = k̂
(0, b, c)
S1
n̂ = −̂ n̂ = ̂
y
S2
(a, b, 0)
x
n̂ = −k̂
For S1 , i.e. the z = c face, and S2 , i.e. the z = 0 face,
Z Z Z

F · n̂ dS = xı̂ı + y ̂ + c k̂ · k̂ dx dy =c dx dy = abc
z=c z=c z=c
face face face
Z Z

F · n̂ dS = xı̂ı + y ̂ + 0 k̂ · (−k̂) dx dy = 0
z=0 z=0
face face
because the z = c face has area ab. Similarly,

Z Z Z Z
F · n̂ dS = F · n̂ dS = 0 F · n̂ dS = F · n̂ dS = abc
x=0 y=0 x=a y=b
face face face face
The total flux is 3abc.

(c) The base of the cone is (x, y, z) x2 + y 2 ≤ 1, z = 0 and has
(outward) normal n̂ = −k̂. So The flux through the base is
Z ZZ
F · n̂ dS = (yı̂ı) · (−k̂) dx dy = 0
base x2 +y 2 ≤1
p coordinates x = r cos θ, y = r sin θ, z = z and the equation

In cylindrical
z = 1 − x2 + y 2 of the top part of the cone becomes z = 1 − r. So we
may parametrize the top part of the cone by
r(r, θ) = r cos θ ı̂ı + r sin θ ̂ + (1 − r) k̂ with 0 ≤ θ ≤ 2π, 0 ≤ r ≤ 1

Then
∂r
= cos θ ı̂ı + sin θ ̂ − k̂
∂r
∂r
= −r sin θ ı̂ı + r cos θ ̂
∂θ
 
ı̂ı ̂ k̂
∂r ∂r
× = det  cos θ sin θ −1
∂r ∂θ
−r sin θ r cos θ 0
= −r cos θ ı̂ı − r sin θ ̂ + r k̂
∂r ∂r
=⇒ n̂ dS = × dr dθ
∂r ∂θ

= − r cos θ ı̂ı − r sin θ ̂ + rk̂ dr dθ
by (3.1.1). We have the orientation correct because the k̂ component of n̂

is positive. The flux through the top, as well as the total flux, is
Z
F · n̂ dS
top
y z
Z 1 Z 2π z }| { z }| {
= dr dθ r sin θ ı̂ı + (1 − r) k̂ · − r cos θ ı̂ı − r sin θ ̂ + r k̂
0 0
Z 1 Z 2π
dθ − r2 sin θ cos θ + r(1 − r)

= dr
0 0
Z 1 Z 2π Z 1
1
dr r2 dr r − r2

=− dθ sin(2θ) + 2π
0 0 2 0
1 h1 1i π
= − × 0 + 2π − =
3 2 3 3
3.3.6.17. ∗. Solution. Let G(x, y, z) = x2 + y 2 + 2z. Then, by (3.3.3),
∇G 2xı̂ı + 2y̂ + 2k̂
n̂ dS = dxdy = dxdy = (xı̂ı + y̂ + k̂) dxdy
∇ G · k̂ 2
so that
x2 + y 2 x2 + y 2 p
p dS = p 1 + x2 + y 2 dxdy = (x2 + y 2 ) dxdy
1 + x2 + y 2 1 + x2 + y 2
and
F · n̂ dS = xı̂ı + y̂ + z k̂ · xı̂ı + y̂ + k̂ dxdy = x2 + y 2 + z dxdy

h 1 i
= 1 + (x2 + y 2 ) dxdy
2
since z = 1 − 21 (x2 + y 2 ) on S.
(a)
Z 1 Z 1
x2 + y 2
ZZ ZZ
p dS = (x2 + y 2 ) dxdy = 4 dx dy (x2 + y 2 )
2
1+x +y 2
S S 0 0
Z 1
1 1 1 8
dx x2 +

=4 =4 + =
0 3 3 3 3
(b)
ZZ Z 1 Z 1 h 1 i 1 8 16
F · n̂ dS = dx dy 1 + (x2 + y 2 ) = 2 × 2 + × =
S −1 −1 2 2 3 3
3.3.6.18. ∗. Solution. Let G(x, y, z) = z − xy. Then, using (3.3.3),
∇G −yı̂ı − x̂ + k̂
n̂ dS = dxdy = dxdy = (−yı̂ı − x̂ + k̂) dxdy
∇ G · k̂ 1
so that
x2 y x2 y p
p dS = p y 2 + x2 + 1 dxdy = x2 y dxdy
1+ x2 + y2 1+ x2 + y2
and

F · n̂ dS = xı̂ı + y̂ + k̂ · − yı̂ı − x̂ + k̂ dxdy = 1 − 2xy dxdy
(a)
1 1
x2 y
ZZ ZZ Z Z
2
p dS = x y dxdy = dx dy x2 y
S 1 + x2 + y 2 S 0 0
Z 1
1
= dx 21 x2 =
0 6
(b)
ZZ Z 1 Z 1 Z 1
1 1
F · n̂ dS = dx dy [1 − 2xy] = dx [1 − x] = 1 − =
S 0 0 0 2 2
3.3.6.19. ∗. Solution.For the surface z = f (x, y) = y 3/2 ,
r 3 √ 2 r
q
2 2
9
dS = 1 + fx + fy dxdy = 1 + y dxdy = 1 + y dxdy
2 4
by (3.3.2). So the area is
Z 1 Z 1
r Z 1
9 8 h 9 3/2 i1
dx dy 1+ y = 1+ y
dx
0 0 4 0 27 4 0
Z 1
8 h 13 3/2
i
= dx −1
0 27 4
" #
3/2
8 13
= −1
27 4
3.3.6.20. ∗. Solution. The surface is a sphere of radius 2 centered on

(0, 0, 2), The plane z = 1 intersects the sphere on the circle x2 + y 2 = 3.
Let F (x, y, z) = x2 + y 2 + (z − 2)2 . Then, by (3.3.3),
∇F 2xı̂ı + 2y̂ + 2(z − 2)k̂
dS = dxdy = dxdy

∇ F · k̂ 2(z − 2)
xı̂ı + y̂ + (z − 2)k̂
= dxdy

(z − 2)
p
x2 + y 2 + (z − 2)2
= dxdy
|z − 2|
2
= dxdy
|z − 2|
since x2 + y 2 + (z − 2)2 = 4 on S. On S, z ≤ 2, so |z − 2| = 2 − z and

ZZ ZZ
2
f (x, y, z) dS = (2 − z)(x2 + y 2 ) dxdy
S 2 2
x +y ≤3 |z − 2|
ZZ
=2 (x2 + y 2 ) dxdy
x2 +y 2 ≤3
Switching to polar coordinates

√ √
3 2π
r4 3
ZZ Z Z
f (x, y, z) dS = 2 dr r dθ r2 = 2(2π) = 9π
S 0 0 4 0
3.3.6.21. ∗. Solution. (a) Each (horizontal) constant z cross-section
is a circle centred on the z-axis. The radius varies linearly from 2, when
z = 0 to 0, when z = 3. So the radius at height z is 23 (3 − z) and we can
use
2 2
r(θ, z) = (3 − z) cos θ ı̂ı + (3 − z) sin θ ̂ + z k̂
3 3
0 ≤ θ < 2π, 0 ≤ z ≤ 3
as the parametrization.
(b) By symmetry the centre of mass will lie on the z-axis. We are only
asked for the z-coordinate anyway. The z-coordinate of the centre of mass
is the weighted average of z over the cone. Since a density has not been
specified, we assume that it is a constant. We RR may takeRR the density to be
1, so the z-coordinate of the centre of mass is S z dS/ S dS.
Since
∂r 2 2

∂θ = − 3 (3 − z) sin θ , 3 (3 − z) cos θ , 0
∂r 2 2

∂z = − 3 cos θ , − 3 sin θ , 1
∂r ∂r 2 2 4

∂θ × ∂z = 3 (3 − z) cos θ , 3 (3 − z) sin θ , 9 (3 − z)
the element of surface area for this parametrization is

∂r ∂r
× ∂z dθdz = 23 (3 − z)(cos θ , sin θ , 32 dθdz

dS = ∂θ
√
2 13
= 9 (3 − z)dθdz
RR
So the surface area, S
dS, of the cone is
Z 3 Z 2π √ √
Z 3
2 13 4 13
dz dθ 9 (3 − z) = 9 π dz (3 − z)
0 0 0
√ 3
= − 2 913 π(3 − z)2

0
√
= 2 13π
and the z-coordinate of the centre of mass is

Z 3 Z 2π √
1
z̄ = √ dz dθ 2 913 (3 − z)z
2 13π 0 0
2 3
Z
= dz (3z − z 2 )
9 0
3
2 h 3z 2 z3 i
= −
9 2 3 0
2 27
= =1
9 6
This is a little less than half way up the cone, which is reasonable since the
cone is “bottom heavy”.
3.3.6.22. ∗. Solution. Each constant z cross-section of the cone is a
circle. When z = 0, that circle has radius a. When z = a that circle has
radius 0. Thus the radius decreases linearly from a to 0 as z increases from
0 to a. So the radius at height z is a − z and we can parametrize the cone
by
r(θ, z) = (a − z) cos θ ı̂ı + (a − z) sin θ ̂ + z k̂ 0 ≤ θ < 2π, 0 ≤ z ≤ a
Since
∂r

∂θ = − (a − z) sin θ , (a − z) cos θ , 0
∂r

∂z = − cos θ , − sin θ , 1
∂r ∂r

∂θ × ∂z = (a − z) cos θ , (a − z) sin θ , a − z
the element of surface area for this parametrization is

∂r ∂r
dS = ∂θ × ∂z dθdz = (a − z)(cos θ , sin θ , 1 dθdz
√
= 2(a − z) dθdz
by (3.3.1). So the surface area of the cone is

ZZ Z a Z 2π √
dS = dz dθ 2(a − z)
S 0
√ Z 0a √ a
= 2 2π dz (a − z) = − 2 π(a − z)2

0 0
√ 2
= 2 πa
and the z-coordinate of the centre of mass is

Z a Z 2π √ Z a
RR
z dS 1 2
z̄ = RRS =√ dz dθ 2(a − z)z = 2 dz (az − z 2 )
S
dS 2 πa2
0 0 a 0
a
2 h az 2 z3 i 2 a3 a
= 2 − = 2 =
a 2 3 0 a 6 3
This is a little less than half way up the cone, which is reasonable since the
cone is “bottom heavy”.
3.3.6.23. ∗. Solution. Parametrize the surface by
x(θ, z) = cos θ y(θ, z) = 2 sin θ z(θ, z) = z
with (θ, z) running over 0 ≤ θ ≤ 2π, 0 ≤ z ≤ 1. Then, by (3.3.1),

∂x ∂y ∂z
, , = (− sin θ, 2 cos θ, 0)
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (0, 0, 1)
∂z ∂z ∂z
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , dθdz
∂θ ∂θ ∂θ ∂z ∂z ∂z
= (2 cos θ, sin θ, 0) dθ dz (+ for outward normal)
F x(θ, z), y(θ, z), z(θ, z) = cos θ ı̂ı + 2z sin θ cos θ ̂ + 16z sin4 θ k̂

So
ZZ Z 1 Z 2π
dθ 2 cos2 θ + 2z sin2 θ cos θ

F · n̂ dS = dz
S 0 0
Z 1 Z 2π
dθ 1 + cos(2θ) + 2z sin2 θ cos θ

= dz
0 0
Z 1 i2π
h 1 2
= dz θ + sin(2θ) + z sin3 θ = 2π
0 2 3 0
R 2π
cos2 θ dθ, see Example 2.4.4.
3.3.6.24. ∗. Solution. By (3.3.2), with f (x, y) = 4 − x2 − y 2 ,

n̂ dS = ± − fx , −fy , 1 dxdy

= ± 2x , 2y , 1 dxdy
To get the downward pointing normal, we want the minus sign. Set

T = (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1 − x
Then
ZZ ZZ z
z }| {
x + 1 , y + 1 , 2 (4 − x2 − y 2 ) · 2x , 2y , 1 dxdy

F · n̂ dS = −
S
Z ZT

=− 8 + 2x + 2y dxdy
T
Z 1 Z 1−x
=− dx dy 8 + 2x + 2y
0 0
Z 1
dx 8(1 − x) + 2x(1 − x) + (1 − x)2

=−
0
Z 1
dx 9 − 8x − x2

=−
0
1 14
=− 9−4− =−
3 3
3.3.6.25. ∗. Solution. First we have √ to parametrize S. It is natural
to use spherical coordinates with ρ = 2. However if we use the standard
spherical coordinates
√ √ √
x = 2 sin ϕ cos θ y = 2 sin ϕ sin θ z = 2 cos ϕ
√ 2 2 √ 2
p y +z sin ϕ sin2 θ+cos2 ϕ
the condition x ≥ y 2 + z 2 , i.e. x ≤ 1, becomes sin ϕ cos θ ≤
1, which is very complicated. So let’s back up and think a bit before we
compute. From the sketch below
(x, y, z)
y
p
(x, 0, 0) y2 + z2
x
√
y 2 +z 2
we see that x is the tangent of the angle between the radius vector
(x, y, z) and the x-axis. The angle between the radius vector (x, y, z) and
the z-axis (not the x-axis) is exactly spherical coordinate ϕ. So let’s modify
spherical coordinates to make the x-axis play the role of the z-axis. The
RR x = Z, y = X, z = Y . Then the
easy way to do is to just rename pintegral
we are to compute becomes S ZX 2 dS, and the condition x ≥ y 2 + z 2
√
becomes Z ≥ X 2 + Y 2 . Under the parametrization
√ √ √
X = 2 sin ϕ cos θ Y = 2 sin ϕ sin θ Z = 2 cos ϕ
√ √
2 2 sin ϕ
the condition Z ≥ X 2 + Y 2 is X Z+Y = cos ϕ ≤ 1, which is turn is
0 ≤ ϕ ≤ π4 . As dS = 2 sin ϕ dθdϕ (see Appendix A.6.3 and recall that
√
ρ = 2) the specified integral is
ZZ ZZ
xy 2 dS = ZX 2 dS
S S
Z π/4 √
Z 2π √
2
=2 2 cos ϕ
dϕ sin ϕ 2 sin ϕ cos θ
dθ
0 0
( ) Z
√ Z π/4 2π
3 2
=4 2 dϕ cos ϕ sin ϕ dθ cos θ
0 0
(Z ) Z
√ π/4 2π
cos(2θ) + 1
=4 2 dϕ cos ϕ sin3 ϕ dθ
0 0 2
π/4 2π
√ sin ϕ 4

sin(2θ) θ
=4 2 +
4 0 4 2 0
√
2π
=
4
R 2π
3.3.6.26. ∗. Solution. Here is a sketch of the part of S that is in the
first octant.
z
x2 + z 2 = sin2 y
(x, y, z)
θ y
x
For each fixed y, x2 + z 2 = sin2 y is a circle of radius sin y. (It’s the
blue circle in the sketch above.) So we may parametrize the surface by

r(θ, y) = sin y cos θ , y , sin y sin θ 0 ≤ θ < 2π, 0 ≤ y ≤ π
Then, by (3.3.1),
∂r
= − sin y sin θ , 0 , sin y cos θ
∂θ
∂r
= cos y cos θ , 1 , cos y sin θ
∂y
∂r ∂r
× = − sin y cos θ , sin y cos y , − sin y sin θ
∂θ ∂y

∂r ∂r p
dS =
× dθ dy = sin y 1 + cos2 y dθ dy
∂θ ∂y
So the specified integral is

ZZ p Z π Z 2π
dθ sin y 1 + cos2 y

1 + cos2 y dS = dy
S 0 0
Z π
dy sin y 1 + cos2 y

= 2π
0
Z −1
du 1 + u2

= −2π
1
with u = cos y, du = − sin y dy
Z 1
du 1 + u2

= 4π
0
1
u3

= 4π u +
3 0
16
= π
3
3.3.6.27. ∗. Solution. The paraboloid is
S = (x, y, z) z = 1 − x2 − y 2 , z ≥ 0

= (x, y, z) z = 1 − x2 − y 2 , x2 + y 2 ≤ 1

By (3.3.2), the paraboloid has

q
dS = 1 + fx (x, y)2 + fy (x, y)2 dxdy with z = f (x, y) = 1 − x2 − y 2
p
= 1 + 4x2 + 4y 2 dxdy
By symmetry, the centre of mass will lie on the z-axis. By definition, the
z-coordinate of the centre of mass is the weighted average of z over S,
which is RR
z ρ(x, y, z) dS
z̄ = RRS
S
ρ(x, y, z) dS
On S,
z 1 − x2 − y 2
ρ(x, y, z) = √ =p
5 − 4z 1 + 4x2 + 4y 2
so that
ρ(x, y, z) dS = (1 − x2 − y 2 ) dxdy
So, using polar coordinates, the denominator of z̄ is

ZZ ZZ Z 1 Z 2π
2 2
ρ(x, y, z) dS = (1 − x − y ) dxdy = dr r dθ (1 − r2 )
S x2 +y 2 ≤1 0 0
Z 1
= 2π r(1 − r2 ) dr
0
2
hr r4 i1
= 2π −
2 4 0
π
=
2
and the numerator of z̄ is
ZZ ZZ
2
zρ(x, y, z) dS = (1 − x2 − y 2 ) dxdy
S x2 +y 2 ≤1
Z 1 Z 2π
2
= dr r dθ (1 − r2 )
0 0
Z 1
2
= 2π r(1 − r2 ) dr
0
h r2 r4 r 6 i1
= 2π −2 +
2 4 6 0
π
=
3
and
π/3 2
z̄ = =
π/2 3
3.3.6.28. ∗. Solution. The equation of the plane is z = f (x, y) =
2 − x − y. So by (3.3.2),

n̂ dS = − fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ dxdy = ı̂ı + ̂ + k̂ dxdy
A point (x, y, z) on the plane lies in the first octant if and only if
x≥0 and y ≥ 0 and z = 2 − x − y ≥ 0
So the domain of integration is the triangle

y
(0, 2)

T = (x, y) x ≥ 0, y ≥ 0, x +
x+y =2
y≤2
T
(2, 0)
x
and
ZZ ZZ z
z }| {
F · n̂ dS = xı̂ı + y ̂ + (2 − x − y) k̂ · ı̂ı + ̂ + k̂ dxdy
S
ZTZ
=2 dxdy
T
1
= 2 (2)(2) = 4
2

∂r
= v 2 , 2uv , v

∂u
∂r
= 2uv , u2 , u

∂v
∂r ∂r
= u2 v , uv 2 , −3u2 v 2

∂u × ∂v
(3.3.1) gives
n̂ dS = ± u2 v , uv 2 , −3u2 v 2 dudv

We are told that n̂ should have a positive z-component, so
n̂ dS = − u2 v , uv 2 , −3u2 v 2 dudv = − u2 v , −uv 2 , 3u2 v 2 dudv

and
ZZ ZZ z F
}| {
(uv 2 , u2 v , uv) · − u2 v , −uv 2 , 3u2 v 2 dudv

F · n̂ dS =
S S
Z 1 Z 3 Z 1 Z 3
3 3 3 3
= du dv u v = du u dv v
0 0 0 0
1 34 81
= =
4 4 16
3.3.6.30. ∗. Solution. (a) We start by just sketching the curve z = ey ,
considering the yz-plane as the plane x = 0 in R3 . This curve is the red
curve in the figure below. Concentrate on any one point on that curve. It
is the blue dot at (0, Y, eY )
z
p0, Y, eY q
x
in the figure. When our curve is rotated about the y-axis, the blue dot
sweeps out a circle. The circle that the blue dot sweeps out
• lies in the vertical plane y = Y and
• is centred on the y-axis and

• has radius eY .
We can parametrize the circle swept out in the usual way. Here is an end
view of the circle (looking down the y-axis), with the parameter, named θ,
indicated.
z
(0, Y, eY )
Y Y
(e sin θ, Y, e cos θ)
θ
eY
x
end view

The coordinates of the red dot are e sin θ , Y , eY cos θ . This also
Y
gives a parametrization of the surface of revolution
x(Y, θ) = eY sin θ
y(Y, θ) = Y
z(Y, θ) = eY cos θ
0 ≤ Y ≤ 1, 0 ≤ θ < 2π
Finally here is a sketch of the part of the surface in the first octant, x, y, z ≥
0.
z
x2 +z 2 = ey
x
(b) We are using the parametrization
r(Y, θ) = eY sin θ ı̂ı + Y ̂ + eY cos θ k̂ 0 ≤ Y ≤ 1, 0 ≤ θ ≤ 2π
so that
 
ı̂ı ̂ k̂
∂r ∂r
eY cos θ  = − eY sin θ, e2Y , −eY cos θ ,

× = det  eY sin θ 1
∂Y ∂θ
eY cos θ 0 −eY sin θ
and, by (3.3.1),
∂r ∂r p p
dS = × dY dθ = e2Y + e4Y dY dθ = eY 1 + e2Y dY dθ

∂Y ∂θ
So the integral is
ZZ Z 1 Z 2π p Z 1 p
ey dS = dY dθ e2Y 1 + e2Y = 2π dY e2Y 1 + e2Y
S 0 0 0
2π h i3/2 1
= 1 + e2Y
3
0
2π h i
= (1 + e2 )3/2 − 23/2
3
(c) Again, we are using the parametrization
r(Y, θ) = eY sin θ ı̂ı + Y ̂ + eY cos θ k̂ 0 ≤ Y ≤ 1, 0 ≤ θ ≤ 2π
so that
∂r ∂r
= − eY sin θ, e2Y , −eY cos θ ,

×
∂Y ∂θ
and, by (3.3.1),
∂r ∂r
dY dθ = ± − eY sin θ, e2Y , −eY cos θ dY dθ

n̂ dS = ± ×
∂Y ∂θ
We choose the “+” sign so that n̂ points towards the y-axis. As an example,
when 0 ≤ θ ≤ π2 , then z = eY cos θ > 0 while the z-coordinate of n̂ is
−ey cos θ < 0. So the integral is
ZZ
F · n̂ dS
S
x z
Z 1 Z 2π z }| { z }| {
dθ eY sin θ, 0, eY cos θ · − eY sin θ, e2Y , −eY cos θ

= dY
0 0
Z 1 Z 2π Z 1
2Y
=− dY = −2π dθ e dY e2Y
0 0 0
= −π e2 − 1 = π 1 − e2

3.3.6.31. ∗. Solution. Write
V = (x, y, z) 1 ≤ x2 + y 2 + z 2 ≤ 4

The boundary of V consists of two parts — the sphere, S2 , of radius 2,

r
centred on the origin, with (outward) normal n̂ = |r| = 2r , and the sphere
S1 of radius 1, centred on the origin, with (inward) normal n̂ = −r, So,
ZZ ZZ ZZ
r r r
F · n̂ dS = · dS − · r dS
∂V |r| 2 S1 |r|
Z ZS2 ZZ
= dS − dS
S2 S1
2 2
= 4π(2) − 4π(1)
= 12π
3.3.6.32. ∗. Solution. The part of the cone that has somefixed value,
Z, of z with 0 ≤ Z ≤ 1 is the part of the circle (x, y, z) x2 + y 2 =
4Z 2 , z = Z of radius 2Z that has 0 ≤ x ≤ y. Here is a sketch of the top
view of that part of that circle.
x2 + y 2 = 4Z 2
θ y
x
So we can parametrize S by
π
r(θ, Z) = 2Z sin θ ı̂ı + 2Z cos θ ̂ + Z k̂ 0≤θ≤ , 0≤Z≤1
4
So
∂r
= 2Z cos θ ı̂ı − 2Z sin θ ̂
∂θ
∂r
= 2 sin θ ı̂ı + 2 cos θ ̂ + k̂
∂Z
so that
 
ı̂ı ̂ k̂
∂r ∂r
× = det 2Z cos θ −2Z sin θ 0  = − 2Z sin θ, −2Z cos θ, 4Z ,
∂θ ∂Z
2 sin θ 2 cos θ 1
and, by (3.3.1),
√

∂r ∂r
dS =
× dθdZ = 20 Z dθdZ
∂θ ∂Z
and
√
ZZ √ Z 1 Z π/4 √ π 1 5π
z 2 dS = 20 dZ dθ Z 3 = 20 =
S 0 0 4 4 8
3.3.6.33. ∗. Solution. We’ll start by parametrizing S. Note that as
x2 + y 2 runs from 0 to 4, z runs from 5 to 1, and that, for each fixed
1 ≤ Z ≤ 5, the cross-section of S with z = Z is the circle x2 + y 2 = 5 − Z,
z = Z. So we may parametrize S by
√ √
r(θ, Z) = 5 − Z cos θ ı̂ı + 5 − Z sin θ ̂ + Z k̂
0 ≤ θ ≤ 2π, 1 ≤ Z ≤ 5
Since
∂r √ √
= − 5 − Z sin θ ı̂ı + 5 − Z cos θ ̂
∂θ
∂r 1 1
=− √ cos θ ı̂ı − √ sin θ ̂ + k̂
∂Z 2 5−Z 2 5−Z
so that
 
ı̂ı ̂ k̂
∂r ∂r  √ √
× = det − 5 − Z sin θ 5 − Z cos θ 0

∂θ ∂Z 1 1
− 2√5−Z cos θ − 2√5−Z sin θ 1
√ √
= 5 − Z cos θ , 5 − Z sin θ , 1/2 ,
(3.3.1) gives
∂r ∂r √ √
n̂ dS = ± × dθdZ = ± 5 − Z cos θ , 5 − Z sin θ , 1/2 dθdZ
∂θ ∂Z
Choosing the minus sign to give the downward pointing normal
ZZ
F · n̂ dS
S
x3 xy 2
Z 5 Z 2π
1
z }| { z }| {
=− dZ dθ − [5 − Z]3/2 cos3 θ − [5 − Z]3/2 cos θ sin2 θ ,
1 0 2
y3 z 2
1
z }| { z}|{
3/2 3
− [5 − Z] sin θ , Z 2
√ 2 √
· 5 − Z cos θ , 5 − Z sin θ , 1/2
Z 5 Z 2π 1
=− dZ dθ − [5 − Z]2 cos4 θ − [5 − Z]2 cos2 θ sin2 θ
1 0 2
1 1
− [5 − Z]2 sin4 θ + Z 2
2 2
Since
1 1 1 2 1
cos4 θ + cos2 θ sin2 θ + sin4 θ = cos2 θ + sin2 θ =
2 2 2 2
the flux
ZZ Z 5 Z 2π 1
1
F · n̂ dS = dZ [5 − Z]2 − Z 2
dθ
S 1 0 2 2
Z 5
=π dZ [5 − Z]2 − Z 2
1
5
Z3 43 53

1 1
= π − [5 − Z]3 − =π − +
3 3 1 3 3 3
= −20 π
√
3.3.6.34. ∗. Solution. The
q surface is z = f (x, y) with f (x, y) = 2xy.
py x
Since fx = 2x and fy = 2y , (3.3.2) gives
r
q
2 2
y x
dS = 1 + fx + fy dxdy = 1 + + dxdy
2x 2y
s
2xy + y 2 + x2
= dxdy
2xy
x+y
=√ dxdy
2xy
On the shell, z 2 = 2xy ≤ 4. So the x and y components of points (x, y, z)

on the shell run over the region x ≥ 1, y ≥ 1, xy ≤ 2, which is sketched
below
y=1
xy = 2
x=1
x
So the mass is
ZZ Z 2 Z 2/x
x+y
ρ(x, y, z) dS = dx dy 3f (x, y) √
S 1 1 2xy
Z 2 Z 2/x
= dx dy 3(x + y)
1 1
Z 2 2/x Z 2
1 2 1
=3 xy + y 2
dx =3 dx 2 + 2 − x −
1 2 1 1 x 2
( 2 2 )
2

3 2 x 3 1
=3 − − =3 −1+2−2+
2 x 1 2 1 2 2
=3
3.3.6.35. ∗. Solution. Since x = g(y, z) with g(x, y) = y 2 + z 2 , (3.3.2)

gives
n̂ dS = ±(1, −gy , −gz ) dydz = ±(1, −2y, −2z) dydz
We choose the + sign so that n̂ · ı̂ı > 0. Furthermore
S = (x, y, z) x = y 2 + z 2 , x ≤ 2y

= (x, y, z) x = y 2 + z 2 , y 2 + z 2 ≤ 2y

= (x, y, z) x = y 2 + z 2 , (y − 1)2 + z 2 ≤ 1

= (x, y, z) x = y 2 + z 2 , (y, z) in D

where D = (x, y) (y − 1)2 + z 2 ≤ 1 is a disk with radius 1. Hence
ZZ ZZ ZZ
F · n̂ dS = (2, z, y) · (1, −2y, −2z) dydz = (2 − 4yz) dydz
S D D
Since −4yz is odd under z → −z the integral of −4yz is zero and

ZZ
F · n̂ dS = 2 Area(D) = 2π
S
3.3.6.36. ∗. Solution. For the specified F and the surface x = f (x, y) =

1 − 41 x2 − y 2 , by (3.3.2),
x
n̂ dS = − fx ı̂ı − fy ̂ + k̂ dxdy = ı̂ı + 2y ̂ + k̂ dxdy
  2
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z 
3y 2 + z x − x2 1
= ̂ + (1 − 2x − 6y) k̂
∇ × F · n̂ dS = (2y + 1 − 2x − 6y) dxdy = (1 − 2x − 4y) dxdy
The domain of integration is 1 − 41 x2 − y 2 ≥ 0 or 14 x2 + y 2 ≤ 1. This is an

ellipse. Call it D. So
ZZ ZZ
∇ × F · n̂ dS = (1 − 2x − 4y) dxdy
S D
The integrals over D of x, which is odd under x → −x, and of y, which is

odd under y → −y, are both zero. As the ellipse D has area A = π×2×1 =
2π ZZ ZZ
∇ × F · n̂ dS = (1 − 2x − 4y) dxdy = A = 2π
S D
3.3.6.37. Solution. Due to the symmetry of the surface and the vector
field under reflection in the xy-plane, i.e. under z → −z, it is sufficient to
compute the integral over the upper half of the surface, where z ≥ 0, and
then multiply the result by 2. The upper half of the surface consists of two
pieces, S1 and S2 , where S1 is the part on the sphere and S2 is the part
on the hyperboloid. S1 and S2 intersect on a circle. The circle is obtained
by imposing the two equations x2 + y 2 + z 2 = 16 and x2 + y 2 − z 2 = 8
simultaneously. √ Thus we have x2 + y 2 = 12 and z = 2, or in cylindrical
coordinates r = 12, z = 1, on the circle. Here is a sketch of a cross-section
of the apple core.
z
S1
√
r= 12, z = 2
φ1
S2
r
Let φ1 be the angle between z-axis and√the cone √formed by connecting

the circle to the origin. We have tan φ1 = 12/2 = 3. Thus φ1 =RRπ/3.
We’ll use spherical coordinates to compute the flux integral S1
F·
n̂ dS . As the spherical coordinate ρ = 4 on all of S1 , we can paramerize
S1 by
r(θ, ϕ) = 4 cos θ sin ϕı̂ı + 4 sin θ sin ϕ ̂ + 4 cos ϕ k̂

0 ≤ θ ≤ 2π, 0 ≤ φ ≤ π/3
So
∂r
= −4 sin θ sin ϕı̂ı + 4 cos θ sin ϕ ̂
∂θ
∂r
= 4 cos θ cos ϕı̂ı + 4 sin θ cos ϕ ̂ − 4 sin ϕ k̂
∂ϕ
 
ı̂ı ̂ k̂
∂r ∂r
× = det −4 sin θ sin ϕ 4 cos θ sin ϕ 0 
∂θ ∂ϕ
4 cos θ cos ϕ 4 sin θ cos ϕ −4 sin ϕ
= −16 cos θ sin2 ϕ , sin θ sin2 ϕ , sin ϕ cos ϕ

= −4 (sin ϕ) r(θ, ϕ)
and, by (3.3.1),
∂r ∂r
n̂ dS = ± × dθdϕ = ∓4 (sin ϕ) r(θ, ϕ) dθdϕ
∂θ ∂ϕ
To get the outward pointing normal, i.e. the normal point in the same
direction as r(θ, ϕ), we take the plus sign. As F = r(θ, ϕ),
42
z }| {
F · n̂ dS = 4 |r(θ, ϕ)|2 sin ϕ dθ dϕ = 64 sin ϕ dθ dϕ
and
ZZ Z 2π Z π/3 h iπ/3
F · n̂ dS = 64 dθ dϕ sin ϕ = 64 · 2π − cos ϕ = 64π
S1 0 0 0
The surface S2 can be parametrized using the cylindrical coordinates θ

and z. Indeed, we have
p
r = x2 + y 2 = (8 + z 2 )1/2
for the hyperboloid and we always have x = r cos θ and y = r sin θ. Thus
the hyperboloid has the following parametrization:
R(θ, z) = (8 + z 2 )1/2 cos θ ı̂ı + (8 + z 2 )1/2 sin θ ̂ + z k̂
The range for the parameters of S2 is 0 ≤ θ ≤ 2π and 0 ≤ z ≤ 2. We have

∂R
= −(8 + z 2 )1/2 sin θ ı̂ı + (8 + z 2 )1/2 cos θ ̂ + 0 k̂
∂θ
∂R
= z(8 + z 2 )−1/2 cos θ ı̂ı + z(8 + z 2 )−1/2 sin θ ̂ + k̂
∂z
and
 
ı̂ı ̂ k̂
∂R ∂R
× = det −(8 + z )1/2 sin θ
 2 2 1/2
(8 + z ) cos θ 0
∂θ ∂z
z(8 + z 2 )−1/2 cos θ z(8 + z 2 )−1/2 sin θ 1
= (8 + z 2 )1/2 cos θ ı̂ı + (8 + z 2 )1/2 sin θ ̂ − z k̂
= xı̂ı + y ̂ − z k̂
Note that ∂R ∂R
∂θ × ∂z is pointing downward (since z > 0) and hence outward.
Since F · ∂θ × ∂R
∂R

∂z = (x, y, z) · (x, y, −z) = x2 + y 2 − z 2 = 8 on S2 , we
have
ZZ ZZ Z 2 Z 2π
F · n̂ dS = F · (Rθ × Rz ) dθ dz = dz dθ 8 = 32π
S2 S2 0 0
Finally, the flux integral over the whole apple core surface is
Z Z ZZ

2 F · n̂ dS + F · n̂ dS = 2 64π + 32π = 192π
S1 S2
3.3.6.38. ∗. Solution. (a) The specified surface is of the form
G(x, y, z) = x2 + z 2 − cos2 y = 0
So one normal vector at the point 21 , π4 , 21 is

1 π 1
∇G , , = 2x , 2 sin y cos y , 2z 1 π 1 = (1, 1, 1)
2 4 2 2, 4 ,2)
and an equation for the tangent plane at 21 , π4 , 12 is

(1, 1, 1) · x − 1/2 , y − π/4 , z − 1/2 = 0 or x + y + z = 1 + π/4
(b) For each fixed y, x2 + z 2 = cos2 y is a circle of radius | cos y|. So we

may parametrize the surface by
π
r(θ, y) = cos y cos θ , y , cos y sin θ 0 ≤ θ < 2π, 0 ≤ y ≤
2
Then
∂r
= − cos y sin θ , 0 , cos y cos θ
∂θ
∂r
= − sin y cos θ , 1 , − sin y sin θ
∂y
∂r ∂r
× = − cos y cos θ , − sin y cos y , − cos y sin θ
∂θ ∂y

∂r ∂r
q
dS = × dθ dy = cos y 1 + sin2 y dθ dy
∂θ ∂y
So the specified integral is

π
ZZ Z 2
Z 2π q
sin y dS = dy dθ cos y 1 + sin2 y sin y
S 0 0
Z π q
2
= 2π dy 1 + sin2 y sin y cos y
0
Z 2 √
=π du u with u = 1 + sin2 y, du = 2 sin y cos y dy
1
2
u3/2

=π
3/2 1
2π √

= 2 2−1
3
3.3.6.39. ∗. Solution. (a) By definition F is a conservative vector field
with potential f . Suppose that the curve C starts at P1 , on S, and ends
at P2 , on S. Then f (P1 ) = f (P2 ) = c and, by Theorem 2.4.2,
Z
F · dr = f (P2 ) − f (P1 ) = c − c = 0
C
(b) Since F = ∇ f , F is normal to the level surfaces of f by Lemma

2.3.6. So, at any point of S, F is a scalar multiple of n̂ and F × G is
perpendicular to n̂. Thus (F × G) · n̂ = 0 and

ZZ
(F × G) · n̂ dS = 0.
S
3.3.6.40. ∗. Solution. (a) (i) Here is a sketch of the part of the plane
in question.
z
(0, 0, 16/3)
y
(0, 4, 0)
(8, 0, 0)
x
We can use x and y as parameters. As we can rewrite the equation of
the plane as z = 13 16 − 2x − 4y , we have the parametrization

1
r(x, y) = xı̂ı + y ̂ +16 − 2x − 4y k̂
3
In terms of x and y, the condition z = 13 16−2x−4y ≥ 0 is 16−2x−4y ≥ 0

or x + 2y ≤ 8. So the domain is

(x, y) x ≥ 0, y ≥ 0, x + 2y ≤ 8
Renaming x to u and y to v, the parametrization is also

1
r(u, v) = u , v , (16 − 2u − 4v) k̂ u ≥ 0, v ≥ 0, u + 2v ≤ 8
3
(a) (ii) Here is a sketch of the part of the cap in the first octant.
z
√ x2 + y 2 + z 2 = 16
z=2 2
S
x
The full sphere can be parametrized, using spherical coordinates with
ρ = 4, by

0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤ π
√
In these coordinates, the condition 4/ 2 ≤ z ≤ 4 is
4 1
√ ≤ 4 cos ϕ ≤ 4 ⇐⇒ √ ≤ cos ϕ ≤ 1
2 2
π
⇐⇒ 0 ≤ ϕ ≤
4
So our parametrization is

π
0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤
4
Renaming θ to u and ϕ to v, the parametrization is also
π
r(u, v) = 4 cos u sin v , 4 sin u sin v , 4 cos v 0 ≤ u ≤ 2π, 0 ≤ v ≤
4
(a) (iii) Here is a sketch of the hyperboloid.
p
If we use x and y as parameters, then, since z = 1 + x2 + y 2 , we have
the parametrization
p
r(x, y) = xı̂ı + y ̂ + 1 + x2 + y 2 k̂
In terms of x and y, the condition 1 ≤ z ≤ 10 is

p
1 ≤ 1 + x2 + y 2 ≤ 10 or 0 ≤ x2 + y 2 ≤ 99
So the domain is
(x, y) x2 + y 2 ≤ 99

Renaming x to u and y to v, the parametrization is also

p
u2 + v 2 ≤ 99

r(u, v) = u , v , 1 + u2 + v 2
Alternatively, if we replace x and y with the polar coordinates r and θ,

we get the parametrization
p √
r(r, θ) = r cos θ ı̂ı + r sin θ ̂ + 1 + r2 k̂ 0 ≤ θ ≤ 2π, 0 ≤ r ≤ 99
Renaming r to u and θ to v, the parametrization is also

p √
r(u, v) = u cos v , u sin v , 1 + u2 0 ≤ v ≤ 2π, 0 ≤ u ≤ 99
(b) Let’s use the parametrization

π
0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤
4
from part (a) (ii), so that
 
ı̂ı ̂ k̂
∂r ∂r
∂θ ∂ϕ
= −16 cos θ sin2 ϕ , sin θ sin2 ϕ , sin ϕ cos ϕ

and, by (3.3.1),

∂r ∂r
dS = × dθdϕ
∂θ ∂ϕ
q
= 16 sin ϕ cos2 θ sin2 ϕ + sin2 θ sin2 ϕ + cos2 ϕ dθdϕ
= 16 sin ϕ dθdϕ
So the area is
ZZ Z π/4 Z 2π Z π/4
Area = dS = dϕ dθ 16 sin ϕ = 32π dϕ sin ϕ
S 0 0 0
h iπ/4
= 32π − cos ϕ
0
h 1 i
= 32π 1 − √
2
3.3.6.41. ∗. Solution. Solution 1 — using tweaked spherical coordi-
nates.
√ have to parametrize S. It is natural to use spherical coordinates
First we
with ρ = 2. However if we use the standard spherical coordinates
√ √ √
x = 2 sin ϕ cos θ y = 2 sin ϕ sin θ z = 2 cos ϕ
the condition y ≥ 1 becomes sin ϕ sin θ ≥ √12 , which is very complicated.

So let’s back up and think a bit before we compute. The condition z ≥ 1,
as opposed to y ≥ 1, is easy to implement in spherical coordinates. It is
cos ϕ ≥ √12 or 0 ≤ ϕ ≤ π4 . So let’s modify spherical coordinates to make
the y-axis play the role of the z-axis, by just exchanging y and z in the
parametrization.
√ √ √
x = 2 sin ϕ cos θ y = 2 cos ϕ z = 2 sin ϕ sin θ
√
The condition y ≥ 1 is then 2 cos ϕ ≥ 1, which is turn is 0 ≤ ϕ ≤ π4 . Since
we have just exchanged y and z we could probably just guess n̂ dS and dS
from√standard spherical coordinates. (See Appendix A.6.3 and recall that
ρ = 2.) But to be on the safe side, let’s derive them. We are using the
parametrization
√ √ √
r(θ, ϕ) = 2 sin ϕ cos θ ı̂ı + 2 cos ϕ ̂ + 2 sin ϕ sin θ k̂
π
0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤
4
Since
∂r √ √
= − 2 sin ϕ sin θ ı̂ı + 2 sin ϕ cos θ k̂
∂θ
∂r √ √ √
= 2 cos ϕ cos θ ı̂ı − 2 sin ϕ ̂ + 2 cos ϕ sin θ k̂
∂ϕ
so that
 
∂r ∂r √ ı̂ı ̂ √ k̂
× = det − 2 sin ϕ sin θ
√0 √2 sin ϕ cos θ

∂θ ∂ϕ √
2 cos ϕ cos θ − 2 sin ϕ 2 cos ϕ sin θ
= 2 sin2 ϕ cos θ ı̂ı + 2 sin ϕ cos ϕ ̂ + 2 sin2 ϕ sin θ k̂
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dθdϕ = ±2 sin ϕ sin ϕ cos θ , cos ϕ , sin ϕ sin θ dθdϕ
∂θ ∂ϕ

∂r ∂r
dS = × dθdϕ = 2 sin ϕ dθdϕ
∂θ ∂ϕ
Choose the plus sign to give the outward pointing normal.

(a) The specified integral is
y3
ZZ Z π/4 Z 2π z√ }| {
3
y 3 dS = 2 dϕ dθ sin ϕ 2 cos ϕ
S 0 0
√ Z π/4
= 8 2π dϕ sin ϕ cos3 ϕ
0
π/4
√ cos4 ϕ

= 8 2π −
4
0
√

1 3
= 2 2π 1 − =√ π
4 2
(b) The specified integral is

ZZ

xy ı̂ı + xz ̂ + zy k̂ · n̂ dS
S
Z π/4 Z 2π
=2 dϕ dθ sin ϕ 2 sin ϕ cos ϕ cos θ , 2 sin2 ϕ sin θ cos θ ,
0 0

2 sin ϕ cos ϕ sin θ ·

· sin ϕ cos θ , cos ϕ , sin ϕ sin θ
Z π/4 Z 2π n
=4 dϕ dθ sin3 ϕ cos ϕ cos2 θ + sin3 ϕ cos ϕ sin θ cos θ
0 0
o
+ sin3 ϕ cos ϕ sin2 θ
"Z # Z
π/4 2π
3

=4 dϕ sin ϕ cos ϕ dθ 1 + sin θ cos θ
0 0
π/4 2π
sin4 ϕ sin2 θ

=4 θ+
4 0 2 0
1 π
=4× × (2π) =
16 2
Solution 2 — parametrizing by x and z.
We can also parametrize S by using x and z as parameters. On S,
√
• y = 2 − x2 − z 2 and
√
• y runs over the range 1 ≤ y ≤ 2. Correspondingly, x2 + z 2 = 2 − y 2
runs over 0 ≤ x2 + z 2 ≤ 1
So we can use the parametrization
p
r(x, z) = xı̂ı + 2 − x2 − z 2 ̂ + z k̂ 0 ≤ x2 + z 2 ≤ 1
Since
∂r x
= ı̂ı − √ ̂
∂x 2 − x2 − z 2
∂r z
= −√ ̂ + k̂
∂z 2 − x2 − z 2
so that
 
ı̂ı ̂ k̂
∂r ∂r
×
1
= det  − √2−xx2 −z2 0
∂x ∂z

0 − √2−xz2 −z2 1
x z
= −√ ı̂ı − ̂ − √ k̂
2
2−x −z 2 2 − x2 − z 2
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dxdz
∂x
∂z
x z
=∓ √ ,1, √ dxdz
2 − x2 − z 2 2 − x2 − z 2

∂r ∂r
dS = × dxdz
∂x ∂z
r √
x2 + z 2 2
= 1+ 2 2
dxdz = √ dxdz
2−x −z 2 − x2 − z 2
Choose the plus sign to give the outward pointing normal.
(a) The specified integral is
y3
ZZ ZZ √ z }| {
2 3/2
y 3 dS = dxdz √ 2 − x2 − z 2
S x2 +z 2 ≤1 2 − x2 − z 2
√ ZZ
dxdz 2 − x2 − z 2

= 2
x2 +z 2 ≤1
Switching to polar coordinates with x = r cos θ and z = r sin θ,

ZZ √ Z 2π Z 1
y 3 dS = 2 dθ dr r(2 − r2 )
S 0 0
1
√ r4 √

2 3 3
= 2 (2π) r − = 2 (2π) =√ π
4 0 4 2
(b) The specified integral is
ZZ

xy ı̂ı + xz ̂ + zy k̂ · n̂ dS
S
ZZ p p
= dxdz x 2 − x2 − z 2 , xz , z 2 − x2 − z 2 ·
2 2
x +z ≤1

x z
√ ,1, √
2 − x2 − z 2 2 − x2 − z 2
ZZ
dxdz x2 + xz + z 2

=
x2 +z 2 ≤1
Switching to polar coordinates with x = r cos θ and z = r sin θ,

ZZ

xy ı̂ı + xz ̂ + zy k̂ · n̂ dS
S
Z 2π Z 1
= dr r(r2 cos2 θ + r2 sin θ cos θ + r2 sin2 θ)
dθ
0 0
Z 1 Z 2π
3

= dr r dθ 1 + sin θ cos θ
0 0
4 1
2π
sin2 θ

r
= θ+
4 0 2 0
1 π
= × (2π) =
4 2
3.3.6.42. ∗. Solution. First observe that,
• because (x + y + 1)2 ≥ 0, all points on (x + y + 1)2 + z 2 = 4 have
|z| ≤ 2 and that,
• for |z0 | ≤ 2, the surface (x + y + 1)2 + z 2 = 4 intersects the horizontal
2 2
p z = z0 on (x + y + 1) = 4 − z0 , i.e. on the two lines x + y =
plane
2
± 4 − z0 − 1, z = z0 .
p
• The line x + y = ± 4 −p z02 − 1, z = z0 intersects the first octant if
and only if z0 ≥ 0 and ± 4 − z02 − 1 ≥ 0.
p
• Thus x + y = − 4 − z02 − 1, z = z0 never intersects the first octant
and
p
• x + y = √4 − z02 − 1, z = z0 intersects the first octant if and only if
0 ≤ z0 ≤ 3.
p
• When z0 = 0, the line x + y = 4 − z02 − 1, z = z0 is x + y = 1,
z = 0.
√ p
• When z0 = 3, the line x + y = 4 − z02 − 1, z = z0 is x + y = 0,
z = 2.
• So as (x, y, z) runs over S, (x, y) runs over the triangle x ≥ 0, y ≥ 0,
x + y ≤ 1.
Let G(x, y, z) = (x + y + 1)2 + z 2 . Then
∇G 2(x + y + 1)ı̂ı + 2(x + y + 1)̂ + 2z k̂
n̂ dS = ± dxdy = ± dxdy
∇ G · k̂ 2z
For the downward normal, we need the minus sign, so
h (x+y+1)ı̂ı + (x+y+1)̂ + z k̂ i
F · n̂ dS = − xy ı̂ı + (z − xy) ̂ · dxdy
z
1
= − xy(x + y + 1) + (z − xy)(x + y + 1) dxdy
z
1
= − z(x + y + 1) dxdy
z
= −(x + y + 1) dxdy
The domain of integration is x ≥ 0, y ≥ 0, x + y ≤ 1, so
ZZ Z 1 Z 1−x
F · n̂ dS = − dx dy (x + y + 1)
S 0 0
Z 1
1
h i
=− dx (1 + x)(1 − x) + (1 − x)2
0 2
Z 1 h3 1 i h3 1 1i 5
=− dx − x − x2 = − − − =−
0 2 2 2 2 6 6
4.1.6 · Exercises
4.1.6.1. ∗. Solution. (a) A. The angle between F and dr is less than
90◦ along the entire path. So F · dr > 0 along the entire path and the work
is positive. R
(b) B. F is perpendicular to dr along all of C2 . So C2 F · dr = 0.
(c) C. It looks like Px = Qy = 0 at N . So ∇ · F = 0 at N .
(d) A. At Q, the vertical component of F is increasing from left to right
(so that Qx > 0) and the horizontal component of F is decreasing from
bottom to top (so that Py < 0). So Qx − Py > 0 at N .
(e) B. At D, the horizontal component of F is increasing from left to
right, so that Px > 0.
4.1.6.2. Solution. No. The vector field F(x, y, z) = ı̂ı + y k̂ has
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z 
1 0 y
= ı̂ı
has dot product 1 with F(x, y, z) (for all x, y, z) and so is not perpendicular
to it.
4.1.6.3. Solution. (a) By the product rule
∂ ∂ ∂
∇ · (f F) = (f F1 ) + (f F2 ) + (f F3 )
∂x ∂y ∂z
∂F1 ∂F2 ∂F3
= f +f +f
∂x ∂y ∂z
∂f ∂f ∂f
+ F1 + F2 + F3
∂x ∂y ∂z
= f ∇ · F + F · ∇f
(b) Again by the product rule
∇ · (F × G)
∂ ∂ ∂
= (F2 G3 − F3 G2 ) + (F3 G1 − F1 G3 ) + (F1 G2 − F2 G1 )
∂x ∂y ∂z
∂F2 ∂F3 ∂F3 ∂F1 ∂F1 ∂F2
= G3 − G2 + G1 − G3 + G2 − G1
∂x ∂x ∂y ∂y ∂z ∂z
∂G3 ∂G2 ∂G1 ∂G3 ∂G2 ∂G1
+ F2 − F3 + F3 − F1 + F1 − F2
∂x ∂x ∂y ∂y ∂z ∂z

∂F3 ∂F2 ∂F1 ∂F3 ∂F2 ∂F1
= − G1 + − G2 + − G3
∂y ∂z ∂z ∂x ∂x ∂y

∂G3 ∂G2 ∂G1 ∂G3 ∂G2 ∂G1
− F1 − − F2 − − F3 −
∂y ∂z ∂z ∂x ∂x ∂y
= G · (∇∇ × F) − F · (∇ ∇ × G)
(c) Recall that ∇ 2 (f g) = ∇ · ∇ (f g) . First

∂ ∂ ∂
∇ (f g) = ı̂ı (f g) + ̂ (f g) + k̂ (f g)
∂x ∂y ∂z
∂f ∂f ∂f
= ı̂ıg + ̂g + k̂g
∂x ∂y ∂z
∂g ∂g ∂g
+ ı̂ıf + ̂f + k̂f
∂x ∂y ∂z
∇f + f∇
= g∇ ∇g
So by part (a), twice,
∇ 2 (f g) = ∇ · g∇

∇f + ∇ · f∇∇g

= g ∇ · ∇f + ∇g · ∇f + f ∇ · ∇g + ∇f · ∇g
= f ∇ 2 g + 2∇
∇f · ∇ g + g ∇ 2 f
4.1.6.4. Solution. (a) By definition

∂ ∂ ∂
∇ · (xı̂ı + y ̂ + z k̂) = x + y + z =3
∂x ∂y ∂z
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × (xı̂ı + y ̂ + z k̂) = det  ∂x ∂y ∂z  = 0
x y z
(b) By definition
∂ ∂ ∂
∇ · (xy 2ı̂ı − yz 2̂ + zx2 k̂) = xy 2 + − yz 2 + zx2

∂x ∂y ∂z
= y 2 − z 2 + x2
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × (xy 2ı̂ı − yz 2̂ + zx2 k̂) = det  ∂x ∂y ∂z 
xy −yz zx2
2 2
= 2yz ı̂ı − 2xz ̂ − 2xy k̂
(c) By definition
!
x y
∇· p ı̂ı + p ̂
x2 + y 2 x2 + y 2
! !
∂ x ∂ y
= p + p
∂x x2 + y 2 ∂y x2 + y 2
1 x2 1 y2
=p − 3/2
+p − 3/2
x2 + y 2 [x2 + y 2 ] x2 + y 2 [x2 + y 2 ]
x2 + y 2 − x2 + x2 + y 2 − y 2
= 3/2
[x2 + y 2 ]
1
=p
x2 + y2
!
x y
∇× p ı̂ı + p ̂
x2 + y 2 x2 + y 2
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
= det  ∂x ∂y ∂z 
√ 2x 2 √ 2y 0
x +y x +y 2
!
xy xy
= − 3/2
+ 3/2
k̂
[x2 + y 2 ] [x2 + y 2 ]
=0
(d) By definition
!
y x
∇· −p ı̂ı + p ̂
x2 + y 2 x2 + y 2
! !
∂ y ∂ x
= −p + p
∂x x2 + y 2 ∂y x2 + y 2
xy xy
= 3/2
− 3/2
2
[x + y ] 2 [x + y 2 ]
2
=0
!
y x
∇ × −p ı̂ı + p ̂
x2 + y 2 x2 + y 2
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
= det 

∂x ∂y ∂z 
y x
−√ 2 2 √ 2 2 0
x +y x +y
!
1 x2 1 y2
= p − 3/2
+p − 3/2
k̂
x2 + y 2 [x2 + y 2 ] x2 + y 2 [x2 + y 2 ]
x2 + y 2 − x2 + x2 + y 2 − y 2
= 3/2
k̂
[x2 + y 2 ]
k̂
=p
x2 + y2
r
4.1.6.5. ∗. Solution. (a) We are to compute the divergence of r =
x ı̂ı+y ̂+z k̂
[x2 +y 2 +z 2 ]1/2
. Since
∂ x 1 1 x(2x)
1/2
= 1/2
−
∂x [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]3/2
y2 + z2
= 3/2
[x2 + y 2 + z 2 ]
∂ y 1 1 y(2y)
1/2
= 1/2
−
∂y [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]3/2
x2 + z 2
= 3/2
[x2 + y 2 + z 2 ]
∂ z 1 1 z(2z)
= −
∂z [x2 + y 2 + z 2 ]1/2 [x2 + y 2 + z 2 ]
1/2 2 [x2 + y 2 + z 2 ]3/2
x2 + y 2
= 3/2
[x2 + y 2 + z 2 ]
the specified divergence is

r 2x2 + 2y 2 + 2z 2 2r2 2
∇ = 3/2
= 3
=
r [x2 + y 2 + z2] r r
(b)
 
ı̂ı ̂ k̂
∇ × yz ı̂ı + 2xz ̂ + exy k̂ = det  ∂x
∂ ∂ ∂ 
∂y ∂z 
xy
yz 2xz e
= xexy − 2x ı̂ı − yexy − y ̂ + z k̂

k/2
4.1.6.6. ∗. Solution. (a) Since rk = x2 + y 2 + z 2 ,
∂ k k 2 k −1
r = 2x x + y 2 + z 2 2 = k (r · ı̂ı) rk−2
∂x 2
∂ k k 2 k −1
r = 2y x + y 2 + z 2 2 = k (r · ̂) rk−2
∂y 2
∂ k k 2 k −1
r = 2z x + y 2 + z 2 2 = k (r · k̂) rk−2
∂z 2
We want k = −3.
(b) Using the computation in part (a)
∂ ∂ ∂
∇ · (rk r) = (xrk ) + (yrk ) + (zrk )
∂x ∂y ∂z
∂ ∂ ∂
= 3rk + x rk + y rk + z rk
∂x ∂y ∂z
= 3rk + x kx rk−2 + y ky rk−2 + z kz rk−2

= 3 + k rk

We want k = 2.
(c) Recalling that ∇ 2 = ∇ · ∇ ,
∇ 2 (rk ) = ∇ · ∇ (rk )

= ∇ · (krk−2 r) by part (a)

= k(3 + k − 2)rk−2 by part (b), but with k replaced by k − 2
We want k = −2.
4.1.6.7. ∗. Solution. (a)
∂x ∂y ∂z
∇·r= + + =3
∂x ∂y ∂z
(b)

2 ∂ ∂ ∂
x2 + y 2 + z 2 = 2xı̂ı + 2y ̂ + 2z k̂ = 2r

∇ (r ) = ı̂ı + ̂ + k̂
∂x ∂y ∂z
(c) Since
 
ı̂ı ̂ k̂
r × a = det  x y z  = ı̂ı a3 y−a2 z + ̂ a1 z−a3 x + k̂ a2 x−a1 y
a1 a2 a3
we have
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (r × a) = det 
 
∂x ∂y ∂z 
a3 y − a2 z a1 z − a3 x a2 x − a 1 y
= −2a1 ı̂ı − 2a2 ̂ − 2a3 k̂

= −2a
(d) Since

∂ ∂ ∂ 1/2
∇ (r) = ı̂ı + ̂ + k̂ x2 + y 2 + z 2
∂x ∂y ∂z
x y x
= ı̂ı 1/2 + ̂ 1/2 + k̂ 1/2
2 2
x +y +z 2 2
x +y +z2 2 x + y + z2
2 2
we have
∂ x ∂ y
∇ · ∇(r) = +
∂x x2 + y 2 + z 2 1/2 ∂y x2 + y 2 + z 2 1/2
∂ z
+
∂z x2 + y 2 + z 2 1/2
3 1 2x2 + 2y 2 + 2z 2
= 1/2 − 2 3/2
x2 + y 2 + z 2 x2 + y 2 + z 2
2
= 1/2
x2 + y 2 + z 2
2
=
r

1 ∂ ∂ ∂ −1/2
∇ = ı̂ ı + ̂ + k̂ x2 + y 2 + z 2
r ∂x ∂y ∂z
x y z
= −ı̂ı 3/2 − ̂ 3/2 − k̂ 3/2
2
x +y +z 2 2 2 2
x +y +z 2 x + y + z2
2 2
x y x
= −ı̂ı 3 − ̂ 3 − k̂ 3
r r r
we have a = −3.
(b) Since
∂ h 2 1/2 i ∂ h 2 1/2 i
x + y2 + z2 x + y2 + z2

∇ · rr = x + y
∂x ∂y
∂ h 2 1/2 i
+ x + y2 + z2 z
∂z
1/2 1 2x2 + 2y 2 + 2z 2
= 3 x2 + y 2 + z 2 +
2 x2 + y 2 + z 2 1/2
1/2
= 4 x2 + y 2 + z 2
= 4r
we have a = 4.
(c) Since

∂ ∂ ∂ 3/2
∇ (r3 ) = ı̂ı + ̂ + k̂ x2 + y 2 + z 2
∂x ∂y ∂z
2 1/2
2 2
1/2 1/2
+ ̂ 3y x2 + y 2 + z 2 + k̂ 3z x2 + y 2 + z 2

= ı̂ı 3x x + y + z
= 3rr
we have
∇ · ∇ (r3 ) = ∇ · 3rr = 3 ∇ · rr = 3 4r

by part (b)
= 12r
so that a = 12.
∂ ∂ ∂
4.1.6.9. Solution. (a) Since ∇ · F = ∂x (1 + yz) + ∂y (2y + zx) + ∂z (3z 2 +
xy) = 2 + 6z 6= 0, F fails the screening test and cannot have a vector
potential.
(b) The vector field A = A1ı̂ı + A2̂ is a vector potential for G if and
only if G = ∇ × A, which is the case if and only if
∂A2 1
− = yz ⇐⇒ A2 = − yz 2 + B2 (x, y)
∂z 2
∂A1 1 2
= zx ⇐⇒ A1 = xz + B1 (x, y)
∂z 2
∂A2 ∂A1 ∂B2 ∂B1
− = xy ⇐⇒ − = xy
∂x ∂y ∂x ∂y
There are infinitely many solutions to ∂B ∂B1

∂x − ∂y = xy. In fact B2 is
2
completely arbitrary. If one chooses B2 = 0, then B1 = − 21 xy 2 does the

job. If one chooses B1 = 0, then B2 = 12 x2 y does the job. Thus two
solutions are A = 12 (z 2 − y 2 )xı̂ı − 21 yz 2̂ and A = 12 xz 2ı̂ı + 12 (x2 − z 2 )y̂.
4.1.6.10. ∗. Solution. (a) F is well-defined wherever the denominator

x2 + z 2 is nonzero. So the (largest possible) domain is
D = (x, y, z) x2 + z 2 6= 0

(b) As preliminary computations, let’s find

−x2 + z 2

∂ −z −1 2z(−z)
= − 2 = 2
∂z x2 + z 2 x2 + z 2 (x2 + z 2 ) (x2 + z 2 )
−x2 + z 2

∂ x 1 2x(x)
2 2
= 2 2
− 2 = 2
∂x x + z x +z (x2 + z 2 ) (x2 + y 2 )
So the curl of F is
 
ı̂ı ̂ k̂ !
∂ ∂ ∂ −x2 + z 2 −x2 + z 2
∇ × F = det  =− − ̂
 
∂x ∂y ∂z  2 2
−z x (x2 + y 2 ) (x2 + y 2 )
x2 +z 2 y x2 +z 2
=0
on the domain of F.
(c) As preliminary computations, let’s find

∂ −z 2x(−z) 2xz
=− 2 = 2
∂x x2 + z 2 2
(x + z ) 2 (x + z 2 )
2

∂ x 2z(x) −2xz
2 2
=− 2 = 2
∂z x + z 2
(x + z ) 2 (x + y 2 )
2
So the divergence of F is

∂ −z ∂ ∂ x
∇·F= + (y) + =1
∂x x2 + z 2 ∂y ∂z x2 + z 2
(d) By part (b), the vector field passes the conservative field screening
test ∇ × F = 0. But we should still be suspicious because of the similarity
of F to the vector field of Examples 2.3.14 and 4.3.8.
So let’s compute the line integral of F around the (closed) circle y = 0,
x2 + z 2 = 1, parametrized by
r(t) = cos tı̂ı + sin t k̂ r0 (t) = − sin tı̂ı + cos t k̂
The line integral is
2
−z
2
x r0 (t)
x2 +y 2
Z Z 2π zx }|
+z
{ z}|{ z }| {
F · dr = − sin t ı̂ı + cos t k̂ · − sin tı̂ı + cos t k̂ dt
C 0
Z 2π
= dt = 2π
0
As the integral of F around the simple closed curve C is not zero, F cannot
be conservative on D. See Theorem 2.4.7 and Examples 2.3.14 and 4.3.8.
4.1.6.11. ∗. Solution. (a) By the vector identity of Theorem 4.1.7.a,
∇·F=∇·∇×G=0
So we must have

0 = ∇ · F = ∇ · (xz + xy)ı̂ı + α(yz − xy)̂ + β(yz + xz)k̂
= (z + y) + α(z − x) + β(y + x)
This is true for all (x, y, z) if and only if α = β = −1.

(b) Since

∇ × G = ∇ × xyzı̂ı − xyz̂ + g(x, y, z)k̂
= (gy + xy)ı̂ı − (gx − xy) hj + (−yz − xz) k̂
we will have that ∇ × G = F if and only if
(gy + xy)ı̂ı − (gx − xy) ̂ + (−yz − xz) k̂

= (xz + xy)ı̂ı − (yz − xy) ̂ − (yz + xz) k̂
which is the case if and only if
gy = xz, gx = yz
The first equation, gy = xz, is satisfied if and only if g = xyz + h(x, z).
The second equation is also satisfied if and only if gx = yz + hx (x, z) = yz.
This is the case if and only if hx (x, z) = 0. That is, if and only if h is
independent of x. Equivalently, if and only if h(x, z) = w(z) for some
function w(z). So, in fact, any function of the form g(x, y, z) = xyz + w(z)
will work.
4.1.6.12. Solution. (a) Denote by θ the angle between â and r. The
point r is a distance ` = |r| sin θ from the axis of rotation. So as the body
rotates, the point sweeps out a circle of radius ` centred on the axis of
rotation.
Ω
â
v
r
θ
0
In one second the point sweeps out an arc of this circle that subtends
Ω
an angle of Ω radians. This arc is the fraction 2π of a full circle and so
Ω
has length 2π 2π` = Ω` = Ω|r| sin θ. Thus the point is moving with speed
Ω|r| sin θ. The velocity vector of the point must have length Ω|r| sin θ and
direction perpendicular to both â and r. The vector Ω × r is perpendicular
to both r and Ω = Ωâ and has length |Ω Ω| |r| sin θ = Ω|r| sin θ as desired.
So the velocity vector is either Ω × r or its negative. By the right hand
rule it is Ω × r.
(b) By vector identities
∇ × F) − F · (∇
∇ · (F × G) = G · (∇ ∇ × G)
∇ × (F × G) = F(∇∇ · G) − (∇
∇ · F)G + (G · ∇ )F − (F · ∇ )G
(which are Theorems 4.1.4(d) and 4.1.5(d)) and the assumption that Ω is
constant
∇ × (Ω ∇ · r) − (∇
Ω × r) = Ω (∇ ∇ · Ω )r + (r · ∇ )Ω
Ω − (Ω
Ω · ∇ )r
∇ · r) − (Ω
= Ω (∇ Ω · ∇ )r
∇ · (Ω ∇ × Ω ) − Ω · (∇
Ω × r) = r · (∇ ∇ × r) = −Ω ∇ × r)
Ω · (∇
Substituting in
∂x ∂y ∂z
∇·r= + + =3
∂x ∂y ∂z
∂z ∂y ∂x ∂z ∂y ∂x
∇×r= − ı̂ı + − ̂ + − k̂ = 0
∂y ∂z ∂z ∂x ∂x ∂y
∂ ∂ ∂
Ω · ∇ )r = Ω1
(Ω + Ω2 + Ω3 xı̂ı + y̂ + z k̂
∂x ∂y ∂z
= Ω1ı̂ı + Ω2̂ + Ω3 k̂ = Ω
gives
∇ × (Ω
Ω × r) = 2 Ω ∇ · (Ω
Ω × r) = 0
(c) The students are a distance 6378 sin(90◦ − 49◦ ) = 6378 cos(49◦ ) =
4184 km from the axis of rotation. The rate of rotation is Ω = 2π
24 radians
per hour. In each hour the students sweep out an arc of 2π
24 radians from a
circle of radius 4184 km. Their speed is 2π
24 × 4184 = 1095km/hr.
∂G3 ∂G2
4.1.6.13. Solution. We shall show that ∂y − ∂z = F1 . The other
components are similar. First we have
dr
t F r(t) × (t) = t F tx, ty, tz × xı̂ı + y ̂ + z k̂
dt
 
ı̂ı ̂ k̂
= t det F1 F2 F3 
x y z
Reading off the k̂ and ̂ components of the determinant gives

Z 1

G3 (x, y, z) = t F1 tx, ty, tz y − F2 tx, ty, tz x dt
0
Z 1
G2 (x, y, z) = t F3 tx, ty, tz x − F1 tx, ty, tz z dt
0
So
Z 1
∂G3 h ∂F1 ∂F2 i
= t F1 tx, ty, tz + tx, ty, tz ty − tx, ty, tz tx dt
∂y 0 ∂y ∂y
Z 1 h
∂G2 ∂F3 ∂F1 i
= t tx, ty, tz tx − tx, ty, tz tz − F1 tx, ty, tz dt
∂z 0 ∂z ∂z
so that
Z 1
∂G3 ∂G2 h
− = 2t F1 tx, ty, tz
∂y ∂z 0
∂F1 ∂F1
+ t2 y tx, ty, tz + t2 z

tx, ty, tz
∂y ∂z
∂F 2 ∂F 3 i
− t2 x tx, ty, tz − t2 x

tx, ty, tz dt
∂y ∂z
∂F1 ∂F2 ∂F3
Since, by hypothesis, ∇ · F = ∂x + ∂y + ∂z = 0, the last two terms
n ∂F ∂F3 o n ∂F o
2 1
−t2 x tx, ty, tz + tx, ty, tz = −t2 x − tx, ty, tz
∂y ∂z ∂x
so that
∂G3 ∂G2
−
∂y ∂z
Z 1h
∂F1 ∂F1
2tF1 tx, ty, tz + t2 x tx, ty, tz + t2 y

= tx, ty, tz
0 ∂x ∂y
∂F1 i
+ t2 z tx, ty, tz dt
∂z
Z 1 h it=1
d 2 i h
= t F1 (tx, ty, tz) dt = t2 F1 (tx, ty, tz)
0 dt t=0
= F1 (x, y, z)

4.2.6 · Exercises
4.2.6.1. Solution. (a) Expressing the left hand side as an iterated inte-
gral, with z as the innermost integration variable, we have
ZZZ Z 1 Z 1 Z 1
∂f ∂f
(x, y, z) dx dy dz = dx dy dz (x, y, z)
V ∂z 0 0 0 ∂z
Z 1 Z 1

= dx dy f (x, y, 1) − f (x, y, 0)
0 0
by the fundamental theorem of calculus

ZZ

= f (x, y, 1) − f (x, y, 0) dx dy
Z ZR ZZ
= f (x, y, 1) dx dy − f (x, y, 0) dx dy
R R
(b) Define the vector field F(x, y, z) = f (x, y, z) k̂. Then the divergence
of F is ∇ ·F(x, y, z) = ∂f
∂z (x, y, z). The boundary of the cube V is the union
of six faces

S1 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, z = 1 with outward normal n̂ = k̂

S2 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ y ≤ 1, z = 0 with outward normal n̂ = −k̂

S3 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ z ≤ 1, y = 1 with outward normal n̂ = ̂

S4 = (x, y, z) 0 ≤ x ≤ 1, 0 ≤ z ≤ 1, y = 0 with outward normal n̂ = −̂

S5 = (x, y, z) 0 ≤ y ≤ 1, 0 ≤ z ≤ 1, x = 1 with outward normal n̂ = ı̂ı

S6 = (x, y, z) 0 ≤ y ≤ 1, 0 ≤ z ≤ 1, x = 0 with outward normal n̂ = −ı̂ı
z
n̂ = k̂
(0, 1, 1)
S1
n̂ = −̂ n̂ = ̂
y
S2
(1, 1, 0)
x
n̂ = −k̂
Observe that

+f
 on S1
F · n̂ = f k̂ · n̂ = −f on S2


0 on S3 , S4 , S5 , S6

ZZZ ZZZ
∂f
(x, y, z) dx dy dz = ∇ · F(x, y, z) dx dy dz
V ∂z V
ZZ X6 ZZ
= F · n̂ dS = F · n̂ dS
∂V j=1 Sj
ZZ ZZ
= f dS − f dS
S1 S2
ZZ ZZ
= f (x, y, 1) dx dy − f (x, y, 0) dx dy
R R
4.2.6.2. Solution. (a) The divergence of φ a is ∇ ·(φ a) = ∇ φ·a+φ∇ ∇ ·a =

∇ φ · a, since a is constant. So, by the divergence theorem,
ZZ ZZZ ZZZ
φ a · n̂ dS = ∇ · (φ a) dV = ∇ φ · a dV
∂V V
Z Z ZZZ V
=⇒ φn̂ dS − ∇ φ dV · a = 0
∂V V
This is true for all vectors a. In particular,

RR applying this with a = ı̂ı, ̂, k̂,
RRR
we have that all three components of ∂V
φn̂ dS − V
∇ φ dV are zero.
So ZZ ZZZ
φ n̂ dS − ∇ φ dV = 0
∂V V
(b) By part (a), with φ = x + y + z 2 and ∇ φ = 2xı̂ı + 2y ̂ + 2z k̂,

2 2
ZZ ZZZ
1 1
(x2 + y 2 + z 2 ) n̂ dS = (2xı̂ı + 2y ̂ + 2z k̂) dV
2|V | ∂V 2|V | V
= (x̄, ȳ, z̄)
4.2.6.3. Solution. (a) We’ll parametrize the sphere using the spherical
coordinates θ and ϕ.
x = sin ϕ cos θ
y = sin ϕ sin θ
z = cos ϕ
with 0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤ π. Since

∂x ∂y ∂z
, , = − sin ϕ sin θ , sin ϕ cos θ , 0
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (cos ϕ cos θ , cos ϕ sin θ , − sin ϕ)
∂ϕ ∂ϕ ∂ϕ
(3.3.1) yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ

= ± − sin ϕ sin θ , sin ϕ cos θ , 0 ×(cos ϕ cos θ , cos ϕ sin θ , − sin ϕ) dθdϕ
= ± − sin2 ϕ cos θ , − sin2 ϕ sin θ , − sin ϕ cos ϕ dθdϕ

= ∓ sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ

= ∓ sin ϕ x(θ, ϕ) , y(θ, ϕ) , z(θ, ϕ) dθdϕ
To get an outward pointing normal we need the + sign, since then n̂(θ, ϕ)
is a positive multiple, namely sin ϕ, times r(θ, ϕ). So, on S,
F
z }| {
F · n̂ dS = sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos2 ϕ

· sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
= sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos3 ϕ

and
ZZ Z π Z 2π
dθ sin ϕ sin2 ϕ cos2 θ + sin2 ϕ sin2 θ + cos3 ϕ

F · n̂ dS = dϕ
S 0 0
Z π 2π Z
dθ sin3 ϕ + sin ϕ cos3 ϕ

= dϕ
0
Z π 0 π
1
= 2π dϕ sin3 ϕ + − cos4 ϕ
0 4
Z π 0 π
1
= 2π dϕ sin ϕ(1 − cos ϕ) = 2π − cos ϕ + cos3 ϕ
2
0 3 0

4
= 2π
3
8π
=
3
(b) Let V be the interior of S. Then, by the divergence theorem,
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (1 + 1 + 2z) dV
S V V
By oddness under z → −z, the z integral vanishes, so that

ZZ ZZZ
4π 8π
F · n̂ dS = 2 dV = 2 Volume(V ) = 2 =
S V 3 3
4.2.6.4. Solution. (a) Let’s use spherical coordinates. As S is the sphere
of radius a centred on the origin, we can parametrize it by
r(θ, ϕ) = a sin ϕ cos θ ı̂ı + a sin ϕ sin θ ̂ + a cos ϕ k̂

∂r
= −a sin ϕ sin θ ı̂ı + a sin ϕ cos θ ̂
∂θ
∂r
= a cos ϕ cos θ ı̂ı + a cos ϕ sin θ ̂ − a sin ϕ k̂
∂ϕ
∂r ∂r
n̂ dS = ± × dθ dϕ
∂θ ∂ϕ
 
ı̂ı ̂ k̂
= det −a sin ϕ sin θ a sin ϕ cos θ 0  dθ dϕ
a cos ϕ cos θ a cos ϕ sin θ −a sin ϕ

= ± − a2 sin2 ϕ cos θ ı̂ı − a2 sin2 ϕ sin θ ̂ − a2 sin ϕ cos ϕ k̂ dθ dϕ

= ∓a2 sin ϕ sin ϕ cos θ ı̂ı + sin ϕ sin θ ̂ + cos ϕ k̂ dθ dϕ
For the outward normal, we want the + sign, so

n̂ dS = a2 sin ϕ sin ϕ cos θ ı̂ı + sin ϕ sin θ ̂ + cos ϕ k̂ dθ dϕ
F · n̂ dS = z(θ, ϕ) k̂ · n̂ dS = a3 sin ϕ cos2 ϕ dθ dϕ
and
ZZ Z π Z 2π
F · n̂ dS = a3 dϕ dθ sin ϕ cos2 ϕ
S 0 0
Z π π
3 2 1 3 3
= 2πa dϕ sin ϕ cos ϕ = 2πa − cos ϕ
0 3 0
4
= πa3
3
(b) Call the solid x2 + y 2 + z 2 ≤ a2 , V . As
∂ ∂ ∂
∇·F= (0) + (0) + (z) = 1
∂x ∂y ∂z
the divergence theorem gives
ZZ ZZZ ZZZ
4 3
F · n̂ dS = ∇ · F dV = dV = Volume(V ) = πa
S V V 3
4.2.6.5. Solution. (a) On D, z = 0 and
n̂ = −k̂ dS = dx dy F · n̂ = −y 2
so that ZZ ZZ
F · n̂ dS = − y 2 dx dy
D D
ZZ Z 3 Z 2π
2
F · n̂ dS = − dr r dθ r sin θ
D 0 0
Z 3 Z 2π
=− dr r3 dθ sin2 θ
0 0
4
Z 2π
1 − cos(2θ)
3
=− dθ
0 4 2
2π
81 θ sin(2θ)
=− −
4 2 4 0
81
=− π
4
R 2π
For an efficient, sneaky, way to evaluate 0 sin2 θ dθ, see Example 2.4.4.
(b) Observe that RRR∇ · F = x + 2. Since x is odd and V is invariant under
x → −x, we have V
x dV = 0 (more details below) so that
ZZZ ZZZ ZZZ
∇ · F dV = (x + 2) dV = 2 dV = 2|V |
V V V
RRR
Here are two more detailed arguments showing that V
x dV = 0.
9−x2 −y 2
Argument 1: We may rewrite the equation z = 9+x2 +y 2 of the curved
boundary of V as
9(1 − z)
z(9 + x2 + y 2 ) = 9 − x2 − y 2 ⇐⇒ x2 + y 2 =
1+z
q
9(1−z)
This is the equation of the circle of radius r(z) = 1+z centred on
x = y = 0. So z runs from 0 to 1, and for each fixed 0 ≤ z p ≤ 1, y runs
2 2
p −r(z) to r(z) and, for each fixed y and z, x runs from − r(z) − y
from
2 2
to r(z) − y . So
ZZZ Z Z 1 Z √ 2 2r(z) Z r(z) −y
Z 1 r(z)
x dV = dz dy √ dx x = dz dy 0 = 0
V 0 −r(z) − r(z)2 −y 2 0 −r(z)
Ra
since −a x dx = 0 for any a > 0.
Argument 2: As we have observed above, the curved boundary of V is
x2 + y 2 = 9(1−z)
1+z which is invariant under rotations about the z--axis. By
that symmetry, the centroid of V lies on the z-axis. Recall that, for any
solid V , the centroid of V is (x̄, ȳ, z̄) with
RRR RRR RRR
V
xdV V
ydV zdV
x̄ = RRR ȳ = RRR z̄ = RRRV
V
dV V
dV V
dV
So
ZZZ ZZZ
x dV = x̄ Volume(V ) = 0 and y dV = ȳ Volume(V ) = 0
V V
(c) By the divergence theorem,

ZZZ ZZ ZZ ZZ
∇ · F dV = F · n̂ dS = F · n̂ dS + F · n̂ dS
V ∂V S D
so that
ZZ ZZZ ZZ
81
F · n̂ dS = ∇ · F dV − F · n̂ dS = 2|V | + π
S V D 4
4.2.6.6. Solution. (a) Let G(x, y, z) = x2 + y 2 + z. Then the surface is
G(x, y, z) = 1 and ∇ G(x, y, z) = 2xı̂ı + 2y ̂ + k̂ is upward pointing (i.e. the
coefficient of k̂ is positive) so that, by (3.3.3),
∇G 2xı̂ı + 2y ̂ + k̂
n̂ dS = dx dy = dx dy
∇ G · k̂ 1
= (2xı̂ı + 2y ̂ + k̂) dx dy

F · n̂ dS = xı̂ı + y ̂ + k̂ · 2xı̂ı + 2y ̂ + k̂ dx dy
= 2x2 + 2y 2 + 1 dx dy


ZZ Z 1 Z 2π 1
2 1
F · n̂ dS = dr r dθ (2r + 1) = 2π r4 + r2
2
= 2π
S 0 0 4 2 0
(b) Call the solid 0 ≤ z ≤ 1 − x2 − y 2 , V .

z
z = 1 − x2 − y 2
y
D
x x2 + y 2 = 1
Let D denote the bottom surface of V . The disk D has radius 1, area
π, z = 0 and outward normal −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dx dy = − dx dy = −π
D D D
As
∂ ∂ ∂
∇·F= (x) + (y) + (1) = 2
∂x ∂y ∂z
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS
S V D
ZZZ
= 2 dV − (−π)
V
ZZZ
=π+2 dV
V
RRR
To evaluate the volume V
dV , we slice the V into thin horizontal pan-
cakes. Here is a sketch of the pancake at height z.
z
x2 + y 2 = 1 − z
y
D
x
√
Its cross-section is a circular disk of radius 1 − z, and hence of area
π(1 − z). As the pancake has thickness dz, it has volume π(1 − z) dz. So
ZZ Z 1 ZZ Z 1
F · n̂ dS = π + 2 dz dx dy = π + 2 dz π(1 − z)
S 0 x2 +y 2 ≤1−z 0
1
1 2
= π + 2π z − z = 2π
2 0
4.2.6.7. ∗. Solution. (a) The divergence is
∂ ∂ ∂
∇·F= z + sin y + (zy) + sin x cos y
∂x ∂y ∂z
=z
(b) Let
(x, y, z) x2 + y 2 + z 2 ≤ 9

V =
By the divergence theorem (note that we are to find the outward flux),
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = z dV = 0
∂V V V
since z is odd.
4.2.6.8. Solution. Call the silo V . Call the sides and top of the silo S.
Call the base of the silo (namely, x2 + y 2 ≤ 1, z = 0) B. By the divergence
theorem,
ZZ ZZ ZZZ
V · n̂ dS + V · (−k̂) dS = ∇ · V dV
S B
ZZ ZZ Z Z ZV
V · n̂ dS − (x2 + y) dx dy = (2xyz + z) dV
S x2 +y 2 ≤1 V
RR RRR
By oddness under Y → −y, x2 +y 2 ≤1
y dx dy = V
xyz dV = 0, so
ZZ ZZ ZZZ
V · n̂ dS = x2 dx dy + z dV
S x2 +y 2 ≤1 V
x
Z 1 Z 2π z }| { ZZZ
2
= dr dθ r (r cos θ) + z dV
0 0 V
We can evaluate the volume integral by decomposing V into thin horizontal
pancakes. See Section 1.6 in the CLP-2 text. For 0 ≤ z ≤ 1, the horizontal
cross-section of the silo at height z is a circle of radius 1 and hence of area

π. For z ≥ 1, the horizontal cross-section of the silo at height z is again a
circle. Its radius is determined
√ by the equation x2 + y 2 + z 2 = 2 of the top
2 2
of the silo. The radius is 2 − √z , so the cross-section has area π(2 − z ).
The biggest that z can get is 2. Thus
√
ZZ Z 1 Z 2π Z 1 Z 2
2
V · n̂ dS = dθ r (r cos θ) +
dr dz πz + dz π(2 − z 2 )z
S 0 0 0 1
Z 1 Z 2π Z 1
cos(2θ) + 1
= dr r3 dθ + dz πz
0 0 2 0
Z √2
+ dz π(2 − z 2 )z
1
4 1
2 1
4
√2
r z z
= π+ π + π z2 −
4 0 2 0 4 1

π π 3
= + +π 1−
4 2 4
=π
R 2π
4.2.6.9. Solution. Apply the divergence theorem. The divergence of F
is
∂ 2 ∂ ∂
∇·F= (x ) + (xy) + (3z − yz) = 3 + 3x − y
∂x ∂y ∂z
So
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (3 + 3x − y) dV
S B B
To evaluate the integrals of x and y we use that, for any solid V in R3 ,

ZZZ
dV = Volume(V)
V
and
RRR RRR RRR
x dVV
y dV
V
z dV
V
x̄ = ȳ = z̄ =
Volume(V) Volume(V) Volume(V)
where (x̄, ȳ, z̄) is the centroid of V. Our ball has volume V and centroid
(x̄, ȳ, z̄) = (x0 , y0 , z0 ). So
ZZ
F · n̂ dS = V [3 + 3x̄ − ȳ] = [3 + 3x0 − y0 ] V
S
4.2.6.10. ∗. Solution. Let
V = (x, y, z) x2 + y 2 ≤ 1 − z 4 , 0 ≤ z ≤ 1

Then the boundary, ∂V , of V , with the orientation that is used in the

divergence theorem, consists of two parts
• the surface S, but with the upward pointing normal, and

• the disk D = (x, y, z) x2 + y 2 ≤ 1, z = 0 , with normal −k̂.

ZZZ ZZ ZZ ZZ
∇ · F dV = F · n̂ dS = − F · n̂ dS + F · (−k̂) dS
V ∂V S D

As ∇ · F = 0 and F(x, y, 0) = 1 , 1 , 1
ZZ ZZ ZZ
F · n̂ dS = F · (−k̂) dS = − dS = −π
S D D
4.2.6.11. ∗. Solution. Let V be the solid x +y 2 + 2z 2 ≤ 2

2, z ≥ 0.
The surface of V consists of the half-ellipsoid S = (x, y, z) x2 + y 2 +
2z 2 = 2, z ≥ 0 , on top with upward pointing normal, and the disk
D = {(x, y, z)} z = 0, x2 + y 2 ≤ 2, on the bottom with normal −k̂. Call
the vector field F. By the divergence theorem
ZZ ZZ ZZZ ZZZ
F · n̂ dS + F · (−k̂) dS = ∇ · F dV = 4 dV
S D V V
√ √
The ellipsoid has a = 2, b = 2, c = 1 and volume 43 πabc = 83 π. So
ZZZ
16π
4 dV = 4 × 12 (Volume of the ellipsoid) =
V 3
RR RR
On D, z = 0 and D x dS = D y dS = 0 because x and y are odd. So
ZZ ZZ
F · (−k̂) dS = (xı̂ı + y̂ + 0k̂) · (−k̂) dS = 0
D D
and the desired flux is

ZZ ZZZ
16
F · n̂ dS = 4 dV = π
S V 3
4.2.6.12. ∗. Solution. (a) If (x, y, z) 6= 0,
∂ x ∂ y
∇ · F(x, y, z) = +
∂x x2 + y 2 + z 2 3/2 ∂y x2 + y 2 + z 2 3/2
∂ z
+
∂z x2 + y 2 + z 2 3/2

x + y 2 + z 2 − x 23 (2x) x + y 2 + z 2 − y 23 (2y)
2 2
= 5/2 + 5/2
x2 + y 2 + z 2 x2 + y 2 + z 2
x + y 2 + z 2 − z 23 (2z)
2
+ 5/2
x2 + y 2 + z 2

3 x2 + y 2 + z 2 − 3x2 − 3y 2 − 3z 2
= 5/2
x2 + y 2 + z 2
=0
If (x, y, z) = 0, F(x, y, z) is not defined and hence ∇ · F(x, y, z) is also not

defined.
(b) Let a > 0. Write σa = (x, y, z) x2 +y 2 +z 2 = a2 . The outward
r
unit normal to σa is n̂ = |r| so that
Z Z Z Z
r r 1 1
F · n̂ dS = 3
· dS = 2
dS = 2 dS
σa |r|=a |r| |r| |r|=a |r| a |r|=a
1
= 2 4πa2 = 4π 6= 0

a
(c) No, the results of (a) and (b) do not contradict the divergence
theorem. One hypothesis of the divergence theorem is that ∇ · F (in fact
all first order derivatives of F) be defined and continuous throughout the
solid that ∇ · F is to be integrated over. That hypothesis is violated in this
case.
(d) Let’s first figure out what the surface z 2 − x2 − y 2 + 1 = 0, i.e. the
surface x2 + y 2 = 1 + z 2 , looks like. For each z0 , the z = z0 cross-section
of this surface is the circle x2 + y 2 = 1 + z02 . The radius of this circle is 1
when z0 = 0 and grows as |z0 | increases. So the solid region E looks like
an hourglass drum, as sketched in the figure on the left below.
σa
σ
Ea
σ
We are going to use the divergence theorem to compute the flux of F
out through the surface σ of E. However we cannot apply the divergence
theorem using E as the solid, because F is not defined at the origin, (0, 0, 0),
which is a point in E. So we pick any 0 < a < 1, and define the auxiliary
solid
Ea = (x, y, z) x2 + y 2 + z 2 ≥ a2 , x2 + y 2 ≤ 1 + z 2 , −1 ≤ z ≤ 1

The solid Ea is constructed from the solid E by removing the ball x2 +

y 2 + z 2 ≤ a2 from it. A side view of Ea is sketched in the figure on the
right above. As in part (b), denote by σa the surface x2 + y 2 + z 2 = a2
with outward pointing normal. Then the boundary of Ea is ∂Ea = σ − σa ,
meaning that it consists of two parts. One part is the boundary, σ, of E,
with outward pointing normal. The other part is the surface x2 + y 2 + z 2 =
a2 , but with normal pointing into the sphere, opposite to the normals for
σa . Consequently the divergence theorem gives
ZZZ ZZ ZZ ZZ
0= ∇ · F dV = F · n̂ dS = F · n̂ dS − F · n̂ dS
Ea ∂Ea σ σa
so that, by part (b)

ZZ ZZ
F · n̂ dS = F · n̂ dS = 4π
σ σa
(e) The equation z 2 − x2 − y 2 + 4y − 3 = 0 can be rewritten as
x2 + (y − 2)2 = 1 + z 2
As is part (d), for each z0 , the z = zp 0 cross-section of this surface is a

circle x2 + (y − 2)2 = 1 + z02 of radius 1 + z02 . But this circle is centred
at (0, 2, z0 ), whereas the corresponding circle in part (d) was centred at
(0, 0, z0 ). The solid R again has the shape of an hourglass drum. But
while the origin (0, 0, 0) was in E, it is not in
R = (x, y, z) x2 + (y − 2)2 ≤ 1 + z 2 , −1 ≤ z ≤ 1

So ∇ · F = 0 throughout all of R and the divergence theorem gives

ZZ ZZ ZZZ
F · n̂ dS = F · n̂ dS = ∇ · F dV = 0
Σ ∂R R
4.2.6.13. ∗. Solution. (a) If the surface were the sphere x2 +y 2 +z 2 = 1,

we could parametrize it using the spherical coordinates θ and ϕ (with the
radial spherical coordinate ρ = 1).
x = sin ϕ cos θ
y = sin ϕ sin θ
z = cos ϕ
with 0 ≤ θ < 2π, 0 ≤ ϕ ≤ π. Our surface is not a sphere, but the

equation looks like the equation of the sphere with the units of the y- and
z-coordinates changed. In particular, if we define ỹ = y/2 and z̃ = z/2, so
that y = 2ỹ and z = 2z̃, then on our surface
y2 z2 (2ỹ)2 (2z̃)2
1 = x2 + + = x2 + + = x2 + ỹ 2 + z̃ 2
4 4 4 4
and we can parametrize
x = sin ϕ cos θ
ỹ = sin ϕ sin θ
z̃ = cos ϕ
and then
x = sin ϕ cos θ
y = 2ỹ = 2 sin ϕ sin θ
z = 2z̃ = 2 cos ϕ
or
r(θ, ϕ) = sin ϕ cos θ ı̂ı + 2 sin ϕ sin θ ̂ + 2 cos ϕ k̂

0 ≤ θ < 2π, 0 ≤ ϕ ≤ π
(b) Considering part (c) in this question, we are presumably to evaluate

the flux integral directly. Since
∂x ∂y ∂z
, , = − sin ϕ sin θ , 2 sin ϕ cos θ , 0
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (cos ϕ cos θ , 2 cos ϕ sin θ , −2 sin ϕ)
∂ϕ ∂ϕ ∂ϕ
(3.3.1) yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ

= ± − sin ϕ sin θ , 2 sin ϕ cos θ , 0
× (cos ϕ cos θ , 2 cos ϕ sin θ , −2 sin ϕ) dθdϕ

= ± − 4 sin2 ϕ cos θ , −2 sin2 ϕ sin θ , −2 sin ϕ cos ϕ dθdϕ

= ∓2 sin ϕ 2 sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
To get an outward pointing normal we need the + sign. For example, with
the + sign, the z-component is 2 sin ϕ cos ϕ = sin(2ϕ) so that the normal
is pointing upward when 0 < ϕ < π2 , i.e. in the northern hemisphere, and
is pointing downward when π2 < ϕ < π, i.e. in the southern hemisphere.
So
F · n̂dS = (sin ϕ cos θ)(4 sin2 ϕ cos θ) + (2 sin ϕ sin θ)(2 sin2 ϕ sin θ)

+ (2 cos ϕ)(2 sin ϕ cos ϕ) dθdϕ
= 4 sin3 ϕ cos2 θ + 4 sin3 ϕ sin2 θ + 4 sin ϕ cos2 ϕ dθdϕ

= 4 sin ϕ sin2 ϕ + cos2 ϕ dθdϕ

= 4 sin ϕ dθdϕ
and the flux is

ZZ Z π Z 2π Z π
F · n̂ dS = dϕ dθ 4 sin ϕ = 8π dϕ sin ϕ = 16π
S 0 0 0
(c) Set
y2 z2
(x, y, z) x2 +

V = 4 + 4 ≤1
Since ∇ · F = 3, the divergence theorem gives
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 3Volume(V )
S V
2 2 2
The volume contained in the ellipsoid, xa2 + yb2 + zc2 = 1, of semiaxes a, b
and c is 34 πabc. In our case a = 1, b = c = 2, so
ZZ
4
F · n̂ dS = 3Volume(V ) = 3 × π(1)(2)(2) = 16π
S 3
which is exactly what we found in part (b).
The volume of the ellipsoid V can also be found by observing that, in
V,
• x runs from −1 to 1 and
• for each fixed −1 ≤ x ≤ 1, (y, z) runs over the disk y 2 +z 2 ≤ 4(1−x2 ),
which has area 4π(1 − x2 ).
That is
(x, y, z) − 1 ≤ x ≤ 1, y 2 + z 2 ≤ 4(1 − x2 )

V =
so that
Z 1 ZZ
Volume(V ) = dx dy dz
−1 y 2 +z 2 ≤4(1−x2 )
Z 1 Z 1
2 2 1
= dx 4π(1 − x ) = 2 × 4π dx (1 − x ) = 8π 1 −
−1 0 3
16π
=
3
4.2.6.14. ∗. Solution. Set
V = (x, y, z) x2 + y 2 ≤ 2, 0 ≤ z ≤ 2x + 3

Let’s try the divergence theorem. Since

∂ 3 ∂ 3 ∂ 2
x + cos(y 2 ) + y + zex +

∇·F= z + arctan(xy)
∂x ∂y ∂z
= 3x2 + 3y 2 + 2z
the divergence theorem (Theorem 4.2.2) gives

ZZ ZZZ
F · n̂ dS = ∇ · F dV
S V
Z Z 2x+3
dz 3x2 + 3y 2 + 2z

= dxdy
x2 +y 2 ≤2 0
Z
dxdy 3(x2 + y 2 )(2x + 3) + (2x + 3)2

=
x2 +y 2 ≤2
Z
dxdy 9 + 12x + 13x2 + 9y 2 + 6x3 + 6xy 2

=
x2 +y 2 ≤2
Z
dxdy 13x2 + 9y 2

= 9(2π) +
x2 +y 2 ≤2
because 12x, 6x3 and 6xy 2 are all odd under x → −x. To evaluate the
final remaining integral, let’s switch to polar coordinates.
√
ZZ Z 2 Z 2π
2 2
dθ 13 r cos θ)2 + 9 r sin θ)2

13x + 9y dxdy = dr r
x2 +y 2 ≤2 0 0
√
Z 2 Z 2π
dr r3 dθ 13 cos2 θ + 9 sin2 θ

=
0 0
Since
Z 2π Z 2π 2π
2 cos(2θ) + 1 sin(2θ) θ
cos θ dθ = dθ = + =π
0 0 2 4 2 0
2π 2π 2π
1 − cos(2θ)
Z Z
θ sin(2θ)
sin2 θ dθ = dθ = − =π
0 0 2 2 4 0
we finally have
√
( 2)4
ZZ

F · n̂ dS = 18π + π 13 + 9 = (18 + 22)π = 40π
S 4
R 2π R 2π
cos2 θ dθ and 0
sin2 θ dθ, see
Example 2.4.4.
4.2.6.15. ∗. Solution 1 (By divergence theorem). Set F = (x + y, x +
z, y + z). Then ∇ · F = 2. That’s really simple. So let’s try using the
divergence theorem.
2 2

RR S = (x, y, z) x + z = 4, 0 ≤ y ≤ 3 . We are to compute
• Set
S
F · n̂ dS, with n̂ denoting the outward normal
RR to S. S is not the
boundary of a solid, so we cannot compute S F · n̂ dS by applying
the divergence theorem directly. The figure on the left below shows
the part of S that is in the first octant.
z y=3 z y=3
S S
−̂ Dl ̂
2
x +z =42 2 2 V
y x +z =4 y
Dr
x x
• On the other hand S, is “almost” the boundary of
V = (x, y, z) x2 + z 2 ≤ 4, 0 ≤ y ≤ 3

The boundary, ∂V of V consists of three pieces — S and the two

disks
Dl = (x, y, z) x2 + z 2 ≤ 4, y = 0

Dr = (x, y, z) x2 + z 2 ≤ 4, y = 3

The figure on the right above shows the parts of S, V , Dl and Dr

that are in the first octant.
The outward normal to Dr is ̂ and the outward normal to Dl is −̂, to the
divergence theorem gives
ZZZ ZZ
∇ · F dV = F · n̂ dS
V
Z Z∂V ZZ ZZ
= F · n̂ dS + F · ̂ dS + F · (−̂) dS
S Dr Dl
Since ∇ · F = 2 and F · ̂ = x + z,

ZZ ZZZ ZZ ZZ
F · n̂ dS = 2 dV − (x + z) dxdz − (−x − z) dxdz
S V x2 +z 2 ≤4 x2 +z 2 ≤4
ZZZ
= 2 dV
V
= 2 volume(V ) = 2(π22 )3 = 24π
Solution 2 (By direct evaluation). Let’s parametrize the surface by
r(θ, y) = 2 cos θ ı̂ı + y ̂ + 2 sin θ k̂ 0 ≤ θ < 2π, 0 ≤ y ≤ 3
Then
∂r

∂θ = − 2 sin θ , 0 , 2 cos θ
∂r

∂y = 0, 1, 0
∂r ∂r

n̂dS = ± ∂θ × ∂y dθdy = ± − 2 cos θ , 0 , −2 sin θ dθdy
To get the outward normal, we want the minus sign. So

n̂dS = 2 cos θ , 0 , 2 sin θ dθdy
and, since

F r(θ, y) = 2 cos θ + y , 2 cos θ + 2 sin θ , y + 2 sin θ
the specified flux is

ZZ Z 2π Z 3
F · n̂ dS = dθ dy 2 cos θ + y , 2 cos θ + 2 sin θ , y + 2 sin θ
S 0 0

· 2 cos θ , 0 , 2 sin θ
Z 2π Z 3
dy 4 cos2 θ + 2y cos θ + 2y sin θ + 4 sin2 θ

= dθ
0 0
Z 2π Z 3
= dθ dy 4 + 2y cos θ + 2y sin θ
0 0
R 2π R 2π
Since 0
dθ cos θ = 0
dθ sin θ = 0,
ZZ Z 2π Z 3
F · n̂ dS = 4 dθ dy = 4(2π)3 = 24π
S 0 0
4.2.6.16. ∗. Solution. The question highlights that the vector field has
divergence 0. That strongly suggests that we use the divergence theorem.
Set
n̂
V S

V = (x, y, z) 0 ≤ z ≤ 1 − x2 +
2
y2
n̂ D
Then the boundary, ∂V , of V consists of two parts, namely S (with
normal pointing upwards) and the disk
D = (x, y, 0) x2 + y 2 ≤ 1

(with normal pointing downwards). The divergence theorem (Theorem

4.2.2) gives
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V D
ZZ
= (x2 + y 2 ) dxdy
D
Switching to polar coordinates, the flux is

ZZ Z 1 Z 2π Z 1
F · n̂ dS = dr r dθ r2 = 2π dr r3 = 2π 41 = π
2
S 0 0 0
4.2.6.17. ∗. Solution. As F looks complicated, we will probably want

to avoid evaluating the flux integral directly. Let’s first compute the diver-
gence of F, to see if it looks wise to use the divergence theorem instead.
∂ √ ∂ −x2 ∂
tan z + sin(y 3 ) +

∇·F= e + z =1
∂x ∂y ∂z
Looks good! We cannot yet apply the divergence theorem, since S is not
the boundary of a solid region V . To help us choose a solid V whose
boundary at least includes S, here is a sketch. S is the top of the “ice

cream cone”
z = 2 − x2 − y 2 S
p
z= x2 + y 2
p
Note that the the paraboloid z = 2−x2 −y 2 and the cone z = x2 + y 2
intersect along the circle x2 + y 2 = 1, z = 1. Probably the simplest solid
whose boundary includes S is
V = (x, y, z) 1 ≤ z ≤ 2 − x2 − y 2 , x2 + y 2 ≤ 1

The boundary ∂V of V consists of S (with upward pointing normal) and

the disk
D = (x, y, z) x2 + y 2 ≤ 1, z = 1

with normal −k̂. So the divergence theorem gives

ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V D
Z Z Z z}|{ ∇ ·F
ZZ F·k̂=z
z}|{
= 1 dV + 1 dS
V D
RR
As D is a disk of radius 1, D dS = π. To compute the volume of V ,
we’ll slice it intopa stack of horizontal
√ “pancakes”. Since z = 2 − x2 − y 2
2 2
is equivalent to√ x + y = 2 − z, the pancake at height z is a circular
disk of radius 2 − z and hence of cross-sectional area π(2 − z). So the
volume of V is
ZZZ Z 2 2
π π
dV = π(2 − z) dz = − (2 − z)2 =

V 1 2 1 2
and the flux

ZZ
π 3
F · n̂ dS = +π = π
S 2 2
4.2.6.18. ∗. Solution. As F looks complicated, we will probably want
to avoid evaluating the flux integral directly. Let’s first compute the diver-
gence of F, to see if it looks wise to use the divergence theorem instead.
∂ ∂ ∂
cos z + xy 2 + xe−z + sin y + x2 z = y 2 + x2

∇·F=
∂x ∂y ∂z
Looks promising. Furthermore S is the boundary of the solid region
z
z=4
V =

(x, y, z) x2 + y 2 ≤ z ≤ 4
z = x2 + y 2
x
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (x2 + y 2 ) dV
S V V
To compute the triple integral, we’ll use the cylindrical coordinates (r, θ, z).
The z-coordinate runs from 0 to 4. For each fixed 0 ≤ z ≤ 4 (see the blue
disk in the figure below — which shows the part of V in the first octant),
z
z=4
z = x2 + y 2 = r 2
x
2 2
(x, y) runs over 0 ≤
√ x + y ≤ z, which in cylindrical coordinates is
2
0 ≤ r ≤ z or 0 ≤ r ≤ z. So the flux and the triple integral are
ZZ ZZZ
F · n̂ dS = (x2 + y 2 ) dV
S V
√
Z 4 Z z Z 2π
= dz dr r dθ r2
0 0 0
√
Z 4 Z z
= 2π dz dr r3
0 0
4 2
43
Z
z
= 2π dz = 2π
0 4 3×4
32
= π
3
4.2.6.19. ∗. Solution. If we were to evaluate this integral directly using,
for example, spherical coordinates, our integrand would contain

sin(x) = sin 2 sin ϕ cos θ
That’s not very friendly looking. So let’s consider using the divergence
theorem instead. To start,
∂ y ∂ ∂ 2
∇·F= e + xz + zy + sin(x) + z − 1 = 4z
∂x ∂y ∂z
That’s nice and simple. So let’s move on to consideration of S. The part

of S in the first octant is outlined in red in the figure on the left below.
z z
z=1
Dt
V
S z=0 y S y
Db
x x
The surface S is not closed, and so is not the boundary of a solid, so
we cannot apply the divergence theorem directly. But we can easily come
up with a solid whose boundary contains S. Let
V = (x, y, z) x2 + y 2 + z 2 ≤ 4, 0 ≤ z ≤ 1

The boundary ∂V of V consists of three parts — S, the bottom disk
Db = (x, y, z) x2 + y 2 ≤ 4, z = 0

and the top disk
(x, y, z) x2 + y 2 ≤ 3, z = 1

Dt =
The outward normal to Dt is k̂ and the outward normal to Db is −k̂. So

ZZZ ZZ
∇ · F dV = F · n̂ dS
V
Z Z∂V ZZ ZZ
= F · n̂ dS + F · k̂ dS + F · (−k̂) dS
S Dt Db
On Db , z = 0 so that F · (−k̂) = −(02 − 1) = 1 and on Dt , z = 1 so that

F · k̂ = 12 − 1 = 0. So
ZZ ∇·F
Z Z Z z}|{ ZZ
F · n̂ dS = 4z dV − dS
S V Db
√
The constant z cross-section of V is a disk of radius 4 − z 2 and hence of
area π(4 − z 2 ) and Db is a disk of radius 2 and hence of area 4π. So
1
z 4 i1
ZZ Z h
F · n̂ dS = (4z) π(4 − z 2 ) dz − 4π = 4π 2z 2 − − 4π = 3π
S 0 4 0
4.2.6.20. ∗. Solution. The divergence of F, namely,
∂ 2 ∂ ∂
x − y2

∇·F= x z + cos πy + yz + sin πz +
∂x ∂y ∂z
= 2xz + z
is a lot simpler than F itself. So let’s use the divergence theorem (Theorem
4.2.2).
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (2xz + z) dV
S B B
invariant under x → −x while 2xz is odd under x → −x, the

As B is RRR
integral
RRR B
2xz dV is zero. To help set up the limits of integration for
B
z dV , note that, in B,
• (x, y) runs over the rectangle −1 ≤ x ≤ 1, 0 ≤ y ≤ 2 and
• for each fixed (x, y), z runs over 0 ≤ z ≤ 3 − y.
So
ZZ Z 1 Z 2 Z 3−y
F · n̂ dS = dx dy dz z
S −1 0 0
Z 1 Z 2
1
= dx dy (3 − y)2
2 −1 0
Z 1 Z 1
1
=− dx du u2 with u = 3 − y, du = −dy
2 −1 3
h 13 Z 1
1 33 i1
=−
dx −
2 −1 3 3 −1
26
=
3
4.2.6.21. ∗. Solution. The vector field F looks very complicated. That
strongly suggests that we not evaluate the integral directly. So let’s start
by computing
∂ ∂ ∂ p 2
x + cos(z 2 ) + y + ln(x2 + z 5 ) +

∇·F= x + y2
∂x ∂y ∂z
=2
That’s really simple, which suggest that we use the divergence theorem.
But the surface S is not closed, and so is not the boundary of a solid. So
we cannot apply the divergence theorem directly. But we can easily come
up with a solid whose boundary contains S. Let
p
V = (x, y, z) 0 ≤ z ≤ 1 − x2 − y 2 , x2 + y 2 ≤ 1

n̂
V S
n̂ D
Then the boundary, ∂V , of V consists of two parts, namely S (with
normal pointing upwards) and the disk
D = (x, y, 0) x2 + y 2 ≤ 1

(with normal −k̂). The divergence theorem (Theorem 4.2.2) gives

ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V D
ZZZ ZZ p
= 2 dV + x2 + y 2 dxdy
V D
ZZ p
1 4 3
=2 π1 + x2 + y 2 dxdy
2 3 D

ZZ Z 1 Z 2π Z 1
4 4
F · n̂ dS = π + dr r dθ r = π + 2π dr r2
S 3 0 0 3 0
4 1
= π + 2π = 2π
3 3
4.2.6.22. ∗. Solution. (a) By the divergence theorem (Theorem 4.2.2),
the outward flux of F through the boundary of E is
ZZ ZZZ
F · n̂ dS = ∇ · F dV
∂E
Z Z ZE
− x2 − y 2 + 4 dV

=
E
To evaluate this integral we switch to cylindrical coordinates. In cylindrical

coordinates
E = (r cos θ , r sin θ , z) 0 ≤ z ≤ 4, r2 ≤ z

So
√
ZZ Z 4 Z z Z 2π
− r2 + 4

F · n̂ dS = dz dr r dθ
∂E 0 0 0
√
Z 4 Z z
dr 4r − r3

= 2π dz
0 0
4
z2
Z
= 2π dz 2z −
0 4
3 i4
h z h 16 i 64
= 2π z 2 − = 2π 16 − = π
12 0 3 3
(b) The boundary of S consists of two parts — S, but with downward
pointing normal, on the bottom and the disk
D = (x, y, z) z = 4, x2 + y 2 ≤ 4

with normal k̂, on top.
z k̂
z=4
D
S z = x2 + y 2
n̂
y
x
So, by part (a),
ZZ ZZ ZZ
64
π= F · n̂ dS = − F · n̂ dS + F · k̂ dS
3 ∂E S D
ZZ F·k̂
Z Z z}|{
=− F · n̂ dS + 4z dS
S D
Since z = 4 on D, and D is a disk of radius 2,

ZZ ZZ
64 64 128
F · n̂ dS = − π + 16 dS = − π + 16(4π) = π
S 3 D 3 3
∂ x 1 3 x(2x)
3/2
= 3/2
−
∂x [x + y + z ]
2 2 2 2 2
[x + y + z ] 2 2 [x + y 2 + z 2 ]5/2
2
−2x2 + y 2 + z 2
= 5/2
[x2 + y 2 + z 2 ]
∂ y 1 3 y(2y)
3/2
= 3/2
−
∂y [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]5/2
x2 − 2y 2 + z 2
= 5/2
[x2 + y 2 + z 2 ]
∂ z 1 3 z(2z)
3/2
= 3/2
−
∂z [x2 + y 2 + z 2 ] [x2 + y 2 + z 2 ] 2 [x2 + y 2 + z 2 ]5/2
x2 + y 2 − 2z 2
= 5/2
[x2 + y 2 + z 2 ]
(−2x2 + y 2 + z 2 ) + (x2 − 2y 2 + z 2 ) + (x2 + y 2 − 2z 2 )
∇·F= 5/2
=0
[x2 + y 2 + z 2 ]
if (x, y, z) 6= 0 and is not defined if (x, y, z) = 0.
(b), (c) Set
V1 = (x, y, z) x2 + (y − 2)2 + z 2 ≤ 9

V2 = (x, y, z) x2 + (y − 2)2 + z 2 ≤ 1

Here are side views of both V1 and V2 .

z z
S1
S2
(0, 2, 0)
y y
V2
V1
Both V1 and V2 are spherical balls centred on (0, 2, 0). The difference
between them is that V1 has radius 3 while V2 has radius 1. In particular
(0, 0, 0) is not in V2 . So ∇ · F is well-defined and zero throughout V2 and,
by the divergence theorem (Theorem 4.2.2),
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S2 V2
On the other hand, (0, 0, 0) is in V1 . We cannot blindly apply the divergence

theorem to V1 — ∇ ·F(x, y, z) is not defined at the point (x, y, z) = (0, 0, 0)
in V1 . We can work around this obstruction by
• choosing a number ρ > 0 that is small enough that the sphere
Sρ = (x, y, z) x2 + y 2 + z 2 = ρ2

1
is completely contained inside V1 (for example, ρ = 2 is fine)
• and then removing the interior of Sρ from V1 .

This produces
V3 = (x, y) x2 + (y − 2)2 + z 2 ≤ 9, x2 + y 2 + z 2 ≥ ρ2

whose side view is sketched below.

z
S1
Sρ (0, 2, 0)
y
V3
The boundary of V3 consists of two parts

• the sphere S1 , with outward normal and
• the sphere Sρ with inward normal n̂ = − |r|
r
The divergence ∇ · F is well-defined and zero throughout V3 so that, by

the divergence theorem,
ZZZ ZZ ZZ r
0= ∇ · F dV = F · n̂ dS + F· − dS
V3 S1 Sρ |r|
So
ZZ ZZ r ZZ
r r
F · n̂ dS = F· dS = 3
· dS
S1 Sρ |r| Sρ |r| |r|
ZZ ZZ
1 1
= dS = dS
Sρ |r|2 Sρ ρ 2
1
= 2 4πρ2 = 4π
ρ
2
since Sρ is a sphere of radius
RR ρ and hence ofRRsurface area 4πρ .
(d) The flux integrals S1 F · n̂ dS and S2 F · n̂ dS are different, be-
cause the one point, (0, 0, 0), where ∇ · F fails to be well-defined and zero,
is contained inside S1 but is not contained inside S2 .
4.2.6.24. ∗. Solution. The vector field F looks pretty complicated. But
its divergence
∇·F=2+3+1=6
is very simple. So let’s use the divergence theorem (Theorem 4.2.9). It
says
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 6 dV = 6 Volume(E)
S E E
For any fixed 0 ≤ X ≤ 2, the cross-section of E with x = X has side view

z z =2+y
y=2
E
y
That cross-section has area 2 × 2+4
2 = 6. Consequently the volume of
E is 2 × 6 = 12 and
ZZ
F · n̂ dS = 6 × 12 = 72
S
4.2.6.25. ∗. Solution. (a) The divergence is

∂ ∂ 3 ∂
z arctan(y 2 ) + z ln(x2 + 1) +

∇·F= 3z = 3
∂x ∂y ∂z
(b) The complexity of F and the simplicity of RR ∇ · F strongly suggest
that we use the divergence theorem to evaluate S F · n̂ dS. However, S
is not a closed surface and is not the boundary of a solid. The figure on
the left below is a sketch of the part of S in the first octant.
z z
(0,0,2) (0,0,2)
n̂
S ∂V
n̂ V
D
∂V
n̂
y y
x x
On the other hand S is part of the surface of the solid
V = (x, y, z) x2 + y 2 + z 2 ≤ 4, z ≥ 1

which is sketched on the right above. The boundary of V consists of two

parts:
• the original surface S, but with upward, rather than downward, nor-
mal and

• the disk D = (x, y, z) x2 + y 2 ≤ 3, z = 1 with normal −k̂.
So the divergence theorem (Theorem 4.2.9) gives
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 3 dV
∂V V V
ZZ ZZ
=⇒ − F · n̂ dS + F · (−k̂) dS = 3 Volume(V )
S D
Thus
ZZ Z Z −F·k̂
z}|{
F · n̂ dS = −3 Volume(V ) + −3 dS
S D
= −3 Volume(V ) − 3 Area(D)
= −3 Volume(V ) − 9π
√
since D is a circular disk of radius 3. To compute the volume of V , we
slice V into thin horizontal pancakes each of thinkness dz. The pancake at
height z has cross-section the circular disk x2 + y 2 ≤ 4 − z 2 . As this disk
has area π(4 − z 2 ), the pancake has volume π(4 − z 2 ) dz. All together
2 2
z3
Z
2 7 5π
Volume(V ) = dz π(4 − z ) = π 4z − =π 4− =
1 3 1 3 3
and ZZ
5π
F · n̂ dS = −3 − 9π = −14π
S 3
4.2.6.26. ∗. Solution. Let’s try the divergence theorem. Set
(x, y, z) x2 + y 2 + z 2 ≤ 3

V =
Then the boundary of V is S, but with outward pointing normal. Since

∂ ∂ ∂
xy 2 + y 4 z 6 + yz 2 + x4 z + zx2 + xy 4

∇·F=
∂x ∂y ∂z
2 2 2
=y +z +x
and because S is oriented inward, the divergence theorem (Theorem 4.2.2)

gives
ZZ ZZZ ZZZ
F · n̂ dS = − ∇ · F dV = − (x2 + y 2 + z 2 ) dV
S V V
Switching to spherical coordinates (see Appendix A.6.3)

√
ZZ Z 3 Z π Z 2π
F · n̂ dS = − dρ dϕ dθ ρ4 sin ϕ
S 0 0 0
"Z √ # Z
3 π
= −2π dρ ρ4 dϕ sin ϕ
0 0
√3 h
ρ5
iπ
= −2π − cos ϕ
5 0 0
√
36 3
=− π
5
4.2.6.27. ∗. Solution. (a) The divergence of F is
∂ ∂ 2 ∂ 2
x + y2

∇·F= − 2xy + y + sin(xz) +
∂x ∂y ∂z
= −2y + 2y + 0 = 0
(b) Call the specified surface S and set
V = (x, y, z) x2 + y 2 + (z − 12)2 ≤ 132 , z ≥ 0

The boundary, ∂V , of V consists of two parts — S, with outward normal,

and the disk
D = (x, y, z) x2 + y 2 ≤ 132 − 122 = 52 , z = 0

with normal −k̂. By the divergence theorem, the desired flux is

ZZ ZZZ ZZ
F · n̂ ds = ∇ · F dV − F · (−k̂) dS
S
Z Z ZV ZZ D
= 0 dV + (x2 + y 2 )dxdy
V D
Z 5 Z 2π
=0+ dr r dθ r2
0 0
54 625
= 2π = π
4 2
4.2.6.28. ∗. Solution. The boundary of the solid V enclosed by S and
z = ±1 consists of three pieces: S, the top disk
S1 = (x, y, z) x2 + y 2 ≤ 2, z = 1

and the bottom disk

(x, y, z) x2 + y 2 ≤ 2, z = −1

S2 =
On S1 , n̂ = k̂ and

F · n̂ = F · k̂ = xy − z − z 2 = xy − 2

z=1

so that, denoting D = (x, y) x2 + y 2 ≤ 2 ,
ZZ ZZ
F · n̂ dS = (xy − 2) dxdy = −2 Area(D) = −4π
S1 D
RR
Here we have used that the integral D
xy dxdy = 0 because xy is odd
under x → −x. On S2 , n̂ = −k̂ and

F · n̂ = −F · k̂ = −(xy − z − z 2 ) = −xy

z=−1
so that ZZ ZZ
F · n̂ dS = (−xy) dxdy = 0
S2 D
By the divergence theorem (Theorem 4.2.2),
ZZ ZZ ZZ ZZ
F · n̂dS = ∇ · F dV − F · n̂ dS − F · n̂ dS
S V S1 S2
= 0 − (−4π) − 0 = 4π
since
∂ ∂ ∂
(x + eyz ) + (xy − z − z 2 )

∇·F= 2yz + sin(xz) +
∂x ∂y ∂z
= 1 + 2z − 1 − 2z
=0
4.2.6.29. ∗. Solution. Direct Solution. The surface is given by the

implicit equation f (x, y, z) = 0 with f (x, y, z) = x2 + y 2 + 2z 2 − 1. Hence,
by (3.3.3),
∇f 2xı̂ı + 2y̂ + 4z k̂
n̂ dS = dxdy = dxdy
∇f · k̂ 4z
This n̂ has positive k̂ component. Assume that it is the desired n̂, though
this was not specified in the question. Since
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
2 cos y 3 5
x −y−1 e + z 2xz + z
= −3z 2 ı̂ı − 2z ̂ + k̂
we have
ZZ
∇ × F · n̂ dS
S
ZZ
2xı̂ı + 2y ̂ + 4z(x, y) k̂
= − 3z(x, y)2 ı̂ı − 2z(x, y) ̂ + k̂ · dxdy
x2 +y 2 ≤1 4z(x, y)
ZZ 3
= − x z(x, y) − y + 1 dxdy
x2 +y 2 ≤1 2
q
Since y is an odd function of y and x z(x, y) = x 12 (1 − x2 − y 2 ) is an odd
function of x, they both integrate to zero. Hence
ZZ ZZ
∇ × F · n̂ dS = 1 dxdy = π
S x2 +y 2 ≤1
Tricky Solution. Let V be the solid x2 + y 2 + 2z 2 ≤ 1, z ≥ 0. The

surface
of V consists of S with upward
pointing normal and the disk D =
(x, y, z) z = 0, x2 + y 2 ≤ 1 with normal −k̂. By the divergence
theorem, Theorem 4.2.2,
ZZ ZZ ZZZ
∇ × F · n̂ dS + ∇ × F · (−k̂) dS = ∇ · ∇ × F dV
S D V
ZZZ
= 0 dV = 0
V
Hence
ZZ ZZ ZZ
∇ × F · n̂ dS = ∇ × F · k̂ dS = dS = π
S D D
4.2.6.30. ∗. Solution. Let S 0 be the disk x2 + y 2 ≤ 3, z = 0 (with

n̂ the downward pointing normal) and let V be the portion of the ball
x2 + y 2 + (z − 1)2 ≤ 4 with z ≥ 0. Then, by the divergence theorem,
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V S0
ZZZ ZZ
= (2x + 2y) dV + (4 + 5x) dxdy
V S0
Because x is odd under x → −x and y is odd under y → −y,

ZZZ ZZZ ZZ
x dV = y dV = x dxdy = 0
V V S0
so that
ZZ ZZ √ 2
F · n̂ dS = 4 dxdy = 4 Area(S 0 ) = 4 × π 3 = 12π
S S0
p
4.2.6.31. ∗. Solution. Call the hemisphere 0 ≤ z ≤ 4 − x2 − y 2 , H.
Call the bottom surface of the hemisphere D and the top surface S. The
disk D has radius 2, area 4π, z = 0 and the outward normal −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dxdy = − dxdy = −4π
D D D
As
∂ ∂ 2 ∂
∇·F= (xy 2 ) + (x y) + (1) = x2 + y 2
∂x ∂y ∂z
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS
S
Z Z ZH D
= (x2 + y 2 ) dV − (−4π)
R
To evaluate the remaining integral, let’s switch to the cylindrical coordi-

nates (r, θ, z). In cylindrical coordinates, the equation x2 + y 2 + z 2 = 4
becomes r2 + z 2 = 4. So
√
ZZ Z 2 Z 4−z 2 Z 2π
F · n̂ dS = 4π + dz dr r dθ r2
S 0 0 0
Z 2
1 p 4
= 4π + 2π dz 4 − z2
0 4
π 2
Z
dz (16 − 8z 2 + z 4

= 4π +
2 0
πh 8 1 i2
= 4π + 16z − z 3 + z 5
2 3 5 0
188
= π ≈ 39.37
15
4.2.6.32. ∗. Solution. Let St , Sb and Sc denote the top, bottom and
curved surfaces of D respectively. On the top surface, z = 5 and the
outward normal to D is k̂, so that
ZZ ZZ ZZ
5
F · n̂ dS = (15 − 5ye ) dxdy = 15 dxdy = 15π
St x2 +y 2 ≤1 x2 +y 2 ≤1
The integral over y was zero because y is odd under y → −y. On the
bottom surface, z = 0 and the outward normal to D is −k̂, so that
ZZ ZZ
F · n̂ dS = − (3 × 0 − 0 × ye0 ) dxdy = 0
Sb x2 +y 2 ≤1
Again, the integral over y was zero because y is odd under y → −y. As
∂ ∂ 2 z ∂
∇·F= (x + xyez ) + 12 y ze + (3z − yzez )
∂x ∂y ∂z
= (1 + yez ) + yzez + (3 − yzez − yez ) = 4

ZZ ZZZ ZZ ZZ
Sc D St Sb
ZZZ
= 4 dV − 15π − 0
D
= 4 × π12 × 5 − 15π = 5π

4.2.6.33. Solution. Let V = (x, y, z) x2 + y 2 + z 2 ≤ a2 , x ≥ 0, y ≥
0, z ≥ 0 .
z
x2 + y 2 + z 2 = a2
x
Then ∂V consists of

• the x = 0 face (x, y, z) y 2 + z 2 ≤ a2 , x = 0, y ≥ 0, z ≥ 0 with
normal n̂ = −ı̂ı,

• the y = 0 face (x, y, z) x2 + z 2 ≤ a2 , x ≥ 0, y = 0, z ≥ 0 with
normal n̂ = −̂,

• the z = 0 face (x, y, z) x2 + y 2 ≤ a2 , x ≥ 0, y ≥ 0, z = 0 with
normal n̂ = −k̂,
• and the first octant part of the sphere. Call it S.
Then
ZZZ ZZZ ZZZ
1 4 3
∇ · F dV = z + 1 + z − 2z dV = dV = πa
V V V 8 3
1
= πa3
6
ZZ ZZ Z a Z π/2
F · (−k̂) dx dy = (2x + 02 ) dx dy = 2 dr r dθ r cos θ
z=0 z=0
face face 0 0
a
2a3
Z
=2 r2 dr =
3
ZZ Z0Z
F · (−̂) dx dz = − (0 + 0z) dx dz = 0
y=0 y=0
face face
ZZ ZZ Z a Z π/2
F · (−ı̂ı) dy dz = −(y + 0z) dy dz = − dr r dθ r sin θ
x=0 x=0
face face 0 0
a
a3
Z
=− r2 dr = −
0 3
By the divergence theorem
ZZ ZZZ ZZ
F · n̂ dx dy = ∇ · F dV − F · (−ı̂ı) dy dz
x=0
S V face
ZZ ZZ
− F · (−̂) dx dz − F · (−k̂) dx dy
y=0 z=0
face face

π 1 3
= − a
6 3
4.2.6.34. Solution. (a) On the cylindrical surface √ S1 , use (surprise!)
cylindrical coordinates. Since the cylinder has radius 2, we may parametrize
it by
√ √
r(θ, z) = 2 cos θ ı̂ı + 2 sin θ ̂ + z k̂
∂r √ √
(θ, z) = − 2 sin θ ı̂ı + 2 cos θ ̂
∂θ
∂r
(θ, z) = k̂
∂z
∂r ∂r
n̂ dS = ± (θ, z) × (θ, z) dθ dz
∂θ ∂z
 
√ ı̂ı √ ̂ k̂
= ± det − 2 sin θ ı̂ı 2 cos θ 0  dθ dz
0 0 1
√ √
= ± 2 cos θ ı̂ı + 2 sin θ ̂ dθ dz
To get the inward pointing normal, choose the minus sign. So

√ √
F · n̂ dS = 2 cos θ − z sin θ ı̂ı + 2 sin θ + z cos θ ̂ + (· · · )k̂
√ √
· − 2 cos θ ı̂ı − 2 sin θ ̂ dθ dz

= −2 cos θ − z sin θ cos θ + sin θ + z cos θ sin θ dθ dz
= −2 dθ dz
On the intersection of the sphere and cylinder
z 2 = 4 − x2 − y 2 = 4 − 2 = 2
√ √
so z runs from − 2 to 2 (see the figure below) and
√
ZZ Z 2 Z 2π √
F · n̂ dS = −2 √ dz dθ = −8 2π
S1 − 2 0
(b) Observe that ∇ · F = 3. So

ZZZ ZZZ
∇ · F dV = 3 dV
V V
√ horizontal cross-section of V at height z is a washer with outer radius

The √
4 − z 2 (determined by the equation of the sphere) and inner radius 2
(determined by the equation of the cylinder).
z
√ √
(0, 2, 2)
√
(0, 4 − z 2 , z)
x2 +y 2 =2
x x2 + y 2 + z 2 = 4
√ 2 √ 2
So the cross-section has area π 4 − z2 −π 2 = π 2 − z 2 and
√
ZZZ ZZZ Z 2
π 2 − z 2 dz

∇ · F dV = 3 dV = 3 √
V V − 2
√
Z 2 √ 23/2
2 − z 2 dz = 6π 2 2 −

= 6π
3
√ 0
= 8 2π
(c) By the divergence theorem

ZZ ZZZ ZZ √
F · n̂ dS = ∇ · F dV − F · n̂ dS = 16 2π
S2 V S1
4.2.6.35. Solution. By the divergence theorem

ZZ ZZZ
E · n̂ dS = ∇ · E dV
∂V V
So by Gauss’ law
ZZZ ZZZ ZZZ

∇ · E dV = 4π ρ dV =⇒ ∇ · E − 4πρ dV = 0
V V V
This is true for all solids V for which the divergence theorem applies. If
there were some point in R3 for which ∇ · E − 4πρ were, say, strictly
bigger than zero, then, by continuity, we could find a ball B centered on
that
RRR point
with ∇ · E − 4πρ > 0 everywhere B . This would force
RRR on
B
∇ · E − 4πρ dV > 0, which violates V
∇ · E − 4πρ dV = 0 with
V set equal to B . Hence ∇ · E − 4πρ must be zero everywhere.
4.2.6.36. Solution. By the divergence theorem
ZZ ZZZ ZZZ
r · n̂ dS = ∇ · r dV = ∇ · (xı̂ı + y ̂ + z k̂) dV
∂V
Z Z ZV V
= 3 dV = 3 Volume(V )
V
Our geometric explanation starts with the observation that the volume of
the cone with vertex (0, 0, 0) and base a tiny piece of surface dS is 31 times
the area of the base times the height of the cone. The height of the cone
is |n̂ · r|, where r is a point in dS. So the volume of the cone is 13 |n̂ · r| dS.
n̂
dS
(0, 0, 0)
First assume that (0, 0, 0) is in V and V is convex. Then
• n̂ · r > 0, and the volume is 13 n̂ · r dS.

• the cone is contained in V and
• V is the union of all the tiny conical pieces with dS running over ∂V .
So ZZ
1
Volume(V ) = r · n̂ dS
3 ∂V
To generalise to the case that V is not convex or (0, 0, 0) is not in V , write

V as the difference between a large convex solid and one or more smaller
convex solids.
4.2.6.37. ∗. Solution. (a) We’ll parametrize the sphere using the spher-
ical coordinates θ and ϕ.
x = 3 sin ϕ cos θ
y = 3 sin ϕ sin θ
z = 3 cos ϕ
with 0 ≤ θ ≤ 2π, 0 ≤ ϕ ≤ π. Since
∂x ∂y ∂z
, , = − 3 sin ϕ sin θ , 3 sin ϕ cos θ , 0
∂θ ∂θ ∂θ
∂x ∂y ∂z
, , = (3 cos ϕ cos θ , 3 cos ϕ sin θ , −3 sin ϕ)
∂ϕ ∂ϕ ∂ϕ
(3.3.1) yields
∂x ∂y ∂z ∂x ∂y ∂z
n̂ dS = ± , , × , , dθdϕ
∂θ ∂θ ∂θ ∂ϕ ∂ϕ ∂ϕ

= ± − 3 sin ϕ sin θ , 3 sin ϕ cos θ , 0
× (3 cos ϕ cos θ , 3 cos ϕ sin θ , −3 sin ϕ) dθdϕ

= ± − 9 sin2 ϕ cos θ , −9 sin2 ϕ sin θ , −9 sin ϕ cos ϕ dθdϕ

= ∓9 sin ϕ sin ϕ cos θ , sin ϕ sin θ , cos ϕ dθdϕ
To get an outward pointing normal we need the + sign. For example, with
the + sign, the z-component is 9 sin ϕ cos ϕ = 29 sin(2ϕ) so that the normal
is pointing upward when 0 < ϕ < π2 , i.e. in the northern hemisphere, and
is pointing downward when π2 < ϕ < π, i.e. in the southern hemisphere.
(As a further consistency check, note that n̂(θ, ϕ) is parallel to r(θ, ϕ).) So
ZZ Z 2π Z π
F · n̂ dS = 9 dθ dϕ sin ϕ (0, 0, 3 sin ϕ cos θ + 3 cos ϕ)
S 0 0

· sin ϕ cos θ , sin ϕ sin θ , cos ϕ
Z 2π Z π
dϕ sin2 ϕ cos ϕ cos θ + sin ϕ cos2 ϕ

= 27 dθ
0 0
Z π Z 2π
= 54π dϕ sin ϕ cos2 ϕ since cos θ dθ = 0
0 0
π
= −18π cos3 ϕ 0

= 36π
(b) Set
(x, y, z) ∈ R3 x2 + y 2 + z 2 ≤ 9

V =
Since
∂
x+z =1 ∇·F=
∂z
ZZ ZZZ ZZZ
4
F · n̂ dS = ∇ · F dV = dV = π33 = 36π
S V V 3
4.2.6.38. ∗. Solution. Denote by V the cube specified in the problem.
Then ∂V consists of S together with the face F in the plane z = 0, oriented
with the normal being −k̂.
z
n̂ = k̂
(0, 1, 1)
y
F
(1, 1, 0)
x
As

∂ ∂ z ∂ 2
y cos(y 2 ) + z − 1 + xyez

∇·F= +1 +
∂x ∂y x+1 ∂z
∂ 2
= xyez
∂z
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
S V F
Z 1 Z 1 Z 1 Z Z 1
1
∂ 2

2
= dx xyez + dy dx dy xyez
dz
0 0 0 ∂z 0 0 z=0
Z 1 Z 1 z=1
Z 1 Z 1
2 2
= dx dy xyez + dx dy xyez
0 0 z=0 0 0 z=0
Z 1 Z 1
2
= dx dy xyez
0 0 z=1
2 1 2 1

x y
=e
2 0 2 0
e
=
4
4.2.6.39. ∗. Solution. (a) The equation of the surface is G(x, y, z) =

z − xy = 0. So one normal to the surface at (1, 1, 1) is (∇ ∇G)(1, 1, 1) =
(−y, −x, 1)(x,y,z)=(1,1,1) = (−1, −1, 1) and a unit upward pointing normal
(−1,−1,1)
at (1, 1, 1) is |(−1,−1,1)| = √13 (−1, −1, 1).
(b) For the surface G(x, y, z) = z − xy, so that, by (3.3.3),
∇ G(x, y, z)
n̂ dS = ± dxdy = ±(−y, −x, 1)dxdy
∇ G(x, y, z) · k̂
The “+” sign gives the upward normal, so the specified upward flux is
ZZ ZZ
F · n̂ dS = (y, x, 3) · (−y − x, 1) dxdy
S x2 +y 2 ≤9
ZZ
= (3 − x2 − y 2 ) dxdy
x2 +y 2 ≤9

ZZ Z 3 Z 2π Z 3
F · n̂ dS = dr r dθ (3 − r2 ) = 2π dr (3r − r3 )
S 0 0 0
= 2π 23 32 − 14 34 = − 27π

2
(c) by direct evaluation: Parametrize the specified surface using the

cylindrical coordinates θ and z.
x = 3 cos θ
y = 3 sin θ
z=z
with 0 ≤ θ ≤ 2π and 9 sin θ cos θ ≤ z ≤ 10. Then, using (3.3.1),

∂r
= − 3 sin θ , 3 cos θ , 0
∂θ
∂r
= 0, 0, 1
∂z
∂r ∂r
× = 3 cos θ , sin θ , 0
∂θ ∂z
∂r ∂r
n̂ dS = × dθ dz = 3 cos θ , sin θ , 0 dθ dz
∂θ ∂z
∂r ∂r
(We have taken the + sign in n̂ dS = ± ∂θ × ∂z dθ dz to give the outward
pointing normal.) So the specified flux is
F=(y,x,3)
ZZ Z 2π Z 10 z }| {
F · n̂ dS = 3 dθ dz 3 sin θ , 3 cos θ , 3 · cos θ , sin θ , 0
0 9 cos θ sin θ
Z 2π Z 10
= 18 dθ dz sin θ cos θ
0 9 cos θ sin θ
Z 2π
= 18 dθ 10 − 9 cos θ sin θ sin θ cos θ
0
Z 2π
= −9 × 18 dθ sin2 θ cos2 θ
0
Z 2π Z 2π
1
since sin θ cos θ dθ = sin(2θ) dθ = 0
0 2 0
1 2π
Z
= −9 × 18 × dθ sin2 (2θ)
4 0
81 2π 1 − cos(4θ)
Z
=− dθ
2 0 2
2π
81 θ sin(4θ)
=− −
2 2 8 0
81
=− π
2
R 2π
For an efficient, sneaky, way to evaluate 0 dθ sin2 (2θ) see Example 2.4.4.
(c) using the divergence theorem: Note that if x2 + y 2 ≤ 9, then |x| ≤ 3
and y ≤ 3 so that |xy| ≤ 9 < 10. Set
S̃ = (x, y, z) x2 + y 2 = 9, xy ≤ z ≤ 10

V = (x, y, z) x2 + y 2 ≤ 9, xy ≤ z ≤ 10

Note that the boundary, ∂V , of V consists of three parts:

• the side S̃, with outward pointing normal (which is the surface and
the normal specified in part (c) of the question)
• the bottom, which is the surface S of part (b), with downward point-
ing normal (which is opposite the normal specified in part (b)) and

• the top, which is the surface ST = (x, y, z) x2 + y 2 ≤ 9, z = 10 ,
with normal n̂ = k̂.
Here is a sketch of the part of ∂V that is in the first octant.
z ST
S̃
S
y
x
Note that ∇ · F = 0. So the divergence theorem yields
ZZZ
0= ∇ · F dV
ZZ V
= F · n̂ dS
˜
∂V
ZZ ZZ ZZ
= F · n̂ dS − F · n̂ dS + F · k̂ dS
S̃ S ST
This implies
ZZ ZZ ZZ
F · n̂ dS = F · n̂ dS − F · k̂ dS
S̃ S ST
ZZ
= − 27π
2 − 3 dS
x2 +y 2 ≤9
= − 27π
2 − 3π32 = − 81π 2
4.2.6.40. ∗. Solution. (a) The divergence of F is

∂ ∂ ∂ 2
∇·F= ∂x (x + sin y) + ∂y (z + y) + ∂z (z )
= 2 + 2z
(b) Set
p
(x, y, z) x2 + y 2 ≤ 25, 0 ≤ z ≤ 25 − x2 − y 2

V =
ST = (x, y, z) x2 + y 2 + z 2 = 25, z ≥ 0

SB = (x, y, z) x2 + y 2 ≤ 25, z = 0

Note that the boundary, ∂V , of V consists ot two parts — ST with upward

normal, and SB with normal −k̂. We are to find the flux through ST with
upward normal. By the divergence theorem, it is
ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · (−k̂) dS
ST V SB
ZZZ
= (2 + 2z) dV
V
since F · k̂ = z 2 = 0 on SB . We’ll compute the volume integral by ex-

pressing it as an iterated integral, with the z integration on the outside.
In V , z ranges
for 0 to 5. The set of points at exactly height z in V is
(x, y, z) x2 + y 2 ≤ 25 − z 2 . So
ZZ Z 5 ZZ
F · n̂ dS = dz dx dy (2 + 2z)
ST 0 x2 +y 2 ≤25−z 2
Z 5 ZZ
= dz (2 + 2z) dx dy
0 x2 +y 2 ≤25−z 2
Z 5
= dz π(25 − z 2 )(2 + 2z)
0
RR √
since x2 +y 2 ≤25−z 2
dx dy is the area of a disk of radius 25 − z 2 . Contin-
uing,
ZZ Z 5
F · n̂ dS = π dz (50 − 2z 2 + 50z − 2z 3 )
ST 0
53 52 54
= π 50 × 5 − 2 + 50 − 2
3 2 4
2 5
= π53 2 − + 5 −
3 2
3 4 5

= π5 +
3 2
23 2875
= π 53 = π
6 6
(c) To start, consider any closed surface S that is the boundary of a
solid V . Use
• the outward pointing normal for S,
• |V | to denote the volume of V , and
• z̄ = |V1 |
RRR
V
z dV to denote the z-component of the centroid (i.e.
centre of mass with constant density) of V .
Then, by the divergence theorem

ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (2 + 2z) dV
S
Z ZVZ ZZZ V
=2 dV + 2 z dV
V V
= 2|V | + 2|V |z̄
This takes the value −9 if and only if

9
2|V |z̄ = −9 − 2|V | ⇐⇒ z̄ = − −1
2|V |
One surface which obeys this condition is the unit cube (with outward
normal) centred on 0, 0, − 11
2 .
4.2.6.41. ∗. Solution. (a) The constant z cross-section of the cone at

height 0 ≤ z ≤ 1 is a circle of radius 2z. So we may parametrize the cone
by
r(θ, z) = 2z cos θ ı̂ı + 2z sin θ ̂ + z k̂ 0 ≤ θ < 2π, 0 ≤ z ≤ 1
Since
∂r

∂θ = − 2z sin θ , 2z cos θ , 0
∂r

∂z = 2 cos θ , 2 sin θ , 1
∂r ∂r

∂θ × ∂z = 2z cos θ , 2z sin θ , −4z
(3.3.1) yields that the element of surface area for this parametrization is
∂r ∂r
dS = ∂θ × ∂z dθdz = 2z (cos θ , sin θ , −2 dθdz
√
= 2 5z dθdz
In our parametrization the condition x ≤ y becomes 2z cos θ ≤ 2z sin θ,

which, for z > 0, is equivalent to tan θ ≥ 1. So the specified integral is
√ Z 1 √
ZZ √ Z 1 Z π/2 5π 5π
z 2 dS = 2 5 dz dθ z 3 = dz z 3 =
S 0 π/4 2 0 8
(b) Let’s first do some strategizing. We have to compute a flux integral

over a surface that is not closed. There are two potential sneaky attacks
that come to mind.
• The first uses Stokes’RR theorem. But the flux integral in Stokes’ theo-
rem is of the form S ∇ × A · n̂ dS. So to be able to apply Stokes’
theorem in the current problem, F has to be of the form ∇ ×A. That
is, F has to have a vector potential. We know that in order for F to
have a vector potential, it must pass the screening test ∇ ·F = 0. Our
F = z k̂ fails this screening test. So we can’t use Stokes’ theorem.
• The second uses the divergence theorem. But the flux integral in
the divergence theorem is over the boundary of a solid. That is not
the case for our S. So in order to apply the divergence theorem
in the current problem, we have to enlarge S to the boundary of a
solid. There are many ways to do this. But they all appear fairly
complicated. So it does not seem wise to use the divergence theorem.
So it looks like we have to evaluate the flux integral directly. To do so, we

have to determine n̂ dS for the specified rectangle. Look at the sketch of
S below. It is part of a plane,
z
(0, 0, 4)
(5, 0, 4)
(0, 2, 0)
y
x
(5, 2, 0)
and that plane is invariant under translations parallel to the x axis. As
the plane does not pass through the origin, the equation of the plane has
to be of the form by + cz = 1. For (0, 0, 4) to be on the plane, we need
c = 14 . For (0, 2, 0) to be on the plane, we need b = 12 . So S is contained
in the plane G(x, y, z) = y2 + z4 = 1 and equation (3.3.3) gives that
∇ G(x, y, z) (0, 21 , 41 )
n̂ dS = ± dxdy = ± 1 dxdy = ±(0, 2, 1) dxdy
∇ G(x, y, z) · k̂ 4
The problem specifies that the normal is to be upward, i.e. is to have a

positive z-component. So
n̂ dS = (0, 2, 1) dxdy
Again looking at the sketch of S above we see, as (x, y, z) runs over S,
(x, y) runs over

R = (x, y) 0 ≤ x ≤ 5, 0 ≤ y ≤ 2
Thus our flux integral is
z n̂ dS
ZZ Z Z z }|
{ z }| { Z 2 Z 5
F · n̂ dS = 4 − 2y k̂ · (0, 2, 1) dxdy = dy dx (4 − 2y)
S R 0 0
Z 2 h i2
= dy 5(4 − 2y) = 5 4y − y 2 = 20
0 0
(c) The divergence of the given vector field is ∇ ·F =

2z, which is pretty
simple. So let’s use the divergence theorem. If V = (x, y, z) 0 ≤ x ≤
1, 0 ≤ y ≤ 2, 0 ≤ z ≤ 3 , the divergence theorem says that
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 2 z dV
S V V
This integral would be easy enough to evaluate directly, but we don’t need
to. The average value of z (i.e. the z-coordinate of the centre of mass with
constant density) is 23 , by symmetry. Since V has volume 6, that average
value of z is also
ZZZ
1 3
z̄ = z dV =
6 V 2
RRR
So V
z dV = 9
4.2.6.42. ∗. Solution.
ZZ (a) For theZ Zsurface
Z z = f (x, y) = 1 − x2 − y 2 ,
F · n̂ dS = 2 z dV = 18
S V
with an upwards pointing normal,

n̂ dS = − fx (x, y) dx − fy (x, y) + k̂ dxdy = 2xı̂ı + 2y ̂ + k̂ dxdy
by (3.3.2). So the specified upward flux is
ZZ
F · n̂ dS
σ1
ZZ
[a(y 2 + z 2 ) + bxz]ı̂ı + [c(x2 + z 2 ) + dyz] ̂ + x2 k̂

=
x2 +y 2 ≤1

· 2xı̂ı + 2y ̂ + k̂ z=1−x2 −y2 dxdy
ZZ n
= [2ax(y 2 +z 2 ) + 2bx2 z]
x2 +y 2 ≤1
o
+ [2cy(x2 +z 2 ) + 2dy 2 z] + x2 dxdy
z=1−x2 −y 2
Now ZZ
2ax(y 2 + z 2 )

z=1−x2 −y 2
dxdy = 0
x2 +y 2 ≤1
because the integrand is odd under x → −x and
ZZ
2cy(x2 + z 2 ) z=1−x2 −y2 dxdy = 0

x2 +y 2 ≤1
because the integrand is odd under y → −y. So that leaves

ZZ ZZ
2bx2 z + 2dy 2 z + x2 z=1−x2 −y2 dxdy

F · n̂ dS =
σ1 x2 +y 2 ≤1
We’ll switch to polar coordinates to evaluate the remaining integral.

ZZ Z 1 Z 2π
dθ 2br2 z cos2 θ + 2dr2 z sin2 θ + r2 cos2 θ z=1−r2

F · n̂ dS = dr r
σ1 0 0
Now
Z 2π Z 2π 2π
cos(2θ) + 1 sin(2θ) θ
cos2 θ dθ = dθ = + =π
0 0 2 4 2 0
Z 2π Z 2π 2π
2 1 − cos(2θ) θ sin(2θ)
sin θ dθ = dθ = − =π
0 0 2 2 4 0
R 2π R 2π
For an efficient, sneaky, way to evaluate 0 cos2 θ dθ and 0 sin2 θ dθ, see
Example 2.4.4. So, we finally have
ZZ Z 1
dr 2πbr3 (1 − r2 ) + 2πdr3 (1 − r2 ) + πr3

F · n̂ dS =
σ1 0
h1 1i h1 1i 1 π π(b + d)
= 2πb − + 2πd − +π = +
4 6 4 6 4 4 6
(b), (c) Here is a side view of σ1 , σ2 and σ3 .
σ1 , z = 1 − x2 − y 2
σ3
σ2 , z = x2 + y 2 − 1
Set
(x, y, z) 0 ≤ z ≤ 1 − x2 − y 2 , x2 + y 2 ≤ 1

Vb =
Vc = (x, y, z) x2 + y 2 − 1 ≤ z ≤ 1 − x2 − y 2 , x2 + y 2 ≤ 1

Then ∂Vb = σ1 ∪ σ3 and ∂Vc = σ1 ∪ σ2 , all with outward pointing normals.

Since the divergence of F is
∂ ∂ ∂ 2
∇·F= [a(y 2 +z 2 ) + bxz] + [c(x2 +z 2 ) + dyz] + [x ] = (b + d)z
∂x ∂y ∂z
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (b + d) z dV
σ1 ∪σ3 Vb Vb
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = (b + d) z dV
σ1 ∪σ2 Vc Vc
RRR
Now on Vb , z ≥ 0 and z > 0 except on σ3 . So Vb
z dV > 0 and
RR
σ1 ∪σ3
F · n̂ dS is zero if and only if d = −b. That’s the answer to part
(b). RRR
On the other hand, Vc is even under z → −z so that Vc
z dV = 0.
RR
Consequently σ1 ∪σ3 F · n̂ dS is zero for all a, b, c, d. That’s the answer
to part (c).
4.2.6.43. ∗. Solution. We will be using the divergence theorem in both
parts (a) and b. So as a prelimary calculation, let’s find the divergence of
H(x, y, z) = [(x−a)2(x,y,z)−(a,b,c)
+(y−b)2 +(z−b)2 ]3/2
for any (a, b, c). If (x, y, z) 6= (a, b, c),
∇ · H(x, y, z)
∂ x−a
=
∂x [(x − a)2 + (y − b)2 + (z − c)2 ]3/2
∂ y−b
+
∂y [(x − a) + (y − b)2 + (z − c)2 ]3/2
2
∂ z−c
+
∂z [(x − a)2 + (y − b)2 + (z − c)2 ]3/2
(x − a)2 + (y − b)2 + (z − c)2 − (x − a) 23 (2(x − a))

= 5/2
(x − a)2 + (y − b)2 + (z − c)2
(x − a)2 + (y − b)2 + (z − c)2 − (y − b) 23 (2y)

+ 5/2
(x − a)2 + (y − b)2 + (z − c)2
(x − a)2 + (y − b)2 + z 2 − (z − c) 23 (2(z − c))

+ 5/2
(x − a)2 + (y − b)2 + (z − c)2

3 (x−a)2 + (y−b)2 + (z−c)2 − 3(x−a)2 − 3(y−b)2 − 3(z−c)2
= 5/2
(x − a)2 + (y − b)2 + (z − c)2
=0
If (x, y, z) = (a, b, c), H(x, y, z) is not defined and hence ∇ · H(x, y, z) is

also not defined.
(b) By the above preliminary computation with (a, b, c) = (3, 2, 2), ∇ ·G
is defined and zero for all (x, y, z) 6= (3, 2, 2), and, in particular for all
(x, y, z) in
(x, y, z) x2 + 2y 2 + 3z 2 ≤ 16

V =
So, by the divergence theorem,
ZZ ZZZ
G · n̂ dS = ∇ · G dV = 0
S V
(a) Because (1, 1, 2) is inside V , we cannot use the argument of part

(b), to conclude that the integral is zero. Let ε > 0 be small enough that
Sε = (x, y, z) (x − 2)2 + (y − 1)2 + (z − 1)2 = ε2

is completely contained inside V , as in the sketch below.
Sε S
Vε
(2,1,1)
Set
Vε = (x, y, z) x2 + 2y 2 + 3z 2 ≤ 16, (x − 2)2 + (y − 1)2 + (z − 1)2 ≥ ε2

The boundary, ∂Vε , of V consists of two parts — S and Sε , with the

normals as in the figure above. The divergence ∇ · F of F is well-defined
and zero throughout Vε . Consequently, the divergence theorem gives
ZZZ ZZ ZZ
0= ∇ · F dV = F · n̂ dS + F · n̂ dS
V S Sε
So ZZ ZZ
F · n̂ dS = − F · n̂ dS
S Sε
The unit normal to Sε at the point (x, y, z) on Sε is

1
n̂ = − (x − 2)ı̂ı + (y − 1) ̂ + (z − 1) k̂
ε
(Recall that |(2 − 1)ı̂ı + (y − 1) ̂ + (z − 1) k̂| = ε on Sε . So, on Sε ,
!
1 (x, y, z) − (2, 1, 1)
F · n̂ = −
ε (x − 2)2 + (y − 1)2 + (z − 1)2 3/2

· (x − 2)ı̂ı + (y − 1) ̂ + (z − 1) k̂
!
1 (x − 2)2 + (y − 1)2 + (z − 1)2
=−
ε (x − 2)2 + (y − 1)2 + (z − 1)2 3/2
1
=−
ε2
Hence
ZZ ZZ
1 1
F · n̂ dS = − − dS = (4πε2 ) = 4π
S Sε ε2 ε2
4.2.6.44. ∗. Solution. This was part of Theorem 4.2.9. To prove it

apply the divergence theorem, but with F replaced by a × F, where a is
any constant vector.
ZZ ZZZ
(a × F) · n̂ dS = ∇ · (a × F) dV
∂Ω V
ZZZ

= ∇ × a) −a · (∇
F · (∇ ∇ × F) dV
Ω | {z }
=0
ZZZ ZZZ
=− ∇ × F) dV = −a ·
a · (∇ ∇ × F dV
Ω Ω
To get the second line we used the vector identity Theorem 4.1.4.d. To get
the third line, we used that a is a constant, so that all of its derivatives are
zero. For all vectors a · (b × c) = (a × b) · c (in case you don’t remember
this, it was Lemma 4.1.8.a) so that
(a × F) · n̂ = a · (F × n̂)
and
ZZ ZZZ
a· F × n dS = −a · ∇ × F dV
Z∂Ω
Z ZZZ Ω
=⇒ a · F × n dS + ∇ × F dV = 0
∂Ω Ω
RR choosing a = ı̂ıRRR
In particular, , ̂ and k̂, we see that all three components of
the vector ∂Ω F × n dS + Ω
∇ × F dV are zero. So
ZZZ ZZ ZZ
∇ × F dV = − F × n dS = n̂ × F dS
Ω ∂Ω ∂Ω

4.2.6.45. ∗. Solution. Pressure is force per unit surface area acting
normally into a surface. So the force per unit surface area is −pn̂. The
total force acting on S is
ZZ ZZZ
− pn̂ dS = − ∇p dV
S E
We are assuming that p is a constant, so that ∇p = 0 and the total force

is zero.
4.2.6.46. ∗. Solution. Let Sa denote the sphere x2 + y 2 + z 2 = a2 and
Va denote the solid inside it, which is the ball x2 + y 2 + z 2 ≤ a2 . Then, by
the divergence theorem, Theorem 4.2.2,
ZZ ZZZ
π(a3 + 2a4 ) = F · n̂ dS = ∇ · F dV
Sa Va
RRR a, ∇ · F is almost equal to ∇ · F(0, 0, 0) on all of Va ,

Now, for very small
and the integral Va
∇ · F dV will be
4 3
∇ · F(0, 0, 0)Volume(Va ) + O(a4 ) = πa ∇ · F(0, 0, 0) + O(a4 )
3
Here O(a4 ) is an error term that is bounded by a constant times a4 . This
is consistent with the above equation if and only if ∇ · F(0, 0, 0) = 43 .
4.2.6.47. ∗. Solution. Note that, since z 2 − 2az = (z − a)2 − a2 ,
S = (x, y, z x2 + y 2 + (z − a)2 = 4a2 , z ≥ 0

Let V be the solid
V = (x, y, z) x2 + y 2 + (z − a)2 ≤ 4a2 , z ≥ 0

It is the interior of the sphere of radius 2a centred on (0, 0, a). The surface
of V (with outward normal) is the union of S (with normal pointing away
from the origin) and the disk
B = (x, y, 0) x2 + y 2 ≤ 3a2

with normal −k̂. Hence, by the Divergence Theorem

ZZ ZZZ ZZ
F · n̂ dS = n · F dV − F · (−k̂) dS
S
Z Z ZV B
ZZ
= (2x + 2y + 1) dV − (−3 − x) dS
V B
RRR V and RRR

Both RR under x → −x and under y → −y, so
B are invariant
V
x dV = V
y dV = B
x dS = 0 and
ZZ ZZZ ZZ
F · n̂ dS = dV + 3 dS
S V B
To evaluate the integral over V , we note that z runs from 0 to 3a and that
the cross section of
V = (x, y, z) 0 ≤ z ≤ 3a, x2 + y 2 ≤ 4a2 − (z − a)2 , z ≥ 0

with fixed z is the circular disk x2 + y 2 ≤ 4a2 − (z − a)2 = 3a2 + 2az − z 2 ,

√ 2
which has area π 3a2 + 2az − z 2 . So
ZZ Z 3a p 2
F · n̂ dS = π 3a2 + 2az − z 2 dz + 3 Area(B)
S 0
Z 3a
=π (3a2 + 2az − z 2 ) dz + 3π(3a2 )
0
9a2 27a3

2
= π 3a × 3a + 2a × − + 9πa2
2 3
= 9πa3 + 9πa2
4.2.6.48. ∗. Solution. (a) Let S denote the boundary of R. Then
“the total flux of F = ∇u out through the boundary of R” is given by the
integral ZZ
I= F · n̂ dS
S
Thanks to the divergence theorem,
ZZZ 2
∂ u ∂2u ∂2u
ZZZ ZZZ
I= ∇ · F dV = ∇ · ∇u dV = 2
+ 2 + 2 dV
R R R ∂x ∂y ∂z
=0
(b) Similarly, “the total flux of G = u∇u out through the boundary of
R” equals ZZ ZZZ
J= G · n̂ dS = ∇ · G dV
S R
Here G = uF (using the notation from part (a)), so by the vector identity
of Theorem 4.1.4.c,
∇ · G = ∇ · (uF) = (∇u) · F + u(∇ · F)
But F = ∇u, so ∇ · F = ∆u = 0 as in part (a), giving
∇ · G = |∇u|2 + 0
In conclusion,
ZZZ Z Z Z h 2 2 2 i
∂u ∂u ∂u
J= ∇ · G dV = + + dV
R R ∂x ∂y ∂z
4.2.6.49. ∗. Solution. (a) This is a classic case for the divergence
theorem. The flux we want equals
ZZ ZZZ ZZZ ZZZ
I= F · n̂ dS = ∇ · F dV = (2x + 2 − 2) dV = 2 x dV
S R R R
The solid R clearly has reflection symmetry across the plane x = 0. So the
x-coordinate of the centre of mass of R, i.e. the average value of x over R,
i.e. RRR RRR
R
x dV R
x dV
x̄ = RRR =
R
dV Vol(R)
is zero. Hence
I = 2x̄Vol(R) = 0
Alternatively, here is a direct evaluation of 2
RRR
R
x dV . The base region
x2 + (y − 1)2 ≤ 1 is the circular disk of radius 1 centred on (0, 1). In polar
coordinates it is
r2 cos2 θ + (r sin θ − 1)2 ≤ 1 or r2 − 2r sin θ + 1 ≤ 1 or r ≤ 2 sin θ
Because the disk is contained in the upper half plane, the polar angle θ
is restricted to 0 ≤ θ ≤ π. So, in cylindrical coordinates, the solid R is
described by
0 ≤ θ ≤ π, 0 ≤ r ≤ 2 sin θ, 0 ≤ z ≤ r2 sin2 θ
Hence
Z π Z 2 sin θ Z r 2 sin2 θ
I=2 (r cos θ) dz r dr dθ
θ=0 r=0 z=0
Z π Z 2 sin θ
=2 r4 sin2 θ cos θ dr dθ
θ=0 r=0
Z π h 25 sin5 θ i
=2 sin2 θ cos θ dθ
θ=0 5
Z π
64
= sin7 θ cos θ dθ
5 θ=0
64 h sin8 θ iπ
=
5 8 θ=0
=0
(b) using part (a): We have

ZZ ZZ ZZ ZZ
F · n̂ dS = F · n̂ dS + F · n̂ dS + F · n̂dS
S Sbottom Stop Sside
On Sbottom , z = 0 and the outward unit normal is n̂ = −k̂, so F · n̂ = 0.

Hence ZZ ZZ
F · n̂ dS = 0 dS = 0
Sbottom Sbot
2 2
On Stop , z = y , so F = (2x, 2y, −2y ) and, by (3.3.2),
n̂dS = (0, −2y, 1) dxdy
Hence (by the Hint)

ZZ ZZ
F · n̂ dS = [−4y 2 − 2y 2 ] dxdy
Sbot D
Z π Z 2 sin θ
r2 sin2 θ r dr dθ

= −6
θ=0 r=0
24 π
Z
= −6 sin6 θ dθ
4 θ=0
5 π
Z
= −24 sin4 θ dθ
6 θ=0
53 π
Z
= −24 sin2 θ dθ
6 4 θ=0
531 π
Z
= −24 dθ
6 4 2 θ=0
h5 3 1 i
= −24 π
642
15
=− π
2
The conclusion is
ZZ ZZ ZZ ZZ
15
F · n̂dS = F · n̂dS − F · n̂dS − F · n̂dS = π
Sside S Stop Sbot 2
(b) by direct evaluation: Use the polar equation r = 2 sin θ to parametrize

Sside :
r(θ, t) = (r cos θ, r sin θ, t) = (2 sin θ cos θ, 2 sin2 θ, t),

0 ≤ θ ≤ π, 0 ≤ t ≤ y 2 = 4 sin4 θ
Then using (3.3.1),

∂r ∂r
F · n̂ dS = F · × dθ dt
∂θ ∂t
4 sin2 θ cos2 θ 4 sin2 θ
 
−2t
= det 2(cos2 θ − sin2 θ) 4 sin θ cos θ 0  dθ dt
0 0 1
2 2 2

4 sin θ cos θ 4 sin θ
= det dθ dt
2(cos2 θ − sin2 θ) 4 sin θ cos θ
= 16 sin3 θ cos3 θ − 8 sin2 θ cos2 θ − sin2 θ dθ dt

= 16 sin3 θ 1 − sin2 θ cos θ − 8 sin2 θ 1 − 2 sin2 θ dθ dt

= 8 2 sin3 θ cos θ − 2 sin5 θ cos θ − sin2 θ + 2 sin4 θ dθ dt

so
ZZ
F · n̂dS
Sside
Z π Z 4 sin4 θ
2 sin3 θ cos θ − 2 sin5 θ cos θ − sin2 θ + 2 sin4 θ dt dθ

=8
θ=0 t=0
Z π
2 sin7 θ cos θ − 2 sin9 θ cos θ − sin6 θ + 2 sin8 θ dθ

= 32
θ=0
h sin8 θ π Z π Z π
sin10 θ i 6
= 32 2 −2 − 32 sin θ dθ + 64 sin8 θ dθ
8 10 0 θ=0 θ=0
531 7531
= −32 π + 64 π (by the Hint as above)
642 8642
15
= π.
2
(b) Offset polar alternative: We can also parametrize S using cylindrical
coordinates translated so that the centre of the base of the cylinder, namely
(0, 1, 0), plays the role of the origin. Then, looking at the figure
z
(x, y, z)
(0, 1, 0) z y
θ r
x (x, y, 0)
we see that
x = r cos θ y = 1 + r sin θ z=z
In these coordinates, the base region, x2 + (y − 1)2 ≤ 1, z = 0, of the

cylinder is 0 ≤ r ≤ 1, z = 0. So we can parametrize S by
x = cos θ, y = 1 + sin θ, z = t, 0 ≤ θ ≤ 2π, 0 ≤ t ≤ (1 + sin θ)2
By (3.3.1),
 
ı̂ı ̂ k̂
∂r ∂r
× = det  0 0 1  = − cos θ, − sin θ, 0 ,
∂t ∂θ
− sin θ cos θ 0
∂r ∂r
n̂ dS = − × dt dθ = (cos θ, sin θ, 0)dt dθ
∂t ∂θ
where we have chosen the sign to give the outward pointing normal. So
ZZ Z 2π Z (1+sin θ)2
cos3 θ + 2(1 + sin θ) sin θ dt dθ

F · n̂dS =
S θ=0 t=0
Z 2π
(1 + sin θ)2 cos3 θ + 2(1 + sin θ)3 sin θ dθ

=
0
Z 2π
2 sin θ cos3 θ + 6 sin2 θ + 2 sin4 θ dθ

=
0
2π Z π Z π
1
= − cos4 θ + 12 sin2 θ dθ + 4 sin4 θ dθ

2 0 0 0
π 3π 15
= 0 + 12 + 4 = π
2 42 2
To get the third line, we used that the integral over 0 ≤ θ ≤ 2π of any odd
power of sin θ or cos θ is zero.
4.2.6.50. ∗. Solution. The circle x2 + y 2 = 4y, or equivalently, x2 +
(y − 2)2 = 4, has radius 2 and centre (0, 2). On the bottom surface, z = 0
and the outward normal is −k̂, so that
ZZ ZZ ZZ
F · n̂ dS = − F · k̂ dxdy = − (2x + 3y) dxdy
D D D
By symmetry, the centre of mass, (x̄, ȳ), of the circle is (0, 2). Here x̄ and
ȳ are the average values
RR RR
D
x dxdy y dxdy
x̄ = RR ȳ = RRD
D
dxdy D
dxdy
of x and y over D. As the disk D has area 4π,

ZZ ZZ
x dxdy = 4πx̄ = 0 y dxdy = 4π ȳ = 8π
D D
and
ZZ
F · n̂ dS = −4π(2x̄ + 3ȳ) = −4π(2 × 0 + 3 × 2) = −24π
D
As
∂ ∂ ∂
∇·F= (x + x2 y) + (y − xy 2 ) + (z + 2x + 3y)
∂x ∂y ∂z
= (1 + 2xy) + (1 − 2xy) + (1)
=3

ZZ ZZZ ZZ
F · n̂ dS = ∇ · F dV − F · n̂ dS
S
Z Z ZR D
= 3 dV − (−24π) = 3Vol(R) + 24π = 3 × 10 + 24π

R
= 30 + 24π

4.3.1 · Exercises
4.3.1.1. Solution. (a) Expressing the left hand side as an iterated inte-
gral, with y as the inner integration variable, we have
ZZ Z 1 Z 1
∂f ∂f
(x, y) dx dy = dx dy (x, y)
R ∂y 0 0 ∂y
Z 1

= dx f (x, 1) − f (x, 0)
0
by the fundamental theorem of calculus
Z 1 Z 1
= f (x, 1) dx − f (x, 0) dx
0 0
(b) Define F1 (x, y) = f (x, y) and F2 (x, y) = 0. Then. by Green’s

theorem
ZZ ZZ h
∂f ∂F2 ∂F1 i
(x, y) dx dy = − (x, y) − (x, y) dx dy
R ∂y ∂x ∂y
Z R

=− F1 (x, y) dx + F2 (x, y) dy
Z∂R
=− f (x, y) dx
∂R
The boundary of R, oriented counterclockwise, is the union of four line

segments.
y
C3 (1, 1)
(0, 1)
C1 from (0, 0) to (1, 0)
C2 from (1, 0) to (1, 1) C4 R C2
C3 from (1, 1) to (0, 1)
C4 from (0, 1) to (0, 0) C1 (1, 0) x
Now x is constant on C2 and C4 so that
Z Z
f (x, y) dx = f (x, y) dx = 0
C2 C4
So, using −C3 to denote the line segment from (0, 1) to (1, 1)
ZZ Z Z
∂f
(x, y) dx dy = − f (x, y) dx + f (x, y) dx
R ∂y
Z C1 Z C3
= f (x, y) dx − f (x, y) dx
−C3 C1
Z1 Z 1
= f (x, 1) dx − f (x, 0) dx
0 0
4.3.1.2. Solution. Let r(s) = x(s)ı̂ı + y(s) ̂ be a counterclockwise

parametrization of C by arc length. Then T̂(s) = r0 (s) = x0 (s)ı̂ı + y 0 (s) ̂ is
the forward pointing unit tangent vector to C at r(s) and n̂(s) = r0 (s)× k̂ =
y 0 (s)ı̂ı − x0 (s) ̂. To see that r0 (s) × k̂ really is n̂(s), note that y 0 (s)ı̂ı − x0 (s) ̂
• has the same length, namely 1, as r0 (s) (recall that r(s) is a parametriza-
tion by arc length),
• lies in the xy-plane and
• is perpendicular to r0 (s). (Check that r0 (s) · y 0 (s)ı̂ı − x0 (s) ̂ = 0.)

• Use the right hand rule to check that r0 (s) × k̂ is n̂ rather than −n̂.
n̂(s)
r′ (s)
R
r′ (s)
C
n̂(s)
So, by Green’s theorem,
I I I
dy dx
F · n̂ ds = F1 − F2 ds = [−F2 dx + F1 dy]
C C ds ds C
ZZ
∂ ∂
= F1 − (−F2 ) dx dy
∂x ∂y
Z ZR
= ∇ · F dx dy
R
4.3.1.3. Solution. (a) Parametrize the circle by x = a cos θ, y = a sin θ,

0 ≤ θ ≤ 2π. Then dx = −a sin θ dθ and dy = a cos θ dθ so that
Z 2π 2 Z 2π
x dy − y dx a cos2 θ dθ + a2 sin2 θ dθ
I
1 1 1
= = dθ
2π C x2 + y 2 2π 0 a2 cos2 θ + a2 sin2 θ 2π 0
=1
(b) The boundary of the square has four sides — one with y = −1, one
with x = 1, one with y = 1 and one with x = −1.
dy=0
(−1, 1) y=1 (1, 1)
dx=0 dx=0
x=−1 x=1
(−1, −1) dy=0 (1, −1)

y=−1
To evaluate the integrals over the four sides
• parametrize the y = −1 part by x so that r(x) = xı̂ı − ̂, r0 (x) = ı̂ı,
with x running from −1 to 1,
• parametrize the x = +1 part by y so that r(y) = ı̂ı + y ̂, r0 (y) = ̂,
with y running from −1 to 1,
• parametrize the y = +1 part by x so that r(x) = xı̂ı + ̂, r0 (x) = ı̂ı,

with x running from 1 to −1, and
• parametrize the x = −1 part by y so that r(y) = −ı̂ı + y ̂, r0 (y) = ̂,
with y running from 1 to −1,
so that the integral

y=−1 part x=+1 part
z }| { z }| {
1 Z 1
x dy − y dx −(−1)dx 1
I Z
1 1 (1)dy
= +
2π C x2 + y 2 2π −1 x2 + 1 2π −1 1 + y 2
y=+1 part x=−1 part

z }| { z }| {
−1 Z −1
−(1)dx 1
Z
1 (−1)dy
+ +
2π 1 x2 + 1 2π 1 1 + y2
1
1 2 hπ πi
=4 arctan x = + =1
2π −1 π 4 4
√
(c) As in part (a) with a = 2, but with θ running from 0 to π, the
outer semicircle gives
Z π 2 Z π
1 a cos2 θ dθ + a2 sin2 θ dθ 1 1
= dθ =
2π 0 a2 cos2 θ + a2 sin2 θ 2π 0 2
y
x2 + y 2 = 2
x2 +y 2 =1
y=0 y=0 x
dy=0 dy=0
As in part (a) with a = 1, but with θ running from π to 0, the inner

semicircle gives
Z 0 2 Z 0
1 a cos2 θ dθ + a2 sin2 θ dθ 1 1
2 2 2 2 = dθ = −
2π π a cos θ + a sin θ 2π π 2
The two flat pieces each give zero, since on them y = 0 and dy = 0. So
x dy − y dx
I
1 1 1
2 2
= +0− +0=0
2π C x + y 2 2
4.3.1.4. Solution. The two partial derivatives
∂ x (x2 + y 2 ) − x(2x) y 2 − x2
= 2 = 2
∂x x2 + y 2 (x2 + y 2 ) (x2 + y 2 )
∂ −y −(x2 + y 2 ) − (−y)(2y) y 2 − x2
2 2
= 2 = 2
∂y x + y (x2 + y 2 ) (x2 + y 2 )
are well-defined and equal everywhere except at the origin (0, 0).
Short discussion: Were it not for the singularity at (0, 0), theR vector
field of the last problem would be conservative and the integral F · dr
around any closed curve would be zero. But as we saw in parts (a) and (b)
of Q[4.3.1.3], this is not the case. On the other hand, by Green’s theorem
(Theorem 4.3.2), the integral around the boundary of any region that does
not contain (0, 0) is zero, as happened in part (c) of Q[4.3.1.3].
Long discussion: First consider part (c) of Q[4.3.1.3]. The curve C is
the boundary of the region
R = (x, y) 1 ≤ x2 + y 2 ≤ 2, y ≥ 0

∂ x ∂ −y
The partial derivatives ∂x x2 +y 2 and ∂y x2 +y 2 are well-defined and
equal everywhere in R. So by Green’s theorem
ZZ
x dy − y dx ∂ −y
I
1 1 ∂ x
= − dx dy
2π C x2 + y 2 2π R ∂x x + y
2 2 ∂y x2 + y 2
=0
which is the answer we got before.

We cannot apply Green’s theorem in this way for parts (a) and (b)
of Q[4.3.1.3] because the singularity at (0, 0) is inside the curve C for
both parts (a) and (b). On the other hand suppose, for simplicity, that
0 < a < 1. Denote by Ca , Cb the curves of parts (a) and (b), respectively.
Define R to be the set of points that are inside Cb and outside Ca . That
is,
R = (x, y) − 1 ≤ x ≤ 1, −1 ≤ y ≤ 1, x2 + y 2 ≥ a2

Then the boundary, ∂R, of R consists of two parts. One part is Cb . The
other part is Ca , but oriented clockwise rather than counterclockwise. We’ll
call it −Ca .
y
R Cb
−Ca
x

∂ x ∂ −y
Again the partial derivatives ∂x 2
x +y 2 and 2
∂y x +y 2 are well-defined
and equal everywhere in R. So by Green’s theorem
ZZ
x dy − y dx ∂ −y
I
1 1 ∂ x
= − dx dy
2π ∂R x2 + y 2 2π R ∂x x + y
2 2 ∂y x2 + y 2
=0
Consequently
x dy − y dx x dy − y dx x dy − y dx
I I I
1 1 1
0= 2 2
= 2 2
+
2π ∂R x + y 2π Cb x + y 2π −Ca x2 + y 2
x dy − y dx x dy − y dx
I I
1 1
= 2 2
−
2π Cb x + y 2π Ca x2 + y 2
and we conclude that the answers to parts (a) and (b) should be the
same.We did indeed see that in Q[4.3.1.3].
4.3.1.5. Solution 1 (Direct evaluation). Here is a sketch of C.

(0, 3) y=3 (3, 3)
x=0 x=3
(0, 0) y=0 (3, 0)

The square consists of four line segments.
• The bottom line segment may be parametrized r(x) = (x, 0), 0 ≤
x ≤ 3. So the line integral along this segment is

Z 3 Z 3
dr
F(r(x)) · dx = (0, 0) · (1, 0) dx = 0
0 dx 0
• The second line segment may be parametrized r(y) = (3, y), 0 ≤ y ≤

3. So the line integral along this segment is
Z 3 Z 3 Z 3
dr 2
F(r(y)) · dy = (9y , 6y) · (0, 1) dy = 6y dy = 27
0 dy 0 0
• The third line segment may be parametrized r(t) = (3 − t, 3), 0 ≤

t ≤ 3. So the line integral along this segment is
Z 3 Z 3
dr
9(3 − t)2 , 6(3 − t) · (−1 , 0) dt

F(r(t)) · dt =
0 dt 0
Z 3
=− 9(3 − t)2 dt = −81
0
• The final line segment may be parametrized r(t) = (0, 3 − t), 0 ≤ t ≤

3. So the line integral along this segment is
Z 3 Z 3
dr
F(r(t)) · dt = (0, 0) · (0, −1) dt = 0
0 dt 0
The full line integral is

I
F · dr = 0 + 27 − 81 + 0 = −54
C
Solution 2 (By Green’s theorem). We apply Green’s Theorem.

I Z 3 Z 3
2 2 ∂ ∂ 2 2
x y dx + 2xy dy = dx dy (2xy) − (x y )
C 0 0 ∂x ∂y
Z 3 Z 3
dy 2y − 2x2 y

= dx
0 0
Z 3
dx 9 − 9x2

=
0
33
= 27 − 9 = −54
3
4.3.1.6. Solution. Call the trapezoid T .
y
(0, 2)
(1, 1)
T
x
(1, −1)
(0, −2)
By Green’s theorem,
I
(x sin y 2 − y 2 ) dx + (x2 y cos y 2 + 3x) dy
C
ZZ
∂ 2 ∂
= (x y cos y 2 + 3x) − (x sin y 2 − y 2 ) dx dy
∂x ∂y
Z ZT
2xy cos y 2 + 3 − 2xy cos y 2 + 2y dx dy

=
Z ZT

= 3 + 2y dx dy
T
RR
The integral T (2y) dx dy vanishes because 2y changes sign under y → −y
while
RR the domain of integration is invariant under y → −y. The integral
T
3 dx dy is 3 times the area of the trapezoid, which is its width (1) times
the average of its heights ( 21 [2 + 4]) = 3. So
I
(x sin y 2 − y 2 ) dx + (x2 y cos y 2 + 3x) dy = 3 × 1 × 3 = 9
C
4.3.1.7. ∗. Solution 1 (Using Green’s theorem). By Green’s

√ theorem
(Theorem 4.3.2), using D to denote the half-disk 0 ≤ y ≤ 4 − x2 ,
I
1 2 3
x y − x4 y dx + xy 4 + x3 y 2 dy

C 3
ZZ h
∂ ∂ 1 2 3 i
xy 4 + x3 y 2 − x y − x4 y dxdy

=
∂x ∂y 3
Z ZD
x4 + 2x2 y 2 + y 4 dxdy

=
D
ZZ
2
= x2 + y 2 dxdy
D

Z 2 Z π 2
r6
I
1 2 3
x y − x4 y dx + xy 4 + x3 y 2 dy = dθ r4 = π

dr r
C 3 0 0 6 0
32
= π
3
Solution 2 (By direct evaluation). Write C as the union of C1 , the straight
line from (−2, 0) to (2, 0), and C2 , the half-circle r(θ) = x(θ)ı̂ Rı + y(θ) ̂ =
2 cos θ ı̂ı + 2 sin θ ̂, 0 ≤ θ ≤ π. As y = 0 at every point of C1 , C1 31 x2 y 3 −

x4 y dx + xy 4 + x3 y 2 dy = 0 and
Z
1 2 3
x y − x4 y dx + xy 4 + x3 y 2 dy

I=
C 3
Z 2π h
1
= x(θ)2 y(θ)3 − x(θ)4 y(θ) x0 (θ)
0 3
i
+ x(θ)y(θ)4 + x(θ)3 y(θ)2 y 0 (θ) dθ

Z π h
1 5
= 2 cos2 θ sin3 θ − 25 cos4 θ sin θ (−2 sin θ)
0 3
i
+ 25 cos θ sin4 θ + 25 cos3 θ sin2 θ (2 cos θ) dθ

Z π
5 4
=2 cos2 θ sin4 θ + 4 cos4 θ sin2 θ dθ
0 3
Z π
1
= 25 sin2 (2θ) sin2 θ + cos2 θ dθ

3
Z0 π 1
= 24 sin2 (2θ) [1 − cos(2θ)] + [1 + cos(2θ)] dθ
0 3
since cos(2θ) = 2 cos2 θ − 1 = 1 − 2 sin2 θ
π
25
Z
sin2 (2θ) 2 + cos(2θ) dθ

=
3 0
25 π
Z
1 − cos(4θ) + sin2 (2θ) cos(2θ) dθ

=
3 0
25 h 1 1 iπ 32
= θ − sin(4θ) + sin3 (2θ) = π
3 4 6 0 3
4.3.1.8. ∗. Solution. Let’s use Green’s theorem. The rectangle, which
we shall denote R, is
y (1, 1) (3, 1)
R =
R
{(x, y)} 1 ≤ x ≤ 3, 0 ≤ y ≤ 1
(1, 0) C (3, 0) x
So Green’s theorem gives
I
2 2
3y 2 + 2xey dx + 2yx2 ey dy
C
ZZ h
∂ 2 ∂ 2
i
= 2yx2 ey − 3y 2 + 2xey dxdy
R ∂x ∂y
ZZ h i
2 2
= 4xyey − 6y − 4xyey dxdy
R
Z 3 Z 1 Z 3
1
= −6 dx dy y = −6 dx
1 0 1 2
= −6
4.3.1.9. ∗. Solution. (a) The curves y = x2 + 4x + 4 and y = 4 − x2
meet when
x2 + 4x + 4 = 4 − x2 ⇐⇒ 2x2 + 4x = 2x(x + 2) = 0
So the curves intersect at (0, 4) and (−2, 0). Here is a sketch.
y y = x2 + 4x + 4
C y = 4 − x2
R
C
(−2, 0) x
(b) Let
(x, y) ∈ R2 x2 + 4x + 4 ≤ y ≤ 4 − x2 , −2 ≤ x ≤ 0

R=
By Green’s theorem (Theorem 4.3.2)
I ZZ n o
y 2 ∂ y 2 ∂
xy dx + (e + x )dy = ∂x (e + x ) − ∂y (xy) dxdy
C R
Z 0 Z 4−x2
= dx dy x
−2 x2 +4x+4
Z 0
= dx (−2x2 − 4x)x
−2
4 0
x 4x3
= − −
2 3 −2
8
=−
3
4.3.1.10. ∗. Solution. The integral that would be used for direct eval-
uation looks very complicated. So let’s try Green’s theorem. The curve C
is the boundary of the triangle
y
(1, 2)
y=2x

T = (x, y) 0 ≤ x ≤ 1, 0 ≤ y ≤ 2x
(0, 0) (1, 0) x
So
Z Z
2 2
y 2 − e−y + sin x dx + 2xye−y + x dy

F · dr =
C
ZCZ
2 2
∂
2xye−y + x − ∂
y 2 − e−y + sin x

= ∂x ∂y dxdy
Z ZT
2 2
2ye−y + 1 − 2y + 2ye−y

= dxdy
T
Z 1 Z 2x
= dx dy 1 − 2y
0 0
Z 1
dx 2x − 4x2

=
0
4 1
=1− =−
3 3
4.3.1.11. ∗. Solution. Here is a sketch of the two curves in question.
y
y=3−x2 +2x
(0, 3)
R
(3, 0)
x
y=x2 −4x+3
Note that the curves y = x2 − 4x + 3 and y = 3 − x2 + 2x intersect

when x2 − 4x + 3 = 3 − x2 + 2x or 2x2 − 6x = 2x(x − 3) = 0 or x = 0, 3.
The integrand for direct evaluation looks complicated. So let’s use
√
Green’s theorem with F1 (x, y) = 2xey + 2 + x2 , F2 (x, y) = x2 (2 + ey )
and
R = (x, y) x2 − 4x + 3 ≤ y ≤ 3 − x2 + 2x, 0 ≤ x ≤ 3

By Green’s theorem, which is Theorem 4.3.2,

Z ZZ
y
p
2
2 y ∂F2 ∂F1
2xe + 2 + x dx + x (2 + e ) dy = − dxdy
C ∂x ∂y
Z ZR
= {2x(2 + ey ) − 2xey } dxdy
R
Z 3 Z 3−x2 +2x
=4 dx dy x
0 x2 −4x+3
Z 3
=4 dx (6x − 2x2 )x
0
3
3 1 4
= 4 2x − x
2 0
= 54
4.3.1.12. ∗. Solution. Direct evaluation will lead to three integrals,
one for each side of the triangle. The integral from (0, 0) and (1, −2) and
the integral from (1, 2) to (0, 0) will each contain six (nonconstant) terms.
This does not look very efficient. So let’s try Green’s theorem. Denote by
T , the triangle
y (1, 2)
y=2x

T = (x, y) 0 ≤ x ≤ 1, −2x ≤ y ≤
2x (0, 0) x
y=−2x
(1, −2)
It has boundary ∂T = C, oriented counterclockwise as desired. So, by
Green’s theorem,
Z Z
−y
+ sin x dx + 21 x2 + x − xe−y dy
3 2
F · dr = 2y + e
C
Z∂T
Z
−y −y
∂ 1 2 ∂ 3 2

= ∂x 2 x + x − xe − ∂y 2y + e + sin x dxdy
Z ZT
x + 1 − e−y − 3y − e−y dxdy

=
Z ZT

= x − 3y + 1 dxdy
T
Now
ZZ
1
dxdy = Area(T ) = (4)(1) = 2
T 2
ZZ
y dxdy = 0 since y is odd under y → −y
T
ZZ Z 1 Z 2x Z 1
4
x dxdy = dx dy x = 4x2 dx =
T 0 −2x 0 3
So Z
4 10
F · dr = −3×0+2=
C 3 3
4.3.1.13. ∗. Solution. Set
−y x
F= ı̂ı + 2 ̂
x2
+y 2 x + y2
(a) Green’s theorem must be applied to a curve that is closed, so that
it is the boundary of a region in R2 . The given curve C is not closed. But
it is part of the boundary of
2
R = (x, y) − 2 ≤ x ≤ 2, x4 + 1 ≤ y ≤ 2

Here is a sketch of R.
y
L
(−2, 2) (2, 2)
R
x2
C y= 4
+1
x
The boundary of R consists of two parts — C on the bottom and the
line segment L from (2, 2) to (−2, 2) on the top. Note that F is well-defined
on all of R and that
∂ ∂ ∂ x ∂ y
F2 − F1 = 2 2
+
∂x ∂y ∂x x + y ∂y x + y 2
2
(x + y ) − x(2x) (x2 + y 2 ) − y(2y)

2 2
= 2 + 2
(x2 + y 2 ) (x2 + y 2 )
=0
on all of R. So, by Green’s theorem (Theorem 4.3.2),

−y
Z ZZ Z
x ∂ ∂
2 2
dx + 2 dy = F2 − F1 dxdy − F · dr
C x +y x + y2 R ∂x ∂y L
Z Z 2
= F · dr = F1 dx
−L −2
Z 2
2
=− dx since y = 2 on L
−2 x2 + 4
Z 1
4
=− 2+4
dx with x = 2u, dx = 2du
−1 4u
1 π
= − arctan u = −

−1 2
In the second line, we used the notation −L for the line segment from
(−2, 2) to (2, 2).
(b) This question looks a lot like that of part (a). But there is a critical
difference. Again C is not closed and again it is part of the boundary of a
simple region in the xy-plane, namely
R = (x, y) − 2 ≤ x ≤ 2, x2 − 2 ≤ y ≤ 2

This R is sketched below.

y
(−2, 2) L (2, 2)
y = x2 − 2
C
We cannot continue as in part (a), using this R, because ∂x

∂ ∂
F2 − ∂y F1
is not zero througout R. In fact, it is not even defined throughout R —
it is not defined at (0, 0), which is a point of R. We can work around this
obstruction by
• choosing a number ρ > 0 that is small enough that the circle Cρ
parametrized by
r(θ) = ρ cos θ ı̂ı + ρ sin θ ̂ 0 ≤ θ ≤ 2π
is completely contained inside R (ror example, ρ = 1 is fine)

• and then removing from R the interior of Cρ .
This produces the “deformed washer”
W = (x, y) − 2 ≤ x ≤ 2, x2 − 2 ≤ y ≤ 2, x2 + y 2 ≥ ρ2

that is sketched below.

y
(−2, 2) L (2, 2)
W
Cρ
y = x2 − 2
C
The boundary of W consists the three parts — the curve of interest

C on the bottom, the line segment L from (2, 2) to (−2, 2) on the top,
and the circle −Cρ (that is Cρ but oriented clockwise, rather than counter-
∂ ∂
clockwise) around the hole in the middle. Now ∂x F2 − ∂y F1 is well-defined
and zero throughout W . So, by Green’s theorem (Theorem 4.3.2),
−y
Z
x
2 + y2
dx + 2 dy
C x x + y2
ZZ Z Z
∂ ∂
= F2 − F1 dxdy − F · dr − F · dr
W ∂x ∂y L −Cρ
Z Z
= F · dr + F · dr
−L Cρ
F · dr = − π2 . So it remains
R
We have already found, in part (a), that −L
only to use
r(θ) = ρ cos θ ı̂ı + ρ sin θ ̂

r0 (θ) = −ρ sin θ ı̂ı + ρ cos θ ̂
to evaluate
Z Z 2π
F r(θ) · r0 (θ) dθ

F · dr =
Cρ 0
F(r(θ))
r0 (θ)
Z 2π z 1 }|
1
{
z }| {
= − sin θ ı̂ı + cos θ ̂ · −ρ sin θ ı̂ı + ρ cos θ ̂ dθ
0 ρ ρ
Z 2π
= dθ
0
= 2π
All together
−y
Z Z Z
x π 3π
2 + y2
dx + 2 2
dy = F · dr + F · dr = − + 2π =
C x x + y −L Cρ 2 2
(c) No, F is not conservative. We found, in parts (a) and (b), two
different values for the integrals along two paths, both of which start at
(−2, 2) and end at (2, 2). So F does not have the “path independence”
property of Theorem 2.4.7.c and cannot be conservative.
R
4.3.1.14. ∗. Solution. The given integral is of the form C F · dr with
√
F = 2xey + 2 + x2 ı̂ı + x2 2 + ey ) ̂

If we were to try to evaluate this integral directly, then on the y = x2 −4x+3

2
part of C, the integrand would contain x2 ey = x2 ex −4x+3 . That looks hard
to integrate, so let’s try Green’s theorem. The parabolas y = x2 − 4x + 3
and y = 3 − x2 + 2x intersect at (x, y) with
x2 − 4x + 3 = 3 − x2 + 2x ⇐⇒ 2x2 − 6x = 0 ⇐⇒ 2x(x − 3) = 0
⇐⇒ x = 0 or x = 3
The curve C is the boundary of
R = (x, y) 0 ≤ x ≤ 3, x2 − 4x + 3 ≤ y ≤ 3 − x2 + 2x

It is sketched below.
y C
(0, 3)
R y = x2 − 4x + 3
(3, 0)
x
C
y = 3 − x2 + 2x
By Green’s theorem (Theorem 4.3.2),
Z √
2xey + 2 + x2 dx + x2 2 + ey )dy

C
ZZ h
∂ 2 ∂ √ i
x 2 + ey ) − 2xey + 2 + x2 dxdy

=
∂x ∂y
Z ZR
2x(2 + ey ) − 2xey dxdy

=
R
Z 3 Z 3−x2 +2x
=4 dx dy x
0 x2 −4x+3
Z 3
dx x (3 − x2 + 2x) − (x2 − 4x + 3)

=4
0
Z 3
1
dx 6x2 − 2x3 = 4 2 × 33 − 34 = 54

=4
0 2
4.3.1.15. ∗. Solution. (a) Denote by
R2 = (x, y) (x − 2)2 + y 2 ≤ 1

the interior of the circle C2 . Note that (0, 0) is not in R2 . Consequently,

Qx − Py = 0 everywhere in R2 and, by Green’s theorem (Theorem 4.3.2),
Z ZZ

I2 = F · dr = Qx − Py dxdy = 0
C2 R2
y y
C3
C2 −C1
(2,0) (2, 0)
x x
R2
R3
R
(b) We cannot blindly apply Green’s theorem to I3 = C3 F · dr because
(0, 0) is in the interior of C3 , so that Qx − Py is not identically zero in the
interior of C3 — it is not even defined throughout the interior of C3 . We
can work around this obstruction by considering the interior of C3 with
the interior of C1 removed. That is, by considering
R3 = (x, y) x2 + (y − 2)2 ≤ 9, x2 + y 2 ≥ 1

It is sketched on the right above. The boundary of R3 consists of two parts

• the circle C3 , oriented counterclockwise, and
• the circle −C1 . That is, the circle C1 but oriented clockwise, rather
than counterclockwise.
Then Qx − Py is well-defined and zero throughout R3 and, by Green’s

theorem,
ZZ Z Z

0= Qx − Py dxdy = F · dr + F · dr
R3 C3 −C1
Z Z
= F · dr − F · dr
C3 C1
Z
= F · dr − π
C3
R
So C3
F · dr = π.
R
(c) Again, we cannot blindly apply Green’s theorem to I4 = C4 F · dr
because (0, 0) is in the interior of C4 . This time we cannot remove the
interior of C1 from the interior of C4 , because C1 is not contained in the
interior of C4 . Instead we pick a number ρ > 0 which is small enough that
the positively oriented circle
Cρ = (x, y) x2 + y 2 = ρ2

is completely inside C4 . Then we can define
R4 = (x, y) (x − 2)2 + (y − 2)2 ≤ 9, x2 + y 2 ≥ ρ2

It is sketched on the left below. We can now argue as in part (b). The
boundary of R4 consists of two parts
• the circle C4 , oriented counterclockwise, and
• the circle −Cρ . That is, the circle Cρ but oriented clockwise, rather
Then Qx − Py is well-defined and zero throughout R4 and, by Green’s

theorem,
ZZ

0= Qx − Py dxdy
R
Z 4 Z
= F · dr + F · dr
C4 −Cρ
Z Z
= F · dr − F · dr
C4 Cρ
R R
So F · dr = Cρ F · dr.
C4 R
To complete our computation, we have to determine Cρ F · dr. We
can do so by repeating the same “removing a small disk containing (0, 0)”
argument for the third time. Set
R5 = (x, y) x2 + y 2 ≤ 1, x2 + y 2 ≥ ρ2

Then the boundary of R5 consists of C1 and −Cρ , and, as Qx − Py is

well-defined and zero throughout R5 ,
ZZ

0= Qx − Py dxdy
Z R5 Z
= F · dr + F · dr
C1 −Cρ
Z
=π− F · dr
Cρ
R R
So C4
F · dr = Cρ
F · dr = π.
y y
C1
−Cρ
(2, 2) x
−Cρ C4 R5
x
R4
4.3.1.16. ∗. Solution. (a) If (x, y) 6= (0, 0), we have

∂ y−x ∂ x+y
Qx − Py = −
∂x x2 + y 2 ∂y x2 + y 2
2 2
−(x + y ) − (y − x)(2x) (x2 + y 2 ) − (x + y)(2y)
= 2 − 2
(x2 + y 2 ) (x2 + y 2 )
=0
(b) Parametrize CR by
r(θ) = R cos θ ı̂ı + R sin θ ̂ 0 ≤ θ ≤ 2π
So
F(r(θ))
Z Z 2π
z }| {
1
F · dr = cos θ + sin θ ı̂ı + sin θ − cos θ ̂
CR 0 R
r0 (θ)
z }| {
· −R sin θ ı̂ı + R cos θ ̂ dθ
Z 2π
= (−1) dθ
0
= −2π
R
(c) If F were conservative, the line integral C F · dr would be 0 for any
closed curve C, by Theorem 2.4.7.b. So F is not conservative. Note that
F is not defined at (x, y) = (0, 0) and so fails the screening test ∇ × F = 0
at (x, y) = (0, 0).
(d) Denote by R the interior of the triangle C. It is the grey region in
the figure
y
(0, 1) (1, 1)
R
C
(1, 0) x
Note that (0, 0) is not in R. So Qx − Py is defined and zero throughout
R. So, by Green’s theorem (Theorem 4.3.2),
Z ZZ

F · dr = Qx − Py dxdy = 0
C R
(e) Note that (0, 0) is in the interior of triangle C specified for this
part. So Qx − Py is not defined in that interior and we cannot apply
Green’s theorem precisely as we did in part (d). We can work around this
obstruction by
• picking a number r > 0 that is small enough that the circle Cr , of
radius r centred on (0, 0), is completely contained in the interior of
the triangle C.
• Then we work with the region R defined by removing the interior of
the circle Cr from the interior of the triangle C. It is the grey region
sketched below.
y
(0, 1)
R C
−Cr
(0, 1)
x
(1, 1)
The boundary of R consists of two parts
• the triangle C, oriented counterclockwise, and

• the circle −Cr . That is, the circle Cr , but oriented clockwise, rather
Then Qx − Py is well-defined and zero throughout R and, by Green’s
theorem,
ZZ

0= Qx − Py dxdy
Z R Z
= F · dr + F · dr
ZC Z−Cr
= F · dr − F · dr
C Cr
R R R
So C
F · dr = Cr F · dr. By part (b), with R = r, Cr
F · dr = −2π, so
R
C
F · dr = −2π
R
4.3.1.17. ∗. Solution. (a) The given integral is of the form C
F1 (x, y) dx+
F2 (x, y) dy with
p ∂F2 ∂F1
F1 (x, y) = 1 + x3 F2 (x, y) = 2xy 2 + y 2 − = 2y 2
∂x ∂y
As C is ∂R with
(x, y) x2 + y 2 ≤ 1

R=
Green’s theorem (Theorem 4.3.2) gives

Z p Z
1 + x3 dx + 2xy 2 + y 2 dy =

F1 (x, y) dx + F2 (x, y) dy
C
ZCZ
∂F2 ∂F1
= − dxdy
∂x ∂y
ZRZ
=2 y 2 dxdy
R

Z p Z 2π Z 1
2 2
2
3
1 + x dx + 2xy + y dy = 2 dθ dr r r sin θ
C 0 0
Z 2π Z 1
2 3
=2 dθ sin θ dr r
0 0
1 π
=2π =
4 2
To do the θ integral, we have used
Z 2π Z 2π
1 − cos(2θ) h θ − sin(2θ)/2 i2π
sin2 θ dθ = dθ = =π
0 0 2 2 0
R 2π
For an efficient, sneaky, way to evaluate 0 sin2 θ dθ, see Example 2.4.4.
(b) It is again natural to use Green’s theorem. But Green’s theorem
must be applied to a curve that is closed, so that it is the boundary of
a region in R2 . The given curve C is not closed. But it is part of the
boundary of
R = (x, y) x2 + y 2 ≤ 1, x ≥ 0

Here is a sketch of R.
y
(0, 1)
C
L R
x
x2 + y 2 = 1
(0, −1)
The boundary of R consists of two parts — C on the right and the line
segment L from (0, 1) to (0, −1) on the left. Note that F = F1 ı̂ı + F2 ̂ is
well-defined on all of R and that we still have, from part (a),
∂F2 ∂F1
− = 2y 2
∂x ∂y
on all of R. So, by Green’s theorem (Theorem 4.3.2),
Z p ZZ Z
∂F2 ∂F2
1 + x3 dx + 2xy 2 + y 2 dy =

− dxdy − F · dr
C ∂x ∂y
Z ZR Z L
= 2 2
2y 2 dxdy + F2 dy
x +y ≤1
x≥0
−L
ZZ Z 1
2
= y dxdy + y 2 dy
x2 +y 2 ≤1 −1
by symmetry for the first integral and
since x = 0 and dx = 0 in the
second integral
π 2
= +
4 3
In the second line, we used the notation −L for the line segment from
(0, −1) to (0, 1).
4.3.1.18. ∗. Solution. First, here is a sketch of the curve C.

y
x = cos y
x
C
We’ll evaluate this integral in three different ways.

(1) Direct evaluation: To evaluate the integral directly, we’ll parametrize
C using y as the parameter. That is, we’ll make y(t) = t:
π π
r(t) = x(t)ı̂ı + y(t) ̂ = cos tı̂ı + t ̂ − ≤t≤
2 2
r0 (t) = x0 (t)ı̂ı + y 0 (t) ̂ = − sin tı̂ı + ̂
So the integral is
Z
C
Z π/2
n dx
= x(t)2 + y(t)ex(t) (t)
−π/2 dt
dy o
+ x(t) cos(y(t)) + ex(t)

(t) dt
dt
Z π/2 n
o
− cos2 t + tecos t sin t + cos2 t + ecos t dt

=
−π/2
Z π/2 n o
= − cos2 t sin t + cos2 t − tecos t sin t + ecos t dt
−π/2
π/2
d h cos3 t
Z n io
= cos2 t + + tecos t dt
−π/2 dt 3
Z π/2 n
cos(2t) + 1 d h cos3 t io
= + + tecos t dt
−π/2 2 dt 3
3 iπ/2
h sin(2t) t cos t
= + + + tecos t
4 2 3 −π/2
3π
=
2
R π/2
For an efficient, sneaky, way to evaluate −π/2
cos2 t dt see Example
2.4.4.
(2) Green’s (or Stokes’) theorem: The curve C is not closed so we cannot
apply Green’s theorem directly. However the boundary of the region
π π
R = (x, y) 0 ≤ x ≤ cos y, − ≤ y ≤
2 2
(sketched below) consists
of two parts, one of which is C. The other
is the line L from 0, π2 to 0, − π2 .

x = cos y
L
R
x
C
So Green’s theorem gives

Z
C
ZZ n
∂ ∂ 2 o
= (x cos y + ex ) − (x + yex ) dxdy
R ∂x ∂y
Z
− (x2 + yex ) dx + (x cos y + ex ) dy
ZZ L Z
= cos y dxdy − dy since x = 0 and dx = 0 on L
R L
Z π/2 Z cos y Z −π/2
= dy dx cos y − dy
−π/2 0 π/2
Z π/2
= dy cos2 y + π
−π/2
Z π/2
cos(2y) + 1
= dy + π
−π/2 2
h sin(2y) y iπ/2
= + +π
4 2 −π/2
3π
=
2
(3) (Sort of) conservative fields: The given integral is C F · dr with

R
F = (x2 + yex )ı̂ı + (x cos y + ex ) ̂. The curl of this field is

 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂z  = cos y k̂

∂y
x2 + yex x cos y + ex 0
So F violates our screening test and consequently is not conservative.

But it violates the screening test only because of the term x cos y ̂.
This suggests that we split up
F=G+H with G = (x2 + yex )ı̂ı + ex ̂, H = x cos y ̂

3
Then G is conservative with potential g R= x3 + yex and H is pretty
simple, so that it is not hard to evaluate C H · dr directly. Using the
parametrization r(t) = cos tı̂ı + t ̂, − π2 ≤ t ≤ π2 as above,
Z Z Z
F · dr = G · dr + H · dr
C C C
Z Z
= ∇ g · dr + H · dr
C C
Z π/2
π π dy
= g r( ) − g r(− ) + x(t) cos(y(t)) (t) dt
2 2 −π/2 dt
Z π/2
π π
= g 0, − g 0, − + cos2 t dt
2 2 −π/2
Z π/2
π π cos(2t) + 1
= − − + dt
2 2 −π/2 2
h sin(2t) t iπ/2
=π+ +
4 2 −π/2
3π
=
2
4.3.1.19. ∗. Solution. Call the region enclosed by the curve R. By
Green’s theorem, Theorem 4.3.2,
I ZZ ZZ
1 1 ∂ ∂ 1
x dy − y dx = x− (−y) dxdy = 2 dxdy
2 C 2 R ∂x ∂y 2 R
=A
as desired. The curve x2/3 + y 2/3 = 1 may be parametrized in the coun-

terclockwise orientation by x(θ) = cos3 θ, y(θ) = sin3 θ, 0 ≤ θ ≤ 2π. Then
I
1
A= x dy − y dx
2 C
Z 2π
1
x(θ)y 0 (θ) − y(θ)x0 (θ) dθ

=
2 0
1 2π
Z
3 cos4 θ sin2 θ + 3 sin4 θ cos2 θ dθ

=
2 0
3 2π 2 3 2π 2
Z Z
= sin θ cos2 θ dθ = sin (2θ) dθ
2 0 8 0
Z 2π i2π
3 3h 1 3π
= 1 − cos(4θ) dθ = θ − sin(4θ) =
16 0 16 4 0 8
4.3.1.20. ∗. Solution. If we use D to denote the disk inside the circle
C then we want
I I I
F · dr − A G · dr = (F − AG) · dr
C C
ZCZ h
∂ ∂ i
= (F − AG)2 − (F − AG)1 dxdy
D ∂x ∂y
to vanish for all disks D. We used Green’s theorem, which is Theorem
4.3.2, in the last step. This is the case if and only if
∂ ∂
(F − AG)2 = (F − AG)1
∂x ∂y
∂ ∂
⇐⇒ [(x + y) − A(2x − 3y)] = [(x + 3y) − A(x + y)]
∂x ∂y
⇐⇒ 1 − 2A = 3 − A
⇐⇒ A = −2
4.3.1.21. ∗. Solution. (a) Parametrize the circle r(θ) = (cos θ, sin θ).
Then
F r(θ) = sin3 θ ı̂ı − cos θ sin2 θ ̂

dr
(θ) = − sin θ ı̂ı + cos θ ̂
dθ
dr
F r(θ) · (θ) = − sin4 θ − cos2 θ sin2 θ = − sin2 θ
dθ
I Z 2π Z 2π
dr
F · dr = F r(θ) · (θ) dθ = − sin2 θ dθ
C 0 dθ 0
Z 2π
1 − cos(2θ)
=− dθ
0 2
h θ sin(2θ) i2π
=− − = −π
2 4 0
R 2π
For an efficient, sneaky, way to evaluate 0 sin2 θ dθ see Example 2.4.4.
(b) Denote by W the washer shaped region between the circle x2 +y 2 =
2 2
1 and the ellipse x16 + y25 = 1. It is sketched below. By Green’s theorem
I I ZZ h
∂ ∂ i
F · dr − F · dr = F2 − F1 dxdy
C0 C W ∂x ∂y
For the specified F
∂ ∂ ∂ xy 2 ∂ y3
F2 − F1 = − −
∂x ∂y ∂x (x2 + y 2 )2 ∂y (x2 + y 2 )2
y2 xy 2 (2x) 3y 2 y 3 (2y)
=− 2 +2 3 − 2 +2 3
(x2 + y 2 ) (x2 + y 2 ) (x2 + y 2 ) (x2 + y 2 )
−y 2 (x2 + y 2 ) + 4x2 y 2 − 3y 2 (x2 + y 2 ) + 4y 4
= 3
(x2 + y 2 )
=0
Consequently
I I I I
F · dr − F · dr = 0 =⇒ F · dr = F · dr = −π
C0 C C0 C
W
C C0
4.3.1.22. ∗. Solution. Observe that

∂F2 ∂F1 ∂ x ∂ −y
− = −
∂x ∂y ∂x x2 + y 2 ∂y x2 + y 2
2 2
(x + y ) − x(2x) (x2 + y 2 ) − y(2y)
= 2 + 2 =0
(x2 + y 2 ) (x2 + y 2 )
except at (0,H 0), where F is not defined. Hence by Green’s theorem (Theo-
rem 4.3.2), C F · dr = 0 forH any closed curve that does not contain (0, 0) in
its interior. In particular, C1 F · dr = 0. On the other hand, (0, 0) is con-
tainedH in the interior of C2 , so we cannot use Green’s theorem to conclude
that C2 F · dr = 0.
Let C3 be the circle of radius one centred on (0, 0) and denote by W the
washer shaped region between the circle C2 and the circle C3 . It is sketched
below.
W
C3 C2
By Green’s theorem (Theorem 4.3.2),

I I ZZ h
∂ ∂ i
F · dr − F · dr = F2 − F1 dxdy = 0
C2 C3 W ∂x ∂y
H H
So C2 F · dr = C3 F · dr. Parameterize C3 by x = cos θ, y = sin θ. Then
r(θ) = cos θ ı̂ı + sin θ ̂ 0 ≤ θ ≤ 2π

r0 (θ) = − sin θ ı̂ı + cos θ ̂

F r(θ) = − sin θ ı̂ı + cos θ ̂
F r(θ) · r0 (θ) = 1

so that I I Z 2π
F · dr = F · dr = dθ 1 = 2π
C2 C3 0
4.3.1.23. ∗. Solution. (a) Let C1 be the line segment from (0, 1) to

(0, 0), C2 be the line segment from (0, 0) to (1, 0) and C3 be the curve
y = 1 − x2 from (1, 0) to (0, 1).
y y = 1 − x2
C1 C3
x
C2
Then
Z Z
x ds = 0 ds = 0
C1 C1
Z Z 1
1
x ds = x dx =
C2 0 2
dy
On C3 , y = 1 − x2 so that dx = −2x and
r dy 2
p p
ds = dx2 + dy 2 = 1+ dx = 1 + 4x2 dx
dx
and
Z Z 1 p
x ds = x 1 + 4x2 dx
C3 0
h1 i1 1 3/2
(1 + 4x2 )3/2 =

= 5 −1
12 0 12
All together Z
1 1 3/2
x ds = + 5 − 1 ≈ 1.3484
C 2 12
(b) By either Stokes’ theorem or Green’s theorem
Z ZZ h
∂ 2 ∂ i
x + cos(y 2 ) − sin(x2 ) − xy dxdy

F · dr =
C ∂x ∂y
Z ZR
= 3x dxdy
R
Z 1 Z 1−x2
=3 dx dy x
0 0
Z 1
=3 dx (1 − x2 )x
0
h1 1i 3
=3 − =
2 4 4
4.3.1.24. ∗. Solution. (a) If (x, y, z) is on the curve, it must obey both
z = x + y and z = x2 + y 2 and hence it must also obey x2 + y 2 = x + y or
(x − 21 )2 + (y − 12 )2 = 12 . That’s a circle. We can parametrize the curve by
1 1
x(θ) = + √ cos θ
2 2
1 1
y(θ) = + √ sin θ
2 2
1
z(θ) = x + y = 1 + √ cos θ + sin θ
2

with 0 ≤ θ < 2π. As θ runs from 0 to 2π, x(θ), y(θ) runs once around the
circle without crossing itself so that x(θ), y(θ), z(θ) runs once around the
curve without crossing itself. As x(2π), y(2π), z(2π) = x(0), y(0), z(0) ,
C is a simple closed curve.
(b) (i) The vector field F =H x2 ı̂ı + y 2 ̂ + 3ez k̂ is conservative (with
potential 31 x3 + 13 y 3 + 3ez ). So C F · dr = 0.
(b) (ii) Note that the question did not specify the orientation of C.
It should have. We’ll stick with the most commonly used orientation —
counterclockwise when viewed from high on the z-axis. The vector field
G = 3ez k̂ is conservative (with potential 3ez ). So C G · dr = 0 and, using
H
the parametrization
h1 1 i h1 1 i h 1 1 i
r(θ) = + √ cos θ ı̂ı + + √ sin θ ̂ + 1+ √ sin θ+ √ cos θ k̂
2 2 2 2 2 2
1 1 h 1 1 i
r0 (θ) = − √ sin θ ı̂ı + √ cos θ ̂ + √ cos θ − √ sin θ k̂
2 2 2 2
of part (a), we have
I I
F · dr = (F − G) · dr
C C
Z 2π
y(θ)2 x0 (θ) + x(θ)2 y 0 (θ) dθ

=
0
Z 2π h1 i2 1 i2 1
n 1 h1 1 o
= − + √ sin θ √ sin θ + + √ cos θ √ cos θ dθ
0 2 2 2 2 2 2
Because the integral of any odd power of sin θ or cos θ over 0 ≤ θ ≤ 2π is
zero (see Example 4.4.6 in the text),
I Z 2π n 1 1 o
F · dr = − sin2 θ + cos2 θ dθ
c 0 2 2
=0
since (see Example 2.4.4 in the text)

Z 2π Z 2π
2
cos θ dθ = sin2 θ dθ = π
0 0
4.3.1.25. Solution. By Green’s Theorem

I ZZ h
3 3 ∂ ∂ 3 i
(y − y) dx − 2x dy = (−2x3 ) − (y − y) dx dy
C ∂x ∂y
Z ZR h i
= 1 − 6x2 − 3y 2 dx dy
R
where R is the region in the xy-plane whose boundary is C. Observe

that the integrand 1 − 6x2 − 3y 2 is positive in the elliptical region
RR 6x
2
+
2
3y ≤ 1 and negative outside of it. To maximize the integral R 1 −
6x2 − 3y 2 dx dy we should choose R to contain all points (x, y) with the
integrand 1 − 6x2 − 3y 2 ≥ 0 and to exclude all points (x, y) with the
integrand 1 − 6x2 − 3y 2 < 0. So we choose
R = (x, y) 6x2 + 3y 2 ≤ 1

The corresponding C is 6x2 + 3y 2 = 1.

4.4.3 · Exercises
4.4.3.1. Solution. One approach is to first consider
S
∂S
The correct normal to this surface is sketched in
n̂
S
∂S
It is correct because
• if you walk along ∂S in the direction of the arrow on ∂S,
• with the vector from your feet to your head having direction n̂
• then S is on your left hand side.

Now pretend that the surface S is made of rubber and that n̂ is glued to
S. We can push on this S to deform it to the S of part (a) or to the S of
part (b). This gives the solutions to parts (a) and (b).
(a) n̂ (b)
∂S
S n̂ S
∂S
To deal with part (c), we can first rotate the flat disk that we considered
above to get
n̂
∂S
S
We can push on this S to deform it to the S of part (c). This gives the
solution to part (c).
(c)
∂S
n̂
S
4.4.3.2. Solution. Think of the xy-plane as being the plane z = 0 in

R3 .
z
n̂
y
C R
x
We are going to apply Stokes’ theorem (Theorem 4.4.1) with S being the
given region R in the xy-plane and with F(x, y, z) = F1 (x, y)ı̂ı + F2 (x, y) ̂.
Then
• the unit normal vector to S specified in Stokes theorem is k̂ (if you
walk along ∂S = C in the direction of the arrow on C with the vector
from your feet to your head having direction k̂ then S = R is on your
left hand side) and
• dS = dx dy and
• the curl of F is
 
ı̂ı ̂ k̂ ∂F
 ∂ ∂ ∂  2 ∂F1
∇ × F = det  ∂x ∂y ∂z  = − k̂
∂x ∂y
F1 (x, y) F2 (x, y) 0

I I ZZ

F1 (x, y) dx + F2 (x, y) dy = F · dr = ∇ × F · n̂ dS
C
Z∂S
Z S
∂F2 ∂F1
= − dxdy
R ∂x ∂y
H
4.4.3.3. Solution. We are to show that C [φ∇ ∇ψ+ψ∇ ∇φ]·dr = 0. Suppose
that C = ∂S. Then, by Stokes’ theorem
I ZZ
∇ψ + ψ∇
[φ∇ ∇φ] · dr = ∇ × [φ∇ ∇ψ + ψ∇∇φ] · n̂ dS
C S
HWe will show below that ∇ × [φ∇ ∇ψ + ψ∇ ∇φ] = 0. This will imply that
C
∇
[φ∇ ψ + ∇
ψ∇ φ] · dr = 0. One way to see ∇ψ + ψ∇
that ∇ × [φ∇ ∇φ] = 0 is
∇ψ + ψ∇
∇ × [φ∇ ∇φ] = ∇ × [∇
∇(φψ)] (by part (c) of Theorem 4.1.3)
=0 (by part (b) of Theorem 4.1.7)
∇ψ + ψ∇
Another way to see that ∇ × [φ∇ ∇φ] = 0 is
∇ψ + ψ∇
∇ ×[φ∇ ∇φ] = ∇ φ×∇
∇ψ + φ∇
∇ ×(∇
∇ψ) + ∇ ψ×∇
∇φ + ψ∇
∇ ×(∇
∇φ)
= ∇φ × ∇ψ + ∇ψ × ∇φ
∇ × (∇
since φ∇ ∇ψ) = ψ∇
∇ × (∇
∇φ) = 0
=0
4.4.3.4. Solution. (a) Observe that x(t) = cos t and y(t) = sin t obey
x(t)2 + y(t)2 = 1. Then z(t) = y(t)2 = sin2 t. So we may parametrize the
curve by r(t) = (cos t, sin t, sin2 t) with 0 ≤ t ≤ 2π. Then
r0 (t) = (− sin t , cos t , 2 sin t cos t)

F r(t) = cos2 t − sin t , sin2 t + cos t , 1

F r(t) · r0 (t) = − sin t cos2 t + sin2 t + sin2 t cos t + cos2 t + 2 sin t cos t

1 d
=1+ [cos3 t + sin3 t] + sin(2t)
3 dt
I Z 2π
1 d 3 3
F · dr = 1+ [cos t + sin t] + sin(2t) dt
C 0 3 dt
2π
1 3 3 1
= t + [cos t + sin t] − cos(2t)
3 2 0
= 2π
(b) Let S be the surface z = f (x, y) with f (x, y) = y 2 and x2 + y 2 ≤ 1.

Since C is oriented counter clockwise when viewed from high on the z-axis,
Stokes’ theorem requires that we use the normal n̂ to S with positive z
component. Hence
h ∂f ∂f i h i
n̂ dS = − ı̂ı − ̂ + k̂ dx dy = − 2y ̂ + k̂ dx dy
∂x ∂y
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = 2k̂
x2 − y y 2 + x 1
∇ × F · n̂ dS = 2 dx dy
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = 2 dx dy
C S x2 +y 2 ≤1
= 2π
4.4.3.5. Solution. We apply Stokes’ theorem. First,
 
ı̂ı ̂ k̂
 ∂ ∂ ∂  x x

∇ × F = det  ∂x ∂y ∂z  = 1 + e − e k̂ = k̂
x x 2
ye x+e z
Note that r(t) = x(t)ı̂ı +y(t) ̂ +z(t) k̂ obeys x(t)+y(t)+z(t) = 3, for every
t, and that x(t)ı̂ı + y(t) ̂ = (1 + cos t)ı̂ı + (1 + sin t) ̂ runs counterclockwise
around the circle of radius 1 centered on (1, 1). So we choose S to be the
part of the plane G(x, y, z) = x + y + z = 3 with (x − 1)2 + (y − 1)2 ≤ 1.
Then, by Stokes’ Theorem,
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = k̂ · n̂ dS
C S S
with
∇G
n̂ dS = ± dxdy = ± ı̂ı + ̂ + k̂ dxdy
∇ G · k̂
As (1 + cos t)ı̂ı + (1 + sin t) ̂ runs counterclockwise around the circle (x −
1)2 + (y − 1)2 ≤ 1, Stokes’ theorem specifies the plus sign and
I ZZ
F · dr = dx dy = π
C (x−1)2 +(y−1)2 ≤1
4.4.3.6. ∗. Solution. The boundary of S is
∂S = (x, y, z) z = 0, x2 + y 2 = 4

and can be parametrized
r(θ) = 2 cos θ ı̂ı + 2 sin θ ̂ 0 ≤ θ ≤ 2π

n̂
V S
∂S
So, by Stokes’ theorem (Theorem 4.4.1)
ZZ I
∇ × F · n̂ dS = F · dr
S ∂S
F(r(θ)) r (θ) 0
Z 2π z }| { z }| {
= (−2 sin θ ı̂ı + 2 cos θ ̂ − 2 cos θ k̂) · (−2 sin θ ı̂ı + 2 cos θ ̂) dθ
0
Z 2π
=4 dθ
0
= 8π
4.4.3.7. ∗. Solution 1. The boundary of S is the circle x2 + y 2 = 4,

z = 0. Let C be this circle, oriented by the parametrization x(t) = 2 cos t,
y(t) = 2 sin t, z(t) = 0. By Stokes’ theorem
ZZ Z Z 2π
dr
∇ × F · n̂ dS = F · dr = F(2 cos t, 2 sin t, 0) · (t) dt
S C 0 dt
Z 2π
= 0ı̂ı + 2 cos t(3 + 2 sin t) ̂ + 2 sin t k̂ · − 2 sin tı̂ı + 2 cos t̂ dt
0
Z 2π
12 cos2 t + 8 cos2 t sin t dt

=
0
Z 2π
6 + 6 cos(2t) + 8 cos2 t sin t dt

=
0
h 8 i2π
= 6t + 3 sin(2t) −cos3 t = 12π
3 0
R 2π
For an efficient, sneaky, way to evaluate 0 cos2 t dt, see Example 2.4.4.
Solution 2.
• S be the surface specified in the question, with upward pointing nor-
mal, and

• D be the disk (x, y, z) x2 + y 2 ≤ 4, z = 0 , with normal n̂ = k̂,
and

• C be the circle (x, y, z) x2 + y 2 = 4, z = 0 , oriented as in the
figure below.
n̂
n̂ S
D
C
Note that C is the boundary curve for both S and D. So, by Stokes’
theorem, twice
ZZ Z ZZ
∇ × F · n̂ dS = F · dr = ∇ × F · n̂ dS
S
ZCZ D
= ∇ × F · k̂ dS
D
Now the k̂ component of ∇ × F is

∂ ∂
∇ × F · k̂ = x(3 + y) − x ln(1 + z) = 3 + y
∂x ∂y
RR As y is odd under y → −y and D is invariant under y → −y, we have

D
y dS = 0 and
ZZ ZZ ZZ
∇ × F · n̂ dS = (3 + y) dS = 3 dS = 3 Area(S) = 3 π22 = 12π
S D D
4.4.3.8. ∗. Solution. Let S be the portion of the paraboloid z =

f (x, y) = 4 − x2 − y 2 with x2 + (y − 1)2 ≤ 1 and let n̂ be the upward
normal to S. For this surface

n̂ dS = − fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ dxdy = 2xı̂ı + 2y ̂ + k̂ dxdy
by (3.3.2). As (x, y, z) runs over S, (x, y) runs over the circular disk
D = (x, y) x2 + (y − 1)2 ≤ 1

For the given vector field

 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z 
xz x yz
= z ı̂ı + x ̂ + k̂
so that, by Stokes’ theorem (Theorem 4.4.1),

I ZZ
F · dr = ∇ × F · n̂ dS
C S
z=f (x,y)
ZZ
z }| {
2x (4 − x2 − y 2 ) +2xy + 1 dxdy

=
D
By oddness under x → −x, all terms integrate to zero except for the last.
So I ZZ
F · dr = dxdy = Area(D) = π
C D
4.4.3.9. Solution. The surface
S = (x, y, z) − 1 ≤ x ≤ 1, −1 ≤ y ≤ 1, z ≥ 0, z = (1 − x2 )(1 − y 2 )

= (x, y, z) − 1 ≤ x ≤ 1, −1 ≤ y ≤ 1, z = (1 − x2 )(1 − y 2 )

Note that when x = 1 or x = −1 or y = 1 or y = −1, we have z =

(1 − x2 )(1 − y 2 ) = 0. So the boundary of S, call it C, is the boundary
of the square −1 ≤ x, y ≤ 1, z = 0, oriented counterclockwise. Here is a
sketch of C.
dy=0
(−1, 1) y=1 (1, 1)
dx=0 dx=0
x=−1 x=1
(−1, −1) dy=0 (1, −1)

y=−1
Apply Stokes’ theorem. Observing that z = 0 on C so that F = −y ı̂ı +

x3 ̂,
ZZ I I
∇ × F · n̂ dS = F · dr = [−y ı̂ı + x3̂] · dr
S C C
Z 1 Z Z −1
1 Z −1
= −(−1) dx + (1)3 dy + −(1) dx + (−1)3 dy
−1 −1 1
| {z } | {z } | {z } |1 {z }
y=−1 side x=1 side y=1 side x=−1 side
=8
4.4.3.10. Solution. We shall apply Stokes’ Theorem. The curl of F is

ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z
2
ex − yz sin y − yz xz + 2y

= (2 + y)ı̂ı − (z + y) ̂ + (0 + z) k̂
The curve C is a triangle. All three vertices of the triangle

obey x+y +z =
1. So the triangle is the
boundary of the surface S = (x, y, z) x ≥ 0, y ≥
0, z = 1 − x − y ≥ 0 .
z
n̂
x
The equation of the surface is z = f (x, y) = 1 − x − y. So, by (3.3.2),

n̂ dS = − fx ı̂ı − fy ̂ + k̂ dx dy
= (ı̂ı + ̂ + k̂) dx dy
Here n̂ is the upward pointing unit normal. The set (x, y) for
of points
which there is a corresponding (x, y, z) in S is T = (x, y) x ≥ 0, y ≥
0, x + y ≤ 1 , which is a triangle of area 12 . Since
∇ × F · n̂ dS = [(2 + y)ı̂ı − (z + y)̂ + (0 + z)k̂] · (ı̂ı + ̂ + k̂) dx dy

= 2 dx dy
we have
I ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
= 2 dx dy = 2 Area(T ) = 1
T
4.4.3.11.
H ∗. Solution.
RR Stokes’ theorem, which is Theorem 4.4.1, says
that C F · dr = S ∇ × F · n̂ dS for any surface S whose boundary is C.
For the given vector field
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × F(x, y, z) = det  ∂x ∂y ∂z 
−z x y
= ı̂ı − ̂ + k̂
Choose
x2 y2 z2

S= (x, y, z) z = y, 4 + 2 + 2 ≤1
x2 2

= (x, y, z) z = y, 4 +y ≤1
to be the part of the plane z = y bounded by the ellipsoid.

z
(0, 1, 1)
x2 y2 z2
y=z 4
+ 2
+ 2
=1
y
(2, 0, 0)
x
As S is part of the plane z = f (x, y) = y, (3.3.2), gives that

n̂ dS = ± − fx , −fy , 1 dxdy
= ±(0 , −1 , 1)dxdy
As C has the standard orientation (counter-clockwise when viewed from

high on the z-axis), we want n̂ to have a positive z-component. So n̂ dS =
(0 , −1 , 1)dxdy. From the second form of S given above, we see that as
(x, y, z) runs over S, (x, y) runs over
x2
+ y2 ≤ 1

D= (x, y) 4
Consequently, Stokes’ theorem gives that
I Z Z z ∇×F}| { z
n̂ dS
}| {
F · dr = (1, −1, 1) · (0 , −1 , 1)dxdy
C
ZD
Z
=2 dxdy = 2 Area(D)
D
2
The ellipse D, that is x4 + y 2 ≤ 1, has semi-axes a = 2 and b = 1 and
hence area πab = 2π. Finally
I
F · dr = 2 Area(D) = 4π
C
4.4.3.12. ∗. Solution. Note that the curve of part (a) is a simple

closed curve that lies in the plane x + y + z = 2 and is oriented in a
counterclockwise direction as observed from the positive x-axis. The curve

of part (a) encloses a triangle.
z
(0,0,2)
L2
L3
(0,2,0)
y
(2,0,0)
L1
x
Two of the sides of the triangle are (0, 2, 0) − (2, 0, 0) = (−2, 2, 0) and
(0, 0, 2) − (0, 2, 0) = (0, −2, 2) so the area of the triangle is
 
ı̂ı ̂ k̂ √
1 1 1
(−2, 2, 0) × (0, −2, 2) = det −2 2 0  = (4, 4, 4) = 2 3
2 2 2
0 −2 2
So let’s do part (b) first.

(b) We are not told explicitly what C2 is, so we certainly can’t do a
direct evaluation. Instead, let’s use Stokes’ theorem (Theorem 4.4.1). The
curl of F is
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z 
z 2 x2 y 2
= 2y ı̂ı + 2z ̂ + 2x k̂
ı̂ı+̂
+k̂
The upward pointing unit normal to E is n̂ = √
3
. So, by Stokes’
theorem,
Z ZZ
I2 = F · dr = ∇ × F · n̂ dS
C2 R
ZZ
ı̂ı + ̂ + k̂
= 2y ı̂ı + 2z ̂ + 2x k̂ · √ dS
R 3
Z Z z=2 }| on R
2 { 4 √
= √ y + z + x dS = √ Area(R) = 4 3
3 R 3
(a) Denote by T the triangle enclosed by C1 . By the computation that
we have just done in part (b)
Z
4
I1 = F · dr = √ Area(T ) = 8
C1 3
4.4.3.13. ∗. Solution. (a) Observe that
• the curve C1 is one quarter of a circle in the xy-plane, centred on the
origin, of radius 2, starting at (2, 0, 0) and ending at (0, 2, 0) and
• the curve C2 is one quarter of a circle in the yz-plane, centred on the

origin, of radius 2, starting at (0, 2, 0) and ending at (0, 0, 2) and
• the curve C3 is one quarter of a circle in the xz-plane, centred on the

origin, of radius 2, starting at (0, 0, 2) and ending at (2, 0, 0).
Here is a sketch.
z
(0,0,2)
C2
C3
(0,2,0)
(2,0,0) y
x C1
(b) C lies completely on the sphere x2 + y 2 + z 2 = 4. So it is natural
to choose
S = (x, y, z) x2 + y 2 + z 2 = 4, x ≥ 0, y ≥ 0, z ≥ 0

and to parametrize S using spherical coordinates
r(θ, ϕ) = 2 cos θ sin ϕ ı̂ı + 2 sin θ sin ϕ ̂ + 2 cos ϕ k̂

π π
0≤θ≤ , 0≤ϕ≤
2 2
Since
∂r
= −2 sin θ sin ϕı̂ı + 2 cos θ sin ϕ ̂
∂θ
∂r
= 2 cos θ cos ϕı̂ı + 2 sin θ cos ϕ ̂ − 2 sin ϕ k̂
∂ϕ
so that
 
ı̂ı ̂ k̂
∂r ∂r
∂θ ∂ϕ
= −4 cos θ sin2 ϕ ı̂ı − 4 sin θ sin2 ϕ ̂ − 4 sin ϕ cos ϕ k̂
(3.3.1) gives
∂r ∂r
n̂ dS = ± × dθdϕ
∂θ ∂ϕ

= ∓4 cos θ sin ϕ ı̂ı + sin θ sin ϕ ̂ + cos ϕ k̂ sin ϕ dθdϕ
We want n̂ to point outward, for compatibility with the orientation of C.

So we choose the + sign.

n̂ dS = 4 cos θ sin ϕ ı̂ı + sin θ sin ϕ ̂ + cos ϕ k̂ sin ϕ dθdϕ
= 2 r(θ, ϕ) sin ϕ dθdϕ
(c) The vector field F looks too complicated for a direct evaluation of
the line integral. So, in preparation for an application of Stokes’ theorem,
we compute
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
2
y + sin(x2 ) z − 3x + ln(1 + y 2 ) y + ez
= −4 k̂
So, by Stokes’ theorem (Theorem 4.4.1),

Z ZZ
F · dr = ∇ × F · n̂ dS
C S
Z π/2 Z π/2
= dϕ dθ − 4k̂ · cos θ sin ϕ ı̂ı + sin θ sin ϕ ̂ + cos ϕ k̂ 4 sin ϕ
0 0
π/2 π/2 π/2
π cos2 ϕ
Z Z
= −16 dϕ dθ cos ϕ sin ϕ = 16
0 0 2 2 0
= −4π
4.4.3.14. ∗. Solution. (a) The boundary, p ∂S1 , of S1 as specified in
Stokes’ theorem (Theorem 4.4.1) is the circle x2 + y 2 = 4, z = 4 oriented
clockwise when viewed from high on the z-axis. That is, we can parametrize
∂S1 by
r(t) = 4 cos tı̂ı − 4 sin t ̂ + 4 k̂, 0 ≤ t ≤ 2π
So

F r(t) · dr = 16 sin t , 16 cos t , −16 sin t cos t cos(−16 sin t)

· − 4 sin t , −4 cos t , 0 dt
= −64 dt
and, by Stokes’ theorem,
ZZ I

Z 2π
∇ × F · n̂ dS = F r(t) · dr = −64 dt = −128π
S1 ∂S1 0
∂S1 ∂S2a
S2 ∂S2b
S1
n̂ n̂
(b) The boundary, ∂S2 , of S2 consists of two parts, a circle in the plane
z = 4 and a circle in the plane z = 1. We’ll call the first part p
∂S2a . It is the
same as ∂S1 . We’ll call the second part ∂S2b . It is the circle x2 + y 2 = 1,
z = 1 oriented counterclockwise when viewed from high on the z-axis. We
can parametrize it
r(t) = cos tı̂ı + sin t ̂ + k̂, 0 ≤ t ≤ 2π
So, on ∂S2b ,

F r(t) · dr = − sin t , cos t , sin t cos t cos(sin t) · − sin t , cos t , 0 dt
= dt
ZZ I I

∇ × F · n̂ dS = F r(t) · dr + F r(t) · dr
S2 ∂S2a ∂S2b
Z 2π
= −128π + dt
0
= −126π
4.4.3.15. ∗. Solution. Denote by
S = (x, y, z) z = x + 4, x2 + y 2 ≤ 4

the part of the plane z = x+4 that is contained in the cylinder x2 +y 2 = 4.

Orient S by the downward pointing normal n̂ = √12 (1, 0, −1). Then C is
the boundary of S. The part of C and S that are in the first octant are
sketched below.
z
C x2 + y 2 = 4
S
n̂
y
x
We may parametrize S by
r(x, y) = (x, y, x + 4) with x2 + y 2 ≤ 4
So,  
ı̂ı ̂ k̂
∂r ∂r
× = det 1 0 1  = − 1, 0, 1
∂x ∂y
0 1 0
and, by (3.3.1),
∂r ∂r
n̂ dS = − × dxdy = (1, 0, −1) dxdy
∂x ∂y
∂r ∂r
We have chosen to “−” sign in n̂ dS = ± ∂x × ∂y dxdy to give the downward
pointing normal. As the curl of F is
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
x3 + 2y sin(y) + z x + sin(z 2 )
= −ı̂ı − ̂ − 2k̂
Stokes’ theorem (Theorem 4.4.1) gives

I ZZ ZZ
F · dr = ∇ × F · n̂ dS = (−1, −1, −2) · (1, 0, −1) dxdy
C
Z ZS S
= dxdy = 4π
S
4.4.3.16. ∗. Solution. (a) Note that all three vertices, (2, 0, 0), (0, 2, 0)
and 0, 0, 2), lie in the plane x + y + z = 2. So the entire path lies in that
plane too.
z
(0, 0, 2)
n̂
y
(0, 2, 0)
x (2, 0, 0)
In part (b) we will need to evaluate a line integral that clearly cannot
be computed directly — we will need to use Stokes’ theorem. So let’s use
Stokes’s theorem in part (a) too. First, we find
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y
ı  ı
∂z  = 2y ı̂ + 2z ̂ + 2xı̂
z 2 x2 y 2
Let S be the triangular surface that is contained in the plane x + y + z = 2

and is bounded by L1 , L2 and L3 . Orient S by the normal vector n̂ =
√1 (ı̂ı + ̂
 + k̂). Then,
3
1 2
∇ × F · n̂ = √ 2y ı̂ı + 2z ̂ + 2xı̂ı · (ı̂ı + ̂ + k̂) = √ (x + y + z)
3 3
I ZZ ZZ ZZ
2 4
F · dr = ∇ × F · n̂ dS = √ (x + y + z) dS = √ dS
C S 3 S 3 S
4
= √ Area(S)
3
The triangle S is half of the prallelogram with sides (0, 2, 0) − (2, 0, 0) =
(−2, 2, 0) and (0, 0, 2) − (2, 0, 0) = (−2, 0, 2). The area of the parallelogram
is
√
(−2, 2, 0) × (−2, 0, 2) = (4, 4, 4) = 4 3
So
4 √
I
F · dr = √ 2 3 = 8
C 3
(b) Let S̃ be the specified surface. Then, as in part (a),

I ZZ ZZ ZZ
2 4
F · dr = ∇ × F · n̂ dS = √ (x + y + z) dS = √ dS
C S̃ 3 S̃ 3 S̃
4
= √ Area(S̃)
3
√
=4 3
4.4.3.17. ∗. Solution. Let’s try Stokes’ theorem with

1 x
F= z+ ı̂ı + xz ̂ + 3xy − k̂
1+z (z + 1)2
The curl of F is
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
1 x
z + 1+z xz 3xy − (z+1)2

1 1
= (3x − x)ı̂ı − 3y − − 1 + ̂ + z k̂
(z + 1)2 (1 + z)2
= 2xı̂ı + (1 − 3y) ̂ + z k̂
Write
(x, y, z) z = f (x, y) = 1 − x2 y, x2 + y 2 ≤ 1

S=
For S, with the upward pointing normal, by (3.3.2),

n̂ dS = − fx , −fy , 1 dxdy
= 2xy , x2 , 1 dxdy

so that
z
z }| {
∇ × F · n̂ dS = 4x2 y + (x2 − 3x2 y) + (1 − x2 y) dxdy


Z ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
2
4x y + x2 − 3x2 y + 1 − x2 y dxdy

=
x2 +y 2 ≤1
So
Z ZZ ZZ
x2 + 1 dxdy = π + x2 dxdy

F · dr =
C x2 +y 2 ≤1 x2 +y 2 ≤1
To evaluate the final remaining integral, let’s switch to polar coordinates.

ZZ Z 1 Z 2π
x2 dxdy = dr r dθ r cos θ)2
x2 +y 2 ≤1 0 0
Z 1 Z 2π
= dr r3 dθ cos2 θ
0 0
Since
Z 2π Z 2π 2π
1 + cos(2θ) θ sin(2θ)
cos2 θ dθ = dθ = + =π
0 0 2 2 4 0
R1 R 2π π
we finally have 0
dr r3 0
dθ cos2 θ = 4 and
Z
π 5π
F · dr = π + =
C 4 4
R 2π
4.4.3.18. ∗. Solution. We are to evaluate a line integral around a curve
C. We are told that C is the boundary of a surface S that is contained in
the plane x + y + z = 1, but we are not told precisely what C is. So we are
going to have to use Stokes’ theorem. The curl of F is

 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y
ı 
∂z  = 2y ı̂ + 2z ̂ + 2x k̂
z 2 x2 y 2
and, by (3.3.3) with G(x, y, z) = x + y + z,

√

∇G
dS =
dxdy = 3 dxdy
∇ G · k̂
∇G 1
n̂ dS = ± dxdy = ± ı̂ı + ̂ + k̂) dxdy = ± √ ı̂ı + ̂ + k̂) dS
∇ G · k̂ 3
Because C is oriented in a clockwise direction as
z
C
n̂
y
x
observed from the positive z-axis looking down at the plane, n̂ is to
point downwards, so that
1
n̂ dS = − √ ı̂ı + ̂ + k̂) dS
3
On S we have x + y + z = 1, so that Stokes’ theorem gives
I ZZ
F · dr = ∇ × F) · n̂ dS
(∇
C S
ZZ
1
= 2(y ı̂ı + z ̂ + x k̂) · − √ ı̂ı + ̂ + k̂) dS
S 3
ZZ ZZ
2 2
= −√ (y + z + x) dS = − √ dS
3 S 3 S
10
= −√
3
since S has area 5.
4.4.3.19. ∗. Solution. We are to evaluate the line integral of a com-
plicated vector field around a relatively complicated closed curve. That
To see if Stokes’ theorem looks promising, let’s compute the curl
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det   = k̂
 
∂x ∂y √ ∂z
x 4
−y + e sin x y z tan z
That’s suggestive. Next we need to find a surface whose boundary is C.

First, here is a sketch of C.
z
(0, 1, 2)
(0, 1, 1)
Q
(0, 2, 0)
(0, 0, 0)
y
T
(2, 2, 0)
x
We can choose the surface S to be the union of two flat parts:
• the quadralateral Q in the yz-plane with vertices (0, 0, 0), (0, 1, 1),
(0, 1, 2) and (0, 2, 0) and
• the triangle T in the xy-plane with vertices (0, 0, 0), (0, 2, 0) and
(2, 2, 0).
The normal to Q is −ı̂ı and the normal to T is −k̂. Then Stokes’ theorem
gives
Z ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS ZZ
= k̂ · (−ı̂ı) dS + k̂ · (−k̂) dS
Q T
ZZ
=− dS
T
= −Area(T )
base height
z}|{ z}|{
= − 12 2 2
= −2
4.4.3.20. ∗. Solution. The integral looks messy. Let’s compute the
curl of
F = (z + sin z)ı̂ı + (x3 − x2 y) ̂ + (x cos z − y) k̂
to help gauge if Stokes’ theorem would be easier.
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det  ∂x
 
∂y ∂z 
z + sin z x3 − x2 y x cos z − y
= −ı̂ı + ̂ + (3x2 − 2xy) k̂
That’s a lot simpler than F. For the surface z = f (x, y) = xy 2 , with

downward pointing normal (since C is traversed clockwise)
n̂ dS = − − fx , −fy , 1 dxdy = y 2 , 2xy, −1 dxdy

by (3.3.2), So, writing
S = (x, y, z) z = xy 2 , x2 + y 2 ≤ 1

D = (x, y) x2 + y 2 ≤ 1

Stoke’s theorem gives

Z ZZ ZZ
− y 2 + 2xy − 3x2 + 2xy dxdy

F · dr = ∇ × F · n̂ dS =
C
ZSZ D
2
3x + y 2 − 4xy dxdy

=−
x2 +y 2 ≤1
To evaluate this integral, switch to polar coordinates.

Z Z 1 Z 2π
dθ 3r2 cos2 θ + r2 sin2 θ − 4r2 sin θ cos θ

F · dr = − dr r
C 0 0
Z 1
= −4π dr r3 = −π
0
R 2π 1
R 2π R 2π R 2π
since 0 sin θ cos θ dθ = 2 0
sin(2θ) dθ = 0 and 0
sin2 θ dθ = 0
cos2 θ dθ =
π. (See Example 2.4.4.)
4.4.3.21. ∗. Solution. Here is a sketch of the part of S in the first
octant.
z n̂
x2 + y 2 + z 2 = 2
S z=1
∂S
y
x
The boundary, ∂S, of S is the circle x2 + y 2 = 1, z = 1, oriented
counterclockwise when viewed from above. It is parametrized by
r(θ) = cos θ ı̂ı + sin θ ̂ + k̂ 0 ≤ θ ≤ 2π

ZZ I
∇ × F · n̂ dS = F · dr
S ∂S
F(r(t)) r (t) 0
Z 2π z }| { z }| {
2 3
= − sin θ ı̂ı + cos θ ̂ + (mess)k̂ · − sin θ ı̂ı + cos θ ̂ dθ
0
Z 2π
sin3 θ + cos4 θ dθ

=
0
The integral of any odd power of sin θ or cos θ over 0 ≤ θ ≤ 2π is zero.

R 2π
(See Example 4.4.6.) In particular, 0 sin3 θ dθ = 0. To integrate cos4 θ
we use the trig identity
cos(2θ) + 1
cos2 θ =
2
2
cos (2θ) + 2 cos(2θ) + 1
=⇒ cos4 θ =
4
1 cos(4θ) + 1 cos(2θ) 1
= + +
4 2 2 4
3 cos(4θ) cos(2θ)
= + +
8 8 2
Finally
ZZ Z 2π 3 cos(4θ) cos(2θ) 3π
∇ × F · n̂ dS = + + dθ =
S 0 8 8 2 4
4.4.3.22. ∗. Solution. We are to evaluate the line integral of a com-
plicated vector field around a relatively complicated closed curve. That
As we are to use Stokes’ theorem, let’s compute the curl
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x

∂y ∂z 
2
x sin y −y sin x (x − y)z
= −z 2 ı̂ı − z 2 ̂ − (y cos x + x cos y)k̂
Next we need to find a surface whose boundary is C. First, here is a sketch

of C.
z
(0, 0, 1) (0, π/2, 1)
(π/2, 0, 1) Sy
Sx
(0, π/2, 0) y
(π/2, 0, 0)
x
We can choose the surface S to be the union of two flat parts:
• the rectangle Sx in the xz-plane with vertices (0, 0, 0), ( π2 , 0, 0), ( π2 , 0, 1)
and (0, 0, 1) and
• the rectangle Sy in the yz-plane with vertices (0, 0, 0), (0, 0, 1), (0, π2 , 1)
and (0, π2 , 0)
The normal to Sx is −̂ and the normal to Sy is −ı̂ı. Then Stokes’ theorem
gives
Z ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS ZZ
= ∇ × F · (−̂) dS + ∇ × F · (−ı̂ı) dS
Sx Sy
π π
Z 2
Z 1 Z 2
Z 1
= dx dz z 2 + dy dz z 2
0 0 0 0
Z π Z π
2 1 2 1
= dx + dy
0 3 0 3
π
=
3
4.4.3.23. ∗. Solution. (a) Here is a sketch.
z
(0, 0, 2)
Sy
Sx y
(0, 3, 0)
(2, 0, 0)
x
(b) We are to evaluate the line integral of a complicated vector field
around a relatively complicated closed curve. That certainly suggests that
we should not try to evaluate the integral directly. Let’s try Stokes’ theo-
rem. First, we compute the curl
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det 
 
∂x ∂y ∂z 
2z 2 3z 2
1+y + sin(x ) 1+x + sin(y ) 5(x + 1)(y + 2)
3 2
= 5(x + 1) − ı̂ı − 5(y + 2) − ̂
1+x 1+y
3z 2z
+ − 2
+ k̂
(1 + x) (1 + y)2
Next we need to find a surface S whose boundary is C. We can choose the

surface S to be the union of two flat parts:
• the triangle Sx in the xz-plane with vertices (0, 0, 0), (2, 0, 0), and
(0, 0, 2) and
• the triangle Sy in the yz-plane with vertices (0, 0, 0), (0, 0, 2), and
(0, 3, 0)
Note that
• The normal to Sx specified by Stokes’ theorem is −̂. On Sx we have
2
y = 0, so that ∇ × F · ̂ simplifies to − 5(0 + 2) − 1+0 = −8.
• The normal to Sy specified by Stokes’ theorem is −ı̂ı.On Sy we have
3
x = 0, so that ∇ × F · ı̂ı simplifies to 5(0 + 1) − 1+0 = 2.
Z ZZ
F · dr = ∇ × F · n̂ dS
C S
ZZ 8 ZZ −2
z }| { z }| {
= ∇ × F · (−̂) dS + ∇ × F · (−ı̂ı) dS
Sx Sy
1 1
= 8 Area(Sx ) − 2 Area(Sy ) = 8 (2)(2) − 2 (3)(2)
2 2
= 10
4.4.3.24. ∗. Solution. The boundary, ∂S, of S is the circle x2 + y 2 = 1

oriented counter clockwise as usual. It may be parametrized by r(θ) =
cos θ ı̂ı + sin θ̂, 0 ≤ θ ≤ 2π. By Stokes’ theorem

ZZ I Z 2π
dr
∇ × F · n̂ dS = F · dr = F r(θ) · (θ) dθ
S ∂S 0 dθ
Z 2π
= (sin θ, 0, 3 cos θ) · (− sin θ, cos θ, 0) dθ
0
2π 2π
1 − cos(2θ)
Z Z
=− sin2 θ dθ = − dθ
0 0 2
2π
θ sin(2θ)
=− −
2 4 0
= −π
R 2π
dθ sin2 θ, see Example 2.4.4.
4.4.3.25. ∗. Solution. The given surface is an ellipsoid centred at
(x, y, z) = (0, 0, 1). It caps a curve C in the plane z = 0, given by x2 + y 2 =
4. This is a circle of radius 2 centred at the origin, oriented counterclockwise
when viewed from the positive z-axis.
Method I — double Stokes’: Let D denote the plane disk x2 + y 2 ≤ 4,
z = 0. Using Stokes’ theorem twice gives
ZZ ZZ I ZZ
G · n̂ dS = ∇ × F · n̂ dS = F · dr = ∇ × F · n̂ dS
S
Z ZS C D
= G · n̂ dS
D
Now in D we have n̂ = k̂ and z = 0, so on this surface,

 
0 0 1
∂ ∂ ∂
G · n̂ = (∇ × F) · k̂ = det 
 
∂x ∂y ∂z 
2 2 2
(xz − y 3 cos z) x3 ez xyzex +y +z z=0
= 3x2 ez + 3y 2 cos z z=0 = 3(x2 + y 2 )

Hence, using polar coordinates,

ZZ ZZ Z 2π Z 2
G · n̂ dS = 3(x2 + y 2 ) dxdy = 3 (r2 ) r dr dθ
S D θ=0 r=0
= 3(2π)(4) = 24π
Method II — single Stokes’: By Stokes’ theorem
ZZ ZZ I
G · n̂ dS = ∇ × F · n̂ dS = F · dr
S S C
Parametrize the circle C using

r(θ) = 2 cos θ ı̂ı + 2 sin θ ̂, 0 ≤ θ ≤ 2π
to obtain
dr
dr = dθ = (−2 sin θ ı̂ı + 2 cos θ ̂) dθ.
dθ
Then since z = 0 on C,

F r(θ) r (θ) 0
ZZ Z 2π z }| { z }| {
3 3
G · n̂ dS = −(2 sin θ) ı̂ı +(2 cos θ) ̂ · −2 sin θ ı̂ı +2 cos θ ̂ dθ
S 0
Z 2π
sin4 θ + cos4 θ dθ

= 16
0
By the double angle trig identities
1 + cos(2θ) 1 − cos(2θ)
cos2 θ = sin2 θ =
2 2
we have
2 2
4 1 − cos(2θ)
4 1 + cos(2θ)
sin θ + cos θ = +
4 4
2
1 cos (2θ) 1 1 + cos(4θ)
= + = +
2 2 2 4
So
ZZ Z 2π
3 1 3
G · n̂ dS = 16 + cos(4θ) dθ = 16 × × (2π) = 24π
S 0 4 4 4
4.4.3.26. ∗. Solution. Note that

ı̂ı ̂ k̂

∂ ∂ ∂
∇ × F = det ∂x = 2y ı̂ı + 2z ̂ + 2x k̂
2 ∂y2 ∂z2
z x y
Let D be the disk in the plane x + y + z = 3 whose boundary is C and

let n̂ = √13 (ı̂ı + ̂ + k̂) be the upward unit normal to D. If the circle
is oriented counterclockwise, when viewed from above, then, by Stokes’
theorem (Theorem 4.4.1),
I ZZ
F · dr = ∇ × F · n̂ dS
C D
ZZ
1
=√ 2y ı̂ı + 2z ̂ + 2x k̂ · (ı̂ı + ̂ + k̂) dS
3 D
ZZ =3 on D
1 z }| {
=√ 2 x + y + z dS
3 D
√ ZZ √
=2 3 dS = 2 3πR2
D
4.4.3.27. Solution 1. Let S 0 be the bottom surface of the cube, oriented

with normal k̂. Then, by Stokes’ theorem, since ∂S = ∂S 0 ,
ZZ I I ZZ
∇ × F · n̂ dS = F · dr = F · dr = ∇ × F · n̂ dS
S ∂S ∂S 0 S0
Since
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 2

∇ × F = det  ∂x  = · · · , · · · , y − xz

∂y ∂z
2 2
xyz xy x yz
and n̂ = k̂ on S 0 and z = −1 on S 0
ZZ Z 1 Z 1
∇ × F · n̂ dS = dx dy · · · , · · · , y 2 − xz · k̂
S0 −1 −1 z=−1
Z 1 Z 1
= dx dy (y 2 + x)
−1 −1
Z 1 Z 1 Z 1
= dx dy y 2 = 2 × 2 dy y 2
−1 −1 0
4
=
3
Solution 2. The boundary of S is the square C, with sides C1 , · · ·, C4 ,

in the sketch
(−1, 1, −1) C3 (1, 1, −1)
C4 C2
(−1, −1, −1) C1 (1, −1, −1)

By Stokes’ theorem,
ZZ I
∇ × F · n̂ dS = F · dr
S C
Parametrize C1 by x. That is, r(x) = xı̂ı − ̂ − k̂, −1 ≤ x ≤ 1. Since

r0 (x) = ı̂ı, and y = z = −1 on C1 ,
Z Z 1 Z 1
0

F · dr = F r(x) · r (x) dx = F r(x) · ı̂ı dx
C1 −1 −1
xyz
Z 1 z }| {
= x(−1)(−1) dx
−1
=0 (since x is odd)
Parametrize C2 by y. That is, r(y) = ı̂ı + y ̂ − k̂, −1 ≤ y ≤ 1. Since

r0 (y) = ̂, and x = 1 on C2 ,
xy 2 3 1
Z Z 1 Z 1
z}|{ y 2
y 2 dy =

F · dr = F r(y) · ̂ dy = =
C2 −1 −1 3 −1 3
Parametrize C3 by x. That is, r(x) = xı̂ı + ̂ − k̂ with x running from

1 to −1. (If you’re nervous about this, parametrize by t = −x. That is
r(t) = −tı̂ı + ̂ − k̂, −1 ≤ t ≤ 1.) Since r0 (x) = ı̂ı, and y = 1, z = −1 on C3 ,
xyz
Z Z −1
Z −1 z }| {
F · dr = F r(x) · ı̂ı dx = x (1)(−1) dx = 0 (since x is odd)
C3 1 1
Parametrize C4 by y. That is, r(y) = −ı̂ı + y ̂ − k̂, with y running from 1
to −1. Since r0 (y) = ̂, and x = −1 on C4 ,
xy 2 3 −1
Z Z −1 Z −1 z }| { y 2
(−1)y 2 dy = −

F · dr = F r(y) · ̂ dy = =
C4 1 1 3 1 3
All together
ZZ Z Z Z Z
4
∇ × F · n̂ dS = F · dr + F · dr + F · dr + F · dr =
S C1 C2 C3 C4 3
4.4.3.28. Solution. Let’s try Stokes’ Theorem. Call F = y ı̂ı −x ̂ +xy k̂.
Then
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = xı̂ı − y ̂ − 2 k̂
y −x xy
Now compute n̂ dS in the (u, v)-parametrization.

r(u, v) = u cos v, u sin v, v
∂r
(u, v) = cos v, sin v, 0
∂u
∂r
(u, v) = − u sin v, u cos v, 1
∂v
 
ı̂ı ̂ k̂
∂r ∂r
× = det  cos v sin v 0  = sin v, − cos v, u
∂u ∂v
−u sin v u cos v 1

n̂ dS = sin v, − cos v, u du dv
Since u ≥ 0, we do indeed have the upward pointing normal. So, Stokes’

theorem tells us
Z Z ZZ
y dx − x dy + xy dz = F · dr = ∇ × F · n̂ dS
C C S
Z 1 Z 2π
= du dv u cos v, −u sin v, −2 · sin v, − cos v, u
0 0
Z 1 Z 2π
= dv 2u sin v cos v − 2u
du
0 0
Z 1 Z 2π

= du u dv sin 2v − 2
0 0
= 12 (−4π) = −2π
4.4.3.29. Solution. Given the form of F, direct evaluation looks hard.

So let’s try Stokes’ theorem, using as S the part of the plane G(x, y, z) =
x + 2y − z = 7 that is inside x2 − 2x + 4y 2 = 15. Then
∇G
n̂ dS = ± dx dy = ± ı̂ı + 2̂ − k̂ dx dy
∇ G · k̂
As C is oriented counterclockwise when viewed from high, Stokes’ theorem
specifies the upward pointing normal so that n̂ dS = − ı̂ı + 2̂ − k̂ dx dy.
From the observations that
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det  ∂x  = xı̂ı − (z + 2x) k̂
 
∂y ∂z
2
ex + yz cos(y 2 ) − x2 sin(z 2 ) + xy
and that we can rewrite x2 − 2x + 4y 2 = 15 as (x − 1)2 + 4y 2 = 16, we have

I ZZ
F · dr = ∇ × F · n̂ dS
C
Z ZS
= [xı̂ı − (z + 2x)k̂] · (−1, −2, 1) dx dy

(x−1)2 +4y 2 ≤16 z=−7+x+2y
ZZ
= [−x − (−7 + x + 2y + 2x)] dx dy
(x−1)2 +4y 2 ≤16
ZZ
= [7 − 4x − 2y] dx dy
(x−1)2 +4y 2 ≤16
To evaluate the integrals of x and y we use that, for any region R in the
xy--plane, RR RR
R
x dx dy y dx dy
x̄ = ȳ = R
Area(R) Area(R)
(x−1)2 y2
Our ellipse is 42 + = 1 and so has area πab = π × 4 × 2 = 8π and
22
2 2
centroid (x̄, ȳ) = (1, 0). So, using R = (x, y) (x−1) + y22 ≤ 1 ,

42
I ZZ
F · dr = [7 − 4x − 2y] dx dy
C R

= Area(R) 7 − 4x̄ − 2ȳ}
= 8π[7 − 4 × 1 − 2 × 0] = 24π
4.4.3.30. ∗. Solution. (a) The curl is
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × F = det   = (1 − 2xz) ̂
 
∂x ∂y ∂z
2 + x2 + z 0 3 + x2 z
(b) We are going to use Stokes’ theorem. The specified curve C is not
closed and so is not the boundary of a surface. So we extend C to a closed
curve C̃ by appending to C the line segment L from (2, 0, 0) to (0, 0, 0). In
the figure below, C is the red curve and C̃ is C plus the blue line segment.
z
(0, 0, 3)
−ı̂ı
C C
T1
(0, 1, 0)
T2 y
L
(2, 0, 0) C
x −k̂
The closed curve C̃ is boundary of the surface S that is the union of
• the triangle T1 in the yz-plane with vertices (0, 0, 0), (0, 0, 3) and
(0, 1, 0) and with normal vector −ı̂ı and
• the triangle T2 in the xy-plane with vertices (0, 0, 0), (0, 1, 0) and
(2, 0, 0) and with normal vector −k̂.
So, by Stokes’ theorem

Z Z Z ZZ
F · dr + F · dr = F · dr = ∇ × F · n̂ dS
C
Z ZL ∂S
ZZ S
= ∇ × F · (−ı̂ı) dS + ∇ × F · (−k̂) dS
T T2
ZZ 1 ZZ
= (1 − 2xz) ̂ · (−ı̂ı) dS + (1 − 2xz) ̂ · (−k̂) dS
T1 T2
=0
Consequently the integral of interest

Z Z Z 0
F · dr = − F · dr = − (2 + x2 ) dx since dy = dz = z = 0 on L
C L 2
2 2
x3
Z
20
= (2 + x2 ) dx = 2x + =
0 3 0 3
4.4.3.31. ∗. Solution. (a) by direct evaluation: The curl of G is
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × G = det  ∂x ∂y
ı
∂z  = 2ı̂
x −z y
The part of S in the first octant is sketched in the figure on the left below.
S consists of two parts — the cylindrical surface
S1 = (x, y, z) y 2 + z 2 = 9, 0 ≤ x ≤ 5

and the disc

(x, y, z) x = 0, y 2 + z 2 ≤ 9

S2 =
The normal n̂ to S1 always points radially outward from the cylinder and
so always has ı̂ı component zero. The normal of S2 is −ı̂ı. So the flux is
ZZ ZZ ZZ
∇ × G · n̂ dS = 2ı̂ı · n̂ dS + 2ı̂ı · (−ı̂ı) dS
S S1 S2
ZZ
= −2 dS
S2
= −2 π32 = −18π

(a) using Stokes’ theorem: Let’s use Stokes’ theorem. The boundary ∂S
of S is the cirlce y 2 + z 2 = 9, x = 5, oriented clockwise when viewed from

far down the x-axis. We’ll parametrize it by r(θ) = 5, 3 cos θ, −3 sin θ .
Then Stokes’ theorem gives
ZZ I
∇ × G · n̂ dS = G · dr
S ∂S
Z 2π
= 5 , 3 sin θ, 3 cos θ · 0 , −3 sin θ , −3 cos θ dθ
0
Z 2π
− 9 sin2 θ − 9 cos2 θ dθ

=
0
= −18π
z z
S S
y y
∂S n̂ T n̂
x x ı̂ı
(b) This time we’ll use the divergence theorem. The surface S is not
closed. So we’ll use the auxilary surface formed by “topping S off” with
the cap T = (5, y, z) y 2 + z 2 ≤ 9 . If we give T the normal vector
ı̂ı, this auxiliary
surface, the union of S and T , is the boundary of V =
(x, y, z) y 2 + z 2 ≤ 9, 0 ≤ x ≤ 5 . So the divergence theorem gives
ZZ ZZ ZZ
F · n̂ dS + F · n̂ dS = F · n̂ dS
S T
Z Z∂V
Z
= ∇ · F dV
V
=0
since ∇ · F = 0. Thus the flux of interest is

ZZ ZZ ZZ ZZ
F · n̂ dS = − F · n̂ dS = − F · ı̂ı dS = − (2 + z) dS
S T T T
ZZ ZZ
= −2 dS since z dS = 0, because z is odd
T T
= −18π since T has area 9π
y
• x is defined when x 6= 0 and
2 2
• x1+x = e(1+x ) ln x is defined when ln x is defined, which is when
x > 0 (assuming that we are not allowed to use complex numbers)
and
2 2
) ln y
• y 1+y = e(1+y is defined when ln y is defined, which is when
y > 0 and
• cos5 (ln z) is defined when ln z is defined, which when z > 0
the domain of F is

D= (x, y, z) x > 0, y > 0, z > 0
(b) The domain D is both connected (any two points in D can be joined
by a curve that lies completely in D) and simply connected (any simple
closed curve in D can be shrunk to a point continuously in D).
(c) The curl of F is
 
ı̂ı ̂ k̂
∂ ∂ ∂ 1
∇ × F = det   = 2x − k̂
 
∂x ∂y ∂z
y 1+x2 2 x
x +x x2 − y 1+y cos5 (ln z)
(d) The integrand for direct evaluation looks very complicated. On the
other hand ∇ × F is quite simple. So let’s try Stokes’ thoerem. Denote

S = (x, y, z) 2 ≤ x ≤ 4, 2 ≤ y ≤ 4, z = 2
The boundary of S is C. Because of the clockwise orientation of C, we

assign the normal vector −k̂ to S. See the sketch below
z
(2,2,2) (2,4,2)
C
(4,2,2) (4,4,2)
n̂
x
Then, by Stokes’ theorem,
I ZZ ZZ
F · dr = ∇ × F · n̂ dS = ∇ × F · (−k̂) dS
C
ZSZ S
2x − x1 dS

=−
S
Z 4 Z 4 Z 4
1 1

=− dx dy 2x − x =− dx 2 2x − x
2 2 2
h i4
= −2 x2 − ln x = −2 12 − ln 2

2
= 2 ln 2 − 24
(e) Since ∇ × F is not 0, F cannot be conservative.

4.4.3.33. ∗. Solution. (a) By the vector identity ∇ · ∇ × G = 0
(Theorem 4.1.7.a). So we must have
∂ ∂ ∂ 2
axey z + byz + y − xey z 2

0=∇·F= xz +
∂x ∂y ∂z
y y

= z + axe z + bz + − 2xe z
= (1 + b)z + (a − 2)xey z
So we need a = 2 and b = −1.
(b) Note that the boundary, ∂S, is the circle x2 +y 2 = 1, z = 0, oriented
counter-clockwise. Also note that, if we knew what G was, we would be
able to use Stokes’ theorem to give
ZZ ZZ I
F · n̂ dS = ∇
(∇ × G) · n̂ dS = G · dr
S S ∂S
So let’s find a vector potential G. That is, let’s try and find a vector field
G = G1 ı̂ı + G2 ̂ + G3 k̂ that obeys ∇ × G = F, or equivalently,
∂G3 ∂G2
− = F1 = xz
∂y ∂z
∂G3 ∂G1
− + = F2 = 2xey z − yz
∂x ∂z
∂G2 ∂G1
− = F3 = y 2 − xey z 2
∂x ∂y
Let’s also require that G3 = 0. (If this is mysterious to you, review §4.1.2.)
Then the equations above simplify to
∂G2
− = xz
∂z
∂G1
= 2xey z − yz
∂z
∂G2 ∂G1
− = y 2 − xey z 2
∂x ∂y
Now the first equation contains only a single unknown, namely G2 and we
can find all G2 ’s that obey the first equation simply by integrating with
respect to z:
xz 2
G2 = − + N (x, y)
2
∂
Note that, because ∂z treats x and y as constants, the constant of integra-
tion N is allowed to depend on x and y.
Similarly, the second equation contains only a single unknown, G1 , and
is easily solved by integrating with respect to z. The second equation is
satisfied if and only if
1
G1 = xey z 2 − yz 2 + M (x, y)
2
for some function M .
Finally, the third equation is also satisfied if and only if M (x, y) and
N (x, y) obey
∂ xz 2 ∂ y 2 yz 2
− + N (x, y) − xe z − + M (x, y) = y 2 − xey z 2
∂x 2 ∂y 2
which simplifies to
∂N ∂M
(x, y) − (x, y) = y 2
∂x ∂y
This is one linear equation in two unknowns, M and N . Typically, we can
easily solve one linear equation in one unknown.
So we are free to eliminate one of the unknowns by setting, for example,
M = 0, and then choosing any N that obeys
∂N
(x, y) = y 2
∂x
Integrating with respect to x gives, as one possible choice, N (x, y) = xy 2 .
So we have found a vector potential. Namely
1 xz 2
G = xey z 2 − yz 2 ı̂ı + xy 2 − ̂
2 2
We can now evaluate the flux. Parametrize ∂S by
r(θ) = cos θ ı̂ı + sin θ ̂

r0 (θ) = − sin θ ı̂ı + cos θ ̂
with 0 ≤ θ ≤ 2π. So
ZZ I
F · n̂ dS = G · dr
S ∂S
G(r(θ)) r0 (θ)
Z 2π z }| { z }| {
= cos θ sin2 θ ̂ · − sin θ ı̂ı + cos θ ̂ dθ
0
Z 2π
= sin2 θ cos2 θ dθ
0
2π
1 − cos(2θ) 1 + cos(2θ)
Z
= dθ
0 2 2
1 2π
Z
1 − cos2 (2θ) dθ

=
4 0
1 2π n
Z
1 + cos(4θ) o
= 1− dθ
4 0 2
Z 2π
1 1
= 2π since cos(4θ) dθ = 0
4 2 0
π
=
4
4.4.3.34. ∗. Solution. Considering that there are ten line segments in
C, it is probably not very efficient to use direct evaluation. Two other pos-
sible methods come to mind. If F is conservative, we can use F’s potential.
Even if F is not conservative, it may be possible to efficiently use Stokes’
(or Green’s) theorem. So let’s compute
 
ı̂ı ̂ k̂
∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = k̂
y 2x − 10 0
As ∇ × F 6= 0, the vector field F is not conservative. As ∇ × F 6= 0 is
very simple, it looks like Stokes’ theorem could provide an efficient way to
compute the integral. The left figure below contains a sketch of C.
y (4, 5) y (4, 5)
(5, 5) (5, 5)
(3, 4) (3, 4)
(4, 4) (4, 4)
(2, 3) (2, 3)
(3, 3) (3, 3)
(1, 2) (1, 2) L
(2, 2) (2, 2)
C C
(0, 1) (1, 1) (0, 1) (1, 1)
(0, 0) x (0, 0) x
The curve C is not closed, and so is not the boundary of a surface, so
we cannot apply Stokes’ theorem directly. But we can easily come up with
a surface whose boundary contains C. Let R be the shaded region in the
figure on the right above. The boundary ∂R of R consists of two parts —
C and the line segment L. The normal of R for −k̂ (since ∂R is oriented
clockwise). So Stokes’ theorem gives
Z Z ZZ ZZ
F · dr + F · dr = ∇ × F · (−k̂) dS = (k̂) · (−k̂) dS
C L R R
= −Area(R)
R is the union of 5 triangles, each of height 1 and base 1. So
1 5
Area(R) = 5 × ×1×1=
2 2
If we denote by −L the line segment from (0, 0) to (5, 5), we can parametrize
−L by r(t) = t(5, 5), 0 ≤ t ≤ 1 and
F(r(t)) r0 (t)
Z Z 1 z }| { z }| {
Z 1

F · dr = 5tı̂ı + (10t − 10) ̂ · (5ı̂ı + 5 ̂) dt = 5 15t − 10 dt
−L 0 0

3 25
= 25 −2 =−
2 2
All together
Z Z Z
5 25
F · dr = −Area(R) − F · dr = −Area(R) + F · dr = − −
C L −L 2 2
= −15
4.4.3.35. ∗. Solution. If we parametrize the curve as
x = 2 cos θ y = 2 sin θ z = x2 = 4 cos2 θ

0 ≤ θ ≤ 2π
then the term sin x(θ)2 x0 (θ) in the integral will be sin 4 cos2 θ (−2 sin θ).

That looks hard to integrate. So let’s try Stokes’ theorem. The curl of F
is
 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × F = det  ∂x ∂y ∂z  = −xı̂ + z k̂
ı
2 2
sin x xz z
The curve C is the boundary of the surface
S = (x, y, z) x2 + y 2 ≤ 4, z = x2

with upward pointing normal. For the surface z = f (x, y) = x2 , (3.3.2)

gives

n̂ dS = ± − fx (x, y)ı̂ı − fy (x, y) ̂ + k̂ dxdy

= ± − 2xı̂ı + k̂ dxdy
Since we want the upward pointing normal

n̂ dS = − 2xı̂ı + k̂ dxdy
So by Stokes’ theorem (Theorem 4.4.1)

I ZZ
F · dr = ∇ × F · n̂ dS
C S
ZZ z
z}|{
− xı̂ı + x2 k̂ · − 2xı̂ı + k̂ dxdy

=
x2 +y 2 ≤4
ZZ
=3 x2 dxdy
x2 +y 2 ≤4

I Z 2 Z 2π
F · dr = 3 dr r dθ r2 cos2 θ
C 0 0
Z 2 Z 2π
3 2
=3 r dr cos θ dθ
0 0
2π
24
Z
cos(2θ) + 1
=3 dθ
4 0 2
= 12π
R 2π
cos2 t dt see Example 2.4.4.
4.4.3.36. ∗. Solution. By Stokes’ Theorem,
I ZZ
E · dr = ∇ × E) · n̂ dS
(∇
C S
so Faraday’s law becomes

ZZ
1 ∂H
∇×E+ · n̂ dS = 0
S c ∂t
This is true for all surfaces S. So the integrand, assuming that it is con-
tinuous, must be zero.
To see this, let G = ∇ × E + 1c ∂H∂t . Suppose that G(x0 ) 6= 0. Pick
a unit vector n̂ in the direction of G(x0 ). Let S be a very small flat disk
centered on x0 with normal n̂ (the vector we picked). Then G(x0 ) · n̂ > 0
and, by continuity,G(x) · n̂ > 0 for
all x on S, if we have picked S small
1 ∂H
RR
enough. Then S ∇ × E + c ∂t · n̂ dS > 0, which is a contradiction. So
G = 0 everywhere and we conclude that
1 ∂H
∇×E+ =0
c ∂t
4.4.3.37. ∗. Solution. The curl of the specified vector field is
∇ × F = ∇ × z ı̂ı + x ̂ + y 3 z 3 k̂

 
ı̂ı ̂ k̂
∂ ∂ ∂ 
= det  ∂x ∂y ∂z 
z x y3 z3
= 3y 2 z 3 ı̂ı + ̂ + k̂
For every t, we have x(t) = z(t) and x(t)2 + y(t)2 + z(t)2 = 2. So the
specified curve is the intersection of the plane x = z and the sphere x2 +
y 2 + z 2 = 2. This curve is the boundary of the circular disk
D = (x, y, z) x = z, x2 + y 2 + z 2 ≤ 2

√
The curve is oriented so that x(t), y(t) = cos t, 2 sin t runs in the
standard (counterclockwise) direction. So the unit normal to D used in
Stokes’ theorem has positive k̂ component. Since the plane x − z = 0
has unit normal ± √12 (1, 0, −1), the unit normal used in Stokes’ theorem is
n̂ = √12 (−1, 0, 1). By Stokes’ theorem
I ZZ ZZ
1
F · dr = ∇ × F · n̂ dS = √ (3y 2 z 3 , 1, 1) · (−1, 0, 1) dS
C D 2 D
ZZ
1
=√ (1 − 3y 2 z 3 ) dS
2 D
The disk D is invariant under the reflection (x, y, z) → (−x, y, −z). Since
y 2 z 3 is odd under this reflection, D y 2 z 3 ds = 0 and

RR
I ZZ
1 1
F · dr = √ dS = √ Area(D)
2 D 2
Because the centre of the ball x2 +y 2 +z 2 ≤ 2 (namely (0, 0, 0)) is contained

in the plane x = z, the radius of the disk D is √ the same as the radius of
the sphere x2 + y 2 + z 2 = 2. So D has radius 2 and
I
1 1 √ 2 √
F · dr = √ Area(D) = √ π 2 = 2π
2 2
4.4.3.38. ∗. Solution. The curl of the vector field F = z ı̂ı + x ̂ − y k̂ is
∇ × F = −ı̂ı + ̂ + k̂
The unit normal to the plane x + y + z = 1, with positive k̂ component as

required by Stokes’ theorem in this case, is n̂ = √13 (1, 1, 1). If we denote
by D the circular disk x + y + z = 1, x2 + y 2 + z 2 ≤ 1, then Stokes’ theorem
(Theorem 4.4.1) says
I I ZZ
z dx + x dy − y dz = F · dr = ∇ × F · n̂ dS
C C D
ZZ
1
= (−1, 1, 1) · √ (1, 1, 1) dS
D 3
1
= √ Area(D)
3
A reasonable guess for the centre of the disk is 13 (1, 1, 1). (This guess is
just based on symmetry.) To check this we just need to observe that it is
indeed on the plane x + y + z = 1 and that the distance from 13 (1, 1, 1) to
any point (x, y, z) obeying x + y + z = 1 and x2 + y 2 + z 2 = 1, namely
r
1 2 1 2 1 2
x− + y− + z−
3 3
r 3 r
2 3 2 1
= x2 + y 2 + z 2 − (x + y + z) + = 1 − +
3 9 3 3
r
2
=
3
q
is the same. This also tells us that D has radius 23 and hence area 23 π.
2π
So the specified line integral is √ .
3 3
4.4.3.39. ∗. Solution. (a) We parametrize S in cylindrical coordinates:
r(r, θ) = r cos θ ı̂ı + r sin θ ̂ + r k̂ with 0 ≤ r ≤ 1, 0 ≤ θ ≤ π
(b) We compute
∂r
= cos θ ı̂ı + sin θ ̂ + k̂
∂r
∂r
= −r sin θ ı̂ı + r cos θ ̂
∂θ
∂r ∂r
n̂ dS = ± × drdθ = ± − r cos θ ı̂ı − r sin θ ̂ + r k̂ drdθ
∂r ∂θ
To calculate the downward flux, we use the minus sign. We find

ZZ Z π Z 1
v · n̂ dS = dθ dr (r cos θ, r sin θ, −2r) · (r cos θ, r sin θ, −r)
S 0 0
Z π Z 1 1
= dθ dr 3r2 = πr3 =π

0 0 r=0
(c) Solution 1 : Let P be the path along line segments from (1, 0, 1) to
(0, 0, 0) and from (0, 0, 0) to (−1, 0, 1). Here is a sketch. P is in blue.
(−1, 0, 1)
z
(1, 0, 1) p
z= x2 + y 2
x
Then Z Z ZZ
F · dr + F · dr = ∇ × F · n̂ dS
C P S
R Theorem. Along P, the vector field F is orthogonal to the curve

by Stokes’
so that P F · dr = 0. Note that ∇ × F is the vector field v from part (b).
Thus Z ZZ
F · dr = v · n̂ dS = π
C S
(c) Solution 2: Let L be the line segment from (1, 0, 1) to (−1, 0, 1) and
let
R = {(x, y, z)} x2 + y 2 ≤ 1, y ≥ 0, z = 1
Here is a sketch. L is in blue and R is shaded.
(−1, 0, 1)
z
L R C
(1, 0, 1) p
z= x2 + y 2
x
Then Z Z ZZ
F · dr + F · dr = ∇ × F · (−k̂) dS
C L R
by Stokes’ Theorem. Along L, the vector R field F = ̂ is orthogonal to the
curve (which has direction −ı̂ı so that L F · dr = 0. Note that ∇ × F is
the vector field v from part (b). Thus
Z ZZ ZZ ZZ
F · dr = − v · k̂ dS = 2z dS = 2 dS = 2 Area(R) = π
C R R R
4.4.3.40. Solution. Let S 0 be the portion of x + y + z = 1 that is inside

the sphere x2 + y 2 + z 2 = 1. Then ∂S = ∂S 0 , so, by Stokes’ theorem, (with
n̂ always the upward pointing normal)
ZZ I I ZZ
∇ × F) · n̂ dS =
(∇ F · dr = F · dr = ∇ × F) · n̂ dS
(∇
S0 ∂S 0 ∂S S
As
 
ı̂ı ̂ k̂
 ∂ ∂ ∂
∇ × F = det  ∂x  = −2(ı̂ı + ̂ + k̂)

∂y ∂z
y−z z−x x−y
and, on S 0 , n̂ = √1 (ı̂ı
3
+ ̂ + k̂)
ZZ ZZ √ √
∇ × F) · n̂ dS =
(∇ − 2 3 dS = −2 3 × Area(S 0 )
S0 S0
S 0 is the intersection of a plane with a sphere and so is a circular disk. It’s

center (xc , yc , zc ) has to obey xc + yc + zc = 1. By symmetry, xc = yc = zc ,
so xc = yc = zc = 13 . Any point, (x, y, z), which satisfies both x + y + z = 1
and x2 + y 2 + z 2 = 1, obeys
2 2 2
1 1 1
x− + y− + z−
3 3 3
2 1
= x2 + y 2 + z 2 − (x + y + z) + 3
3 9
2 1 2
=1− + =
3 3 3
q
That is, any point on the boundary of S 0 is a distance 23 from 1 1 1

3 , 3 , 3 .
q
So the radius of S 0 is 23 , the area of S 0 is 23 π and
ZZ √ 4
∇ × F) · n̂ dS = −2 3 × Area(S 0 ) = − √ π
(∇
S0 3

5.2 · Exercises
5.2.1. ∗. Solution. (a) True. For any constant vector a = (a1 , a2 , a3 ),
 
ı̂ı ̂ k̂
a × r = det a1
 a2 a3  = (a2 z − a3 y)ı̂ı − (a1 z − a3 x)̂ + (a1 y − a2 x)̂
x y z
This vector field does indeed have divergence 0.

(b) True. This is our conservative field screening condition Theorem 4.1.7.b.
(c) True. This is one of our vector identities, namely Theorem 4.1.4.c.
(d) False. The trap here is that F need not be defined at the origin. We
mr
saw, in Example 3.4.2, that the point source FS = |r| 3 had flux 4πm through
every sphere centred on the origin. We also saw, in Example 4.2.7, that the
divergence ∇ · FS = 0 everywhere except at the origin (where it is not defined).
So if we choose m to be a very big negative number (say −10100 ) and add in
a very small vector field with positive divergence (say 10−100 (xı̂ı + y̂ + z k̂)),
we will get the vector field F = −10100 |r|r 3 + 10−100 (xı̂ı + y̂ + z k̂) which has
divergence ∇ · F = 3 × 10−100 > 0 everywhere except at the origin. The flux
of this field through the specified sphere will be −4π × 10100 plus a very small
positive number.
(e) True. The statement that “the flux out of one hemisphere is equal to
the flux into the opposite hemisphere” is equivalent to the statement that “the
flux out of the sphere is equal to zero”. Since ∇ · F = 0 everywhere, that is
true by the divergence theorem.
(f) That depends.
T̂
If κ = 0, then dds dr
= 0, so that ds = T̂ is a constant. So r(s) = sT̂ + r(0)
is part of a straight line.
If κ > 0, then, because the curve is in a plane, the torsion τ is zero and the
Frenet-Serret formulae reduce to
dT̂ dN̂
= κN̂ = −κT̂
ds ds
Now consider the centre of curvature c(s) = r(s) + κ1 N̂(s). Since
dc dr 1 dN̂ 1
= + = T̂(s) + − κT̂(s) = 0
ds ds κ ds κ
c(s) is a constant and
1
|r(s) − c| =
κ
which says that the curve is part of the circle of radius κ1 centred on c.
(g) False. We saw in Examples 2.3.14 and 4.3.8 that the given vector field
is not conservative. H H
(h) False. For example, if P = −y, then C F · dr = − C y dx is the area
inside C. See Corollary 4.3.5.
(i) False.
If dv
dt = a is a constant, then v(t) = a t + v0 . Integrating a second time,
r(t) = 21 a t2 + v0 t + r0 . This is not a spiral, whether or not the speed is
constant. (In fact, for the speed |v(t)| = |a t + v0 | to be constant, a has to be
0, so that r(t) = v0 t + r0 is a straight line.)
Another way to come to the same conclusion uses
d2 s ds 2
a(t) = (t) T̂(t) + κ(t) (t) N̂(t)
dt2 dt
ds
As the speed dt is a constant, it reduces to
ds 2
a(t) = κ(t) (t) N̂(t)
dt
As a(t) is a constant, its direction, N̂(t), is also a constant. The normal vector
to a spiral is not constant.
5.2.2. ∗. Solution. (a) False. For any constant vector a = (a1 , a2 , a3 ),
 
ı̂ı ̂ k̂
a × r = det a1 a2 a3  = (a2 z − a3 y)ı̂ı − (a1 z − a3 x)̂ + (a1 y − a2 x)̂
x y z
So
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (a × r) = det 
 
∂x ∂y ∂z 
a2 z − a3 y −a1 z + a3 x a1 y − a2 x
= 2a1ı̂ı + 2a2̂ + 2a3 k̂
is nonzero, unless the constant vector a = 0.

(b) False. For example, if f (x) = x2 , then
∇x2 ) = ∇ · (2xı̂ı) = 2
∇f ) = ∇ · (∇
∇ · (∇
(c) False. For example, if F = x2ı̂ı, then
∇ · F) = ∇ ∇ · (x2ı̂ı) = ∇ (2x) = 2ı̂ı

∇ (∇
(d) False. The trap here is that F need not be defined at the origin. We
mr
saw, in Example 3.4.2, that the point source F = |r| 3 had flux 4πm through
every sphere centred on the origin. We also saw, in Example 4.2.7, that the
divergence ∇ · F = 0 everywhere except at the origin (where it is not defined).
(e) True. Any simple, smooth, closed curve in R3 that avoids the origin
is the boundary of a surface S that also avoids the origin. Then, by Stokes’
theorem, I ZZ
F · dr = ∇ × F · n̂ dS = 0
C S

True. Let S =
(f) r |r − c| = R be a sphere. Denote by V =
r |r − c| ≤ R the ball whose boundary is S. Let H be one hemisphere of
S with outward pointing normal and let H 0 be the other hemisphere of S with
inward point normal. Then the boundary of V , with outward pointing normal,
can be viewed as consisting of two parts, namely H and −H 0 , where by −H 0 we
mean H 0 but with outward pointing normal. Then, by the divergence theorem
ZZ ZZ ZZ
F · n̂ dS − F · n̂ dS = F · n̂ dS
H H0
Z Z∂V
Z
= ∇ · F dV > 0
V
RR RR
which implies that H F · n̂ dS > H 0 F · n̂ dS.
(g) False. The trap here is that the curve is in R3 , not R2 . As we saw in
Example 1.4.4, a helix has constant curvature, but does not lie in a plane and
so is not part of a circle.
(h) False. Even if we restrict F to the xy-plane (i.e. to z = 0), this vector
field is not conservative. We saw that in Examples 2.3.14 and 4.3.8.
(i) False. For example, the vector field F = x k̂ is always parallel to the
z-axis. So its flow lines are also all parallel to the z-axis. But if the closed
curve C consists of the line segments
• L1 from (0, 0, 0) to (1, 0, 0), followed by
• L4 from (0, 0, 1) back to (0, 0, 0).

z
L3
(1, 0, 1)
L4
L2
y
L1
x
then
R R1
• L1
F · dr = 0
(xk̂) · ı̂ı dx = 0 since k̂ ⊥ ı̂ı, dr = ı̂ı dx on L1 and
R R1
• L2
F · dr = 0
(1k̂) · k̂ dz = 1 since x = 1 and dr = k̂ dz on L2 and
R R1
• L3
F · dr = − 0
(xk̂) · ı̂ı dx = 0 since k̂ ⊥ ı̂ı and
R R1
• L4
F · dr = − 0
(0k̂) · k̂ dz = 0 since x = 0 on L4 .
All together
Z Z Z Z Z
F · dr = F · dr + F · dr + F · dr + F · dr = 1
C L1 L2 L3 L4
(j) True. If the speed |v| is constant then
d 2 d
0= |v| = (v · v) = 2v · a
dt dt
5.2.3. ∗. Solution. (a) False. r00 (t) is the full acceleration. So |r00 (t)| is
the magnitude of the full acceleration, not just the tangential component of
acceleration. For example, if r(t) = cos tı̂ı + sin t ̂ (i.e. the particle is just going
around in circles), the acceleration r00 (t) = − cos tı̂ı − sin t ̂ is perpendicular to
the direction of motion. So the tangential component of acceleration is zero,
while |r00 (t)| = 1.
(b) T̂(t) is the tangent vector to the curve at r(t). N̂(t) and B̂(t) are both
perpendicular to T̂(t) (and to each other) and so span the plane normal to the
curve at r(t).
(c) True. This is (half of) Theorem 2.4.8.
(d) False. The statement ∇ × (∇ ∇ · F ) = 0 is just plain gibberish, because
∇ · F is a scalar valued function and there is no such thing as the curl of a
scalar valued function.
(e) False. For example if F = ı̂ı, then, by the divergence theorem,
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S V

Here V = x, y, z x2 + y 2 + z 2 ≤ 1 is the inside of the sphere.
(f) True. If S is the boundary of the solid region E, then we can orient S
by always choosing the normal vector that points into E.
5.2.4. ∗. Solution. (a) The helix is approximately a bunch of circles stacked
one on top of each other. The radius of the circles increase as z increases. So
the curvature decreases as z increases.
(b) Here are two arguments both of which conclude that f (x) is D.
• If C were the graph y = f (x), then f 0 (x) would have two points of
discontinuity. The curvature κ(x) would not the defined at those two
points. The function whose graph is D is defined everywhere and so

cannot be the curvature of the function whose graph is C.
• The function whose graph is D has two inflection points. So its curvature
is zero at two points. The function whose graph is C is indeed zero at
two points (that in fact correspond to the inflection points of D). So D
is the graph of f (x) and C is the graph of κ(x).
(c) For any fixed y, x2 +z 2 = 1 is a circle of radius 1. So we can parametrize

it by x(θ) = cos θ, z(θ) = sin θ, 0 ≤ θ < 2π. The y-coordinate of any point on
the intersection is determined by y = xz. So we can use
r(θ) = cos θ ı̂ı + sin θ k̂ + sin θ cos θ ̂ 0 ≤ θ < 2π
(d) We are told that the helical ramp starts starts with the y-axis when
z = 0.
• In the cases of parametrisations (a) and (c), z = 0 forces u = 0 and u = 0
forces x = y = 0. That is only the origin, not the y-axis. So we can rule
out (a) and(c).
• In the case of parametrisation (b), z = 0 forces v = 0 and v = 0 forces
y = 0 and x = u. As u varies that sweeps out the x-axis, not the y-axis.
So we can rule out (b).
• In the case of parametrisation (d), z = 0 forces v = 0 and v = 0 forces
x = 0 and y = u. As u varies that sweeps out the y-axis, which is what
we want.
Furthermore
• we are told that z = v runs from 0 to 5 and that
• x2 + y 2 = u2 ≥ 4
So we want parametrisation (d) with domain |u| ≥ 2, 0 ≤ v ≤ 5.
(e) Straight lines have curvature 0. So one acceptable parametrized curve
is r(t) = tı̂ı, 0 ≤ t ≤ 1.
(f) The cube S has six sides. So the outward flux through ∂S is 6 and, by
ZZ ZZZ ZZZ
6= F · n̂ dS = ∇ · F dV = C dV = C
∂S S S
since S has volume one. So C = 6.

(g) For the vector field F to be conservative, we need
∂F1 ∂F2
=
∂y ∂x
∂ ∂
⇐⇒ (ax + by) = (cx + dy)
∂y ∂x
⇐⇒ b=c
When b = c, an allowed potential is a2 x2 + bxy + d2 y 2 . The specified set is

(a, b, c, d) a, b, c, d all real and b = c
(h) By the definition of arclength parametrisation, the arclength along the

curve between r(0) and r(s) is s. In particular, the arclength between r(0) and
r(3) is 3 and the arclength between r(0) and r(5), which is the same as the
arclength between r(0) and r(3) plus the arclength between r(3) and r(5), is
5. So the arclength between r(3) and r(5) is 5 − 3 = 2.
(i) In this solution, we’ll use, for example −T to refer to the curve T , but
with the arrow pointing in the opposite direction to that of the arrow on T .
In parts (2), (3) and (4) we will choose F to be the vector field
y x
G(x, y) = − ı̂ı + 2 ̂
x2 + y 2 x + y2
We saw, in Example 2.3.14, that ∇ × G = 0 exceptH at the origin where it

is not defined. We also saw, in Example 4.3.8, that C G · dr = 2π for any
counterclockwise oriented circle centred on the origin.
• (1): Let R1 be the region between S and T . It is the shaded region in

the figure on the left below. Note that R1 is contained in the domain
of F, so that ∇ × F = 0 on all of R1 . The boundary of R1 is S − T ,
meaning that the boundary consists of two parts, with one part being S
and the other part being −T . So, by Stokes’ theorem
Z Z Z ZZ
F · dr − F · dr = F · dr = ∇ × F · k̂ dS = 0
S T ∂R1 R1
and (1) is true.
R U
S
R1
T
R2
• (2): False. Choose a coordinate system so that Q is at the origin and

choose F = G. We saw, in Examples 2.3.14 and 4.3.8, that the curl of G
vanished
R everywhere except at the origin, where it was not defined, but
that R G · dr 6= 0.
• (3),(4): False. Here is a counterxample that shows that both (3) and
(4) are false. Choose a coordinate system so that Q is at the origin and
choose F = G. By Stokes’ theorem
Z Z
G · dr = G · dr = 0
S T
because ∇ × G = 0 everywhere inside S, Rincluding atRP . So now both

parts (3) and (4) reduce to the claim that U G · dr = R G · dr.
R
We saw, in Example 4.3.8, that R G · dr = 2π.
R
To finish off the counterexample, we’ll now show that U G · dr = −2π.
Let R2 be the region between U and R. It is the shaded region in the
figure on the right above. Note that ∇ × G = 0 on all of R2 . including at
P . The boundary of R2 is −U − R, meaning that the boundary consists

of two parts, with one part being −U and the other part being −R. So,
by Stokes’ theorem
Z Z Z ZZ
− G · dr − G · dr = G · dr = ∇ × G · k̂ dS = 0
U R ∂R2 R2
R R
and U
G · dr = − R
G · dr = −2π
R
• (5): False. For any conservative vector field F, with potential f , V F · dr
is just the difference of the values of f at the two end points of V . It is
easy to choose an f for which those two values are different. For example
f (x, y) = x does the job.
(j) Let S be any closed surface and denote by V the volume that it encloses.
Presumably the question assumes that S is oriented so that S = ∂V . Then by
the divergence theorem
ZZ ZZ ZZZ
F · n̂ dS = F · n̂ dS = ∇ · F dV
S ∂V V
This is exactly the volume of V if ∇ · F = 1 everywhere. One vector field F

with ∇ · F = 1 everywhere is F = xı̂ı.
(k) Let C be the counterclockwise boundary of a small square centred on
P , like the blue curve in the figure below, but much smaller. Call the square
(the inside of C) S.
P
C
By Stokes’ theorem
ZZ I
∇ × F · k̂ dS = F · dr
S C
H
• The contribution to C F · dr coming from the left and right sides of C
will be zero, because F is perpendicular to dr there.
H
• The contribution to C F · dr coming from the top of C will be negative,
because there F is a positive number times ı̂ı and dr is a negative number
times ı̂ı.
H
• The contribution to C F·dr coming from the bottom of C will be positive,
because there F is a positive number times ı̂ı and dr is a positive number
times ı̂ı.
• The magnitude of the contribution from the top of C will be larger than
the magnitude of the contribution from the bottom of C, because |F| is
larger on the top than on the bottom.
H
So, all together, C F · dr < 0, and consequently (taking a limit as the square
size tends to zero) ∇ × F · k̂ is negative at P .
5.2.5. ∗. Solution. (a) False. We could have, for example, ∇ · F zero
at one point and strictly positive elsewhere. One example would be F =
x3 ı̂ı + y 3 ̂ + z 3 k̂, with S1 and S2 being the upward oriented top and bottom
hemispheres, respectively, of the unit sphere x2 + y 2 + z 2 = 1.
(b) False. The conditions that (1) ∇ × F = 0 and (2) the domain of
F is simply-connected, are sufficient, but not necessary, to imply that F is
conservative. For example the vector field F = 0, with any domain at all,
is conservative with potential 0. Another example (which does not depend
on choosing a domain that is smaller than the largest possible domain) is
1

F = ∇ x2 +y 2 with domain (x, y, z) (x, y) 6= (0, 0) . That is, the domain is
R3 with the z-axis removed.
(c) That’s true. Consider any point r(t0 ) on a parametrized curve r(t).
That’s the blue point in the figure below.
r(t0 )
T̂
N̂
c
The centre of curvature for the curve at r(t0 ) is c = r(t0 ) + ρ(t0 )N̂(t0 ). It
is the red dot in the figure.
• The radius of the osculating circle is the distance from its centre, c, to
any point of the circle, like r(t0 ). That’s |r(t0 ) − c| = |ρ(t0 )N̂(t0 )| =
ρ(t0 ). The curvature of the osculating circle is one over its radius. So its
curvature is ρ(t10 ) = κ(t0 ).
• The unit normal to the osculating circle at r(t0 ) is a unit vector in the
opposite direction to the radius vector from the centre c to r(t0 ). The
radius vector is r(t0 ) − c0 = −ρ(t0 )N̂(t0 ), so the unit normal is N̂(t0 ).
• The osculating circle lies in the plane that best fits the curve near r(t0 ).
(See the beginning of §1.4.) So the unit tangents to the osculating circle
at r(t0 ) are perpendicular to both N̂(t0 ) and B̂(t0 ) and so are either
T̂(t0 ) or −T̂(t0 ), depending on how we orient the osculating circle.
(d) False. Kepler’s third law is that a planet orbiting a sun has the square
of the period proportional to the cube of the major axis of the orbit.
(e) True. That’s part (a) of Theorem 4.1.7.
(f) True. Every domain contains closed surfaces. This has nothing to do
with vector fields.
(g) True. We saw this in Example 2.3.4.
(h) False. Let F be an everywhere defined conservative vector field with
potential ϕ. Then ∇ × F = 0 everywhere. If P and Q are R two points and if
ϕ(P ) − ϕ(Q) = 3 and if C is a curve from Q to P , then C F · dr = 3. One
example would be ϕ(x, y, z) = x, F = ı̂ı, P = (3, 0, 0), Q = (0, 0, 0).
(i) False. The normal component of acceleration depends on speed, as well
as curvature.
(j) False. The curve r1 contains only points in the xy-plane. Every r2 (t)
with t 6= 0 has a nonzero z-coordinate.
5.2.6. ∗. Solution. (a) False. Changing the orientation of a surface does
not change dS at all. (It changes n̂dS by a factor of (−1).) So
ZZ ZZ
f dS = + f dS
S −S
RR
which is not − −S f dS, unless the integral is zero.
(b) False. For every vector field with two continuous partial derivatives,
∇ × F) = 0 (see Theorem 4.1.7.a), so the divergence theorem gives
∇ · (∇
ZZ ZZZ
∇ × F) · n̂ dS =
(∇ ∇ × F) dV = 0
∇ · (∇
S V
whether or not F is conservative.

(c) True. Define the vector field F = f ı̂ı. Then, by Stokes’ theorem,
Z Z ZZ ZZ
∂f ∂f
f dx = F · dr = ∇ × F · n̂ dS = ̂ − k̂ · n̂ dS
C C S S ∂z ∂y
(d) True. The left hand side, (∇∇f ) × (∇∇f ), is zero because (∇
∇f ) is parallel
to itself and the right hand side ∇ × (∇ ∇f ) is zero by Theorem 4.1.7.b (the
screening test for conservative fields).
(e) True. The curve r(t) = 2 , 0 , 1 + t3 4 , −1 , −2 is a straight line.
Straight lines have curvature 0.
(f) True. In general |r0 (t)| = ds
dt . Under arc length parametrization t = s
ds
so that dt = 1.
(g) True. If F is a constant vector fleld, then, by the divergence thoerem,
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = 0 dV = 0
S V V
(h) False. The statement ∇ ×F = (x, y, z) means that F is a vector potential

for the vector field G = (x, y, z). But G fails the screening test ∇ · G = 0 for
vector potentials.
5.2.7. ∗. Solution. (a) P is the x-component of F. As we travel vertically
upward through A, that x-component decreases. Hence Py < 0 at A.
(b) Q is the y-component of F. As we travel horizontally to the right
through A, that y-component increases. Hence Qx > 0 at A.
(c) ∇ × F = (Qx − Py )k̂ and Qx − Py > 0 at A, so that the curl of F at A
is in the direction of +k̂.
(d) Along the curve C1 the magnitude of the angle between F and dr is less
than 90◦ , so that F · dr > 0 and C1 F · dr > 0.
R
(e) Along the curve C2 the magnitude R of the angle between F and dr is
greater than 90◦ , so that F · dr < 0 and C2 F · dr < 0.
R R
(f) If F were conservative, we would have C1 F·dr = C2 F·dr. As these two
integrals have opposite signs F is not conservative. (Since F is not conservative,
it is not the gradient of some function. At A, Px > 0 and Qy > 0. So F is not
divergence free and is not the curl of a vector potential.)
5.2.8. ∗. Solution. (a) False. The curve r1 contains only points with z ≥ 0.
Every r2 (t) with t < 0 has z < 0.
(b) True. r2 (t2 ) = r1 (t) and t2 runs from 0 to 1 as t runs from 0 to 1.
(c) True. In general |r0 (t)| = ds ds
dt . When t = s, dt = 1.
(d) False. The curve need not even lie in a plane. For example, as we saw in
Example 1.4.4, the helix r(t) = a cos tı̂ı + b sin t ̂ + bt k̂ has constant curvature
a
κ = a2 +b 2 but is not a circle.
√
(e) True. If the speed |v| = v · v of a moving object is constant, then
d
0= v · v = 2v · a
dt
(f) False. If the vector field F(x, y, z) = x2−y ı x


+y 2 ı̂ + x2 +y 2 ̂ +z k̂ were conserva-
−y x
tive, its restriction, x2 +y2 ı̂ı + x2 +y2 ̂, to the xy-plane would also be conservative.
But we saw in Examples 2.3.14 and 4.3.8 that the vector field x2−y ı x

+y 2 ı̂ + x2 +y 2 ̂
is not conservative.
(g) False. The vector field of part (f), with domain (x, y, z) x2 +y 2 > 1 ,
provides a counterexample.
(h) False.
2The curve x 2 + y 2 = 2 can not be shrunk to a point continuously
2
in (x, y) x + y > 1 .

2
(i) True. Any curve in 2(x, y) y > x can be shrunk to a point

continuously in (x, y) y > x .
(j) True. By the divergence theorem,
ZZ ZZZ
∇ × F · n̂ dS = ∇ × F) dV = 0
∇ · (∇
S E
∇ × F) = 0 by the vector identity of Theorem 4.1.7.a.

since ∇ · (∇
5.2.9. ∗. Solution. (a) True. In general |r0 (t)| = ds ds

dt . When t = s, dt = 1.
(b) False. The curve need not even lie in a plane. For example, as we saw in
Example 1.4.4, the helix r(t) = a cos tı̂ı + b sin t ̂ + bt k̂ has constant curvature
a
κ = a2 +b 2 but is not a circle.
(c) True. See Theorem 2.4.7.

(d) False. The vector field F(x, y, z) = x2−y ı x

+y 2 ı̂ + x2 +y 2 ̂ , with domain
(x, y, z) x2 + y 2 > 1

provides a counterexample.
(e) False. The curve r1 contains only points with z ≥ 0. Every r2 (t) with
t < 0 has z < 0.
(f) True. r2 (t2 ) = r1 (t) and t2 runs from 0 to 1 as t runs from 0 to 1.
(g) True. ∇ · (∇∇ × F) = 0 by the vector identity of Theorem 4.1.7.a.
(h) False. A counterexample is f (x, y, z) = x2 . It has ∇ f = 2xı̂ı and hence
∇f ) = 2.
∇ · (∇
False.
(i) The curve x 2 + y 2 = 2 can not be shrunk to a point continuously
in (x, y) x + y 2 > 1 .
2

(j) True. Any curve in (x, y) y > x2 can be shrunk to a point

continuously in (x, y) y > x2 .
5.2.10. ∗. Solution. (a)R False. ∇ f = 0 if and only if f is constant. But if
f is the constant K, then C f ds is K times the length of C, which need not
be zero.
(b) False. Any curve which lies in a plane has constant binormal. For
example, the circle r(t) = cos tı̂ı + sin t ̂ + 0 k̂ has constant binormal B̂ = k̂,
but is not a straight line.
2
(c) True. If r(t) has constant speed, the ds dt (t) = r0 (t) · r0 (t) is constant
and
d 0
r (t) · r0 (t) = 2r0 (t) · r00 (t)

0=
dt
R
(d) False. For the line integral C F × G · dr to be independent of the
path C, the vector field F × G has to be conservative and so has to obey
∇ × (F × G) = 0. But
• Not all vector fields are conservative. For example, the vector field H =
x ̂ obeys ∇ × H = k̂ and so is not conservative.
• We can make F×G be any vector field through judicious choices of F and
G. For example, if F = x k̂ and G = ı̂ı, then F × G = x k̂ × ı̂ı = x ̂ = H.
R
(e) True. The contribution to C f ds from an “infinitesmal piece of C” is
the value of f on the piece times the length of the piece. That does not depend
on the orientation of the piece.
∂r ∂r
(f) False. The two vectors in the cross product ∂u × ∂u are identical. So
the cross product is 0.
(g) False. The integral is completely independent of x(u, v) and y(u, v).
In particular if, for example, x(u, v) = 157u, y(u, v) = 157v, z(u, v) = 0 then
RR ∂z 2
1/2
∂z 2
D
1 + ∂u + ∂v dudv is always exactly the area of D, while the
area of S is 1572 times the area of D.
(h) True. If the fluid is incompressible then its flow preserves volumes and
consequently ∇ · F = 0.
(i) Not only False, but Ridiculous. The left had side is scalar valued while
the right hand side is vector valued.
5.2.11. ∗. Solution. (a) True. That ∇ · (∇ ∇ × F) = 0 is the vector identity
of Theorem 4.1.7.a. That identity is the basis of the vector potential screening
test. R
(b) False. If F is not conservative, then C F · dr will depend on the end-
points of C.
(c) True. If ∇ f = 0, then
∂f
(x, y, z) = 0 =⇒ f (x, y, z) = g(y, z)
∂x
∂f ∂g
(x, y, z) = 0 =⇒ (y, z) = 0 =⇒ g(y, z) = h(z)
∂y ∂y
∂f
(x, y, z) = 0 =⇒ h0 (z) = 0 =⇒ h(z) = C
∂z
for some functions g(y, z), h(z) and constant C.
(d) False. The curl ∇ × F is zero for every conservative vector fields F.
There are many nonconstant conservative vector fields, like F(x, y, z) = xı̂ı.
(e) True. As S is closed, it is the boundary of a solid region V . Then, by
ZZ ZZZ
F · n̂ dS = ∇ · F dV = 0
S V
R
(f) True. If C F · dr = 0 for every closed curve C, then F is conservative
by Theorem 2.4.7. Consequently, ∇ × F = 0 by Theorem 2.3.9.
(g) True. If the speed |v| is constant then
d 2 d
0= |v| = (v · v) = 2v · a
dt dt
v
Since T̂ = |v| , T̂ · a = 0 too. Here, we have assumed that the constant |v| is
not zero. If the constant |v| is zero, then T̂ is not defined at all (and a = 0).
(h) False. The trap here is that the curve is in R3 , not R2 . As we saw in
(i) False. The trap here is that we are told nothing about ∇ · F. As an
example, let S1 be the hemisphere
S1 = (x, y, z) x2 + y 2 + z 2 = 1, z ≥ 0

with upward pointing normal and S2 be the disk
S2 = (x, y, 0) x2 + y 2 ≤ 1

also with upward pointing normal.

n̂
V n̂ S1
S2
Set p
(x, y, z) 0 ≤ z ≤ x2 + y 2 , x2 + y 2 ≤ 1

V =
Then the boundary, ∂V , of V consists of two parts, namely S1 (with nor-
mal pointing upwards) and S2 (but with normal pointing downwards). The
divergence theorem (Theorem 4.2.2) gives
ZZ ZZ ZZZ
F · n̂ dS − F · n̂ dS = ∇ · F dV
S1 S2 V
RR RR
If ∇ · F > 0 (as is the case, for example, if F = xı̂ı) then S1
F · n̂ dS − S2
F·
n̂ dS is definitely nonzero.
(j) True. This is one of Kepler’s laws. See §1.9.
5.2.12. ∗. Solution. It’s (b). (a) is gibberish — the left hand side is a scalar
while the right hand side is a vector. (c) is also gibberish — the left hand side
is a vector while the right hand side is a scalar. (b) is the vector identity of
Theorem 4.1.4.c.
5.2.13. ∗. Solution. (a) False. For example, if f (x, y, z) = x2 , then ∇ f =
2xı̂ı and ∇ · ∇ f = 2.
(b) Not only false, but ridiculous. The left hand side is a vector while the
right hand side is a scalar.
(c) Not only false, but ridiculous. The right hand side is a vector while the
left hand side is a scalar.
(d) True. That’s the screening test for conservative fields, Theorem 4.1.7.b.
(e) Not only false, but ridiculous. The curl of a scalar function is not
defined.
(f) True. That’s the screening test for vector potentials, Theorem 4.1.7.a.
(g) False.
r ∂ x ∂ y ∂ z
∇· = + +
|r|2 ∂x x2 + y 2 + z 2 ∂y x2 + y 2 + z 2 ∂z x2 + y 2 + z 2
1 2x2 1 2y 2
= − 2 + − 2
x2 + y 2 + z 2 [x2 + y 2 + z 2 ] x2 + y 2 + z 2 [x2 + y 2 + z 2 ]
1 2z 2
+ 2 − 2
x + y2 + z2 [x2 + y 2 + z 2 ]
3[x2 + y 2 + z 2 ] − 2x2 − 2y 2 − 2z 2
= 2
[x2 + y 2 + z 2 ]
1
=
x2 + y 2 + z 2
(h) False. For any constant vector ω = (ω1 , ω2 , ω3 ),

 
ı̂ı ̂ k̂
ω × r = det ω1 ω2 ω3 
x y z
= (ω2 z − ω3 y)ı̂ı − (ω1 z − ω3 x)̂ + (ω1 y − ω2 x)̂
So
 
ı̂ı ̂ k̂
∂ ∂ ∂
∇ × (ω
ω × r) = det 
 
∂x ∂y ∂z 
ω2 z − ω3 y −ω1 z + ω3 x ω1 y − ω2 x
= 2ω1ı̂ı + 2ω2̂ + 2ω3 k̂
is nonzero, unless the constant vector ω = 0.

(i) True. The given equation is equivalent (by the vector identity Theorem
4.1.4.c) to ZZZ ZZ

∇ · f F dV = f F · n̂ dS
Ω ∂Ω
which is true by the divergence theorem.
(j) False. One of the variants of the divergence theorem given in Theorem
4.2.9 is ZZ ZZZ
f n̂ dS = ∇ f dV
∂Ω Ω
Note that the sign on the right hand side is “+”, not “−”. RRR In order for the
equation given in part (j) to be true, it would be necessary that Ω
∇ f dV = 0
for all smooth scalar functions f . That’s silly. One counterexample is
Ω = (x, y, z) x2 + y 2 + z 2 ≤ 1

f (x) = x
Then
n̂
ZZ z ZZ }| { ZZ
f n̂ dS = x(xı̂ı + y ̂ + z k̂) dS = ı̂ı x2 dS
∂Ω ∂Ω ∂Ω
ZZZ ZZZ ZZZ
− ∇ f dV = − ı̂ı dV = −ı̂ı dV
Ω Ω Ω
The coefficient of ı̂ı is obviously strictly positive in the upper integral and
strictly negative in the lower integral.
5.2.14. ∗. Solution. (a) True. If the vector field is F = aı̂ı + b ̂ + c k̂, then
f (x, y, z) = ax + by + cz obeys F = ∇ f and so is a potential for F.
(b) False. For example the vector field F = xı̂ı − y ̂ obeys ∇ · F = 0 but is
not a constant vector field.

dr
(c) True, assuming that r(t) is not indentically 0. If r(t) and dt are orthog-
onal at all points of the curve C, then
d dr
r(t) · r(t) = 2r(t) · (t) = 0
dt dt
So x(t)2 + y(t)2 + z(t)2 = r(t) · r(t) is a constant. If r(t) is not indentically 0,
that constant must be strictly positive. That is x(t)2 + y(t)2 + z(t)2 = a2 for
some constant a > 0.
(d) False. The curvature (see §1.5) is

dT̂
dt (t)
κ(t) = 0
|r (t)|
Changing the orientation of the curve amounts to replacing t by −t. This

changes the signs of T̂ and r0 , but does not change κ, because the absolute
values eliminate the signs.
(e) False.
For example, the vector field F = 0, with domain (x, y, z) x2 +
y 2 > 0 is a conservative vector field (with potential 0) whose domain is not
1
simply connected. As a less nitpicky example, let F = ∇ f with f = x2 +y 2 . The
2 2

biggest possible domain for this vector field is also (x, y, z) x + y > 0 .

5.2.15. ∗. Solution. (a) We are to compute the divergence of x2 y ı̂ı +

ey sin x ̂ + ezx k̂. Since
∂ 2
x y = 2xy
∂x
∂ y
e sin x = ey sin x

∂y
∂ zx
e = xexz
∂z
∇ · x2 y ı̂ı + ey sin x ̂ + ezx k̂ = 2xy + ey sin x + xexz

(b) The specified curl is

 
ı̂ı ̂ k̂
 ∂ ∂ ∂ 
∇ × cos x2 ı̂ı − y 3 z ̂ + xz k̂ = det  ∂x = y 3 ı̂ı − z ̂

∂y ∂z 
cos x2 3
−y z xz

(c) In principle, the domain could be any subset of (x, y, z) x2 +y 2 > 0 .
We are not told which subset to use, so, by default, D is the maximal domain
D = (x, y, z) x2 + y 2 > 0 = (x, y, z) (x, y) 6= (0, 0)

This D is connected (any two points in D can be joined by a curve that

lies completely in D) but is not simply connected (the simple closed curve
r(θ) = cos θ ı̂ı + sin θ ̂, 0 ≤ θ ≤ 2π lies in D but cannot be shrunk to a point
continuously in D). So (I) and (IV) are true. That’s (iii).
(d) False. If the position of the particle at time t is r(t) = cos tı̂ı + sin t ̂,
then its speed is the constant 1 but its acceleration is − cos tı̂ı − sin t ̂, which
is nonzero.
5.2.16. ∗. Solution. (a) True. By the vector identity of Theorem 4.1.5.c,
∇f ) = (∇
∇ × (f∇ ∇f ) × (∇
∇f ) + f ∇ × (∇
∇f ) = 0
The second term vanished because of the screening test vector identity of The-
orem 4.1.7.b.
(b) True. That’s the vector identity of Theorem 4.1.4.c.
(c) True. To have constant curvature 0 the curve must have unit tangent
vector T̂(s) obeying
dT̂
(s) = 0
ds
(See §1.5.) So r0 (s) = T̂(s) must be a constant vector. Call it T̂0 . Integrating
gives
r(s) = sT̂0 + r0
for some constant vector r0 . So r(s) lies on the same straight line for all s.
(d) False. The trap here is that the curve is in R3 , not R2 . As we saw in
(e) True. The vector field F = ∇ f is conservative. So, by Theorem 2.4.7.b,
the work integral Z Z
∇ f · dr = F · dr = 0
C C
for any closed curve C, and, in particular, for any circle C.
(f) True. The statement that “the flux out of one hemisphere is equal to
the flux into the opposite hemisphere” is equivalent to the statement that “the
flux out of the sphere is equal to zero”. Since ∇ · F = 0 everywhere, that is
true by the divergence theorem.
(g) True. Let S be the boundary of the solid region V . Then, by the
divergence theorem (Theorem 4.2.9),
ZZ ZZZ

∇ × F · n̂ dS = ∇ · ∇ × F dV
S V

But ∇ · ∇ × F is identically zero, by the screening test vector identity of
Theorem 4.1.7.a. So the integral is zero.
5.2.17. ∗. Solution. (a) True. Let F be the vector field. We are assuming
that ∇ × F = 0 on all of R3 . As a result, F = ∇φ for some potential function
∂2 ∂2 ∂2
φ. We are also assuming that 0 = ∇ · F = ∇ · ∇φ = ∂x 2 + ∂y 2 + ∂z 2 φ. This
is the definition of “φ is harmonic”.

(b) False. Let F be the vector field. We are assuming that F = ∇φ for
some potential function φ. If S is any smooth closed surface, with S being the
boundary of the solid V , then, by the divergence theorem, the outward flux of
F through S is
ZZ ZZZ ZZZ
F · n̂ dS = ∇ · F dV = ∇ · ∇φ dV
S
Z Z ZV V
∂2 ∂2 ∂2

= ∂x2 + ∂y 2 + ∂z 2 φ dV
V
2
∂ ∂2 ∂2

If, for example, φ = x2 , then ∂x 2 + ∂y 2 + ∂z 2 φ = 2 and the flux of F through
S is twice the volume of V , which is not zero.

5.2.18. ∗. Solution. (a) True. The vector field ∇ f is conservative and the
work done by a conservative field around any closed curve is zero.
(b) False. By the vector identity Theorem 4.1.7.a, we have
∇ × F) = 0
∇ · (∇
for all vector fields F. But ∇ · (xı̂ı + y ̂ + z k̂) = 3.

R
5.2.19. ∗. Solution. (a) C ∇ f · dr = 0 is the work done along the curve
using the conservative force ∇ f . That work is difference between the potential
f at the final point minus the potential f at the initial point. If the final and
initial points are both on the level surface f (x, y, z) = 0, that difference is zero.
(b) The rate of change of the specified vector is
d
v(t) × r(t) = v0 (t) × r(t) + v(t) × v(t)
dt
The first term vanishes because v0 (t) = a(t) = f (t)r(t) is parallel to r(t). The
second term vanishes because v(t) = v(t).
(c) Call the constant vector v × r of part (b) N. This vector is a constant
and is perpendicular to both v(t) and r(t). In particular
N · r(t) = 0
Assuming that N is nonzero, this is the equation of the plane through the
origin with normal vector N.
(d) Yes, as long as T̂, N̂, and B̂ are well-defined, since B̂ = T̂ × N̂.
2 2
(e) No. When the maximum speed occurs ddt2s = 0 so that a = κ(t) ds dt (t) N̂(t).
If the speed and (constant) curvature are nonzero, the acceleration is nonzero.
5.2.20. ∗. Solution. We apply Green’s Theorem:
Z ZZ
∂F2 ∂F1
F1 dx + F2 dy = − dxdy
C R ∂x ∂y
Z ZZ
1 1
(a) −y dx + x dy = 1 − (−1) dx dy = Area(R)
2 ZC 2 Z ZR
1 1
(b) −x dx + y dy = 0 dx dy = 0 6= Area(R)
2
Z C ZZ 2 R

(c) y dx = − 1 dx dy = −Area(R) 6= Area(R)
ZC R ZZ

(d) 3y dx + 4x dy = 4 − 3 dx dy = Area(R)
C R
5.2.21. ∗. Solution. (a) True . Since v = |v| = 1 is constant, we have
dv
a= T̂ + v 2 κN̂ = 0T̂ + κN̂.
dt
Thus 1 = |a| = κ|N̂|, i.e., κ = 1.

(a) (Again.) Since v · v = |v|2 = 1 for all t, differentiation gives v · a = 0,
i.e., v ⊥ a always. It follows that |v × a| = |v| |a| sin θ = 1 always, because the
angle θ here is always π/2. Thus, for all t,
|v × a| 1
κ= = = 1.
|v|3 1
(b) True . By the divergence theorem, if V is the solid bounded by S,

ZZ ZZZ

∇ × F · n̂ dS = ∇ · ∇ × F dV = 0
S V

since ∇ · ∇ × F = 0.
(c) False . If F = 0 and G is any nonzero, conservative field, like G = 2xı̂ı =
∇(x2 ), then I I
F · dr = G · dr = 0
C C
for every closed curve C.
5.2.22. ∗. Solution. (a) Define Ω (t) = r(t) × v(t). Then by the product
rule,
Ω
dΩ dr dv
= ×v+r×
dt dt dt
= v × v + r × f (r, v)r .

= 0 + f (r, v) r × r = 0.
It follows that Ω is constant.

(b) By the divergence theorem, where R is the solid cylinder as described,
ZZ ZZZ ZZZ
2

(xı̂ı − y ̂ + z k̂) · n̂dS = 1 − 1 + 2z dV = 2 z dV
S R R
The solid R clearly has reflection symmetry across the plane z = 2. So the
z-coordinate of the centre of mass of R, i.e. the average value of z over R, i.e.
RRR RRR
R
z dV R
z dV
z̄ = RRR =
R
dV Vol(R)
is 2. Hence
ZZ
(xı̂ı − y ̂ + z 2 k̂) · n̂dS = 2z̄ Vol(R) = 4 Vol(R)
S
By basic geometry, Vol(R) = πr2 h = πb2 2. Hence

ZZ
(xı̂ı − y ̂ + z 2 k̂) · n̂dS = 8πb2
S
(c) By Stokes’ theorem (Theorem 4.4.1),

I ZZ ZZ ZZ
F · dr = G · n̂ dS =⇒ ∇ × F · n̂ dS = G · n̂ dS
∂D D D D
ZZ

=⇒ ∇ × F − G · n̂ dS = 0
D
for all disks D. Because this is true for all disks D, the integrand must be
zero. To see this, let H = ∇ × F − G. Suppose that H(x0 ) 6= 0. Pick a unit
vector n̂ in the direction of H(x0 ). Let D be a very small flat disk centered
on x0 with normal n̂ (the vector we picked). Then H(x0 ) · n̂ > 0 and, by
RR H(x) · n̂ >0 for all x on D, if we have picked D small enough.
continuity,
Then D ∇ × F − G · n̂ dS > 0, which is a contradiction. So we conclude
that ∇ × F − G = 0 and hence G = ∇ × F.

Combined CLP 4

Uploaded by

Document Information

Original Title

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

Combined CLP 4

Uploaded by

Copyright:

Available Formats

CLP-4 Vector Calculus

CLP-4 Vector Calculus

August 23, 2022

Edition: CLP4 Vector Calculus: May 2021

2. Is the problem in the online version or the PDF version or both?

Feedback about the text vi

3 Surface Integrals 123

4 Integral Theorems 178

4.3 Green’s Theorem . . . . . . . . . . . . . . . . . . 235

5 True/False and Other Short Questions 299

B Hints for Exercises 339

C Answers to Exercises 359

D Solutions to Exercises 392

might be the position of a particle at time t. As t varies, r(t) sweeps out a

it is clearer to use arrows, like ~

is a parametrization of the circle (x − h)2 + (y − k)2 = a2 .

is a curve. We can choose to use y as the parameter and think of

is a parametrization for the given curve.

to do so. A more realistic example is the set of all (x, y, z) obeying

which is the blue curve in the figure above. Substituting x2 = z − 2y 2 (from

Then x2 = z − 2y 2 also gives x as a function of z. If x ≥ 0,

1.1 Derivatives, Velocity, Etc.

r0 (t) = x0 (t) , y 0 (t) , z 0 (t)

That is, to differentiate a vector valued function of t, just differentiate each of

a(t) = t2 ı̂ı + t4 ̂ + t6 k̂

We are about to compute some derivatives. To make it easier to follow what

γ(t) b(t) = t2 e−t ı̂ı + t2 e−3t ̂ + t2 e−5t k̂

= γ 0 (t)b(t) + γ(t)b0 (t)

a(t) · b(t) = t2 e−t + t4 e−3t + t6 e−5t

= 2tı̂ı + 4t3 ̂ + 6t5 k̂ · e−t ı̂ı + e−3t ̂ + e−5t k̂

+ t2 ı̂ı + t4 ̂ + t6 k̂ · − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂

= a0 (t) · b(t) + a(t) · b0 (t)

+ t2 ı̂ı + t4 ̂ + t6 k̂ × − e−t ı̂ı − 3e−3t ̂ − 5e−5t k̂

= a0 (t) × b(t) + a(t) × b0 (t)

a s(t) = (sin t)2 ı̂ı + (sin t)4 ̂ + (sin t)6 k̂

• γ(t) and s(t) be real valued differentiable functions of t ∈ R

As an application, we have the

position at time t = r(t) = x(t) , y(t) , z(t)

and the distance travelled between times T0 and T is

Example 1.1.6 Circumference of a circle. In general it can be quite

r(t) = 6 sin(2t)ı̂ı + 6 cos(2t)̂ + 5tk̂

Our curve is the helix

r(t) = 6 sin(2t)ı̂ı + 6 cos(2t)̂ + 5tk̂

is perpendicular to the radius vector, and hence parallel to the tangent

Before we go on to sketch the astroid and compute its perimeter, we can

r1 (t) = (t5 , t2 ) r01 (t) = (5t4 , 2t)

requires an infinite force. You might not look so healthy afterwards

Example 1.1.11 Corkscrew. We’ll find the arc length of

We’ll first sketch it, by observing that

r(t) = t cos tı̂ı + t sin t ̂ + t k̂

Our goal, stated at the beginning of this example, was to compute

We recognize that this is a trigonometric integral that contains an odd power

The integrand is now a rational function, i.e. a ratio of polynomials. So we

A circle of radius a rolls along the x-axis in the positive direction,

b Give the position of P as a function of θ.

5. The curve C is defined to be the intersection of the hyperboloid

Select the correct answer:

r(t) = a cos tı̂ı + a sin t ̂ + ct k̂

Describe the path of the particle.

Find the unit tangent vector to this parametrized curve at t = 1,

x = a cos t sin t y = a sin2 t z = bt

between t = 0 and t = T > 0.

b Compute the arc length of the curve from t = 0 to t = π.