Multivariable Calculus
George Cain & James Herod
(c)Copyright 1996, 1997 by George Cain & James Herod
All rights reserved
Table of Contents
Chapter One  Euclidean Three Space
1.1 Introduction
1.2 Coordinates in ThreeSpace
1.3 Some Geometry
1.4 Some More GeometryLevel Sets
Chapter Two  VectorsAlgebra and Geometry
2.1 Vectors
2.2 Scalar Product
2.3 Vector Product
Chapter Three  Vector Functions
3.1 Relations and Functions
3.2 Vector Functions
3.3 Limits and Continuity
Chapter Four  Derivatives
4.1 Derivatives
4.2 Geometry of Space CurvesCurvature
4.3 Geometry of Space CurvesTorsion
4.4 Motion
Chapter Five  More Dimensions
5.1 The space Rn
5.2 Functions
Chapter Six  Linear Functions and Matrices
6.1 Matrices
6.2 Matrix Algebra
Chapter Seven  Continuity, Derivatives, and All That
7.1 Limits and Continuity
7.2 Derivatives
7.3 The Chain Rule
Chapter Eight  f:Rn› R
8.1 Introduction
8.2 The Directional Derivative
8.3 Surface Normals
8.4 Maxima and Minima
8.5 Least Squares
8.6 More Maxima and Minima
8.7 Even More Maxima and Minima
Chapter Nine  The Taylor Polynomial
9.1 Introduction
9.2 The Taylor Polynomial
9.3 Error
Supplementary material for Taylor polynomial in several variables.
Chapter Ten  Sequences, Series, and All That
10.1 Introduction
10.2 Sequences
10.3 Series
10.4 More Series
10.5 Even More Series
10.6 A Final Remark
Chapter Eleven  Taylor Series
11.1 Power Series
11.2 Limit of a Power Series
11.3 Taylor Series
Chapter Twelve  Integration
12.1 Introduction
12.2 Two Dimensions
Chapter Thirteen  More Integration
13.1 Some Applications
13.2 Polar Coordinates
13.3 Three Dimensions
Chapter Fourteen  One Dimension Again
14.1 Scalar Line Integrals
14.2 Vector Line Integrals
14.3 Path Independence
Chapter Fifteen  Surfaces Revisited
15.1 Vector Description of Surfaces
15.2 Integration
Chapter Sixteen  Integrating Vector Functions
16.1 Introduction
16.2 Flux
Chapter Seventeen  Gauss and Green
17.1 Gauss's Theorem
17.2 Green's Theorem
17.3 A Pleasing Application
Chapter Eighteen  Stokes
18.1 Stokes's Theorem
18.2 Path Independence Revisited
Chapter Ninteen  Some Physics
19.1 Fluid Mechanics
19.2 Electrostatics
George Cain & James Herod
School of Mathematics
Georgia Institute of Technology
Atlanta, Georgia 303320160
1.1
Chapter One
Euclidean ThreeSpace
1.1 Introduction.
Let us briefly review the way in which we established a correspondence between
the real numbers and the points on a line, and between ordered pairs of real numbers and
the points in a plane. First, the line. We choose a point on a line and call it the origin.
We choose one direction from the origin and call it the positive direction. The opposite
direction, not surprisingly, is called the negative direction. In a picture, we generally
indicate the positive direction with an arrow or a plus sign:
Now we associate with each real number r a point on the line. First choose some
unit of measurement on the line. For r > 0, associate with r the point on the line that is a
distance r units from the origin in the positive direction. For r < 0, associate with r the
point on the line that is a distance r units from the origin in the negative direction. The
number 0 is associated with the origin. A moments reflection should convince you that
this procedure establishes a socalled onetoone correspondence between the real
numbers and the points on a line. In other words, a real number determines exactly one
point on a line, and, conversely, a point on the line determines exactly one real number.
This line is called a real line.
Next we establish a onetoone correspondence between ordered pairs of real
numbers and points in a plane. Take a real line, called the first axis, and construct another
real line, called the second axis, perpendicular to it and passing through the origin of the
first axis. Choose this point as the origin for the second axis. Now suppose we have an
ordered pair ( , ) x x
1 2
of reals. The point in the plane associated with this ordered pair is
1.2
found by constructing a line parallel to the second axis through the point on the first axis
corresponding to the real number x
1
, and constructing a line parallel to the first axis
through the point on the second axis corresponding to the real number x
2
. The point at
which these two lines intersect is the point associated with the ordered pair ( , ) x x
1 2
. A
moments reflection here will convince you that there is exactly one point in the plane thus
associated with an ordered pair (a, b), and each point in the plane is the point associated
with some ordered pair (a, b):
It is traditional to assume the point of view we have taken in this picture, in which
the first axis is horizontal, the second axis is vertical, the positive direction on the first
axis is to the right, and the positive direction on the second axis is up. We thus usually
speak of the horizontal axis and the vertical axis, rather than the first axis and the second
axis. We also frequently abuse the language by speaking of a point ( , ) x x
1 2
when, of
course, we actually mean the point associated with the ordered pair ( , ) x x
1 2
. The
numbers x
1
and x
2
are called the coordinates of the point x
1
is the first coordinate and
x
2
is the second coordinate.
Given any collection of ordered pairs( A collection of ordered pairs is called a
relation.), a picture of the collection is obtained by simply looking at the set of points in
the plane corresponding to the pairs in the given collection. Suppose we have an equation
1.3
involving two variables, say x and y. Then this equation defines a collection of ordered
pairs of numbers, namely all ( , ) x y that satisfy the equation. The corresponding picture
in the plane is called the graph of the equation. For example, consider the equation
y x
2 4
· . Let’s take a look at the graph of this equation. A little algebra (very little,
actually), convinces us that
{( , ): } {( , ): } {( , ): } x y y x x y y x x y y x
2 4 2 2
· · · ∪ · − ,
and we remember from the sixth grade that each of the sets on the right hand side of this
equation is a parabola:
What do we do with all this? These constructions are, of course, the bases of
analytic geometry, in which we join the subjects of algebra and geometry, to the benefit of
both. A geometric figure (a subset of the plane ) corresponds to a collection of ordered
pairs of real numbers. Algebraic facts about the collection of ordered pairs of real are
reflected by geometric facts about the subset of the plane, and, conversely, geometric
1.4
facts about the plane subset are reflected by algebraic facts about the collection of pairs of
reals.
Exercises
Draw a picture of the given relation:
1. R x y x y · ≤ ≤ ≤ ≤ {( , ): , } 0 1 1 4 and
2. R x y x x y x · − ≤ ≤ − ≤ ≤ {( , ): , } 4 4
2 2
and
3. R x y y x y y x · ≤ ≤ ∩ ≥ {( , ): } {( , ): } 1 2
2
4. S x y x y x · + · ≥ {( , ): }
2 2
1 0 , and
5. S x y x y x y y x · + ≤ ∩ ≤ {( , ): } {( , ): }
2 2 2
1
6. E r s r s · · {( , ):   }
7. T u v u v · + · {( , ):    } 1
8. R u v u v · + ≤ {( , ):    } 1
9. T x y x y · · {( , ): }
2 2
10. A x y x y · ≤ {( , ): }
2 2
1.5
11. G s t s t · · {( , ): { , } } max 1
12. B s t s t · ≤ {( , ): { , } } max 1
1.2 Coordinates in ThreeSpace
Now let’s see what’s doing in three dimensions. We shall associate with each
ordered triple of real numbers a point in three space. We continue from where we left off
in the previous section. Start with the plane constructed in the previous section, and
construct a line perpendicular to both the first and second axes, and passing through the
origin. This is the third, axis. Now we must be careful about which direction on this
third axis is chosen as the positive direction; it makes a difference. The positive direction
is chosen to be the direction in which a righthand threaded bolt would advance if the
positive first axis is rotated to the positive second axis:
We now see how to define a onetoone correspondence between ordered triples
of real numbers ( , , ) x x x
1 2 3
and the points in space. The association is a simple extension
of the way in which we established a correspondence between ordered pair and points in
1.6
a plane. Here’s what we do. Construct a plane perpendicular to the first axis through the
point x
1
, a plane perpendicular to the second axis through x
2
, and a plane perpendicular
to the third axis through x
3
. The point at which these three planes intersect is the point
associated with the ordered triple ( , , ) x x x
1 2 3
. Some meditation on this construction
should convince you that this procedure establishes a onetoone correspondence between
ordered triples of reals and points in space. As in the two dimensional, or plane, case, x
1
is called the first coordinate of the point, x
2
is called the second coordinate of the point,
and x
3
is called the third coordinate of the point. Again, the point corresponding to
( , , ) 0 00 is called the origin, and we speak of the point ( , , ) x x x
1 2 3
, when we actually mean
the point which corresponds to this ordered triple.
The three axes so defined is called a coordinate system for three space, and the
three numbers x, y, and z , where ( , , ) x y z is the triple corresponding to the point P, are
called the coordinates of P. The coordinate axes are sometimes given labelsmost
commonly, perhaps, the first axis is called the x axis, the second axis is called the y axis,
and the third axis is called the z axis.
1.7
1.3 Some Geometry
Suppose P and Q are two points, and suppose space is endowed with a
coordinate system such that P x y z · ( , , ) and Q u v w · ( , , ) . How do we find the distance
between P and Q ? This simple enough; look at the picture:
We can see that d h z w
2 2 2
· + − ( ) and h x u y v
2 2 2
· − + − ( ) ( ) . Thus we have
d x u y v z w
2 2 2 2
· − + − + − ( ) ( ) ( ) , or
d x u y v z w · − + − + − ( ) ( ) ( )
2 2 2
.
We saw that in the plane an equation in two variables defines in a natural way a collection
of ordered pairs of numbers. The analogous situation obtains in threespace: an equation
in three variables defines a collection of ordered triples. We thus speak of the collection
of triples ( , , ) x y z which satisfy the equation
x y z
2 2 2
1 + + ·
1.8
The collection of all such points is the graph of the equation. In this example, it is easy
to see that the graph is precisely the set of all points at a distance of 1 from the origina
sphere of radius 1 and center at the origin.
The graph of the equation x · 0 is simply the set of all points with first
coordinate 0, and this is clearly the plane determined by the second axis and the third axis,
or the y axis and the z axis. When the axes are labeled x, y, and z, this is known as the yz
plane. . Similarly, the plane y · 0 is the xz plane, and z · 0 is the xy plane. These
special planes are also called the coordinate planes.
More often than not, it is difficult to see exactly what a graph of equation looks
like, and even more difficult for most of us to draw it. Computers can help, but they
usually draw rather poor pictures whose main application is in stimulating your own
imagination sufficiently to allow you to see the graph in your mind’s eye. An example:
This picture was drawn using Maple.
Let’s look at a more complicated example. What does the graph of
x y z
2 2 2
1 + − ·
look like? We’ll go after a picture of this one by slicing the graph with the coordinate
planes. First, let’s slice through it with the plane z · 0 ; then we see
1.9
x y
2 2
1 + · ,
a circle of radius 1 centered at the origin. Next, let’s slice with the plane y · 0 . Here we
see x z
2 2
1 − · , a hyperbola:
We, of course, see the same hyperbola when we slice the graph with the plane
x · 0 . What the graph looks like should be fairly clear by now:
1.10
This graph has a name; it is called a hyperboloid.
Exercises
13. Describe the set of points S x y z x y z · ≥ ≥ ≥ {( , , ): , } 0 0 0 , and .
14. Describe the following sets
a) S x y z · ≥ {( , , ): z 0} b) S x y z x · ≥ {( , , ): } 5
c) R x y z x y · + ≤ {( , , ): }
2 2
1 d) T r s t r s t · + + ≤ {( , , ): }
2 2 2
4
15. Let G be the graph of the equation x y z
2 2 2
4 9 36 + + · .
a)Sketch the graphs of the curves sliced from G by the coordinate planes x · 0 ,
y · 0 , and z · 0 .
b)Sketch G. (This graph is called an ellipsoid.)
16. Let G be the graph of the equation x y z
2 2 2
3 4 12 − + · .
a)Sketch the graphs of the curves sliced from G by the coordinate planes x · 0 ,
y · 0 , and z · 0 .
b)Sketch G. (Does this set look at all familiar to you?.)
1.4 Some More GeometryLevel Sets
The curves that result from slicing the graphs with the coordinate planes are
special cases of what are called level sets of a set. Specifically, if S is a set, the
intersection of S with a plane z · constant is called a level set. In case the level set is a
1.11
curve, it is frequently called a level curve. (The slices by planes x = constant, or y =
constant are also level sets.) A family of level sets can provide a nice stimulant to your
powers of visualization. Everyday examples of the use of level sets to describe a set are
contour maps, in which the contours are, of course, just level curves ; and weather maps,
in which, for instance, the isoclines on a 500mb chart are simply level curves for the
500mb surface. Let’s illustrate with an example.
Let S be the graph of
z y x
2 2 2
1 − − ·
Now we look at the level set z c · :
c y x
2 2 2
1 − − · , or
x y c
2 2 2
1 + · − .
Notice first that we have the same curve for z = c and z = c. The graph is symmetric
about the plane z = 0. We shall thus look at just that part of the graph that is above the
xy plane.
It is clear that these curves are concentric circles of radius c
2
1 − centered at the
origin. There are no level sets for   c < 1, and for c = 1 or 1, the level set is a single point,
the origin.
1.12
Next, slice with the planes x = 0 and y = 0 to get a better idea of what this thing
looks like. For x = 0, we see
z y
2 2
1 − · ,
a hyperbola:
The slice by y = 0, of course, is the same. It is rather easy to visualize this graph. Here is
a Maple drawn picture:
1.13
This also is called a hyperboloid. This is a hyperboloid of two sheets, while the
previously described hyperboloid is a hyperboloid of one sheet.
Exercises
17. Let S x x x x x x · + + · {( , , ):      }
1 2 3 1 2 3
1 .
a)Sketch the coordinate plane slices of S.
b)Sketch the set S.
18. Let C be the graph of the equation z x y
2 2 2
4 · + ( ) .
a)Sketch some level sets z = c.
b)Sketch the slices by the planes x = 0 and y = 0.
c)Sketch C. What does the man on the street call this set?
19. Using level sets, coordinate plane slices, and whatever, describe the graph of the
equation z x y · +
2 2
. (This one has a name also; it is a paraboloid.).
1.14
20. Using level sets, coordinate plane slices, and whatever, describe the graph of the
equation z x y · −
2 2
.
2.1
Chapter Two
VectorsAlgebra and Geometry
2.1 Vectors
A directed line segment in space is a line segment together with a direction. Thus
the directed line segment from the point P to the point Q is different from the directed
line segment from Q to P. We frequently denote the direction of a segment by drawing an
arrow head on it pointing in its direction and thus think of a directed segment as a spear.
We say that two segments have the same direction if they are parallel and their directions
are the same:
Here the segments L1 and L2 have the same direction. We define two directed segments L
and M to be equivalent ( L M ≅ ) if they have the same direction and have the same
length. An equivalence class containing a segment L is the set of all directed segments
equivalent with L. Convince yourself every segment in an equivalence class is equivalent
with every other segment in that class, and two different equivalence classes must be
disjoint. These equivalence classes of directed line segments are called vectors. The
members of a vector v are called representatives of v. Given a directed segment u, the
vector which contains u is called the vector determined by u. The length, or magnitude,
of a vector v is defined to be the common length of the representatives of v. It is generally
designated by v. The angle between two vectors u and v is simply the angle between the
directions of representatives of u and v.
2.2
Vectors are just the right mathematical objects to describe certain concepts in
physics. Velocity provides a ready example. Saying the car is traveling 50 miles/hour
doesn’t tell the whole story; you must specify in what direction the car is moving. Thus
velocity is a vectorit has both magnitude and direction. Such physical concepts abound:
force, displacement, acceleration, etc. The real numbers (or sometimes, the complex
numbers) are frequently called scalars in order to distinguish them from vectors.
We now introduce an arithmetic, or algebra, of vectors. First, we define what we
mean by the sum of two vectors u and v. Choose a spear u from u and a spear v from v.
Place the tail of v at the nose of u. The vector which contains the directed segment from
the tail of u to the nose of v is defined to be u v + , the sum of u and v. An easy
consequence of elementary geometry is the fact that  u + v  <  u  +  v . Look at the
picture and convince yourself that the it does not matter which u spear or v spear you
choose, and that u v v u + · + :
Convince yourself also that addition is associative: u + (v + w) = (u + v) + w.
Since it does not matter where the parentheses occur, it is traditional to omit them and
write simply u + v + w.
Subtraction is defined as the inverse operation of addition. Thus the difference uv
of two vectors is defined to be the vector you add to v to get u. In pictures, if u is a
representative of u and v is a representative of v, and we put the tails of u and v together,
the directed segment from the nose of v to the nose of u is a representative of u  v:
2.3
Now, what are we to make of u  u ? We define a special vector with 0 length,
called the zero vector and denoted 0. We may think of 0 as the collection of all degenerate
line segments, or points. Note that the zero vector is special in that it has no direction (If
you are going 0 miles/hour, the direction is not important!). To make our algebra of
vectors nice , we make the zero vector behave as it should:
u  u = 0 and u + 0 = u
for all vectors u.
Next we define the product of a scalar r (i.e., real number) with a vector u. The
product ru is defined to be the vector with length ru and direction the same as the
direction of u if r > 0, and direction opposite the direction of u if r < 0. Convince
yourself that all the following nice properties of this multiplication hold:
(r + s)u = ru + su ,
r ( u + v ) = ru + rv.
0u = 0, and
u + (1)v = u  v.
It is then perfectly safe to write  u to stand for (1)u.
Our next move is to define a onetoone correspondence between vectors and
points in space (This will, of course, also establish a onetoone correspondence between
vectors and ordered triples of real numbers.). The correspondence is quite easy; simply
take a representative of the vector u and place its tail at the origin. The point at which is
2.4
found the nose of this representative is the point associated with u. We handle the vector
with no representatives by associating the origin with the zero vector. The fact that the
point with coordinates (a, b, c) is associated with the vector u in this manner is
shorthandedly indicated by writing u = (a, b, c). Strictly speaking this equation makes
no sense; an equivalence class of directed line segments cannot possible be the same as a
triple of real numbers, but this shorthand is usually clear and saves a lot of verbiage (The
numbers a, b, and c are called the coordinates, or components, of u.). Thus we frequently
do not distinguish between points and vectors and indiscriminately speak of a vector
(a,b,c) or of a point u.
Suppose u = (a, b, c) and v = (x, y, z). Unleash your vast knowledge of
elementary geometry and convince yourself of the truth of the following statements:
u = a b c
2 2 2
+ + ,
u + v = (a + x, b + y, c + d),
u  v = (a  x, b  y, c  d), and
ru = (ra, rb, rc).
Let i be the vector corresponding to the point (1, 0, 0); let j be the vector
corresponding to (0, 1, 0); and let k be the vector corresponding to (0, 0, 1). Any vector
u can now be expressed as a linear combination of these special socalled coordinate
vectors:
u i j k · · + + ( , , ) x y z x y z .
Example
Let’s use our newfound knowledge of vectors to find where the medians of a
triangle intersect. Look at the picture:
2.5
We shall find scalars s and t so that
a
b
a a
b a
+ − · +
−
t s ( ) ( )
2 2
.
Tidying this up gives us
( ) ( ) 1
2 2 2
− − · − t
s s t
a b .
This means that we must have
s t
t
s
2 2
0
1
2
0
− ·
− − ·
, and
.
Otherwise, a and b would be nonzero scalar multiples of one another, which would mean
they have the same direction. It follows that
s t · ·
2
3
.
This is, no doubt, the result you remember from Mrs. Turner’s high school geometry
class.
Exercises
2.6
1. Find the vector such that if its tail is at the point ( , , ) x y z
1 1 1
its nose will be at the
point ( , , ) x y z
2 2 2
.
2. Find the midpoint of the line segment joining the points (1, 5, 9) and (3, 2, 3).
3. What is the distance between the points ( , , ) x y z
1 1 1
and ( , , ) x y z
2 2 2
?
4. Describe the set of points L t t · −∞ < < ∞ { : } i .
5. Let u = (2,3,8). Describe the set of points L t · ∞ ∞ { : u  < t < }.
6. Describe the set of points M t t · + −∞ < < ∞ { } 3k i: .
7. Let u = (2,3,8).and v = (1,5,7). Describe the set of points
M t t · + −∞ < <∞ { : } v u .
8. Describe the set P t s t s · + −∞ < < ∞ −∞ < <∞ { , } i j: and .
9. Describe the set P t s t s · + + −∞ < <∞ −∞ < < ∞ { , } 5k i j: and .
10. Let u = (2,4,1) and v = (1,2,3). Describe the set
P t s t s · + −∞ < < ∞ −∞< < ∞ { , } u v: and .
11. Let u = (2,4,1), v = (1,2,3), and w = (3,6,1). Describe the set
P t s t s · + + −∞ < < ∞ −∞< < ∞ { , } w u v: and .
12. Describe the set C t t t · + ≤ ≤ {cos sin } : i j 0 2π .
2.7
13. Describe the set E t t t · + ≤ ≤ { cos sin } 4 3 0 2 : i j π .
14. Describe the set P t t · + ≤ ≤ { : i j
2
1 t 2} .
15. Let T be the triangle with vertices (2, 5, 7), (1, 2, 4), and (4, 2, 6). Find the point at
which the medians intersect.
2.2 Scalar Product
You were perhaps puzzled when in grammar school you were first told that the
work done by a force is the product of the force and the displacement since both force
and displacement are, of course, vectors. We now introduce this product. It is a scalar
and hence is called the scalar product. This scalar product u v ⋅ is defined by
u v u v ⋅ ·  cosθ ,
where θ is the angle between u and v. The scalar product is frequently also called the dot
product. Observe that u u u ⋅ · 
2
, and that u v ⋅ · 0 if and only if u and v are
perpendicular (or orthogonal ), or one or the other of the two is the zero vector. We
avoid having to use the latter weasel words by defining the zero vector to be
perpendicular to every vector; then we can say u v ⋅ · 0 if and only if u and v are
perpendicular.
Study the following picture to see that if u = 1, then u v ⋅ is the length of the
projection of v onto u. (More precisely, the length of the projection of a representative of
v onto a representative of u. Generally, where there is no danger of confusion, we omit
mention of this, just as we speak of the length of vectors, the angle between vectors, etc.)
2.8
It is clear that ( ) ( ) ( ) a b ab u v u v ⋅ · ⋅ . Study the following picture until you believe that
u v w u v u w ⋅ + · ⋅ + ⋅ ( ) for any three vectors u, v, and w.
Now let’s get a recipe for the scalar product of u · ( , , ) a b c and v · ( , , ) x y z :
u v i j k i j k
i i i j i k j i j j j k k i k j k k
⋅ · + + ⋅ + +
· ⋅ + ⋅ + ⋅ + ⋅ + ⋅ + ⋅ + ⋅ + ⋅ + ⋅
· + +
( ) ( )
,
a b c x y z
ax ay az bx by bz cx cy cz
ax by cz
since i i j j k k ⋅ · ⋅ · ⋅ · 1 and i j i k j k ⋅ · ⋅ · ⋅ · 0.
We thus see that it is remarkably simple to compute the scalar product of two
vectors when we know their coordinates.
Example
2.9
Again, let’s see how vectors can make geometry easy by using them to find the
angle between a diagonal of a cube and the diagonal of a face of the cube.
Suppose the cube has edge length s. Introduce a coordinate system so that the
faces are parallel to the coordinate planes, one vertex is the origin and the vertex at the
other end of the diagonal from the origin is (s, s, s). The vector determined by this
diagonal is thus d · si + sj + sk and the vector determined by the diagonal of the face in
the horizontal coordinate plane is f i j · + s s . Thus
d f d f ⋅ · · +   cosθ s s
2 2
,
where θ is the angle we seek. This gives us
cos
  
θ · · ·
2 2
3 2
2
3
2 2
2 2
s s
s s
d f
.
Or,
θ ·

.
`
,
−
Cos
1
2
3
.
Exercises
16. Find the work done by the force F i j k · − + 6 3 2 in moving an object from the point
(1, 4, 2) to the point (3, 2, 5).
17. Let L be the line passing through the origin and the point (2, 5), and let M be the line
passing through the points (3, 2) and (5, 3). Find the smaller angle between L and M.
18. Find an angle between the lines 3 2 1 x y + · and x y − · 2 3.
2.10
19. Suppose L is the line passing through (1, 2) having slope 2, and suppose M is the
line tangent to the curve y x ·
3
at the point (1, 1). ). Find the smaller angle between
L and M.
20. Find an angle between the diagonal and an adjoining edge of a cube.
21. Suppose the lengths of the sides of a triangle are a, b, and c; and suppose γ is the
angle opposite the side having length c. Prove that
c a b ab
2 2 2
2 · + − cosγ .
(This is, of course, the celebrated Law of Cosines.)
22. Let v = (1, 2, 5). . What is the graph of the equation v ⋅ · ( , , ) x y z 0 ?
2.3 Vector Product
Hark back to grammar school physics once again and recall what you were taught
about the velocity of a point at a distance r from the axis of rotation; you were likely told
that the velocity is rω , where ω is the rate at which the turntable is rotatingthe so
called angular velocity. We now know that these quantities are actually vectorsω is the
angular velocity, and r is the position vector of the point in question. The grammar school
quantities are the magnitudes of ω (the angular speed) and of r . The velocity of the point
is the socalled vector product of these two vectors. The vector product of vectors u and
v is defined by
u v u v n × ·  sin  θ ,
2.11
where θ is the angle between u and v and n is a vector of length 1 (such vectors are called
unit vectors) which is orthogonal to both u and v and which points in the direction a right
hand threaded bolt would advance if u were rotated into the direction of v.
Note first that this is a somewhat more exciting product than you might be used
to:
the order of the factors makes a difference. Thus u v v u × · − × .
Now let’s find a geometric construction of the vector product u v × . Proceed as
follows. Let P be a plane perpendicular to u. Now project v onto this plane, giving us a
vector v * perpendicular to u and having length  sin v θ  . Now rotate this vector v * 90
degrees around u in the “positive direction.” (By the positive direction of rotation about
a vector a, we mean the diction that would cause a righthand threaded bolt to advance in
the direction of a. ) This gives a vector v ** having the same length as v * and having the
direction of u v × . Thus u v u v** × ·  :
2.12
Now, why did we go to all this trouble to construct u v × in this fashion? Simple. It
makes it much easier to see that for any three vectors u, v, and w, we have
u v w u v u w × + · × + × ( ) .
(Draw a picture!)
We shall see how to compute this vector product u v × for
u i j k · · + + ( , , ) a b c a b c and v i j k · · + + ( , , ) x y z x y z .
We have
u v i j k i j k
i i i j i k
j i j j j k
k i k j k k
× · + + × + +
· × + × + × +
× + × + × +
× + × + ×
( ) ( )
( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( )
a b c x y z
ax ay az
bx by bz
cx cy cz
This looks like a terrible mess, until we note that
i i j j k k 0 × · × · × · ,
i j j i k × · − × · ( ) ,
j k k j i × · − × · ( ) , and
k i i k j × · − × · ( ) .
Making these substitutions in the above equation for u v × gives us
2.13
u v i j k × · − + − + − ( ) ( ) ( ) bz cy cx az ay bx .
This is not particularly hard to remember, but there is a nice memory device using
determinants:
u v
i j k
× · a b c
x y z
.
Example
Let’s find the velocity of a point on the surface of the Earth relative to a
coordinate system whose origin is fixed at its centerwe thus shall consider only motion
due to the Earth’s rotation, and neglect its motion about the sun, etc. For our point on the
Earth, choose Room 254, Skiles Classroom Building at Georgia Tech. The latitude of the
room is about 33.75 degrees (North, of course.), and it is about 3960 miles from the center
of the Earth. As we said, the origin of our coordinate system is the center of the Earth.
We choose the third axis to point through the North Pole; In other words, the coordinate
vector k points through the North Pole. The velocity of our room, is of course, not a
constant, but changes as the Earth rotates. We find the velocity at the instant our room is
in the coordinate plane determined by the vectors i and k.
The Earth makes one complete revolution every 24 hours, and so its angular
velocity ω is ω· ≈
2
24
02618
π
k k . radians/hour. The position vector r of our room is
r i k i k · + ≈ + 3960 3375 3375 32926 22001 (cos( . ) sin( . ) ) . . miles. Our velocity is thus
2.14
ω× · ≈ r
i j k
j 0 0 02618
32926 0 22001
862 .
. .
miles/hour.
Suppose we want to find the area of a parallelogram, the nonparallel sides of
which are representatives of the vectors a and b:
The area A is clearly A · · ×   sin   a b a b θ .
Example
Find the are of the parallelogram with a vertex (1,4,2) and the vertices at the other
ends of the sides adjoining this vertex are (4, 7, 8), and (6, 10, 20). This is easy. This is
just as in the above picture with a i j k i j k · − + − + − − + + ( ) ( ) ( ( )) 4 1 7 4 8 2 3 10 = 3 and
b i j k i j k · − + − + − − · + + ( ) ( ) ( ( )) 6 1 10 4 20 2 5 6 22 . So we have
a b
i j k
i j k × · · − + 3 3 10
5 6 22
6 16 3 ,
and so,
Area =  a b × · + + · 6 16 3 301
2 2 2
.
Exercises
2.15
23. Find a a vector perpendicular to the plane containing the points (1,4,6), (1,2,7), and
(3,6,10).
24. Are the points (0,4,7), (2, 6, 8), and (5, 10, 20) collinear? Explain how you know?
25. Find the torque created by the force f i j k · + − 3 2 3 acting at the point
a i j k · − − 2 7 .
26. Find the area of the triangle whose vertices are (0,0,0), (1,2,3), and (4,7,12).
27. Find the volume of the parallelepiped
3.1
Chapter Three
Vector Functions
3.1 Relations and Functions
We begin with a review of the idea of a function. Suppose A and B are sets. The
Cartesian product A B × of these sets is the collection of all ordered pairs ( , ) a b such
that a A ∈ and b B ∈ . A relation R is simply a subset of A B × . The domain of R is
the set dom R = { :( , ) } a A a b R ∈ ∈ . In case A = B and the domain of R is all of A, we call
R a relation on A. A relation R A B ⊂ × such that ( , ) a b R ∈ and ( , ) a c R ∈ only if b =
c is called a function. In other words, if R is a function, and a dom R ∈ , there is exactly
one ordered pair ( , ) a b R ∈ . The second “coordinate” b is thus uniquely determined by a.
It is usually denoted R a ( ) . If R A B ⊂ × is a relation, the inverse of R is the relation
R B A
−
⊂ ×
1
defined by R b a a b R
−
· ∈
1
{( , ):( , ) } .
Example
Let A be the set of all people who have ever lived and let S A A ⊂ × be the relation
defined by S a b b a · {( , ): } is the mother of . The S is a relation on A, and is, in fact, a
function. The relation S
−1
is not a function, and domS A
−
≠
1
.
The fact that f A B ⊂ × is a function with dom f = A is frequently indicated by
writing f A B : → , and we say f is a function from A to B. Very often a function f is
defined by specifying the domain, and giving a recipe for finding f(a). Thus we may
define the function f from the interval [0,1] to the real numbers by f x x ( ) ·
2
. This says
that f is the collection of all ordered pairs ( , ) x x
2
in which x ∈[ , ] 01 .
Exercises
3.2
1. Let A be the set of all Georgia Tech students, and let B be the set of real numbers.
Define the relation W A B ⊂ × by W a b b a · {( , ): } is the weight (in pounds) of . Is
W a function? Is W
−1
a function? Explain.
2. Let X be set of all states of the U. S., and let Y be the set of all U. S. municipalities.
Define the relation c X Y ⊂ × by c x y y x · {( , ): } is the capital of . Explain why c is
a function, and find c(Nevada), c(Missouri), and c(Kentucky).
3. With X, Y as in Exercise 2, let b be the function b x y y x · {( , ): } is the largest city in .
a)What is b(South Carolina)?
b)What is b(California)?
c)Let f c b · ∩ , where c is the function defined in Exercise 2. Find dom f.
4. Suppose f X Y ⊂ × and g X Y ⊂ × . If f is a function, is it necessarily true that
f g ∩ is a function? Prove your answer.
5. Suppose f X Y ⊂ × and g X Y ⊂ × . If f and g are both functions, is it necessarily
true that f g ∪ is a function? Prove your answer.
6. Suppose f X Y : → is a function and the inverse f
−1
is also a function.
a)What is f f x
−1
( ( )) ? Explain.
b)If y dom f ∈
−
1
, what is f f y ( ( ))
−1
? Explain.
3.2 Vector Functions
Our interest now will be focused on functions f X Y : → in which Y is a set of
vectors. These are called vector functions, or sometimes, vectorvalued functions.
Initially, we shall be solely interested in the special case in which X is a “nice” set of real
numbers, such as an interval. As the drama unfolds, we shall see that such functions
provide just the right tool for describing curves in space.
Let’s begin with a simple example. Let X be the entire real line and let the function
f be defined by f i j ( ) t t t · +
2
. It should be reasonably clear that if we place the tail of
3.3
f ( ) t (actually,. a representative of f ( ) t ) at the origin, the nose will lie on the curve
y x ·
2
. In fact, as t varies over the reals, the nose traces out this curve. The function f is
called a vector description of the curve. Let’s look at another example. This time, let
g i j ( ) cos sin t t t · + for 0 4 ≤ ≤ t π . What is the curve described by this function? First,
note that for all t, we have  ( ) g t · 1. The nose of g thus always lies on the circle of radius
one centered at the origin. It’s not difficult to see that, in fact, as t varies from 0 to 2π, the
nose moves around the circle once, and as t varies on from 2π to 4π, the nose traces out the
circle again.
The real usefulness of vector descriptions is most evident when we consider curves
in space. Let f i j k ( ) cos sin t t t t · + + , for all t ≥ 0 . Now, what curve is followed by the
nose of f(t)? Notice first that if we look down on this curve from someplace up the
positive third axis (In other words, k is pointing directly at us.), we see the circle described
by cos sin t t i j + . As t increases, we run around this circle and the third component of our
position increases linearly. Convince yourself now that this curve looks like this:
This curve is called a helix, or more precisely, a right circular helix. The picture was
drawn by Maple. Let’s draw another. How about the curve described by the vector
function g i j k (t t t t ) cos sin sin( ) · + + 2 ? This one is just a bit more exciting. Here’s a
computer drawn picture:
3.4
(This time we put the axes where they are “supposed to be.”)
Observe that in giving a vector description, we are in effect specifying the three
coordinates of points on the curves as ordinary real valued functions defined on a subset of
the reals. Assuming the axes are labeled x, y, and z, the curve described by the vector
function
r i j k ( ) ( ) ( ) ( ) t f t g t h t · + +
is equivalently described by the equations
x f t
y g t
z h t
·
·
·
( )
( )
( )
These are called parametric equations of the curve (The variable t is called the
parameter.).
Exercises
7. Sketch or otherwise describe the curve given by f i k ( ) t t t · +
3
for − ≤ ≤ 1 3 t .
8. Sketch or otherwise describe the curve given by f i j ( ) ( ) ( ) t t t · − + + 2 3 3 1 .
[Hint: Find an equation in x and y, the graph of which is the given curve.]
3.5
9. Sketch or otherwise describe the curve given by c i j k ( ) cos sin t t t · + + 7 .
10. Sketch or otherwise describe the curve given by
c i j k ( ) cos( ) sin( ) t t t · + +
2 2
7 .
11. Find an equation in x and y, the graph of which is the curve
g i j ( ) cos sin t t t · + 3 4 .
12. a)Find a vector equation for the graph of y x x x · + + +
3 2
2 5.
b)Find a vector equation for the graph of x y y y · + + +
3 2
2 5.
13. Find a vector equation for the graph of x y
2
3
2
3
1 + · .
14. a)Sketch or otherwise describe the curve given by the function r a b ( ) t t · + , where
a i j k · − + 2 3 and b i j k · + − 3 5 .
b)Express r(t) in the form r i j k ( ) ( ) ( ) ( ) t f t g t h t · + + .
15. Describe the curve given by L i j k ( ) ( ) ( ) t t t t · + + − + 3 1 1 2 .
16. Find a vector function for the straight line passing through the point (1,4,2) in the
direction of the vector v i j k · − + 2 .
17. a)Find a vector function for the straight line passing through the points (1,2,4) and
(3,1,5).
b)Find a vector function for the line segment joining the points (1,2,4) and (3,1,5).
18. Let L be the line through the points (1,5,2) and (2,2,4); and let M be the line
through the points (2,4,6) and (3,1,2). Find a vector description of the line which
passes through the point (1,1,2) and is perpendicular to both L and M.
3.3 Limits and Continuity
3.6
Recall from grammar school what we mean when we say the limit at t
0
of a real
valued, or scalar, function f is L. The definition for vector functions is essentially the
same. Specifically, suppose f is a vector valued function, t
0
is a real number, and L is a
vector such that for every real number ε > 0, there is a δ > 0 such that  ( )  f L t − < ε
whenever 0
0
< − <   t t δ and t is in the domain of f. This is traditionally written
lim ( )
t t
t
→
·
0
f L.
The vector L is called a limit of f at a.
Suppose α( ) t is a scalar function for which lim ( )
t t
t a
→
·
0
α , and f is a vector function
for which lim ( )
t t
t
→
·
0
f L. It is but a modest exercise to show that
lim( ( ) ( ))
t t
t t a
→
·
0
α f L .
To see this, we use the “behold!” method. Let ε > 0 be given. Choose δ δ δ
1 2 3
, , , and
δ
4
so that
 ( ) 
(  )
f L t
a
− <
+
ε
3 1
for 0
0 1
< − <   t t δ ;
 ( )  f L t − <
ε
3
for 0
0 2
< − <   t t δ ;
 ( ) 
(  )
α
ε
t a − <
+ 3 1 L
for 0
0 3
< − <   t t δ ; and
 ( )  α
ε
t a − <
3
for 0
0 4
< − <   t t δ .
Now let δ δ δ δ δ · min{ , , , }
1 2 3 4
suppose t is such that 0
0
< − <   t t δ . Then
 ( ) ( )   ( ( ) ) ( ( ) ) ( ( ) )( ( ) )
 ( ( ) )  ( ( ) ) ( ( ) )( ( ) )
 
(  )
α α α
α α
ε ε
ε ε ε ε ε
ε
t t a a t t a t a t
a t t a t a t
f L f L L f L
f L L f L
L
L
− · − + − + − −
≤ − + − + − −
+
+
+ < + + ·
<
a
3(1+a)
3 1 3 3 3 3 3
3.7
Or, in other words,
lim( ( ) ( ))
t t
t t a
→
·
0
α f L ,
which is what we set out to show.
Now suppose f i j k ( ) ( ) ( ) ( ) t x t y t z t · + + and L i j k · + + a b c . Then we see
that lim ( )
t t
t
→
·
0
f L if and only if
lim ( ) ,
lim ( )
lim ( ) .
t t
t t
t t
x t a
y t b
z t c
→
→
→
·
·
·
0
0
0
, and
It is now easy to show that all the usual nice properties of limits are valid for vector
functions:
lim( ( ) ( )) lim ( ) lim ( ).
t t t t t t
t t t t
→ → →
+ · +
0 0 0
f g f g
lim( ( ) ( )) (lim ( )) (lim ( ))
t t t t t t
t t t t
→ → →
⋅ · ⋅
0 0 0
f g f g .
lim( ( ) ( )) (lim ( )) (lim ( ))
t t t t t t
t t t t
→ → →
× · ×
0 0 0
f g f g .
We are now ready to say what we mean by a vector function’s being continuous at
a point of its domain. Suppose t
0
is in the domain of the vector function f. Then we say f
is continuous at t
0
if it is true that lim
t t
0
→
· f f ( ) ( ) t t
0
. It is easy to see that if
f i j k ( ) ( ) ( ) ( ) t x t y t z t · + + ,
then f is continuous at t
0
if and only if each of the everyday scalar functions x t y t ( ), ( ), and
z t ( ) is continuous at t
0
. This shows there is nothing particularly mysterious or exotic
about continuity of vector functions.
If f is continuous at each point of its domain, then we say simply that f is
continuous,
Exercises
3.8
19. Is it possible for a function f to have more than one limit at t t ·
0
? Prove your
answer.
20. Suppose m is a continuous realvalued function and f is a continuous vectorvalued
function. Is the vector function h defined by h f ( ) ( ) ( ) t m t t · also continuous?
Explain.
21. Let f and g be continuous at t t ·
0
. Is the function h defined by
h f g ( ) ( ) ( ) t t t · × continuous? Explain. How about the function r t t t ( ) ( ) ( ) · ⋅ f g ?
22. Let r i j k ( ) t t t
t
· + +
2
1
. Is r a continuous function? Explain.
23. Suppose r is a continuous function. Explain how you know that the length
function n t t ( )  ( ) · r is continuous.
4.1
Chapter Four
Derivatives
4.1 Derivatives
Suppose f is a vector function and t
0
is a point in the interior of the domain of f
( t
0
in the interior of a set S of real numbers means there is an interval centered at t
0
that
is a subset of S.). The derivative is defined just as it is for a plain old everyday real
valued function, except, of course, the derivative is a vector. Specifically, we say that f is
differentiable at t
0
if there is a vector v such that
lim [ ( ) ( )]
t t
h
t h t
→
+ − ·
0
1
0 0
f f v .
The vector v is called the derivative of f at t
0
.
Now, how would we find such a thing? Suppose f i j k ( ) ( ) ( ) ( ) t a t b t c t · + + .
Then
1
0 0
0 0 0 0 0 0
h
t h t
a t h a t
h
b t h b t
h
c t h c t
h
[ ( ) ( )]
( ) ( ) ( ) ( ) ( ) ( )
f f i j k + − ·
+ − 
.
`
,
+
+ − 
.
`
,
+
+ − 
.
`
,
.
It should now be clear that the vector function f is differentiable at t
0
if and only if each
of the coordinate functions a t b t ( ), ( ), and c t ( ) is. Moreover, the vector derivative v is
v i j k · + + a t b t c t '( ) '( ) '( ) .
Now we “know” what the derivative of a vector function is, and we know how to
compute it, but what is it, really? Let’s see. Let f i j ( ) t t t · +
3
. This is, of course, a
vector function which describes the graph of the function y x ·
3
. Let’s look at the
4.2
derivative of f at t
0
: v i j · + 3
0
2
t . Convince yourself that the direction of the vector v is
the direction tangent to the graph of y x ·
3
at the point ( , ) t t
0 0
3
. It is not so clear what
we should define to be the tangent to a curve other than a plane curve. Again, vectors
come to our rescue. If f is a vector description of a space curve, the direction of the
derivative f '( ) t vector is the tangent direction at the point f ( ) t the derivative f '( ) t is
said to be tangent to the curve at f ( ) t .
If f ( ) t specifies the position of a particle at time t, then, of course, the derivative
is the velocity of the particle, and its length  ' ( ) f t is the speed. Thus the distance the
particle travels from time t a · to time t b · is given by the integral of the speed:
d t dt
a
b
·
∫
 '( ) f .
If the particle behaves nicely, this distance is precisely the length of the arc of the curve
from f ( ) a to f ( ) b . It should be clear what we mean by “behaves nicely”. . For the
distance traveled by the particle to be the same as the length of its path, there must be no
“backtracking”, or reversing direction. This means we must not allow the velocity to be
zero for any t between a and b.
Example
Consider the function r i j ( ) cos sin t t t · + . Then the derivative, or velocity, is
r i j '( ) sin cos t t t · − + . This vector is indeed tangent to the curve described by r (which
we already know to be a circle of radius 1 centered at the origin.) at r( ) t . Note that the
scalar product r r ( ) '( ) sin cos sin cos t t t t t t ⋅ · − + · 0 , and so the tangent vector and the
vector from the center of the circle to the point on the circle are perpendiculara well
known fact you learned from Mrs. Turner in 4
th
grade. Note that the derivative is never
4.3
zerothere is no value of t for which both cost and sint vanish. The length of a piece of the
curve can thus be found by integrating the speed:
p t dt t tdt dt · · + · ·
∫ ∫ ∫
 '( ) sin cos r
2 2
0
2
0
2
0
2
2
π π π
π .
No surprise here.
Exercises
1. a)Find a vector tangent to the curve f i j k ( ) ( ) t t t t · + + −
2 3
1 at the point (1, 1, 0).
b)Find a vector equation for the line tangent to this same curve at the point (1, 1, 0).
2. The position of a particle is given by r i j ( ) cos( ) sin( ) t t t · +
3 3
.
a)Find the velocity of the particle.
b)Find the speed of the particle.
c)Describe the path of the particle, and find its length.
3. Let L be the line tangent to the curve g i j k ( ) cos sin t t t t · + + 10 10 16 at the point
( , , )
10
2
10
2
4π . Find the point at which L intersects the ij plane.
4. Let L be the straight line passing through the point (5, 0, 3) in the direction of the
vector a i j k · + − 2 , and let M be the straight line passing through the point (0, 0, 6)
in the direction of b i j k · − + 3 2 .
a)Are L and M parallel? Explain.
b)Do L and M intersect? Explain.
4.4
5. Let L be the straight line passing through the point (1, 1, 3) in the direction of the
vector a i j k · + − 2 , and let M be the straight line passing through the point (0, 1, 5)
in the direction of b i j k · − + 3 2 . Find the distance between L and M.
6. Find the length of the arc of the curve R i j k ( ) cos sin t t t t · + + 3 3 4 between the
points (3, 0, 0) and (3, 0, 16π).
7. Find an integral the value of which is the length of the curve y x ·
2
between the
points (1, 1) and (1, 1).
4.2 Geometry of Space CurvesCurvature
Let R( ) t be a vector description of a curve. Then the distance s t ( ) along the
curve from the point R( ) t
0
to the point R( ) t is, as we have seen, simply
s t d
t
t
( )  '( ) ·
∫
R ξ ξ
0
;
assuming, of course, that R'( ) t ≠ 0 . The speed is
ds
dt
t · ' ( ) R .
Now then the vector
T
R
R
R
R
R
· · · ·
'( )
 ' ( )
'( )
/
' ( )
t
t
t
ds dt
t
dt
ds
d
ds
is tangent to R and has length one. It is called the unit tangent vector.
Consider next the derivative
4.5
d
ds
d
ds
d
ds
d
ds
T T T
T T
T T
T
⋅ · ⋅ + ⋅ · ⋅ 2 .
But we know that T T T ⋅ · ·  
2
1. Thus T
T
⋅ ·
d
ds
0, which means that the vector
d
ds
T
is
perpendicular, or orthogonal, or normal, to the tangent vector T. The length of this vector
is called the curvature and is usually denoted by the letter κ. Thus
κ ·
d
ds
T
.
The unit vector
N
T
·
1
κ
d
ds
is called the principal unit normal vector, and its direction is sometimes called the
principal normal direction.
Example
Consider the circle of radius a and center at the origin: R i j ( ) cos sin t a t a t · + .
Then R i j '( ) sin cos t a t a t · − + , and
ds
dt
t a t a t a a a · · + · · ·  ' ( ) sin cos   R
2 2 2 2 2
.
Thus
T R i j · · − +
1
a
t t t ' ( ) sin cos .
Let’s not stop now.
d
ds
d
dt
dt
ds a
d
dt a
t t
T T T
i j · · · − −
1 1
( cos sin ) .
4.6
Thus κ · ·
d
ds a
T 1
, and N i j · − + (cos sin ) t t . So the curvature is the reciprocal of the
radius and the principal normal vector points back toward the center of the circle.
Another Example
This time let R i j k · + + + ( ) t t t 1 2
2
. First, R i j k '( ) t t · + + 2 2 , and so
ds
dt
t t · · +  ' ( ) R 5 4
2
. The unit tangent is then
T i j k ·
+
+ +
1
5 4
2 2
2
t
t ( ) .
It’s a bit of a chore now to find the curvature and the principal normal, so let’s use a
computer algebra system; viz., Maple:
First, let’s enter the unit tangent vector T:
See if we got it right:
T(t);
Fine. Now differentiate:
A(t);
We need to tidy this up:
4.7
B(t);
This vector is, of course, the normal
d
ds
T
. We continue and find the curvature κ and the
principal normal N.
kappa:=t>simplify(sqrt(dotprod(B(t),B(t))));
kappa(t);
N(t);
So there we have at last the speed
ds
dt
, the unit tangent T, the curvature κ., and the
principal normal N.
Exercises
8. Find a line tangent to the curve R i j k ( ) ( ) ( ) ( ) t t t t t · + + + − +
2 3
1 5 and passing
through the point (5, 2, 15), or show there is no such line.
4.8
9. Find the unit tangent T, the principal normal N, and the curvature κ, for the curves:
a) R i j k ( ) cos( ) sin( ) t t t t · + + 5 5 2
b) R i j ( ) ( ) (5 ) t t t · + + − 2 3
2
c) R i j k ( ) cos sin t e t e t
t t
· + + 6
10. Find the curvature of the curve y f x · ( ) at ( , ( )) x f x
0 0
.
11. Find the curvature of R i j ( ) t t t · +
2
. At what point on the curve is the curvature the
largest? smallest?
12. Find the curvature of R i j ( ) t t t · +
3
. At what point on the curve is the curvature the
largest? smallest?
4.3 Geometry of Space CurvesTorsion
Let R( ) t be a vector description of a curve. If T is the unit tangent and N is the
principal unit normal, the unit vector B T N · × is called the binormal. Note that the
binormal is orthogonal to both T and N. Let’s see about its derivative
d
ds
B
with respect
to arclength s. First, note that B B B ⋅ · ·  
2
1, and so B
B
⋅ ·
d
ds
0 , which means that
being orthogonal to B, the derivative
d
ds
B
is in the plane of T and N. Next, note that B is
perpendicular to the tangent vector T, and so B T ⋅ · 0 . Thus
d
ds
B
T ⋅ · 0 . So what have
4.9
we here? The vector
d
ds
B
is perpendicular to both B and T, and so must have the
direction of N (or, of course,  N). This means
d
ds
B
N · −τ .
The scalar τ is called the torsion.
Example
Let’s find the torsion of the helix R i j k ( ) cos sin t a t a t bt · + + . Here we go!
R i j k '( ) sin cos t a t a t b · − + + . Thus
ds
dt
t a b · · +  ' ( ) R
2 2
, and we have
T i j k ·
+
− + +
1
2 2
a b
a t a t b ( sin cos ) .
Now then
d
ds
d
dt
dt
ds
a
a b
t t
T T
i j · ·
−
+
+
( )
(cos sin )
2 2
.
Therefore,
κ ·
+
a
a b ( )
2 2
and N i j · − + (cos sin ) t t .
Let’s don’t stop now:
4.10
B T N
i j k
i j k · × ·
+
−
− −
·
+
− +
1
0
1
2 2 2 2
a b
a t a t b
t t
a b
b t b t a sin cos
cos sin
( sin cos ) ;
and
d
ds
d
dt
dt
ds
b
a b
t t
b
a b
B B
i j N · ·
+
+ ·
−
+ ( )
(cos sin )
( )
2 2 2 2
.
The torsion, at last:
τ ·
b
a
2
+ b
2
.
Suppose the curve R( ) t is such that the torsion is zero for all values of t. In other
words,
d
ds
B
≡ 0 . Look at
d
ds
t t
d
ds
t t
d
ds
[( ( ) ( )) ] ( ( ) ( )) R R B
R
B R R
B
− ⋅ · ⋅ + − ⋅ ·
0 0
0 .
Thus the scalar product ( ( ) ( )) R R B t t − ⋅
0
is constant. It is 0 at t
0
, and hence it is 0 for
all values of t. This means that R R ( ) ( ) t t −
0
and B are perpendicular for all t, and so
R R ( ) ( ) t t −
0
lies in a plane perpendicular to B. In other words, the curve described by
R( ) t is a plane curve.
Exercises
13. Find the binormal and torsion for the curve R i k ( ) cos sin t t t · + 4 3 .
14. Find the binormal and torsion for the curve R i j k ( )
sin
cos
sin
t
t
t
t
· + +
2 2
.
4.11
15. Find the curvature and torsion for R i j k ( ) . t t t t · + +
2 3
16. Show that the curve R i j k ( ) t t
t
t
t
t
· +
+
+
− 1 1
2
lies in a plane.
17. What is the vector B T × ? How about N T × ?
4.4 Motion
Suppose t is time and R( ) t is the position vector of a body. Then the curve
described by R( ) t is the path, or trajectory, of the body, v
R
( ) t
d
dt
· is the velocity, and
a
v
( ) t
d
dt
· is the acceleration. We know that v T ( ) t
ds
dt
· , and so the direction of the
velocity is the unit tangent T. Let’s see about the direction of the acceleration:
a
v
T
T
T N
( ) t
d
dt
d s
dt
ds
dt
d
dt
d s
dt
ds
dt
· · +
· +

.
`
,
2
2
2
2
2
κ
,
since
d
dt
ds
dt
T
N · κ . This tells us that the acceleration is always in the plane of the
vectors T and N. The derivative of the speed
d s
dt
2
2
is the tangential component of the
acceleration, and κ
ds
dt

.
`
,
2
is the normal component of the acceleration.
Example
4.12
Suppose a person who weighs 160 pounds moves around a circle having radius 20
feet at a constant speed of 60 miles/hour. What is the magnitude of the force on this
person at any time?
First, we know the force f is the mass times the acceleration: f a ( ) ( ) t m t · . Thus
f T N · +

.
`
,
m
d s
dt
m
ds
dt
2
2
2
κ
also have The speed is a constant 60 miles/hour, or 88 feet/second; in other words,
ds
dt
· 88 and
d s
dt
2
0 · . Hence,
    f N ·

.
`
,
·

.
`
,
m
ds
dt
m
ds
dt
κ κ
2 2
.
The mass m· ·
160
32
5 slugs, and the curvature κ ·
1
20
. The magnitude of the force is
thus  f ·
5 ⋅ 88
2
20
· 1936 pounds.
Exercises
18. The position of an object at time t is given by r i j k ( ) ( ) t t t t · + − +
3
2 2 . Find the
velocity, the speed, and the tangential and normal components of the acceleration.
19. A force f i j k ( ) ( ) t t t · + − +
2
1 newtons is applied to an object of mass 2 kilograms.
At time t = 0, the object is at the origin. Find its position at time t.
4.13
20. A projectile of weight w is fired from the origin with an initial speed v
0
in the
direction of the vector cos sin θ θ i j + , and the only force acting on the projectile is
f j · −w .
a)Find a vector description of the trajectory of the projectile.
b)Find an equation the graph of which is the trajectory.
21. A 16 lb. bowling ball is rolled along a track with a circular vertical loop of radius a
feet. What must the speed of the ball be in order for it not to fall from the track?
What must the speed of an 8 lb. ball be in order for it not to fall?
5.1
Chapter Five
More Dimensions
5.1 The Space R
n
We are now prepared to move on to spaces of dimension greater than three. These
spaces are a straightforward generalization of our Euclidean space of three dimensions. Let
n be a positive integer. The ndimensional Euclidean space R
n
is simply the set of
all ordered ntuples of real numbers x · ( , , , ) x x x
n 1 2
K . Thus R
1
is simply the real
numbers, R
2
is the plane, and R
3
is Euclidean threespace. These ordered ntuples are
called points, or vectors. This definition does not contradict our previous definition of a
vector in case n =3 in that we identified each vector with an ordered triple ( , , ) x x x
1 2 3
and
spoke of the triple as being a vector.
We now define various arithmetic operations on R
n
in the obvious way. If we
have vectors x · ( , , , ) x x x
n 1 2
K and y · ( , , , ) y y y
n 1 2
K in R
n
, the sum x y + is defined
by
x y + · + + + ( , , , ) x y x y x y
n n 1 1 2 2
K ,
and multiplication of the vector x by a scalar a is defined by
a ax ax ax
n
x · ( , , , )
1 2
K .
It is easy to verify that a a a ( ) x y x y + · + and ( ) a b a b + · + x x x .
Again we see that these definitions are entirely consistent with what we have done
in dimensions 1, 2, and 3there is nothing to unlearn. Continuing, we define the length,
or norm of a vector x in the obvious manner
  x · + + + x x x
n 1
2
2
2 2
K .
The scalar product of x and y is
5.2
x y ⋅ · + + + ·
·
∑
x y x y x y x y
n n i i
i
n
1 1 2 2
1
K .
It is again easy to verify the nice properties:
  , x x x
2
0 · ⋅ ≥
     , a a x x ·
x y y x ⋅ · ⋅ ,
x y z x y x z ⋅ + · ⋅ + ⋅ ( ) , and
( ) ( ) a a x y x y ⋅ · ⋅ .
The geometric language of the three dimensional setting is retained in higher
dimensions; thus we speak of the “length” of an ntuple of numbers. In fact, we also speak
of d( , )   x y x y · − as the distance between x and y . We can, of course, no longer rely
on our vast knowledge of Euclidean geometry in our reasoning about R
n
when n > 3.
Thus for n ≤ 3 , the fact that       x y x y + ≤ + for any vectors x and y was a simple
consequence of the fact that the sum of the lengths of two sides of a triangle is at least as
big as the length of the third side. This inequality remains true in higher dimensions, and,
in fact, is called the triangle inequality, but requires an essentially algebraic proof.
Let’s see if we can prove it.
Let x · ( , , , ) x x x
n 1 2
K and y · ( , , , ) y y y
n 1 2
K . Then if a is a scalar, we have
  ( ) ( ) a a a x y x y x y + · + ⋅ + ≥
2
0 .
Thus,
( ) ( ) . a a a a x y x y x x x y y y + ⋅ + · ⋅ + ⋅ + ⋅ ≥
2
2 0
This is a quadratic function in a and is never negative; it must therefore be true that
4 4 0
2
( ) ( )( ) x y x x y y ⋅ − ⋅ ⋅ ≤ , or
     x y x y ⋅ ≤ .
This last inequality is the celebrated CauchySchwarzBuniakowsky inequality. It
is exactly the ingredient we need to prove the triangle inequality.
5.3
  ( ) ( ) x y x y x y x x x y y y + · + ⋅ + · ⋅ + ⋅ + ⋅
2
2 .
Applying the CSB inequality, we have
         (   ) x y x x y y x y + ≤ + + · +
2 2 2 2
2 , or
      x y x y + ≤ + .
Corresponding to the coordinate vectors i , j , and k, the coordinate vectors
e e e
1 2
, , , K
n
are defined in R
n
by
e
e
e
e
1
2
3
10 00 0
0100 0
0010 0
0 00 01
·
·
·
·
( , , , , , )
( , , , , , )
( , , , , , )
( , , , , , )
K
K
K
M
K
n
,
Thus each vector x · ( , , , ) x x x
n 1 2
K may be written in terms of these coordinate vectors:
x e ·
·
∑
x
i i
i
n
1
.
Exercises
1. Let x and y be two vectors in R
n
. Prove that       x y x y + · +
2 2 2
if and only if
x y ⋅ · 0. (Adopting more geometric language from three space, we say that x and y
are perpendicular or orthogonal if x y ⋅ · 0.)
2. Let x and y be two vectors in R
n
. Prove
a)    x y x y x y + − − · ⋅
2 2
4 .
b)    [    ] x y x y x y + + − · +
2 2 2 2
2 .
5.4
3. Let x and y be two vectors in R
n
. Prove that           x y x y − ≤ + .
4. Let P ⊂ R
4
be the set of all vectors x · ( , , , ) x x x x
1 2 3 4
such that
3 5 2 15
1 2 3 4
x x x x + − + · .
Find vectors n and a such that P · ∈ ⋅ − · { : ( ) } x R n x a
4
0 .
5. Let n and a be vectors in R
n
, and let P · ∈ ⋅ − · { : ( ) x R n x a
n
0 .
a)Find an equation in x x
1 2
, , , K and x
n
such that x · ∈ ( , , , ) x x x P
n 1 2
K if and only if
the coordinates of x satisfy the equation.
b)Describe the set P be in case n = 3. Describe it in case n =2.
[The set P is called a hyperplane through a.]
5.2 Functions
We now consider functions F: R R
n p
→ . Note that when n = p = 1, we have the
usual grammar school calculus functions, and when n = 1 and p = 2 or 3, we have the
vector valued functions of the previous chapter. Note also that except for very special
circumstances, graphs of functions will not play a big role in our understanding. The set of
points ( , ( )) x x F resides in R
n p +
since x R
n
∈ and F( ) x R
p
∈ ; this is difficult to “see”
unless n p + ≤ 3 .
We begin with a very special kind of functions, the socalled linear functions. A
function F: R R
n p
→ is said to be a linear function if
i) F F F ( ) ( ) ( ) x y x y + · + for all x y R
n
, ∈ , and
ii) F a aF ( ) ( ) x x · for all scalars a and x R
n
∈ .
Example
Let n = p = 1, and define F by F x x ( ) · 3 . Then
F x y x y x y F x F y ( ) ( ) ( ) ( ) + · + · + · + 3 3 3 and
F ax ax a x aF x ( ) ( ) ( ) · · · 3 3 .
This F is a linear function.
5.5
Another Example
Let F R R
3
: → be defined by F i j k ( ) ( , , ) t t t t t t t · + − · − 2 7 2 7 . Then
F i j k
i j k i j k
F F
( ) ( ) ( ) ( )
[ ] [ ]
( ) ( )
t s t s t s t s
t t t s s s
t s
+ · + + + − +
· + − + + −
· +
2 7
2 7 2 7
Also,
F i j k
i j k F
( )
[ ] ( )
at at at at
a t t t a t
· + −
· + − ·
2 7
2 7
We see yet another linear function.
One More Example
Let F: R R
3 4
→ be defined by
F x x x x x x x x x x x x x x (( , , )) ( , , , )
1 2 3 1 2 3 1 2 3 1 2 3 1 3
2 3 4 5 2 · − + + − − + + + .
It is easy to verify that F is indeed a linear function.
A translation is a function T:R R
p p
→ such that T( ) x a x · + , where a is a
fixed vector in R
n
. A function that is the composition of a linear function followed by a
translation is called an affine function. An affine function F thus has the form
F L ( ) ( ) x a x · + , where L is a linear function.
Example
Let F: R R
3
→ be defined by F t t t t ( ) ( , , ) · + − 2 4 3 . Then F is affine. Let
a · ( , , ) 2 4 0 and L t t t t ( ) ( , , ) · 4 . Clearly F t L t ( ) ( ) · + a .
Exercises
6. Which of the following functions are linear? Explain your answers.
a) f x x ( ) · −7 b) g x x ( ) · − 2 5
c) F x x x x x x x x x ( , ) ( , , , , )
1 2 1 2 1 2 1 1 2
2 3 5 2 · + − − x
1
d) G x x x x x x ( , , )
1 2 3 1 2 3
· + e) F t t t t ( ) ( , , ) · − 2 2 0,
f) h x x x x ( , , , ) ( , , )
1 2 3 4
1 0 · 0 g) f x x ( ) sin ·
5.6
7. a)Describe the graph of a linear function from R to R.
b)Describe the graph of an affine function from R to R.
6.1
Chapter Six
Linear Functions and Matrices
6.1 Matrices
Suppose f : R R
n p
→ be a linear function. Let e e e
1 2
, , , K
n
be the coordinate
vectors for R
n
. For any x R
n
∈ , we have x e e e · + + + x x x
n n 1 1 2 2
K . Thus
f f x x x x f x f x f
n n n n
( ) ( ) ( ) ( ) ( ) x e e e e e e · + + + · + + +
1 1 2 2 1 1 2 2
K K .
Meditate on this; it says that a linear function is entirely determined by its values
f f f
n
( ), ( ), , ( ) e e e
1 2
K . Specifically, suppose
f a a a
f a a a
f a a a
p
p
n n n pn
( ) ( , , , ),
( ) ( , , , ),
( ) ( , , , ).
e
e
e
1 11 21 1
2 12 22 2
1 2
·
·
·
K
K
M
K
Then
f a x a x a x a x a x a x
a x a x a x
n n n n
p p pn n
( ) ( , , ,
).
x · + + + + + +
+ + +
11 1 12 2 1 21 1 22 2 2
1 1 2 2
K K K
K
The numbers a
ij
thus tell us everything about the linear function f. . To avoid labeling
these numbers, we arrange them in a rectangular array, called a matrix:
6.2
a a a
a a a
a a a
n
n
p p pn
11 12 1
21 22 2
1 2
K
K
M M
K
]
]
]
]
]
]
The matrix is said to represent the linear function f.
For example, suppose f : R R
2 3
→ is given by the receipt
f x x x x x x x x ( , ) ( , , )
1 2 1 2 1 2 1 2
2 5 3 2 · − + − .
Then f f ( ) ( , ) ( , , ) e
1
10 213 · · , and f f ( ) ( , ) ( , , ) e
2
01 15 2 · · − − . The matrix representing f
is thus
2
1
3
1
5
2
−
−
]
]
]
]
]
Given the matrix of a linear function, we can use the matrix to compute f ( ) x for
any x. This calculation is systematized by introducing an arithmetic of matrices. First,
we need some jargon. For the matrix
A
a a a
a a a
a a a
n
n
p p pn
·
]
]
]
]
]
]
11 12 1
21 22 2
1 2
K
K
M M
K
,
6.3
the matrices [ ] a a a
i i in 1 2
, , , K are called rows of A, and the matrices
a
a
a
j
j
pj
1
2
M
]
]
]
]
]
]
are called
columns of A. Thus A has p rows and n columns; the size of A is said to be p n × . A
vector in R
n
can be displayed as a matrix in the obvious way, either as a 1×n matrix, in
which case the matrix is called a row vector, or as a n ×1 matrix, called a column vector.
Thus the matrix representation of f is simply the matrix whose columns are the column
vectors f f f
n
( ), ( ), , ( ) e e e
1 1
K .
Example
Suppose f : R R
3 2
→ is defined by
f x x x x x x x x x ( , , ) ( , )
1 2 3 1 2 3 1 2 3
2 3 2 5 · − + − + − .
So f f ( ) ( , , ) ( , ) e
1
10 0 2 1 · · − , f f ( ) ( , , ) ( , ) e
2
010 32 · · − , and f f ( ) ( , , ) ( , ) e
3
001 1 5 · · − .
The matrix which represents f is thus
2
1
3
2
1
5 −
−
−
]
]
]
Now the recipe for computing f(x) can be systematized by defining the product of
a matrix A and a column vector x. Suppose A is a p n × matrix and x is a n ×1 column
6.4
vector. For each i p · 12 , , , , K let r
i
denote the i
th
row of A . We define the product Ax
to be the p ×1column vector given by
Ax
r x
r x
r x
·
⋅
⋅
⋅
]
]
]
]
]
]
1
2
M
p
.
If we consider all vectors to be represented by column vectors, then a linear function
f : R R
n p
→ is given by f ( ) x Ax · , where, of course, A is the matrix representation of
f.
Example
Consider the preceding example:
f x x x x x x x x x ( , , ) ( , )
1 2 3 1 2 3 1 2 3
2 3 2 5 · − + − + − .
We found the matrix representing f to be
A ·
−
−
−
]
]
]
2
1
3
2
1
5
.
Then
Ax x ·
−
−
−
]
]
]
]
]
]
]
]
·
− +
− + −
]
]
]
·
2
1
3
2
1
5
2 3
2 5
1
2
3
1 2 3
1 2 3
x
x
x
x x x
x x x
f ( )
Exercises
6.5
1. Find the matrix representation of each of the following linear functions:
a) f x x x x x x x ( , ) ( , , , )
1 2 1 2 1 2 2
2 4 5 · − + + 7x 3x
1 1
.
b) R i j k ( ) t t t t · − − 4 5 2 .
c) L x x ( ) · 6 .
2. Let g be define by g( ) x Ax · , where A ·
−
−
−
]
]
]
]
]
]
2
2
0
3
1
1
3
5
. Find g( , ) 3 9 − .
3. Let f : R R
2 2
→ be the function in which f(x) is the vector that results from rotating
the vector x about the origin
π
4
in the counterclockwise direction.
a)Explain why f is a linear function.
b)Find the matrix representation for f.
d)Find f(4,9).
4. Let f : R R
2 2
→ be the function in which f(x) is the vector that results from rotating
the vector x about the origin θ in the counterclockwise direction. Find the matrix
representation for f.
5. Suppose g: R R
2 2
→ is a linear function such that g(1,2) = (4,7) and g(2,1) = (2,2).
6.6
Find the matrix representation of g.
6. Suppose f : R R
n p
→ and g: R R
p q
→ are linear functions. Prove that the
composition g f o : R R
n q
→ is a linear function.
7. Suppose f : R R
n p
→ and g: R R
n p
→ are linear functions. Prove that the function
f g + → : R R
n p
defined by ( )( ) ( ) ( ) f g f g + · + x x x is a linear function.
6.2 Matrix Algebra
Let us consider the composition h g f · o of two linear functions f : R R
n p
→
and g: R R
p q
→ . Suppose A is the matrix of f and B is the matrix of g. Let’s see about
the matrix C of h. We know the columns of C are the vectors g f j n
j
( ( )), , , , e · 12 K ,
where, of course, the vectors e
j
are the coordinate vectors for R
n
. Now the columns of
A are just the vectors f j n
j
( ), , , , e · 1 2 K . Thus the vectors g f
j
( ( )) e are simply the
products B e f
j
( ) . In other words, if we denote the columns of A by k
i
i n , , , , · 12 K , so
that A k k k · [ , , , ]
1 2
K
n
, then the columns of C are Bk Bk Bk
1 2
, , , K
n
, or in other words,
C Bk Bk Bk · [ , , , ]
1 2
K
n
.
Example
6.7
Let the matrix B of g be given by B ·
− −
−
−
]
]
]
]
]
]
1
1
2
2
0
5
7
2
2
8
3
1
and let the matrix A of f be
given by A ·
− −
]
]
]
]
]
3
1
4
1
2
3
. Thus f : R R
2 3
→ and g: R R
3 4
→ (Note that for the
composition h g f · o to be defined, it must be true that the number of columns of B be
the same as the number of rows of A.). Now, k
1
3
1
4
·
−
]
]
]
]
]
and k
2
1
2
3
·
−
]
]
]
]
]
, and so
Bk
1
·
−5
−40
25
0
]
]
]
]
]
and Bk
2
·
−5
−35
25
−3
]
]
]
]
]
. The matrix C of the composition is thus
C ·
−5
−40
25
0
−5
−35
25
−3
]
]
]
]
]
.
These results inspire us to define a product of matrices. Thus, if B is an n p ×
matrix, and A is a p q × matrix, the product BA of these matrices is defined to be the
n q × matrix whose columns are the column vectors Bk
j
, where k
j
is the j
th
column of
A. Now we can simply say that the matrix representation of the composition of two
linear functions is the product of the matrices representing the two functions.
6.8
There are several interesting and important things to note regarding matrix
products. First and foremost is the fact that in general BA AB ≠ , even when both
products are defined (The product BA obviously defined only when the number of
columns of B is the same as the number of rows of A.). Next, note that it follows directly
from the fact that h f g h f g o o o o ( ) ( ) · that for C(BA) = (CB)A. Since it does not
matter where we insert the parentheses in a product of three or more matrices, we usually
omit them entirely.
It should be clear that if f and g are both functions from R
n
to R
p
, then the
matrix representation for the sum f g + → : R R
n p
is simply the matrix
A B + ·
+ + +
+ + +
+ + +
]
]
]
]
]
]
a b a b a b
a b a b a b
a b a b a b
n n
n n
p p p p pn pn
11 11 12 12 1 1
21 21 22 22 2 2
1 1 2 2
K
K
M
K
,
where
A ·
]
]
]
]
]
]
a a a
a a a
a a a
n
n
p p pn
11 12 1
21 22 2
1 2
L
L
M
L
is the matrix of f, and
6.9
B ·
]
]
]
]
]
]
b b b
b b b
b b b
n
n
p p pn
11 12 1
21 22 2
1 2
L
L
M
L
is the matrix of g. Meditating on the properties of linear functions should convince you
that for any three matrices (of the appropriate sizes) A, B, and C, it is true that
A B C AB AC ( ) + · + .
Similarly, for appropriately sized matrices, we have ( ) A B C AC BC + · + .
Exercises
8. Find the products:
a)
2 1
0 3
2
1
]
]
]
−
]
]
]
b)
2 1
0 3
1
3
]
]
]
]
]
]
c)
2 1
0 3
2 1
1 3
]
]
]
−
]
]
]
d)[ ] 1 3 2 1
1 5
2 3
0 2
3 4
− −
−
−
]
]
]
]
]
]
9. Find a)
1 0 0
0 1 0
0 0 1
11 12 13
21 22 23
31 32 33
]
]
]
]
]
]
]
]
]
]
a a a
a a a
a a a
b)
0 0 0
0 0 0
0 0 0
11 12 13
21 22 23
31 32 33
]
]
]
]
]
]
]
]
]
]
a a a
a a a
a a a
6.10
10. Let A( ) θ be the 2 2 × matrix for the linear function that rotates the plane θ
counterclockwise. Compute the product A A ( ) ( ) θ η , and use the result to give
identities for cos( ) θ η + and sin( ) θ η + in terms of cosθ , cosη, sinθ , and sinη.
11. a)Find the matrix for the linear function that rotates R
3
about the coordinate vector j
by
π
4
(In the positive direction, according to the usual “right hand rule” for rotation.).
b)Find a vector description for the curve that results from applying the linear
transformation in a) to the curve R i j k ( ) cos sin t t t t · + + .
12. Suppose f : R R
2 2
→ is linear. Let C be the circle of radius 1 and center at the origin.
Find a vector description for the curve f(C).
13. Suppose g: R R
2 n
→ is linear. Suppose moreover that g( , ) ( , ) 11 2 3 · and
g( , ) ( , ) − · − 11 4 5 . Find the matrix of g.
7.1
Chapter Seven
Continuity, Derivatives, and All That
7.1 Limits and Continuity
Let x R
0
n
∈ and r > 0. The set B r r ( ; ) { :  } a x R x a
n
· ∈ − < is called the open
ball of radius r centered at x
0
. The closed ball of radius r centered at x
0
is the set
B r r ( ; ) { :  } a x R x a
n
· ∈ − ≤ . Now suppose D R
n
⊂ . A point a D ∈ is called an
interior point of D if there is an open ball B r ( ; ) a D ⊂ . The collection of all interior
points of D is called the interior of D, and is usually denoted int D. A set U is said to be
open if U = int U.
Suppose f : D R
p
→ , where D R
n
⊂ and suppose a R
n
∈ is such that every
open ball centered at a meets the domain D. If y R
p
∈ is such that for every ε > 0, there
is a δ > 0 so that  ( )  f x y − < ε whenever 0 < − <   x a δ , then we say that y is the limit of
f at a. This is written
lim ( )
x a
x y
→
· f ,
and y is called the limit of f at a.
Notice that this agrees with our previous definitions in case n = 1 and p =1,2, or 3.
The usual properties of limits are relatively easy to establish:
lim( ( ) ( )) lim ( ) lim ( )
x a x a x a
x x x x
→ → →
+ · + f g f g , and
lim ( ) lim ( )
x a x a
x x
→ →
· af a f .
Now we are ready to say what we mean by a continuous function f : D R
p
→ ,
where D R
n
⊂ . Again this definition will not contradict our previous lower dimensional
7.2
definitions. Specifically, we say that f is continuous at a D ∈ if lim ( ) ( )
x a
x a
→
· f f . If f is
continuous at each point of its domain D, we say simply that f is continuous.
Example
Every linear function is continuous. To see this, suppose f : R R
n p
→ is linear
and a R
n
∈ . Let ε > 0. Now let M f f f · max{ ( ), ( ), , ( )} e e e
1 2 n
K and let δ
ε
·
nM
.
Then for x such that 0 < − <   x a δ , we have
 ( ) ( )  ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
  ( )   ( )   ( )
(     )
f f f x x x f a a a
x a f x a f x a f
x a f x a f x a f
x a x a x a M
n n n n
n n n
n n n
n n
x a e e e e e e
e e e
e e e
− · + + + − + + +
· − + − + + −
≤ − + − + + −
≤ − + − + + −
1 1 2 2 1 1 2 2
1 1 1 2 2 2
1 1 1 2 2 2
1 1 2 2
K K
K
K
K
≤ −
<
n M   x a
ε
Thus lim ( ) ( )
x a
x a
→
· f f and so f is continuous.
Another Example
Let f : R R
2
→ be defined by f f x x
x x
x x
x x
( ) ( , )
,
,
x · ·
+
+ ≠
¹
'
¹
¹
¹
1 2
1 2
1
2
2
2 1
2
2
2
0
0
for
otherwise
.
Let’s see about lim ( ).
( , ) x
x
→ 0 0
f Let x · α( , ) 11 . Then for all α ≠ 0, we have
f f ( ) ( , ) x · ·
+
· α α
α
α α
2
2 2
1
2
.
7.3
Now. let x · · α α ( , ) ( , ) 10 0 . It follows that all α ≠ 0, f ( ) x · 0 . What does this tell
us? It tells us that for any δ > 0 , there are vectors x with 0 00 < − <  ( , ) x δ such that
f ( ) x ·
1
2
and such that f ( ) x · 0 . This, of course, means that lim ( )
( , ) x
x
→ 0 0
f does not
exist.
7.2 Derivatives
Let f : D R
p
→ , where D R
n
⊂ , and let x D
0
∈int . Then f is differentiable at
x
0
if there is a linear function L such that
lim
 
[ ( ) ( ) ( )]
h 0
0 0
h
x h x L h 0
→
+ − − ·
1
f f .
The linear function L is called the derivative of f at x
0
. It is usual to identify the linear
function L with its matrix representation and think of the derivative at a p n × matrix.
Note that in case n = p = 1, the matrix L is simply the 1 1 × matrix whose sole entry is the
every day grammar school derivative of f .
Now, how do find the derivative of f ? Suppose f has a derivative at x
0
. First, let
h e · · t t
j
( , , , , , , , ) 00 0 0 0 K K . Then
f f x x x t x
f x x x t x
f x x x t x
f x x x t x
j n
j n
j n
p j n
( ) ( , , , , , )
( , , , , , )
( , , , , , )
( , , , , , )
x h + · + ·
+
+
+
]
]
]
]
]
]
1 2
1 1 2
2 1 2
1 2
K K
K K
K K
M
K K
,
and
7.4
Lh ·
]
]
]
]
]
]
]
]
]
]
]
]
]
]
]
·
]
]
]
]
]
]
m m m
m m m
m m m
t
m t
m t
m t
n
n
p p pn
j
j
pj
11 12 1
21 22 2
1 2
1
2
0
0
0
L
L
M
L
M
M
M
,
where x
0
· ( , , , ) x x x
n 1 2
K , etc.
Now then,
1
 h
[ f (x
0
+ h) − f (x
0
) − L(h)]
·
1
t
f
1
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
1
(x
1
, x
2
,K, x
n
) − m
1j
t
f
2
(x
1
, x
2
, K, x
j
+ t,K, x
n
) − f
2
(x
1
, x
2
,K, x
n
) − m
2j
t
M
f
p
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
p
(x
1
, x
2
,K, x
n
) − m
pj
t
]
]
]
]
]
=
f
1
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
1
(x
1
, x
2
,K, x
n
)
t
− m
1j
f
2
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
2
(x
1
, x
2
,K,, x
n
)
t
− m
2j
M
f
p
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
p
(x
1
, x
2
,K,, x
n
)
t
− m
pj
]
]
]
]
]
]
]
Meditate on this vector. For each component,
lim
t →0
f
i
(x
1
, x
2
,K, x
j
+ t,K, x
n
) − f
i
(x
1
, x
2
, K,, x
n
)
t
·
d
ds
f
i
(x
1
, x
2
,K, s,K, x
n
)
s · x
j
This derivative has a name. It is called the partial derivative of f
i
with respect to the j
th
variable. There are many different notations for the partial derivatives of a function
g x x x
n
( , , , )
1 2
K . The two most common are:
7.5
g x x x
x
g x x x
j n
j
n
,
( , , , )
( , , , )
1 2
1 2
K
K
∂
∂
The requirement that lim
 
[ ( ) ( ) ( )]
h 0
0 0
h
x h x L h 0
→
+ − − ·
1
f f now translates into
m
f
x
ij
i
j
·
∂
∂
,
and, mirabile dictu, we have found the matrix L !
Example
Let f : R R
2 2
→ be given by f x x
x x
x x x
( . )
sin
1 2
1 2
1
3
1 2
2
3
·
+
]
]
]
. Assume f is differentiable
and let’s find the derivative (more precisely, the matrix of the derivative. This matrix will,
of course, be 2 2 × : L ·
]
]
]
m m
m m
11 12
21 22
. Now
f x x x x
f x x x x x
1 1 2 1 2
2 1 2 1
3
1 2
2
3 ( , ) sin ,
( , )
·
· +
and
Compute the partial derivatives:
∂
∂
∂
∂
f
x
x
f
x
x x
1
1
2
2
1
1
2
2
2
3
3
·
· +
sin
,
7.6
and
∂
∂
∂
∂
f
x
x x
f
x
x x
1
2
1 2
2
2
1 2
3
2
·
·
cos
.
The derivative is thus
L ·
+
]
]
]
3 3
3 2
2 1 2
1
2
2
2
1 2
sin cos x x x
x x x x
.
We now know how to find the derivative of f at x if we know the derivative exists;
but how do we know when there is a derivative? The function f is differentiable at x if the
partial derivatives exist and are continuous. It should be noted that it is not sufficient
just for the partial derivatives to exist.
Exercises
1. Find all partial derivatives of the given functions:
a) f x y x y ( , ) ·
2 3
b) f x y z x yz z xy ( , , ) cos( ) · +
2
c) g x x x x x x x ( , , )
1 2 3 1 2 3 2
· + d) h x x x x
x e
x x
x
( , , , )
sin( )
1 2 3 4
3
2 4
1
·
+
2. Find the derivative of the linear function whose matrix is
1
2
3
7
2
0 −
]
]
]
.
3. What is the derivative a linear function whose matrix is A ?
7.7
4. Find the derivative of R i j k ( ) cos sin t t t t · + + .
5. Find the derivative of f x y x y ( , ) ·
2 3
.
6. Find the derivative of
f (x
1
, x
2
, x
3
) ·
x
1
x
3
+ e
x 2
x
3
log(x
1
+ x
2
2
)
x
2
x
1
x
3
2
+ 5
]
]
]
]
]
.
7.3 The Chain Rule
Recall from elementary one dimensional calculus that if a function is differentiable
at a point, it is also continuous there. The same is true here in the more general setting of
functions f : R R
n p
→ . Let’s see why this is so. Suppose f is differentiable at a with
derivative L. Let h x a · − . Then lim ( ) lim ( )
x a h 0
x a h
→ →
· + f f . Now,
f f
f f h
( ) ( )  
( ) ( ) ( )
 
( ) a h a h
a h a L
h
L h + − ·
+ − −
]
]
]
−
Now look at the limit of this as   h→0 :
lim
( ) ( ) ( )
 
h 0
a h a L h
h
0
→
+ − −
]
]
]
·
f f
7.8
because f is differentiable at a, and lim ( ) ( )
h 0
L h L 0 0
→
· · because the linear function L is
continuous. Thus lim( ( ) ( ))
h 0
a h a 0
→
+ − · f f , or lim ( ) ( )
h 0
a h a
→
+ · f f , which means f is
continuous at a.
Next, let’s see what the celebrated chain rule looks like in higher dimensions. Let
f : R R
n p
→ and g: R R
p q
→ . Suppose the derivative of f at a is L and the derivative of
g at f ( ) a is M. We go on a quest for the derivative of the composition g f o : R R
n q
→
at a . Let r g f · o , and look at r r g f g f ( ) ( ) ( ( )) ( ( )) a h a a h a + − · + − . Next, let
k a h a · + − f f ( ) ( ) . Then we may write
r r g f g f
g f g f
g f g f
( ) ( ) ( ) ( ( )) ( ( )) ( )
( ( ) ) ( ( )) ( ) ( ) ( )
( ( ) ) ( ( )) ( ) ( ( ))
a h a ML h a h a ML h
a k a M k M k ML h
a k a M k M k L h
+ − − · + − −
· + − − + −
· + − − + −
.
Thus,
r r g f g f ( ) ( ) ( )
 
( ( ) ) ( ( )) ( )

(
( )
 
)
a h a ML h
h
a k a M k
h
M
k L h
h
+ − −
·
+ − −
+
−

Now we are ready to see what happens as   h→0 . look at the second term first:
lim (
( )
 
) lim
( ) ( ) ( )
 
(lim
( ) ( ) ( )
 
)
( )
h 0 h 0 h 0
M
k L h
h
M
a h a L h
h
M
a h a L h
h
M 0 0
→ → →
−
·
+ − − 
.
`
,
·
+ − − 
.
`
,
· ·
f f f f
since L is the derivative of f at a and M is linear, and hence continuous.
Now we need to see what happens to the term
lim
( ( ) ) ( ( )) ( )

h 0
a k a M k
h
→
+ − − 
.
`
,

g f g f
.
7.9
This is a bit tricky. Note first that because f is differentiable at a , we know that
 
 
 ( ) ( )
 
k
h
a h a
h
·
+ − f f
behaves nicely as   h→0 . Next,
lim
( ( ) ) ( ( )) ( )

 
 
lim
( ( ) ) ( ( )) ( )

 
 
h 0
h 0
a k a M k
h
k
k
a k a M k
k
k
h
0
→
→
+ − −
⋅

.
`
,
·
+ − − 
.
`
,
·


g f g f
g f g f
since the derivative of g at f ( ) a is M, and
 
 
k
h
is wellbehaved. Finally at last, we have
shown that
lim
( ) ( ) ( )
 
h 0
a h a ML h
h
0
→
+ − − 
.
`
,
·
r r
,
which means the derivative of the composition r g f · o is simply the composition, or
matrix product, of the derivatives. What could be more pleasing from an esthetic point of
view!
Example
Let f t t t ( ) ( , ) · +
2 3
1 and g x x x x ( , ) ( )
1 2 1 2
3
2 · − , and let r g f · o . First, we
shall find the derivative of r at t · 2 using the Chain Rule. The derivative of f is
7.10
L ·
]
]
]
2
3
2
t
t
,
and the derivative of g is
[ ]
M · − − − 6 2 3 2
1 2
2
1 2
2
( ) ( ) x x x x .
At t · 2 , L ·
]
]
]
4
12
; and at g f g ( ( )) ( , ) 2 4 9 · , [ ] M · − 6 3 . Thus the derivative of the
composition is [ ] ML · −
]
]
]
· − · − 6 3
4
12
12 12 [ ] .
Now for fun, let’s find an explicit recipe for r and differentiate:
r t g f t g t t t t ( ) ( ( )) ( , ) ( ) · · + · − −
2 3 2 3 3
1 2 1 . Thus r t t t t t '( ) ( ) ( ) · − − − 3 2 1 4 3
2 3 2 2
,
and so r'( ) ( )(8 ) . 2 3 1 12 12 · − · − It is, of course, very comforting to get the same answer
as before.
There are several different notations for the matrix of the derivative of
f : R R
n p
→ at x R
n
∈ The most usual is simply f '( ) x .
Exercises
7. Let g x x x x x x x ( , , ) ( , )
1 2 3 1 3 2 3
1 · + and f x x x x x x x x x ( , ) ( sin , , )
1 2 1 2 1 1 2 2 1
2
3 2 · + − .
Find the derivative of g f o at (2,4).
8. Let u x y z x y xy x y x y ( , , ) ( , , sin , ) · +
2 3 2
2 and v r s t q r s q r t e
s
( , , , ) ( , ( ) ) · + − −
3
.
a)Which, if either, of the composition functions u v o or v u o is defined? Explain.
b)Find the derivative of your answer to part a).
9. Let f x y e e
x y x y
( , ) ( , )
( ) ( )
·
+ −
and g x y x y x y ( , ) ( , ) · − +
3 2
.
7.11
a)Find the derivative of f g o at the point (1,2).
b)Find the derivative of g f o at the point (1,2).
c) Find the derivative of f f o at the point (1,2).
d) Find the derivative of g g o at the point (1,2).
10. Suppose r t t ·
2
cos and t x y · −
2 2
3 . Find the partial derivatives
∂
∂
r
x
and
∂
∂
r
y
.
7.4 More Chain Rule Stuff
In the everyday cruel world, we seldom compute the derivative of the
composition of two functions by explicitly multiplying the two derivative matrices.
Suppose, as usual, we have r g f · → o : R R
n q
. The the derivative is, as we now know,
r r x x x
r r r
r
x
r
x
r
x
r r r
n
n
n
p p p
n
'( ) '( , , , ) x · ·
]
]
]
]
]
]
]
]
]
]
1 2
1
1
1
2
1
2
1
2
2
2
1 2
K
L
L
M
L
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
x
x
x
x
x
x
.
We can thus find the derivative using the Chain Rule only in the very special case in
which the compsite function is real valued. Specifically, suppose g: R R
p
→ and
f : R R
n p
→ . Let r g f · o . Then r is simply a realvalued function of
x · ( , , , ) x x x
n 1 2
K . Let’s use the Chain Rule to find the partial derivatives.
7.12
r
r
x
r
x
r
x
g
y
g
y
g
y
x x x
x x x
x x x
n p
n
n
n
'( ) x ·
]
]
]
·
]
]
]
]
]
]
]
]
]
]
]
]
]
]
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
f
f
f
f
f
f
f
f
f
1 1 1
2 2 2
p p p
1 2 1 2
1 2
1 2
1 2
L L
L
L
M
L
Thus makes it clear that
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
g
g
g
r
x y
f
x y
f
x y
f
x
j j j p
p
j
· + + +
1
1
2
2
L .
Frequently, engineers and other malefactors do not use a different name for the
composition g f o , and simply use the name g to denote both the composition
g f x x x g f x x x f x x x f x x x
n n n p n
o K K K K K ( , , , ) ( ( , , , ), ( , , , ), , ( , , , ))
1 2 1 1 2 2 1 2 1 2
· and the
function g given by g g y y y
p
( ) ( , , , ) y ·
1 2
K . Since y f x x x
j j n
· ( , , , )
1 2
K , these same
folks also frequently just use y
j
to denote the function f
j
. The Chain Rule given above
then looks even nicer:
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
g
g
g
g
x y
y
x y
y
x y
y
x
j j j p
p
j
· + + +
1
1
2
2
L .
Example
Suppose g x y z x y ye
z
( , , ) · +
2
and x s t · + , y st ·
3
, and z s t · +
2 2
3 . Let us
find the partial derivatives
∂
∂
g
r
and
∂
∂
g
t
. We know that
7.13
∂ g
∂ s
·
∂ g
∂x
∂ x
∂ s
+
∂ g
∂y
∂ y
∂ s
+
∂ g
∂z
∂ z
∂ s
· 2xy(1) +(x
2
+ e
z
)t
3
+ ye
z
(2s)
· 2xy +(x
2
+ e
z
)t
3
+ 2sye
z
Similarly,
∂ g
∂ t
·
∂ g
∂x
∂ x
∂ t
+
∂ g
∂y
∂ y
∂ t
+
∂ g
∂z
∂ z
∂ t
· 2xy(1) + (x
2
+ e
z
)3st
2
+ ye
z
(6t)
· 2xy + 3(x
2
+ e
z
)st
2
+ 6tye
z
These notational shortcuts are fine and everyone uses them; you should, however,
be aware that it is a practice sometimes fraught with peril. Suppose, for instance, you
have g x y z x y z ( , , ) · + +
2 2 2
, and x t z · + , y t z · +
2
2 , and z t ·
3
. Now it is not at all
clear what is meant by the symbol
∂
∂
g
z
. Meditate on this.
Exercises
11. Suppose g x y f x y y s ( , ) ( , ) · − − . Find
∂
∂
∂
∂
g
x
g
y
+ .
12. Suppose the temperature T at the point ( , , ) x y z in space is given by the function
T x y z x xyz zy ( , , ) · + −
2 2
. Find the derivative with respect to t of a particle moving
along the curve described by r i j k ( ) cos sin t t t t · + + 3 .
7.14
13. Suppose the temperature T at the point ( , , ) x y z in space is given by the function
T x y z x y z ( , , ) · + +
2 2 2
. A particle moves along the curve described by
r i j k ( ) sin cos ( ) t t t t t · + + − + π π
2
2 2 . Find the coldest point on the trajectory.
14. Let r x y f x g y ( , ) ( ) ( ) · , and suppose x t · and y t · . Use the Chain Rule to find
dr
dt
.
8.1
Chapter Eight
f R R
n
: →
8.1 Introduction
We shall now turn our attention to the very important special case of functions that
are real, or scalar, valued. These are sometimes called scalar fields. In the very, but
important, special subcase in which the dimension of the domain space is 2, we can
actually look at the graph of a function. Specifically, suppose f : R R
n
→ . The
collection S x x x f x x x · ∈ · {( , , ) : ( , ) }
1 2 3 1 2 3
R
3
is called the graph of f. If f is a
reasonably nice function, then S is what we call a surface. We shall see more of this later.
Let us now return to the general case of a function f : R R
n
→ . The derivative of f is just
a row vector f
f
x
f
x
f
x
n
' ( ) x ·
]
]
]
∂
∂
∂
∂
∂
∂
1 2
L . It is frequently called the gradient of f
and denoted grad f or ∇f .
8.2 The Directional Derivative
In the applications of scalar fields it is of interest to talk of the rate of change of the
function in a specified direction. Suppose, for instance, the function T x y z ( , , ) gives the
temperature at points ( , , ) x y z in space, and we might want to know the rate at which the
temperature changes as we move in a specified direction. Let f : R R
n
→ , let a R
n
∈ ,
and let u R
n
∈ be a vector such that   u · 1. Then the directional derivative of f at a in
the direction of the vector u is defined to be
D f
d
dt
f t
t u
a a u ( ) ( ) · +
·0
.
8.2
Now that we are experts on the Chain Rule, we know at once how to compute such a
thing. It is simply
D f
d
dt
f t f
t u
a a u u ( ) ( ) · + · ∇ ⋅
·0
.
Example
The surface of a mountain is the graph of f x y x y ( , ) · − − 700 5
2 2
. In other words, at
the point (x, y), the height is f (x, y). The positive yaxis points North, and, of course,
then the positive xaxis points East. You are on the mountain side above the point (2, 4)
and begin to walk Southeast. What is the slope of the path at the starting point? Are you
going uphill or downhill? (Which!?).
The answers to these questions call for the directional derivative. We know we are at
the point a · ( , ) 2 4 , but we need a unit vector u in the direction we are walking. This is,
of course, just u · −
1
2
1 1 ( , ) . Next we compute the gradient ∇ · − − f x y x y ( , ) [ , ] 2 10 . At
the point a this becomes ∇ · − − f ( , ) [ , ] 2 4 2 40 , and at last we have
∇ ⋅ ·
− +
· f u
2 40
2
38
2
. This gives us the slope of the path; it is positive so we are going
uphill. Can you tell in which direction the path will be level?
Another Example
The temperature in space is given by T x y z x y yz ( , , ) · +
2 3
. From the point (1,1,1), in
which direction does the temperature increase most rapidly?
We clearly need the direction in which the directional derivative is largest. The
directional derivative is simply ∇ ⋅ ·∇ T T u  cosθ , where θ is the angle between ∇T and
u. Anyone can see that this will be largest when θ = 0. Thus T in creases most rapidly in
8.3
the direction of the gradient of T. Here that direction is [ , , ] 2 3
2 3 2
xy x z yz + . At (1,1,1),
this becomes [2, 2, 3].
Exercises
1. Find the derivative of f x y z x z xy ( , , ) log · + 2 at (1, 2, 1) in the direction of the
vector [ , , ] 1 2 2 .
2. Find the derivative of f x y z x y z xz ( , , ) cos · + − 3
3
at (1, π, 1) in the direction of the
vector [ , , ] 3 2 2  .
3. Find the directions in which g x y x y e y
xy
( , ) sin · +
2
increases and decreases most
rapidly from the point (1, 0).
4. The surface of a hill is the graph of the equation z x x y · + − − 1000
2 4 2
. You stand
on the hill above the point (5,3) and pour out a glass of water. In which direct will it
begin to run? Explain.
5. The position of a particle at time t is given by r i j k ( ) ( sin ) cos t t t t t · − + − 3
2
, and
the position of another particle is R i j k ( ) ( ) sin t t t t t · + + +
2 3
. At time t = π, what
is the rate of change of the distance between the two particles? Are they getting
closer to one another, or are they getting farther apart? (Which!) Explain.
8.3 Surface Normals
8.4
Let f : R R
3
→ be a function and let c be some constant. Recall that the set
S x y z f x y z c · ∈ · {( , , ) : ( , , ) } R
3
is called a level set, or level surface, of the function f .
Suppose r i j k ( ) ( ) ( ) ( ) t x t y t z t · + + describes a curve in R
3
that lies on the surface S.
This means, of course, that f t f x t y t z t c ( ( )) ( ( ), ( ), ( )) r · · . Now look at the derivative
with respect to t of this equation:
d
dt
f t f t ( ( )) ' ( ) r r · ∇ ⋅ · 0 .
In other words, the gradient of f and the tangent to the curve are perpendicular. Note there
was nothing special about our choice of r(t); it is any curve on the surface. The gradient
∇f is thus perpendicular, or normal to the surface f x y z c ( , , ) · .
Example
Suppose we want to find an equation of the plane tangent to the surface
x y z
2 2 2
3 2 12 + + ·
at the point (1, 1, 2). For an equation of a plane, we need a point a on the plane and a
vector N normal to the plane. Then the equation we seek is simply N x a ⋅ − · ( ) 0 ,
where x · ( , , ) x y z . In the case at hand, we have a point on the plane: a = (1, 1, 2).
Let’s find a normal vector N. We have just learned that the gradient of
f x y z x y z ( , , ) · + +
2 2 2
3 2 does the job.
∇ · f x y z x y z ( , , ) [ , , ] 2 6 4 ,
8.5
and so N · ∇ − · − f ( , , ) [ , , ] 1 12 2 68 . The tangent plane is thus given by the equation
N x a ⋅ − · ( ) 0 , which in this case is
2 1 6 1 8 2 0 ( ) ( ) ( ) x y z − − + + − · .
You should note that the discussion here didn’t depend on the dimension of the
domain. Thus if f : R R
2
→ , then the set {( , ) : ( , ) } x y f x y c ∈ · R
2
is a level curve of f,
and the gradient of f is normal to such a curve.
Combining these results with what we know about the directional derivative, we see
that at a point the value of a function increases most rapidly in a direction normal to the
level set passing through that point. On a contour map of a portion of the Earth’s
surface, for example, the steepest path is in the direction normal to the contour lines.
Exercises
6. Find an equation for the plane tangent to the surface z x y · +
2 2
2 at the point (1,1,3).
7. Find an equation for the plane tangent to the surface z x y · + log( )
2 2
at the point
( , , ) 10 0 .
8. Find an equation for the plane tangent to the surface cosπ x x y e yz
xz
− + + ·
2
4 at
the point (0,1,2).
9. Find an equation of the straight line tangent to the curve of intersection of the surfaces
x x y y xy z
3 2 2 3 2
3 4 0 + + + − · and x y z
2 2 2
11 + + · at the point (1, 1, 3).
8.6
8.4 Maxima and Minima
Let f : R R
n
→ . A point a in the domain of f is called a local minimum if there is an
open ball B r ( ; ) a centered at a such that f f ( ) ( ) x a − ≥ 0 for all x a ∈B r ( ; ) . If f is a nice
function, then this means the directional derivative D f
u
a ( ) ≥ 0 for all unit vectors u. In
other words, ∇ ⋅ ≥ f ( ) a u 0 . Then it must be true that both ∇ ⋅ ≥ f ( ) a u 0 and
−∇ ⋅ ·∇ ⋅ − ≥ f f ( ) ( ) ( ) a u a u 0 . This can be so for every u only of ∇ · f ( ) a 0 . Thus f has
a local minimum at a point at which it has a derivative only if the derivative is zero there.
You should guess the definition of a local maximum and see why it must be true that
the gradient is zero at such a point. Thus if a is a local minimum or a local maximum of f,
and if f has a derivative at a, then the derivative ∇ · f ( ) a 0. You should be aware of the
fact that here, just as in Mrs. Turner’s elementary calculus class, the converse is not
necessarily true. We may have ∇ · f ( ) a 0 without a being either a local minimum or a
local maximum.
Example
Let us find all local maxima and local minima of the function
f x y x xy y x y ( , ) · + + + − +
2 2
3 3 4 .
Meditate on just how should proceed. This function clearly has a derivative everywhere,
so at any local maximum or minimum, this derivative, or gradient, must be zero. So let’s
begin by finding all points at which ∇ · f ( ) a 0 . In other words, we want (x, y) at which
∂
∂
f
x
· 0 and
∂
∂
f
y
· 0:
8.7
∂
∂
∂
∂
f
x
x y
f
y
x y
· + + ·
· + − ·
2 3 0
2 3 0
We are thus faced with the borderline trivial problem of solving the system of equations
2 3
2 3
x y
x y
+ · −
+ ·
.
There is just one solution: ( , ) ( , ) x y · −3 3 . Now let us reflect on what we have here.
What we have actually found is all the points that cannot possibly be local minima or
maxima. These are all points except (3, 3).. All we know right now is that this point is
the only possible candidate. Let’s find out what we have by the hammer and tongs
method of examining the quantity f x y f ( , ) ( , ) − + + − − 3 3 33 :
f x y f f x y
x x y y x y
x xy y x
y y
( , ) ( , ) ( , ) ( )
( ) ( )( ) ( ) ( ) ( )
− + + − − · − + + − −
· − + + − + + + + + − + − + +
· + + · +

.
`
,
+
3 3 33 3 3 5
3 3 3 3 3 3 3 3 9
2
3
4
2 2
2 2
2 2
It is therefore clear that f x y f ( , ) ( , ) − + + − − ≥ 3 3 33 0, which means that (3, 3) is a local
minimum.
Exercises
In each of the following, find all local maxima and minima:
8.8
10. f x y x xy y x y ( , ) · + + − + −
2 2
3 3 6 3 6
11. f x y x xy x y ( , ) · + + + +
2
3 2 5
12. f x y xy x y x ( , ) · − − + − 2 5 2 4 4
2 2
13. f x y x xy ( , ) · +
2
2
14. f x y y x ( , ) · −
2
8.5 Least Squares
We shall next look at some very simple, yet important, applications in which the
location of a minimum value of a function is sought.
Suppose we have a set of n points in the plane, say ( , ),( , ), ,( , ) x y x y x y
n n 1 1 2 2
K ,
and we seek the straight line that "best" fits this collection of points. We first decide
what we mean by "best". Let's say we mean the line that minimizes the sum of the
squares of the vertical distances from the points to the line. We can describe all
nonvertical lines in the world by means of two variables, traditionally called m and b.
Thus every such line has the form y mx b · + . Our quest is thus for the values of m and
b at which the function
f m b mx b y
i i
i
n
( , ) ( ) · + −
·
∑
2
1
has its minimum value. Knowing these values will give us our line.
We simply apply our vast and growing knowledge of calculus and find where the
gradient of f is 0:
8.9
∇ · · f
f
m
f
b
( , )
∂
∂
∂
∂
0 .
Now,
∂
∂
f
m
x mx b y m x b x x y
i
i
n
i i i
i
n
i
i
n
i i
i
n
· + − · + −
· · · ·
∑ ∑ ∑ ∑
2 2
1
2
1 1 1
( ) [ ], and
∂
∂
f
b
mx b y m x nb y
i i
i
n
i
i
n
i
i
n
· + − · + −
· · ·
∑ ∑ ∑
2 2
1 1 1
( ) [ ].
We are thus faced with solving the 2 x 2 linear system
m x b x x y
m x bn y
i
i
n
i
i
n
i i
i
n
i
i
n
i
i
n
2
1 1 1
1 1
· · ·
· ·
∑ ∑ ∑
∑ ∑
+ ·
+ ·
Meditate sufficiently to convince yourself that there is always exactly one
solution to this system, and continue meditating sufficiently to convince yourself that
there must be an honesttogoodness minimum of the original function at this solution.
Let's have a go at an example. Suppose we have the following table of values:
x y
0 1
1 2
2 4
3 3.5
4 5
8.10
5 4
7 7
8 9
9 12
10 18
12 21
15 29
The linear system for m and b is
718m + 76b · 1156.5
76m +12b · 115.5
Solving this system gives us m ·
255
142
and b · −
993
568
. In other words, the line that best
fits the data in the “sense of least squares” is
y ·
255
142
x −
993
568
Here’s a picture of this line together with the data points:
8.11
Looks pretty good!
Exercises
15. Here is a table of Köchel numbers versus year of composition for the compositions of
W. A. Mozart. Find the "least squares" straight line approximation to this table and
use it to estimate the year in which Mozart's Sinfonia Concertante in Eflat major was
composed.
Köchel
Number
Year
composed
1 1761
75 1771
155 1772
219 1775
271 1777
351 1780
425 1783
503 1786
575 1789
626 1791
[This problem is taken from Calculus and Analytic Geometry (8th Edition), by
Thomas & Finney.]
8.12
16. Find some data somewhere (The Statistical Abstract of the United States is a good
source of interesting data.), find the least squares linear approximation to the data, and
say something intelligent about your results.
8.6 More Maxima and Minima
In real life, one is most likely interested in finding the places at which the largest
and smallest values of a function f : D R → occur, rather than in simply finding local
maxima and minima. (Here D is a subset of R
n
.). To begin, let's think a moment about
how we can tell if there is a maximum or minimum value of f on D. First, we suppose
that f is continuous—otherwise, anything can happen! Next, what properties of D will
insure the existence of a biggest and smallest value of f ? The answer is fairly simple.
Certainly D must be a closed subset of R
n
; consider, for example the function
f :( , ) 01 → R given simply by f x x ( ) · , which has neither a maximum nor a minimum on
D · ( , ) 01 . Having the domain be closed, however, is not sufficient to guarantee the
existence of a maximum and minimum. Consider, for example f : R R → again with
f :( , ) 01 → R given by f x x ( ) · . The domain R is certainly closed, but f has neither a
maximum nor a minimum. We need also to have the domain be bounded. It turns out that
for continuous f , if the domain D is both closed and bounded, then there must necessarily
be a maximum and a minimum value for f on D. Let's think a moment about what the
candidates for such points are. If the biggest or smallest value of f occurs in the interior of
D, then surely the point at which it occurs is a local maximum (or minimum). If f has a
gradient there, then the gradient must be 0 . The points at which the largest or smallest
values occur must therefore be either i)points in the interior of D at which the gradient of f
vanishes, ii)points in the interior at which the gradient of f does not exist, or iii)points in
D but not in the interior of D (that is, points on the boundary of D).
Hark back to Mrs. Turner's third grade calculus class. How did you find the
maximum value of a function f whose domain D is a closed interval [ , ] a b ⊂ R ? Recall
8.13
found all points in the interior (that is, in the open interval (a,b)) at which the derivative
vanishes. You then simply evaluated f at these points, evaluated f at any points in (a,b)
at which there is no derivative, evaluated f at the two end points of the interval (in this
one dimensional case, the boundary of D is particularly simple.), and then picked out the
biggest and smallest numbers you computed. The situation in higher dimensions is a bit
more complicated, mostly because the boundary of even a nice domain D is not a nice
finite set as in the case of an interval, but is an infinite set. Let's look at an example.
Example
A flat circular plate has the shape of the region {( , ) : } x y x y ∈ + ≤ R
2 2 2
1 . The
temperature at the point ( , ) x y on the plate is given by T x y x y x ( , ) · + −
2 2
2 . Our
assignment is to find the hottest and coldest points on the plate. According to our
previous discussion, candidates for the hottest and coldest points are all points inside the
circular boundary at which the gradient of T is 0 and all points on the boundary. (Note
that T has a gradient at all points inside the circle.) First, let's find where among all points
( , ) x y such that x y
2 2
1 + < , the ones at which ∇ · − · T x y ( , ) 2 14 0 . This is easy; it
should be clear there is just one such point: ( , )
1
2
0 . Now for the more difficult part,
finding the candidates on the boundary. Note that the boundary may be described by the
vector equation
r i j ( ) cos sin t t t · + , where 0 2 ≤ ≤ t π .
The temperature on this set is then given by
T t T t t ( ) ( ( )), · ≤ ≤ r 0 2π
[Here we are abusing the notation, as we have done before, by using the same name for
the function T x y ( , ) and the composition T t ( ( )) r .] We are now faced with the one
dimensional problem of finding the maximum and minimum values of a nice differentiable
function of one variable on a closed interval. First, we know the endpoints of the interval
are candidates: t t · · 0 2 , and π . We have at this point added one more point to our list
8.14
of candidates: r r ( ) ( ) ( , ) 0 2 10 · · π . Now for candidates inside the interval, we seek
places at which the derivative
dT
dt
· 0 . From the Chain Rule, we know
dT
dt
T t t t t t t t t t · ∇ ⋅ · − ⋅ − · + ( ( )) '( ) ( cos , sin ) ( sin ,cos ) cos sin sin r r 2 14 2 .
The equation
dT
dt
· 0 now becomes
2 0
2 1 0
cos sin sin ,
sin ( cos )
t t t
t t
+ ·
+ ·
or
Thus sint · 0 , or 2 1 0 cos . t + · We have, in other words, y · 0 , or x · −
1
2
. When
y · 0 , then x · 1 or x · −1; and when x · −
1
2
, then y ·
3
2
or y · −
3
2
. Thus our
new candidates are ( , ), ( , ), , ), 1 0 10
3
2
(
1
2
− and ( , ) − −
1
2
3
2
. These together with the one
we have already found, ( , )
1
2
0 , make up our entire list of possibilities for the hottest and
coldest points on the plate. All we need do now is to compute the temperature at each of
these points:
T
T
T
T T
( , ) .
( , )
( , )
( , ) ( , )
1
2
0
1
4
1
2
1
4
10 1 1 0
10 1 1 2
1
2
3
2
1
2
3
2
1
4
3
2
1
2
9
4
· − · −
· − ·
− · + ·
− · − − · + + ·
Finally, we have our answer. The coldest point is ( , )
1
2
0 , and the hottest points are
( , ) −
1
2
3
2
and ( , ) − −
1
2
3
2
.
8.15
Exercises
17. Find the maximum and minimum value of f x y x xy y ( , ) · − + +
2 2
4 on the closed
area in the first quadrant bounded by the triangle formed by the lines x · 0 , y · 4 ,
and y x · .
18. Find the maximum and minimum values of f x y y y x ( , ) ( )cos · − 4
2
on the closed
area bounded by the rectangle 1 3 ≤ ≤ y , − ≤ ≤
π π
4 4
x .
8.7 Even More Maxima and Minima
It should be clear now that the really troublesome part of finding maxima and
minima is in dealing with the constrained problem; that is, the problem of finding the
maxima and minima of a given function on a set of lower dimension than the domain of the
function. In the problems of the previous section, we were fortunate in that it was easy
to find parametric representations of the these sets; in general, this, of course, could be
quite difficult. Let's see what we might do about this difficulty.
Suppose we are faced with the problem of finding the maximum or minimum value
of the function f : D R → , where D R
2
· ∈ · {( , ) : ( , ) } x y g x y 0 , where g is a nice
function. (In other words, D is a level curve of g .) Suppose r( ) t is a vector description
of the curve D. Now then, we are seeking a maximum or minimum of the function
F t f t ( ) ( ( )) · r . At a maximum or minimum, we must have
dF
dt
· 0 . (Here g is
sufficiently nice to insure that g x y ( , ) · 0 is a closed curve, and so there are no endpoints
to worry about.) The Chain Rule tells us that
dF
dt
f · ∇ ⋅ · r' 0 . Thus at a maximum or
minimum, the gradient of f must be perpendicular to the tangent to g x y ( , ) · 0 . But if
∇f is perpendicular to the tangent to the level curve g x y ( , ) · 0 , then it must have the
8.16
same direction as the normal to this curve. This is just what we need to know, for the
gradient of g is normal to this curve. Thus at a maximum or minimum, ∇f and ∇g must
"line up". Thus ∇ · ∇ f λ g , and there is no need actually to know a vector representation
r for g x y ( , ) · 0 .
Let's see this idea in action. Suppose we wish to find the largest and smallest
values of f x y x y ( , ) · +
2 2
on the curve x x y y
2 2
2 4 0 − + − · .
Here, we may take g x y x x y y ( , ) · − + −
2 2
2 4 . Then ∇ · + f x y 2 2 i j , and
∇ · − + − g x y ( ) ( ) 2 2 2 4 i j , and our equation ∇ · ∇ f λ g becomes
2 2 2
2 2 4
x x
y y
· −
· −
λ
λ
( )
( )
We obtain a third equation from the requirement that the point ( , ) x y be on the curve
g x y ( , ) · 0 . In other words, we need to find all solutions to the system of equations
2 2 2
2 2 4
2 4 0
2 2
x x
y y
x x y y
· −
· −
− + − ·
λ
λ
( )
( )
The first two equations become
x
y
( )
( )
λ λ
λ λ

1
1 2
·
− ·
Thus x ·
−
λ
λ 1
and y ·
−
2
1
λ
λ
. (What about the possibility that λ − · 1 0 ?). The last
equation then becomes
λ
λ
λ
λ
λ
λ
λ
λ
2
2
2
2
1
2
1
4
1
8
1
0
( ) ( ) −
−
−
+
−
−
−
· ; or,
λ λ λ
λ λ
2
2
2 1 0
2 0
− − ·
− ·
( ) ,
We have two solutions: λ · 0 and λ · 2 . What do you make of the solution λ · 0?
These values of λ give us two candidates for places at which extrema occur: x · 0 and
y · 0 ; and x · 2 and y · 4 . Now then f ( , ) 00 0 · , and f ( , ) 2 4 4 16 20 · + · . There
8.17
we have them—the minimum value is 0 and it occurs at (0,0); and the maximum value is
20, and it occurs at (2,4).
This method for finding "constrained" extrema is generally called the method of
Lagrange Multipliers. (The variable λ is called a Lagrange multiplier.)
Exercises
19. Use the method of Lagrange multipliers to find the largest and smallest values of
f x y x y ( , ) · + 4 3 on the circle x y
2 2
1 + · .
20. Find the points on the ellipse x y
2 2
2 1 + · at which f x y xy ( , ) · has its extreme
values.
21. Find the points on the curve x xy y
2 2
1 + + · that are nearest to and farthest from the
origin.
9.1
Chapter Nine
The Taylor Polynomial
9.1 Introduction
Let f be a function and let F be a collection of "nice" functions. The approximation
problem is simply to find a function g ∈F that is "close" to the given function f . There are
two issues immediately. How is the collection F selected, and what do we mean by
"close"? The answers depend on the problem at hand. Presumably we want to do
something to f that is difficult or impossible (This might be something as simple as finding
f x ( ) for some x.). The collection F would thus consist of functions to which it is easy to
do that which we wish to do to f . Our measure of how close one function is to another
would try to reflect the closeness of the results of our operations. Now, what are we
talking about here. Suppose, for example, we wish to find f x ( ) . Our collection F of
functions should include functions that are easy to evaluate at x , and two function would
be "close" simply if there values are close. We might, for instance, want to evaluate sin x
for all x is some interval I. The collection F could be a collection of second degree
polynomials. The approximation problem is then to find elements of F that make the
"distance" max{sin ( ): } x p x x I − ∈ as small as possible. Similarly, we might want to find
the integral of some function f over an interval I . Here we would want F to consist of
functions easily integrated and measure the distance between functions by the difference of
their integrals over I . In the previous chapter, we found the "best" straight line
approximation to a set of data points. In that case, the collection F consisted of all
nonvertical straight lines, and we measured the distance between functions by the sum of
the squares of their differences on a specified set of points { , , , } x x x
n 1 2
K . You can
imagine many other examples.
9.2 The Taylor Polynomial
We look first at a simple but useful problem: Given a nice function f : D R R ⊂ → , a
point a in the interior of the domain D , and an integer n , find a polynomial p of degree
≤ n such that
9.2
p a f a
p a f a
p a f a
p a f a
n n
( ) ( )
' ( ) ' ( )
' ' ( ) ' ' ( )
( ) ( )
( ) ( )
·
·
·
·
M
We solve the problem by the Behold Method. Simply verify that
p x f a f a x a
f a
x a
f a
x a
f a
n
x a
n
n
( ) ( ) ' ( )( )
' ' ( )
!
( )
' ' ' ( )
!
( )
( )
!
( )
( )
· + − + − + − + + −
2 3
2 3
K
does the job! It is also fairly easy to see that this polynomial is the only polynomial of
degree ≤ n that does the job. Suppose q is also a polynomial with degree g ≤ n such that
p a f a
p a f a
p a f a
p a f a
n n
( ) ( )
' ( ) ' ( )
' ' ( ) ' ' ( )
( ) ( )
( ) ( )
·
·
·
·
M
and consider the function r p q · − . Note that r is also a polynomial of degree ≤ n . But
r a r a r a r a
n
( ) ' ( ) ' ' ( ) ( )
( )
· · · · · K 0 .
Or, in other words, r has a zero of order n + 1, and the only way this can happen is if
r x ( ) ≡ 0 for all x . That is, p x q x ( ) ( ) ≡ identically.
Example
Let f x x ( ) sin · and let a · 0. Let's find the Taylor polynomial for a few different
values of n. For n = 1, we have simply p x f a f a x a x x
1
0 0 ( ) ( ) ' ( )( ) sin cos ( ) · + − · + · .
Note that for n = 2, we have p x x x x
2
2
0 0 0 ( ) sin cos ( ) sin ( ) · + − · , also. Let's take a look
at the next Taylor polynomial. Here p x x
x
3
3
6
( ) · − . Let's draw some pictures; we'll look
at the graph of p
3
and f . We shall use Maple.
9.3
What we see is that the Taylor polynomial looks like a pretty good approximation as long
as we don't get too far away from a = 0. Let us continue. Convince yourself that p p
4 3
· ,
and p x x
x x
5
3 5
6 120
( ) · − + . Another picture:
9.4
Exercises
1. Find the Taylor polynomial of degree n for f x e
x
( ) · at a = 0.
2. Find the Taylor polynomial of degree n for f x x ( ) ·
3
at a = 1.
3. Find the Taylor polynomial of degree 3 for f x x ( ) log · at a = 1.
4. Find the Taylor polynomial of degree n for f x x ( ) sin · at a = 0.
5. Find the Taylor polynomial of degree 3 for f x x ( ) · at a = 4.
9.3 Error
Let's see how close the Taylor polynomial is to the function f . To do this, suppose p is
the Taylor polynomial of degree ≤ n for the function f at a , and consider the function
g t f t p t
t a
x a
f x p x
n
n
( ) ( ) ( )
( )
( )
( ( ) ( )) · − −
−
−
−
+
+
1
1
.
(We assume x a ≠ .) Note that g a g x ( ) ( ) · · 0. Now, from the Mean Value Theorem (or
Rolle's Theorem, or whatever.) we know that g' ( ) ξ
1
0 · for some ξ
1
between a and x .
But note also that g a f a p a
n a a
x a
f x p x
n
n
' ( ) ' ( ) ' ( )
( )( )
( )
( ( ) ( )) · − −
+ −
−
− ·
+
1
0
1
. It thus follows
from the Mean Value Theorem that the derivative of g' is zero at some ξ
2
between a and
ξ
1
. Also, g a f a p a
n n a a
x a
f x p x
n
n
' ' ( ) ' ' ( ) ' ' ( )
( ) ( )
( )
( ( ) ( )) · − −
+ −
−
− ·
−
+
1
0
1
1
. Once again, from
the celebrated Mean Value Theorem, we conclude that g' ' ' ( ) ξ
3
0 · for some ξ
3
between a
and ξ
2
. Continuing in this fashion, we are finally able to conclude that g
n ( )
( )
+
·
1
0 ξ for
some ξ . Let's see what this looks like.
g t f t p t
n
x a
f x p x
n n n
n
( ) ( ) ( )
( ) ( ) ( )
( )!
( )
( ( ) ( ))
+ + +
+
· − −
+
−
−
1 1 1
1
1
9.5
and so g
n ( )
( )
+
·
1
0 ξ becomes
f
n
x a
f x p x
n
n
( )
( )
( )!
( )
( ( ) ( ))
+
+
−
+
−
− ·
1
1
1
0 ξ .
(Remember, p is a polynomial of degree ≤ n , and so p t
n ( )
( )
+
≡
1
0 . From this we obtain an
expression for the difference between f and the Taylor polynomial g :
f x p x
f
n
x a
n
n
( ) ( )
( )
( )!
( )
( )
− ·
+
−
+
+
1
1
1
ξ
.
Example
Remember when in 7
th
grade physics class, Mr. Crews replaced the sine of a "small"
angle θ by θ itself ? He assured us that for small angles this was just fine. Well, what was
going on here? Let's see if our newfound knowledge of Taylor polynomials will help.
Observe that p( ) θ θ · is simply the Taylor polynomial of degree ≤ 2 for f ( ) sin θ θ · at
a · 0. Using the result just derived, we have that
sin
sin
θ θ
ξ
θ − ·
−
6
3
.
Now, we don't know what ξ is, but we do know that sin } ξ ≤ 1; thus
sin  θ θ
θ
− ≤
3
6
,
and we have a precise estimate of the error incurred by substituting θ for sinθ . Suppose,
for example, that θ · 10
o
; then what? Well, θ π
π
· ·
10
360
2
18
. Then the error we get when
we use
π
18
instead of sin
π
18
is estimated by
sin .
π π θ
18 18
1
6 18
0 008862
3
− ≤

.
`
,
≤ .
Now we know exactly what "pretty close" means. For 10 degrees, I guess that's "not too
bad."
Exercises
9.6
6. a)Find the Taylor polynomial of degree ≤ 2 for f x e
x
( ) · at a =0.
b)Use the result of part a) to find an approximation for e .
c)Find as small an upper bound as you can for the difference between your
approximation found in part b) and e .
7. Use the Taylor polynomial found in Exercise 3 to approximate log(. ) 11 and find an
upper bound for the magnitude of the difference between your approximation and
log(. ) 11 .
8. For what values of x can you replace sin x by x
x
−
3
6
with an error of magnitude no
greater than 3 10
4
×
−
?
9. Calculate e with an error of less than 10
6
.
Taylor’s Theorem
1. Introduction. Suppose f is a onevariable function that has n + 1 derivatives on an interval
about the point x = a. Then recall from Ms. Turner’s class the single variable version of Taylor’s
Theorem tells us that there is exactly one polynomial p of degree ² n such that pÝaÞ = fÝaÞ,
p
v
ÝaÞ = f
v
ÝaÞ, p
vv
ÝaÞ = f
vv
ÝaÞ, u, p
ÝnÞ
ÝaÞ = f
ÝnÞ
ÝaÞ. This polynomial is given by
pÝxÞ = fÝaÞ + f
v
ÝaÞÝx ? aÞ +
f
vv
ÝaÞ
2!
Ýx ? aÞ
2
+ u +
f
ÝnÞ
n!
Ýx ? aÞ
n
We also know the difference between fÝxÞ and pÝxÞ:
fÝxÞ ? pÝxÞ =
f
Ýn+1Þ
ÝYÞ
Ýn + 1Þ!
Ýx ? aÞ
n+1
,
where Y is somewhere between a and x.
The polynomial p is called the Taylor Polynomial of degree ² n for f at a.
Before we worry about what the Taylor polynomial might be in higher dimensions, we need to be
sure we understand what is a polynomial in more than one dimension. In two dimensions, a
polynomial pÝx, yÞ of degree ² n is a function of the form
pÝx, yÞ =
>
i,j=0
i+j=n
a
ij
x
i
y
j
.
Thus a polynomial of degree ² 2 (perhaps more commonly known as a quadratic) looks like
pÝx, yÞ = a
00
+ a
10
x + a
01
y + +a
11
xy + a
20
x
2
+ a
02
y
2
.
I hope it easy to guess what one means by a polynomial in three variables, Ýx, y, zÞ, or indeed, in
any number of variables.
Now, how might we extend the idea of the Taylor polynomial of degree ² n for a function f at a
point a ? Simple enough. It’s a polynomial pÝxÞ of degree ² n so that
/
i
1
+u+iq
fÝaÞ
/x
1
i
1
/x
2
i
2
u/x
q
iq
=
/
i
1
+u+iq
pÝaÞ
/x
1
i
1
/x
2
i
2
u/x
q
iq
,
for all i
1
, i
2
, u, i
q
such that i
1
+ i
2
+ u + i
q
² n.
This looks pretty ferocious in general, so let’s see what it says for just two variables. In this case,
we have a =Ýa, bÞ and the Taylor polynomial pÝx, yÞ at a becomes the polynomial such that
1
/
i+j
fÝaÞ
/
i
x/
j
y
=
/
i+j
pÝaÞ
/
i
x/
j
y
,
for all i + j ² n.
Example
Let fÝx, yÞ = cosÝx + yÞ, and let pÝx, yÞ = 1 ?
x
2
2
? xy ?
y
2
2
. Let’s verify that p is the Taylor
polynomial of degree ² 2 for f at Ý0, 0Þ. He we go.
fÝ0, 0Þ = 1, and pÝ0, 0Þ = 1;
/f
/x
= ?sinÝx + yÞ, and
/p
/x
= ?x ? y;
/f
/y
= ?sinÝx + yÞ, and
/p
/y
= ?x ? y;
/
2
f
/x
2
= ?cosÝx + yÞ, and
/
2
p
/x
2
= ?1,
/
2
f
/y
2
= ?cosÝx + yÞ, and
/
2
p
/y
2
= ?1,
/
2
f
/x/y
= ?cosÝx + yÞ, and
/
2
p
/x/y
= ?1.
Now it’s easy to see that
fÝ0, 0Þ = 0 = pÝ0, 0Þ;
/f
/x
Ý0, 0Þ = 0 =
/p
/x
Ý0, 0Þ;
/f
/y
Ý0, 0Þ = 0 =
/p
/y
Ý0, 0Þ;
/
2
f
/x
2
Ý0, 0Þ = ?1 =
/
2
p
/x
2
Ý0, 0Þ;
/
2
f
/y
2
Ý0, 0Þ = ?1 =
/
2
p
/y
2
Ý0, 0Þ; and
/
2
f
/x/y
Ý0, 0Þ = ?1 =
/
2
p
/x/y
Ý0, 0Þ.
Exercises
1. Verify that the polynomial in the Example is also the Taylor polynomial for f at (0,0) of degree
² 3.
2. Let fÝx, yÞ = sinÝx + yÞ.Which Which of the following is the Taylor polynomial of degree ² 2 for
f at (0,0)? Explain.
a) pÝx, yÞ = 1 + x
2
+ y
2
b) pÝx, yÞ = xy
2
c) pÝx, yÞ = x
2
+ xy + 2y d) pÝx, yÞ = x + y
2. Derivatives. Prior to finding a general recipe for the Taylor polynomial, we need look at finding
higher order derivatives of certain composite functions. Let f be a realvalued function defined on a
subset of R
q
. Suppose that in a neighborhood of the point x, the function f has a lot of continuous
partial derivatives. Define the function g by
gÝtÞ = fÝa + thÞ,
where a = Ýa
1
, a
2
, u, a
q
Þ and h = Ýh
1
, h
2
, u, h
q
Þ. We know from the chain rule that g
v
ÝtÞ is
given by
g
v
ÝtÞ = 4fÝa + thÞ 6 h
=
/f
/x
1
,
/f
/x
2
, u,
/f
/x
q
6 Ýh
1
, h
2
, u, h
q
Þ
= h
1
/
/x
1
+ h
2
/
/x
2
+ u + h
q
/
/x
q
f
Ýa+thÞ
In keeping with our general practice of restricting ourselves to dimensions one, two, or three, let’s
look first at the case q = 2. As usual, we’ll write x =Ýx, yÞ and h = Ýh, kÞ. The expression for g
v
ÝtÞ
now looks like:
g
v
ÝtÞ = h
/
/x
+ k
/
/y
f
Ýx+thÞ
We are now in business, for we have a nice recipe for higher order derivatives of g :
g
ÝmÞ
ÝtÞ = h
/
/x
+ k
/
/y
m
f
Ýx+thÞ
For example,
g
vv
ÝtÞ = h
/
/x
+ k
/
/y
2
f
= h
2 /
2
/x
2
+ 2hk
/
2
/x/y
+ k
2 /
2
/y
2
f
= h
2 /
2
f
/x
2
+ 2hk
/
2
f
/x/y
+ k
2 /
2
f
/y
2
Example
Suppose fÝx, yÞ = x
2
y
3
+ y
2
. Let’s find the second derivative of the function
gÝtÞ = fÝ1 + 3t, ?2 + tÞ
3
First,
g
vv
ÝtÞ = 3
/
/x
+
/
/y
2
f
= 9
/
2
f
/x
2
+ 6
/
2
f
/x/y
+
/
2
f
/y
2
Now,
/f
/x
= 2xy
3
, and
/f
/y
= 3x
2
y
2
+ 2y, and so
/
2
f
/x
2
= 2y
3
,
/
2
f
/y/x
= 6y
2
, and
/
2
f
/y
2
= 6x
2
y + 2.
Thus,
g
vv
ÝtÞ = 18Ý?2 + tÞ
3
+ 36Ý?2 + tÞ
2
+ 6Ý1 + 3tÞ
2
Ý?2 + tÞ + 2
Exercises
3. Let fÝx, yÞ = xe
y
. Find the derivative of gÝtÞ = fÝ1 + t, 3 ? 4tÞ.
4. Find the second derivative of the function g defined in Problem 3.
5. Let FÝu, vÞ = u
3
v + v
2
. Find the second derivative of RÝzÞ = FÝz, 3zÞ.
6. Find g
vvv
ÝtÞ, where g is the function defined in the Example.
3. The Taylor polynomial. To find the Taylor polynomial for a function f of several variables at a
point a, we shall simply apply the onedimensional results to the function
gÝtÞ = fÝa + thÞ.
Thus,
gÝtÞ =
>
m=0
n
g
ÝmÞ
Ý0Þ
m!
t
m
+
g
Ýn+1Þ
ÝYÞ
Ýn + 1Þ!
t
n+1
,
where Y is a number between 0 and t. Next, substitute t = 1 into the above:
gÝ1Þ = fÝaÞ =
>
m=0
n
g
ÝmÞ
Ý0Þ
m!
+
g
Ýn+1Þ
ÝYÞ
Ýn + 1Þ!
We know the value of g
ÝkÞ
from Section 2:
fÝa + hÞ =
>
m=0
n
1
m!
h
1
/
/x
1
+ h
2
/
/x
2
+ u + h
q
/
/x
q
m
fÝaÞ
4
+
1
Ýn + 1!
h
1
/
/x
1
+ h
2
/
/x
2
+ u + h
q
/
/x
q
n+1
fÝcÞ
The point c lies somewhere on the line segment joining a and a + h.
The polynomial
pÝhÞ =pÝh
1
, h
2
, u, h
q
Þ =
>
m=0
n
1
m!
h
1
/
/x
1
+ h
2
/
/x
2
+ u + h
q
/
/x
q
m
fÝaÞ
is the Taylor polynomial of degree ² n for f at a; the last term is traditionally called the error term
or sometimes, the remainder term. Actually, if we let h = x ? a, then qÝxÞ =pÝx ? aÞ is the thing
we called the Taylor polynomial in the first section.
This is pretty fierce looking. Let’s look at the two variable case:
fÝa
1
+ h, a
2
+ kÞ =
>
m=0
n
1
m!
h
/
/x
+ k
/
/y
m
fÝa
1
, a
2
Þ
+
1
Ýn + 1!Þ
h
/
/x
+ k
/
/y
n+1
fÝc
1
, c
2
Þ
where Ýc
1
, c
2
Þ is on the line joining Ýa
1
, a
2
Þ and Ýa
1
+ h, a
2
+ kÞ.
Example
Let fÝx, yÞ = sinxsiny. For n = 2 and a = Ý0, 0Þ, Taylor’s polynomial becomes
pÝh, kÞ = fÝ0, 0Þ + h
/f
/x
Ý0, 0Þ + k
/f
/y
Ý0, 0Þ +
h
2
2
/
2
f
/x
2
Ý0, 0Þ + hk
/
2
f
/x/y
Ý0, 0Þ +
k
2
2
/
2
f
/y
2
Ý0, 0Þ
We have
/f
/x
= cos xsiny;
/f
/y
= sinxcos y;
/
2
f
/x
2
= ?sinxsiny;
/
2
f
/x/y
= cos xcosy;
/
2
f
/y
2
= ?sinxsiny.
Thus,
pÝh, kÞ = hk.
Let’s get an estimate for how well this approximates sinxsiny near Ý0, 0Þ. We know that
sinxsiny ? xy =
1
3!
x
/
/x
+ y
/
/y
3
fÝY, WÞ
where ÝY, WÞ is one the segment joining Ýx, yÞ and the origin. Now,
5
x
/
/x
+ y
/
/y
3
f = x
3 /
3
f
/x
3
+ 3x
2
y
/
3
f
/x
2
/y
+ 3xy
2 /
3
f
/x/y
2
+ y
3 /
3
f
/x
3
.
Next, let’s suppose that x ² c and y ² c for some constant c. Noting that all the partial
derivatives in the above expression are simply products of sine and cosines, we can estimate
x
/
/x
+ y
/
/y
3
f ² 8c
3
,
and so, at last,
sinxsiny ? xy ²
8c
3
6
=
4
3
c
3
Exercises
7. Find the Taylor polynomial of degree ² 1 for fÝx, yÞ = e
xy
at Ý0, 0Þ.
8. Find the Taylor polynomial of degree ² 2 for fÝx, yÞ = e
xy
at Ý0, 0Þ.
9. Find the Taylor polynomial of degree ² 3 for fÝx, yÞ = e
xy
at Ý0, 0Þ.
10. Find the Taylor polynomial of degree ² 1 for fÝx, yÞ = e
x
cos y at Ý0, 0Þ.
11. Use Taylor’s Theorem to find a quadratic approximation of e
x
cos y at the origin.
12. Estimate the error in the approximation found in Problem 11 if x ² 0.1 and y ² 0.1.
6
10.1
Chapter Ten
Sequences, Series, and All That
10.1 Introduction
Suppose we want to compute an approximation of the number e by using the Taylor
polynomial p
n
for f x e
x
( ) · at a =0. This polynomial is easily seen to be
p x x
x x x
n
n
n
( )
!
· + + + + + 1
2 6
2 3
K .
We could now use p
n
( ) 1 as an approximation to e . We know from the previous chapter
that the error is given by
e p
e
n
n
n
− ·
+
+
( )
( )!
1
1
1
1
ξ
,
where 0 1 < < ξ . Assume we know that e <3, and we have the estimate
0 1
3
1
≤ − ≤
+
e p
n
n
( )
( )!
.
Meditate on this error estimate. It tells us that we can make this error as small as we like by
choosing n sufficiently large. This is expressed formally by saying that the limit of
p
n
( ) 1 as n becomes infinite is e . This is the idea we shall study in this chapter.
10. 2 Sequences
A sequence of real numbers is simply a function from a subset of the nonnegative
integers into the reals. If the domain is infinite, we say the sequence is an infinite
sequence. (Guess what a finite sequence is.) We shall be concerned only with infinite
sequences, and so the modifier will usually be omitted. We shall also almost always
consider sequences in which the domain is either the entire set of nonnegative or positive
integers.
There are several notational conventions involved in writing and talking about
sequences. If f Z :
+
→ R , it is customary to denote f n ( ) by f
n
, and the sequence itself
by ( ) f
n
. (Here Z
+
denotes the positive integers.) Thus, for example,
1
n

.
`
,
is the sequence
10.2
f defined by f n
n
( ) ·
1
. The function values f
n
are called terms of the sequence.
Frequently one sees a sequence described by writing something like
14 9
2
, , , , , K K n .
This is simply another way of describing the sequence ( ). n
2
Let ( ) a
n
be a sequence and suppose there is a number L such that for any ε >0,
there is an integer N such that   a L
n
− < ε for all n > N . Then L is said to be a limit of the
sequence, and ( ) a
n
is said to converge to L . This is usually written lim
n
n
a L
→∞
· . Now,
what does this really mean? It says simply that as n gets big, the terms of the sequence get
close to L . I hope it is clear that 0 is a limit of the sequence
1
n

.
`
,
. From the discussion
in the Introduction to this chapter, it should be reasonably clear that a limit of the sequence
1
1
2
1
6
1
+ + + +

.
`
,
K
n!
is e .
The graph of a sequence is pretty dreary compared with the graph of a function
whose domain is an interval of reals, but nevertheless, a look at some pictures can help
understand some of these definitions. Suppose the sequence ( ) a
n
converges to L . Look at
the graph of ( ) a
n
:
The fact that L is a limit of the sequence means that for any ε >0, there is an N so that to the
right of N , all the spots are in the strip of width 2ε centered at L.
Exercises
10.3
1. Prove that a sequence can have at most one limit (We may thus speak of the limit of
a sequence.).
2. Give an example of a sequence that does not have a limit. Explain.
3. Suppose the sequence ( ) , , , a a a a
n
·
0 1 2
K converges to L. Explain how you know
that the sequence ( ) , , , a a a a
n+
·
5 5 6 7
K also converges to L.
4. Find the limit of the sequence
3
2
n

.
`
,
, or explain why it does not converge.
5. Find the limit of the sequence
3 2 7
2
2
n n
n
+ − 
.
`
,
, or explain why it does not converge.
6. Find the limit of the sequence
5 7 2
3 10
3 2
3 2
n n n
n n n
− + +
+ − +

.
`
,
, or explain why it does not
converge.
7. Find the limit of the sequence
logn
n

.
`
,
, or explain why it does not converge.
10. 3 Series
Suppose ( ) a
n
is a sequence. The sequence ( ) a a a
n 0 1
+ + + K is called a series. It is a
little neater to write if we use the usual summation notation: a
k
k
n
·
∑

.
`
,
0
. We have seen an
example of such a thing previously; viz.,
1
1
2
1
6
1 1
0
+ + + +

.
`
,
·

.
`
,
·
∑
K
n k
k
n
! !
.
It is usual to replace lim
n
k
k
n
a
→∞
·
∑
0
by a
k
k ·
∞
∑
0
. Thus, one would, for example, write
e
k
k
·
·
∞
∑
1
0
!
.
10.4
One also frequently sees the limit a
k
k ·
∞
∑
0
written as a a a
n 0 1
+ + + + K K. And one more word
of warning. Some poor misguided souls also use a
k
k ·
∞
∑
0
to stand simply for the series
a
k
k
n
·
∑

.
`
,
0
. It is usually clear whether the series or the limit of the series is meant, but it is
nevertheless an offensive practice that should be ruthlessly and brutally suppressed.
Example
Let's consider the series
1
2
1
1
2
1
4
1
2
0
k
k
n
n
·
∑

.
`
,
· + + + +

.
`
,
K . Let
S
n n
· + + + + + 1
1
2
1
4
1
8
1
2
K . Then
1
2
1
2
1
4
1
8
1
2
1
2
1
S
n n n
· + + + + +
+
K .
Thus
S
S S
n
n n n
2
1
2
1
1
2
1
· − · −
+
.
This makes it quite easy to see that lim
n
n
S
→∞
·
2
1, or lim
n
n
S
→∞
· 2 . In other words,
1
2
2
0
k
k ·
∞
∑
· .
Observe that for series a
k
k
n
·
∑

.
`
,
0
to converge, it must be true that lim
n
n
a
→∞
· 0. To see
this, suppose L a
k
k
·
·
∞
∑
0
, and observe that a a a
n k
k
n
k
k
n
· −
· ·
−
∑ ∑
0 0
1
. Thus,
lim lim lim lim
.
n
n
n
k
k
n
k
k
n
n
k
k
n
n
k
k
n
a a a a a
L L
→∞ →∞
· ·
−
→∞
·
→∞
·
−
· −

.
`
,
· −
− ·
∑ ∑ ∑ ∑
0 0
1
0 0
1
0 =
10.5
In other words, if lim
n
n
a
→∞
≠ 0, then the series a
k
k
n
·
∑

.
`
,
0
does not have a limit.
Another Example
Consider the series
1
1
k
k
n
·
∑

.
`
,
. First, note that lim
n
k
→∞
·
1
0. Thus we do not know that
the series does not converge; that is, we still don't know anything. Look at the following
picture:
0 2 4 6 8 10 12
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
The curve is the graph of y
x
·
1
. Observe that the area under the "stairs" is simply
1
1
11
k
k·
∑
.
Now convince yourself that
1
1
k
k
n
·
∑
is larger than the area under the curve y
x
·
1
from x =1
to x = n+1. In other words,
1 1
1
1
1
1
k x
dx n
k
n
n
·
+
∑ ∫
> · + log( ) .
10.6
We know that log( ) n + 1 can be made as large as we wish by choosing n sufficiently large.
Thus
1
1
k
k
n
·
∑
can be made as large as we wish by choosing n sufficiently large. From this it
follows that the series
1
1
k
k
n
·
∑

.
`
,
does not have a limit. (This series has a name. It is called
the harmonic series. )
The method we used to show that the harmonic series does not converge can be used
on many other series. We simply consider a picture like the one above. Suppose we have a
series a
k
k
n
·
∑

.
`
,
1
such that a
k
> 0 for all k . Suppose f is a decreasing function such that
f k a
k
( ) · for all k . Then if the limit lim ( )
R
R
f x dx
→∞
∫
1
does not exist, the series is divergent.
Exercises
8. Find the limit of the series
1
3
0
n
k
n
·
∑

.
`
,
, or explain why it does not converge.
9. Find the limit of the series
5
3
0
n
k
n
+

.
`
,
·
∑
, or explain why it does not converge.
10. Find a value of n that will insure that 1
1
2
1
3
1
10
6
+ + + + > K
n
.
11. Let 0 1 ≤ ≤ θ . Prove that sin ( )
( )!
θ
θ
· −
+
+
·
∞
∑
1
2 1
2 1
0
k
k
k
k
.
[Hint: p
k
n
k
k
k
n
2 1
2 1
0
1
2 1
+
+
·
· −
+
∑
( ) ( )
( )!
θ
θ
is the Taylor polynomial of degree < 2n+1 for
the function f ( ) sin θ θ · at a = 0.]
10.7
12. Suppose we have a series a
k
k
n
·
∑

.
`
,
1
such that a
k
> 0 for all k , and suppose f is a
decreasing function such that f k a
k
( ) · for all k . Show that if the limit
lim ( )
R
R
f x dx
→∞
∫
1
exists, then the series is convergent.
13. a) Find all p for which the series
1
1
k
p
k
n
·
∑

.
`
,
converges.
b) Find all p for which the series in a) diverges.
10.4 More Series
Consider a series a
k
k
n
·
∑

.
`
,
0
in which a
k
≥ 0 for all k . This is called a positive
seri es. Let b
k
k
n
·
∑

.
`
,
0
be another positive series. Suppose that b a
k k
≤ for all k > N , where
N is simply some integer. Now suppose further that we know that a
k
k
n
·
∑

.
`
,
0
converges.
This tells us all about the series b
k
k
n
·
∑

.
`
,
0
. Specifically, it tells us that this series also
converges. Let's see why that is. First note the obvious: b
k
k
n
·
∑

.
`
,
0
converges if and only if
b
k
k N
n
·
∑

.
`
,
converges. Next, observe that for all n , we have b a
k
k N
n
k
k N
n
· ·
∑ ∑
≤ , from which it
follows at once that lim
n
k
k N
n
b
→∞
·
∑
exists.
Example
10.8
What about the convergence of the series
1
3 4
3 2
1
n n n
k
n
+ + +

.
`
,
·
∑
? Observe first that
1
3 4
1
3 2 3
n n n n + + +
< . Then observe that the series
1
3
1
n
k
n
·
∑

.
`
,
converges because
lim lim
R R
R
x
dx
R
→∞ →∞
·
−
+

.
`
,
·
∫
1 1
3
1
3
1
3
3
1
2
. Thus
1
3 4
3 2
1
n n n
k
n
+ + +

.
`
,
·
∑
converges.
Suppose that, as before, a
k
k
n
·
∑

.
`
,
0
and b
k
k
n
·
∑

.
`
,
0
are positive series, and b a
k k
≤ for all
k > N , where N is some number. This time, suppose we know that b
k
k
n
·
∑

.
`
,
0
is divergent.
Then it should not be too hard for you to convince yourself that a
k
k
n
·
∑

.
`
,
0
must be
divergent, also.
Exercises
Which of the following series are convergent and which are divergent? Explain your
answers.
14.
1
2
0
e k
k
k
n
+

.
`
,
·
∑
15.
1
2 1
0
k
k
n
+

.
`
,
·
∑
16.
1
2
log k
k
n
·
∑

.
`
,
17.
1
1
2
0
k k
k
n
+ −

.
`
,
·
∑
10.9
10.5 Even More Series
We look at one more very nice way to help us determine if a positive series has a
limit. Consider a series a
k
k
n
·
∑

.
`
,
0
, and suppose a
k
> 0 for all k. Next suppose the
sequence
a
a
k
k
+

.
`
,
1
is convergent, and let
r
a
a
k
k
k
·
→∞
+
lim
1
.
The number r tells us almost everything about the convergence of the series a
k
k
n
·
∑

.
`
,
0
. Let's
see about it.
First, suppose that r < 1. Then the number ρ · +
−
r
r 1
2
is positive and less than 1.
For all sufficiently large k, we know that
a
a
k
k
+
≤
1
ρ . In other words, there is an N so that
a a
k k +
≤
1
ρ for all k N ≥ . Thus
a a a a a
k k k k N
k N
+ − −
+ −
≤ ≤ ≤ ≤ ≤
1 1
2
2
3 1
ρ ρ ρ ρ K .
Look now at the series
( )
a a
N
k N
k N
n
N
n N
ρ ρ ρ ρ
−
·
−
∑

.
`
,
· + + + ( ) 1
2
K .
This one converges because the Geometric series ρ
k
k
n
·
∑

.
`
,
0
converges (Recall that
0 1 < < ρ . ). It now follows from the previous section that our original series a
k
k
n
·
∑

.
`
,
0
has a
limit.
A similar argument should convince you that if r > 1, then the series a
k
k
n
·
∑

.
`
,
0
does
not have a limit.
The "method" of the previous section is usually called the Comparison Test, while
that of this section is usually called the Ratio Test.
10.10
Exercises
Which of the following series are convergent and which are divergent? Explain your
answers.
18.
10
0
k
k
n
k !
·
∑

.
`
,
19.
3
5
2 1
0
k
k
k
n +
·
∑

.
`
,
20.
3
10
2 1
0
k
k
k
n +
·
∑

.
`
,
21.
3
5 1
4
1
k
k
k
n
k k ( ) + +

.
`
,
·
∑
22.
3 1
5
4
1
k
k
k
n
k k ( ) + + 
.
`
,
·
∑
10.6 A Final Remark
The "tests" for convergence of series that we have seen so far all depended on the
series having positive terms. We need to say a word about the situations in which this is
not necessarily the case. First, if the terms of a series a
k
k
n
·
∑

.
`
,
0
alternate in sign, and if it is
true that     a a
k k +
≤
1
for all k , then lim
k
k
a
→∞
· 0 is sufficient to insure convergence of the
series. This is not too hard to see—meditate on it for a while.
The second result is a bit harder to see, and we'll just put out the result as the word,
asking that you accept it on faith. It says simply that if the series   a
k
k
n
·
∑

.
`
,
0
converges,
then so also does the series a
k
k
n
·
∑

.
`
,
0
. Thus, faced with an arbitrary series a
k
k
n
·
∑

.
`
,
0
, we
10.11
may unleash out arsenal of tests on the series   a
k
k
n
·
∑

.
`
,
0
. If we find this one to be
convergent, then the original series is also convergent. If, of course, this series turns out
not to be convergent, then we still do not know about the original series.
11.1
Chapter Eleven
Taylor Series
11.1 Power Series
Now that we are knowledgeable about series, we can return to the problem of
investigating the approximation of functions by Taylor polynomials of higher and higher
degree. We begin with the idea of a socalled power series. A power series is a series of
the form
c
k
(x − a)
k
k· 0
n
∑

.
`
,
.
A power series is thus a sequence of special polynomials: each term is obtained from
the previous one by adding a constant multiple of the next higher power of (x − a).
Clearly the question of convergence will depend on x , as will the limit where there is one.
The k
th
term of the series is c
k
(x − a)
k
so the Ratio Test calculation looks like
r(x) · lim
k →∞
c
k +1
(x − a)
k +1
c
k
(x − a)
k
· x − a lim
k →∞
c
k +1
c
k
.
Recall that our series converges for r(x) < 1 and diverges for r(x) > 1. Thus this
series converges absolutely for all values of x if the number lim
k →∞
c
k +1
c
k
· 0 . Otherwise, we
have absolute convergence for  x − a  < lim
k →∞
c
k
c
k +1
and divergence for
 x − a  > lim
k →∞
c
k
c
k +1
. The number R · lim
k →∞
c
k
c
k +1
is called the radius of convergence,
and the interval  x − a < R is called the interval of convergence. There are thus exactly
three possibilities for the convergence of our power series c
k
(x − a)
k
k· 0
n
∑

.
`
,
:
(i)The series converges for no value of x except x · a ; or
11.2
(ii)The series converges for all values of x ; or
(iii)There is a positive number R so that the series converges for  x − a < R and
diverges for  x − a  > R.
Note that the Ratio Test tells us nothing about the convergence or divergence of the
series at the two points where  x − a · R.
Example
Consider the series k!x
k
k ·0
n
∑

.
`
,
. Then R · lim
k →∞
c
k
c
k +1
· lim
k →∞
k!
(k +1)!
= lim
k →∞
1
k +1
· 0.
Thus this series converges only when x = 0.
Another Example
Now look at the series 3
k
(x −1)
k
k ·0
n
∑

.
`
,
. Here R · lim
k →∞
c
k
c
k +1
= lim
k →∞
3
k
3
k +1
· lim
k →∞
1
3
·
1
3
.
Thus, this one converges for  x − 1<
1
3
and diverges for  x − 1>
1
3
.
Exercises
Find the interval of convergence for each of the following power series:
1. (x + 5)
k
k ·0
n
∑

.
`
,
2.
1
k
(x −1)
k
k ·0
n
∑

.
`
,
11.3
3.
k
3k + 1
(x − 4)
k
k ·0
n
∑

.
`
,
4.
3
k
k!
(x +1)
k
k ·0
n
∑

.
`
,
5.
k!
7(k
2
+1)
(x − 9)
k
k ·0
n
∑

.
`
,
11.2 Limit of a Power Series
If the interval of convergence of the power series c
k
(x − a)
k
k· 0
n
∑

.
`
,
is  x − a < R, then,
of course, the limit of the series defines a function f :
f (x) · c
k
(x − a)
k
k ·0
∞
∑
, for  x − a < R.
It is known that this function has a derivative, and this derivative is the limit of the
derivative of the series. Moreover, the differentiated series has the same interval of
convergence as that of the series defining f . Thus for all x in the interval of convergence,
we have
f ' ( x) · kc
k
(x − a)
k −1
k ·1
∞
∑
.
We can now apply this result to the power series for the derivative and conclude that
f has all derivatives, and they are given by
f
(p)
(x) · k(k −1)K(k − p + 1)c
k
(x − a)
k− p
k · p
∞
∑
.
Example
11.4
We know that
1
1 − x
· x
k
k ·0
∞
∑
for  x < 1. It follows that
1
(1− x)
2
· kx
k −1
k ·1
∞
∑
· 1+ 2x + 3x
2
+ 4x
3
+K
for  x < 1.
It is, miraculously enough, also true that the limit of a power series can be integrated,
and the integral of the limit is the limit of the integral. Once again, the interval of
convergence of the integrated series remains the same as that of the original series:
f (t)dt ·
c
k
k +1
k ·0
∞
∑
a
x
∫
(x − a)
k +1
.
Example
We may simply integrate the Geometric series to get
log(1− x) · −
x
k +1
k + 1
k ·0
∞
∑
, for −1 < x < 1, or 0 < 1 − x < 2.
It is also valid to perform all the usual arithmetic operations on power series. Thus if
f (x) · c
k
k ·0
∞
∑
x
k
and g(x) · d
k
k ·0
∞
∑
x
k
for  x < r , then
f (x) t g(x) · (c
k
t d
k
k ·0
∞
∑
)x
k
, for  x < r .
Also,
f (x)g(x) · c
i
i·0
k
∑
d
k −i

.
`
,
c
k
k· 0
∞
∑
x
k
, for  x < r .
The essence of the story is that power series behave as if they were “infinite degree”
polynomials—the limits of power series are just about the nicest functions in the world.
11.5
Exercises
6. What is the limit of the series x
2k
k ·0
n
∑

.
`
,
? What is its interval of convergence?
7. What is the limit of the series 2(−1)
k
kx
2k −1
k ·1
n
∑

.
`
,
? What is its interval of convergence?
8. Find a power series that converges to tan
−1
x on some nontrivial interval.
9. Suppose f (x) · c
k
(x − a)
k
k ·0
∞
∑
. What is f
(p)
(a) ?
11.3 Taylor Series
Our major interest in finding a power series that converges to a given function. The
obvious candidate for such a series is simply the sequence of Taylor polynomials of
increasing degree. Thus if f is a given function, and a is a point in the interior of the
domain of f, the Taylor Series for f at a is the series
f
(k )
(a)
k!
(x − a)
k
k ·0
n
∑

.
`
,
.
The Taylor Series is thus an “infinite degree” Taylor Polynomial>
In general, the Taylor series for a function may not converge on any nontrivial interval
to f , but, mercifully, for many sufficiently nice functions it does. In such cases, we are
provided with the nice answer to the question proposed back in Chapter Nine: Can we
approximate the function f as well as we like by a Taylor Polynomial for sufficiently large
degree?
11.6
Example
The Taylor series for f (x) · sin x at x = a is simply (−1)
k
x
2k +1
(2k +1)!
k ·0
n
∑

.
`
,
. An easy
calculation shows us that the radius of convergence is infinite, or in other words, this
power series converges for all x . But is the limit sin x ? That’s easy to decide. From
Section 9.3, we know that
sin x − (−1)
k
x
2k+1
(2k +1)!
k ·0
n
∑
≤
 x 
2n+3
(2n + 3)!
,
and we know that
lim
n→∞
 x 
2n+3
(2n + 3)!
· 0,
no mater what x is. Thus we have
sin x · (−1)
k
x
2k+1
(2k +1)!
k ·0
∞
∑
, for all x .
Exercises
10. Find the Taylor Series at a = 0 for f (x) · e
x
. Find the interval of convergence and
show that the series converges to f on this interval.
11. Find the Taylor Series at a = 0 for f (x) · cos x. Find the interval of convergence and
show that the series converges to f on this interval.
12. Find the derivative of the cosine function by differentiating the Taylor Series you
found in Problem #11.
13. Find the Taylor Series at a = 1 for f (x) · logx . Find the interval of convergence and
show that the series converges to f on this interval.
11.7
14. Let the function f be defined by
f (x) ·
0, for x · 0
e
−1/ x
2
, for x ≠ 0
¹
'
¹
.
Find the Taylor Series at a = 0 for f. Find the interval of convergence and the limit of
the series.
12.1
Chapter Twelve
Integration
12.1 Introduction
We now turn our attention to the idea of an integral in dimensions higher than one.
Consider a realvalued function f : D R → , where the domain D is a nice closed subset of
Euclidean nspace R
n
. We shall begin by seeing what we mean by the integral of f over
the set D; then later we shall see just what such an abstract thing might be good for in real
life. Mrs. Turner taught us all about the case n · 1. As it was in extending the definition
of a derivative to higher dimensions, our definition of the integral in higher dimensions will
include the definition for dimension 1 we learned in grammar school—as always, there
will be nothing to unlearn. Let us again hark back to our youth and review what we know
about the integral of f : D R → in case D is a nice connected piece of the real line R.
First, in this context, the only nice closed pieces of R are the closed intervals; we thus
have D is a set [ , ] a b , where b a > . Recall that we defined a partition P of the interval to
be simply a finite subset { , , , } x x x
n 0 1
K of [ , ] a b with a x x x x b
n
· < < < < ·
0 1 2
K .
The mesh of a partition is max{ : , , } x x i n
i i
− ·
−1
12K . We then defined a Riemann sum
S P ( ) for this partition to be a sum
S P f x x
i i
i
n
( ) ( )
*
·
·
∑
∆
1
,
where ∆x x x
i i i
· −
−1
is simply the length of the subinterval [ , ] x x
i i −1
and x
i
*
is any point
in this subinterval. (Thus there is not just one Riemann sum for a partition P; the sum
obviously also depends on the choices of the points x
i
*
. This is not reflected in the
notation.)
Now, if there is a number L such that we can make all Riemann sums as close as we
like to L by choosing the mesh of the partition sufficiently small, then f is said to be
12.2
integrable over the interval, and the number L is called the integral of f over [a, b]. This
number L is almost always denoted f x dx
a
b
( )
∫
. More formally, we say that L is the
integral of f over [ , ] a b if for every ε > 0 , there is a δ so that  ( )  S P L − < ε for every
partition P having mesh < δ. You no doubt remember from your first encounter with this
integral that it initially seemed like an impossible thing to compute in any reasonable
situation, but then some version of the Fundamental Theorem of Calculus came to the
rescue.
12.2 Two Dimensions
Let us begin our study of higher dimensional integrals with the two dimensional
case. As we have seen so often in the past, in extending calculus ideas to higher
dimensions, most of the excitement occurs in taking the step from one dimension to two
dimensions—seldom is the step from 97 to 98 dimensions very interesting. We shall thus
begin by looking at the integral of f : D R → for the case in which D is a nice closed
subset of the plane. Complications appear at once. On the real line, nice closed sets are
simply closed intervals; in the plane, nice closed sets are considerably more interesting:
12.3
A moment's reflection convinces us that the domain D can, even in just two dimensions,
be considerably more complicated than it is in one dimension. First, capture D inside a
rectangle with sides parallel to the coordinate axes; and then divide this rectangle into
subrectangles by partitioning each of its sides:
Now, label the subrectangles that meet D, say with subscripts i n · 12 , , , K . The largest
area of all such rectangles is called the mesh of the subdivision. In each such rectangle,
choose a point ( , )
* *
x y
i i
in D. A Riemann sum S now looks like
S f x y A
i i i
i
n
·
·
∑
( , )
* *
∆
1
,
where ∆A
i
is the area of the rectangle from which ( , )
* *
x y
i i
is chosen. Now if there is a
number L such that we can get as close to L as we like by choosing the mesh of the
subdivision sufficiently small, then f is said to be integrable over D, and the number L is
the integral of f over D. The number L is usually written with two snake signs:
f x y dA ( , )
D
∫∫
.
Such integrals over two dimensional domains are frequently referred to as double
integrals.
12.4
I hope the definition of the integral in case D is a nice subset of R
3
is evident. We
capture D inside a box, and subdivide the box into boxes, etc. , etc. There will be more of
the higher dimensional stuff later.
Let's look a bit at some geometry. For the purpose of drawing a reasonable picture,
let us suppose that f x y ( , ) ≥ 0 everywhere on D.
Each term f x y A
i i i
( , )
* *
∆ is the volume of a box with base the rectangle A
i
and height
f x y
i i
( , )
* *
. The top of the box thus meets the surface z f x y · ( , ) . The Riemann sum is
thus the total volume of all such boxes. Convince yourself that as the size of the bases of
the boxes goes to 0, the boxes "fill up" the solid bounded below by the xy plane, above
by the surface z f x y · ( , ) , and on the sides by the cylinder determined by the region D.
The integral f x y dA ( , )
D
∫∫
is thus equal to the volume of this solid. If f x y ( , ) ≤ 0 , then,
of course, we get the negative of the volume bounded below by the surface z f x y · ( , ) ,
above by the xy plane, etc.
Suppose a and b are constants, and D E F · ∪ , where E and F are nice domains
whose interiors do not meet. The following important properties of the double integral
should be evident:
12.5
[ ( , ) ( , )] ( , ) ( , ) af x y bg x y dA a f x y dA b g x y dA + · +
∫∫ ∫∫ ∫∫
D D D
, and
f x y dA f x y dA f x y dA ( , ) ( , ) ( , )
D E F
∫∫ ∫∫ ∫∫
· + .
Now, how on Earth do we ever find an integral f x y dA ( , )
D
∫∫
? Let's see. Again, we
shall look at a picture, and again we shall draw our picture as if f x y ( , ) ≥ 0 . It should be
clear what happens if this is not the case.
We assume our domain D has a special form; specifically, we suppose it to be
bounded above by the curve y h x · ( ) , below by y g x · ( ) , on the left by x a · , and on
the right by x b · :
It is convenient for us to think of the integral f x y dA ( , )
D
∫∫
as the volume of the blob
bounded below by D in the xy plane and above by the surface z f x y · ( , ) . Think of
finding this volume by dividing the blob into slices parallel to the yaxis and adding up the
volumes of the slices. To approximate the volumes of these slices, we use slabs:
12.6
We partition the x interval [a, b ]: a x x x x b
n n
· < < < < ·
− 0 1 1
K . In each subinterval
[ , ] x x
i i −1
choose a point x
i
*
. Our approximating slab has as its base the rectangle of
"width" ∆x x x
i i i
· −
−1
and height h x g x
i i
( ) ( )
* *
− ; the roof is z f x y
i
· ( , )
*
. The volume
of the slab is the cross section area times the thickness, or [ ( , ) ]
*
( )
( )
*
*
f x y dy x
i
g x
h x
i
i
i
∫
∆ .
The sum of the volumes of the approximating slabs is thus
S f x y dy x
i
g x
h x
i
i
n
i
i
·
∫ ∑
·
[ ( , ) ]
*
( )
( )
*
*
∆
1
.
The double integral we seek is just the "limit" of these as we take thinner and thinner
slabs; or finer and finer partitions of the interval [a, b]. But Lo! The above sums are
12.7
Riemann sums for the ordinary one dimensional integral of the function
F x f x y dy
g x
h x
( ) ( , )
( )
( )
·
∫
, and so the double integral is given by
f x y dA F x dx
f x y dy dx
a
b
g x
h x
a
b
( , ) ( )
[ ( , ) ]
( )
( )
·
·
∫ ∫∫
∫ ∫
D
The double integral is thus equal to an integral of an integral, usually called an iterated
integral. It is traditional to omit the brackets and write the iterated integral simply as
f x y dydx
g x
h x
a
b
( , )
( )
( )
∫ ∫
.
Example
Let's find the double integral [ ] x y dA
2 2
+
∫∫
D
, where D is the area enclosed by the
lines y x · , x · 0 , and x y + · 2 . The first item of business here is to draw a picture of
D (We always need a picture of the domain of integration.):
12.8
It should be clear from the picture that in the language of our discussion, g x x ( ) · ,
h x x ( ) · − 2 , a · 0 , and b · 1. So slice parallel to the y axis:
The lower end of the slice is at y x · and the upper end is at y x · − 2 . The "volume" is
thus
[ ] ( )
( ) ( )
x y dy x y
y
x x
x
x
x
x
x
x
x
x
y x
y x
2 2
2
2
3
2
2
3
3
3
2
3
3
3
2
2
3 3
2
2
3
7
3
+ · + · − +
−
− − · +
−
−
−
·
· −
∫
,
and we have such a slice for all x from x · 0 to x · 1. Thus
[ ] [
( )
]
( )
x y dA x
x
x dx
x x x
2 2 2
3
3
0
1
3 4 4
0
1
2
2
3
7
3
2
3
2
12
7
12
16
12
4
3
+ · +
−
−
· −
−
−
· ·
∫ ∫∫
D
Exercises
1. Find x dA
2
D
∫∫
, where D is the domain bounded by the curves y x · − 4
2
and y x · 3 .
12.9
2. Find ( ) x y dA
2
−
∫∫
D
, where D is the area in the first quadrant enclosed by the
coordinate axes and the line 2 4 x y + · .
3. Use double integration to find the area of the region enclosed by the curves x y − · 2
and y x · −
2
.
4. Find the volume of the solid cut from the first octant by the surface z x y · − − 4
2
.
5. Sketch the domain of integration and evaluate the iterated integral:
y e dydx
xy
x
2
1
0
1
∫ ∫
.
6. Sketch the domain of integration and evaluate the iterated integral:
e dydx
x y
x
+
∫ ∫
0 1
8l og log
.
7. Find the volume of the wedge cut from the first octant by the cylinder z · 12 − 3y
2
and the plane x y + · 2 .
8. Suppose you have a double integral f x y dA ( , )
D
∫∫
in which the domain D is bounded
on the left by the curve x g y · ( ) , on the right by x h y · ( ) , below by y a · , and
above by y b · .
12.10
Give an iterated integral for the double integral in which the first integration is with
respect to x , and explain what's going on.
9. Give a double integral for the area of the region bounded by x y ·
2
and x y y · − 2
2
,
and evaluate the integral.
13.1
Chapter Thirteen
More Integration
13.1 Some Applications
Think now for a moment back to elementary school physics. Suppose we have a
system of point masses and forces acting on the masses. Specifically, suppose that for
each i n · 12 , , , K we have a point mass m
i
whose position in space at time t is given by
the vector r
i
.. Assume moreover that there is a force f
i
acting on this mass. Thus
according to Sir Isaac Newton, we have
f
r
i i
i
m
d
dt
·
2
2
for each i. Now sum these equations to get
F f
r
· ·
· ·
∑ ∑ i
i
n
i
i
i
n
m
d
dt
1
2
2
1
, or
F
r
·

.
`
,
·
·
∑
∑
M
d
dt
m
m
i i
i
n
i
i
n
2
1
1
,
where M m
i
i
n
·
·
∑
1
. Reflect for a moment on this equation. If we define R by
R
r
·
·
·
∑
∑
m
m
i i
i
n
i
i
n
1
1
, then the equation becomes F
R
· M
d
dt
2
2
. Thus the sum of the external
forces on the system of masses is the total mass times the acceleration of the mystical
point R. This point R is called the center of mass of the system.
In case the total mass is continuously distributed in space, the "sum" in the
equation for R becomes an integral. Let's look at what this means in two dimensions.
13.2
Suppose we have a plate and the mass density of the plate at (x,y) is given by ρ( , ) x y .
To find the center of mass of the plate, we approximate its location by chopping it into a
bunch of small pieces and treating each of these pieces as a point mass.
Now choose a point r i j
i i i
x y · +
* *
in each rectangle. The mass of this rectangle will be
approximately ρ( , )
* *
x y A
i i i
∆ , where ∆A
i
is the area of the rectangle. The equation for the
center of mass of this system of rectangles is then
~
( , )
( , )
( , )
( , ) ( , )
* *
* *
* *
* * * * * *
R
r r
i j
· ·
·
]
]
]
+
]
]
]
¹
'
¹
¹
'
¹
·
·
·
·
·
· ·
∑
∑
∑
∑
∑
∑ ∑
m
m
x y A
x y A
x y A
x y x A x y y A
i i
i
n
i
i
n
i i i i
i
n
i i i
i
n
i i i
i
n i i i i
i
n
i i i i
i
n
1
1
1
1
1
1 1
1
ρ
ρ
ρ
ρ ρ
∆
∆
∆
∆ ∆
The three sums in the previous line are Riemann sums for two dimensional integrals!
Thus as we take smaller and smaller rectangles, etc., we obtain for R, the location of the
center of mass
13.3
R i j ·
]
]
]
+
]
]
]
¹
'
¹
¹
¹
¹
'
¹
¹
¹
∫∫
∫∫ ∫∫
1
ρ
ρ ρ
( , )
( , ) ( , )
x y dA
x x y dA y x y dA
P
P P
In other words, the coordinates ( , ) x y of the center of mass of P are given by
x
x x y dA
M
y
y x y dA
M
P P
·
]
]
]
·
]
]
]
∫∫ ∫∫
ρ ρ ( , ) ( , )
, and ,
where M x y dA
P
·
∫∫
ρ( , ) is the total mass of the plate.
Example
Let's find the center of mass of a plate having the shape of the plane region
enclosed by the triangle
and having constant density (In this case, we say the mass is uniformly distributed over
the region. Suppose ρ( , ) x y k · . First,
x x y dA k xdydx k xb x a dx k
a b
b x a a
T
a
ρ( , ) ( / )
( / )
· · − ·
−
∫ ∫ ∫∫ ∫
0
1
0
2
0
1
6
, and then
y x y dA k ydydx
kb
x a dx k
ab
b x a a
T
a
ρ( , ) ( / )
( / )
· · − ·
−
∫ ∫ ∫∫ ∫
0
1
0
2
2
2
0
2
1
6
.
13.4
Also, M kdA k dA k
ab
T T
· · ·
∫∫ ∫∫
2
. Thus,
x
a
y
b
· ·
3 3
, and .
Meditate on the fact that the location of the center of mass does not depend on
the value of the constant k. Note that in general, if the density is constant, then the
constant slips out through the integral signs and cancels top and bottom in the recipe for
the coordinates ( , ) x y . This is what most of our intuitions tell us, I believe. It is,
nevertheless, comforting to see this fact come out in the mathematical wash. In this case
of constant density, the center of mass thus depends only on the geometry of the plate; it
is thus a geometric property of the region. It is called the centroid of the region. One
must never confuse the two concepts; intimately related though they be, they are
different. The center of mass is something a physical body has, while the centroid is an
abstract mathematical something.
Exercises
1. Find the center of mass of a plate of density ρ( , ) x y y · +1 having the shape of the
area bounded by the line y · 1 and the parabola y x ·
2
.
2. Find the center of mass of the smaller of the two regions cut from the elliptical region
x y
2 2
4 12 + · by the parabola x y · 4
2
if the density ρ( , ) x y x · 5 .
3. Find the centroid of the semicircular region {( , ) : , } x y x y a y ∈ + ≤ ≥ R
2 2 2 2
0 and .
13.5
4. Find the centroid of the region bounded by the horizontal axis and one arch of the sine
curve. (That is, the region between x · 0 and x · π bounded above by y x · sin and
below by y · 0.)
5. Find the centroid of the region bounded by the curves y x
2
2 · , x y + · 4 , and
y · 0 .
6. The area of a region A is dydx dydx
x
x
2
4
0
0
2
0 0
4
−
∫ ∫ ∫ ∫
+ . Draw a picture of the region.
7. Let f :D R → be a function defined on a nice subset D R
2
⊂ . The average value A
of f on D is defined to be A f x y dA ·
∫∫
1
area of D
D
( , ) .
a)Find the average depth of a bowl having the shape of the bottom half of the sphere
x y z
2 2 2
1 + + · .
b)Find the average depth of a bowl having the shape of the part of the
paraboloidz x y · + −
2 2
1 below the xy plane.
8. Let D be the region inside the circle x
2
+ (y − a)
2
· a
2
that lies below the line y · a .
a)Find the centroid of D.
b)Find the point on the semicircular boundar of D that is closest to the centroid.
13.2 Polar Coordinates
Now we shall see what happens when we express a double integral as an iterated
integral in some coordinate system other than the usual rectangular, or Cartesian,
13.6
coordinate system. We shall see more of this later; right now, let's look at what happens
in polar coordinates.
Suppose we have the integral f x y dA ( , )
D
∫∫
. In polar coordinates, we know that
we must substitute
x r
y r
·
·
cos
sin .
θ
θ
, and
There is, however, more to it than this. When we divided the plane into regions formed
by the curves x · constant and y · constant , we got rectangles, etc., etc. Now we
divide the plane into regions formed by the curves r · constant and θ · constant ,
where r and θ are the usual polar coordinates. This results in funny shaped regions:
Now, a typical region looks like
13.7
The area of this region is thus something like ∆ ∆ ∆ A r r ≈ θ , and our iterated integral looks
like
f x y dA f r r rdrd ( , ) ( cos , sin )
D
∫∫ ∫∫
· θ θ θ
together with the appropriate limits of integration. (We may, of course, integrate first
with respect to θ and then with respect to r if this is convenient.) We desperately need
to see an example.
Example
Let's find the centroid of the region enclosed by the curve whose equation in polar
coordinates is r · + 1 cosθ . Here is a picture drawn by Maple:
13.8
The centroid ( , ) x y is given by
x
xdA
dA
y
ydA
dA
· ·
∫∫
∫∫
∫∫
∫∫
D
D
D
D
, and .
First. let's find the integral xdA
D
∫∫
. Now, when we hold θ fixed and integrate first with
respect to r, the lower limit is independent of θ and is always r · 0 , while the upper
limit depends, of course on θ and is r · + 1 cosθ . We have a slice for each value of θ
from θ · 0 to θ π · 2 , and so our iterated integral looks like
xdA r rdrd r drd
D
∫∫ ∫ ∫ ∫ ∫
· ·
+ +
cos cos
cos cos
θ θ θ θ
θ π θ π
0
1
0
2
2
0
1
0
2
.
It is downhill all the way now:
r drd d
d
d d
d
2
0
1
0
2
3
0
2
2 3 4
0
2
0
2
2
0
2
2
0
2
1
3
1
1
3
3 3
1
3
0
3
2
1 2 0
1
4
1 2
6
1
12
2
6 12
15
12
5
4
cos ( cos ) cos
[cos cos cos cos ]
[ ( cos ) ( cos ) ]
cos
cos
θ θ θ θ θ
θ θ θ θ θ
θ θ θ θ
π
π
θ θ π
π π π
π
θ π π
π
π π
π
· +
· + + +
· + + + + +
· + + · + + · ·
+
∫ ∫ ∫
∫
∫ ∫
∫
Now for the other integrals.
It should be clear that ydA r drd
D
∫∫ ∫ ∫
· ·
+
2
0
1
0
2
0 sin
cos
θ θ
θ π
. Finally,
13.9
dA rdrd d
d
D
∫∫ ∫ ∫ ∫
∫
· · +
· + +
· + ·
+
θ θ θ
π θ θ
π
π
π
θ π π
π
0
1
0
2
2
0
2
0
2
1
2
1
1
4
1 2
2
3
2
cos
( cos )
( cos )
We are, at last, done.
x y · · ·
5
4
3
2
5
6
0
π
π
, . and
Exercises
9. Find the area of the region enclosed by the curve with polar equation r · sin2θ .
10. Evaluate the integral ( ) x y dA +
∫∫
D
, where D is the region in the first quadrant inside
the circle x y a
2 2 2
+ · and below the line y x · 3 .
11. Find the centroid of the region in the first quadrant inside the circle r a · and between
the rays θ · 0 and θ α · , where 0
2
≤ ≤ α
π
. What is the limiting position of the
centroid as α →0 ?
12. Evaluate e dA
x y
R
2 2
+
∫∫
, where R is the semicircular region bounded above by
y x · − 1
2
and below by the x axis.
13.10
13. Find the area enclosed by one leaf of the rose r · cos3θ .
14. Find the area of the region inside r · + 1 cosθ and outside r · 1.
13. 3 Three Dimensions
We move along to integrals in three dimensions. The idea is quite simple.
Suppose we have a function f : D R → , where D is a nice subset of R
3
. Capture D
inside a big box (i.e., a rectangular parallelepiped). Now subdivide this box by partitioning
each of its sides. The volume of the largest such box is called the mesh of the subdivision.
In each box that meets D, choose a point ( , , )
* * *
x y z
i i i
in D. A Riemann sum S now looks
like
S f x y z V
i i i
i
n
i
·
·
∑
( , , )
* * *
1
∆ ,
where ∆V
i
is the volume of the box from which ( , , )
* * *
x y z
i i i
was chosen. (The
summation is over all boxes that meet D.) If there is a number L such that   S L − can be
made arbitrarily small by choosing a subdivision of sufficiently small mesh, then we say
that f is integrable over D, and the number L is called the integral of f over D. This
integral is usually written with three snake signs:
f x y z dV ( , , )
D
∫∫∫
.
Let's see how to evaluate such a thing by considering iterated integrals. Here's
what we do. First, project D onto a coordinate plane. (We choose the xy plane as an
example.)
13.11
Let A be the region in the xy plane onto which D projects. Assume that a vertical line
through a point ( , ) x y ∈A enters D through the surface z g x y · ( , ) and exits through the
surface z h x y · ( , ) . In other words, the blob D is the solid above the region A between
the surfaces z g x y · ( , ) and z h x y · ( , ) . Now we simply integrate the integral
f x y z dz
g x y
h x y
( , , )
( , )
( , )
∫
over the region A:
f x y z dV f x y z dz dA
g x y
h x y
( , , ) ( , , )
( , )
( , )
·

.
`
,
∫ ∫∫ ∫∫∫
A D
.
Example
Let's find the integral ( ) x y z dV + +
∫∫∫
2
D
, where D is the tetrahedron with vertices
(0,0,0), (1,0,0), (0,2,0), and (0,0,1).
13.12
When we project D onto the xy plane, the bottom of D is the surface z · 0 and the top
of D is x
y
z + + ·
2
1, or z x
y
· − − 1
2
. The projection is simply the triangle
Our iterated integral is thus simply ( )
/
x y z dz dA
x y
+ +

.
`
,
− −
∫ ∫∫
2
0
1 2
A
. We now write the double
integral over A as an iterated integral, and we have
( ) ( )
/
x y z dV x y z dz dA
x y
+ + · + +

.
`
,
− −
∫ ∫∫ ∫∫∫
2 2
0
1 2
A D
· + +
− − −
∫ ∫ ∫
( )
/ ( )
x y z dzdydx
x y x
2
0
1 2
0
2 1
0
1
.
13.13
Again, it is traditional to omit the parentheses in the iterated integral. All we need do now
is integrate three times. Let's use Maple for the calculations, but look at the intermediate
steps, rather than just use one statement. Here we go.
For the first integration, we want ( )
/
x y z dz
x y
+ +
− −
∫
2
0
1 2
:
int(x+2*y+z,z=0..(1xy/2));
Thus,
( )
/
x y z dz x xy y y
x y
+ + · − − + − +
− −
∫
2
1
2
2
3
2
7
8
1
2
0
1 2
2 2
,
and our next integral is
( ) ( )
( ) / ( )
1 2
1
2
2
3
2
7
8
1
2
2 2
0
2 1
0
1 2
0
2 1
+ + · − − + − +
− − − −
∫ ∫ ∫
y z dzdy x xy y y dy
x x y x
.
Maple again:
int((x^2)/22*x*y+(3/2)*y(7/8)*y^2+1/2,y=0..2*(1x));
Thus,
( )
( )
− − + − + · − − + +
−
∫
1
2
2
3
2
7
8
1
2
4
2
3
3
5
3
2 2
0
2 1
3 2
x xy y y dy x x x
x
,
and finally,
int(4*x(2/3)*x^3+3*x^2+(5/3),x=0..1);
13.14
At last!
( )
/ ( )
x y z dzdydx
x y x
+ + ·
− − −
∫ ∫ ∫
2
1
2
0
1 2
0
2 1
0
1
.
We make a few obvious observations. First, if S is a solid, the volume V of the
solid is simply V dV ·
∫∫∫
S
. If the mass density of a blob having the shape of S is
ρ( , , ) x y z , then the mass M of the blob is M x y z dV ·
∫∫∫
ρ( , , )
S
, and the location
( , , ) x y z of the center of mass is given by
x
x x y z dV
M
·
∫∫∫
ρ( , , )
S
y
y x y z dV
M
·
∫∫∫
ρ( , , )
S
z
z x y z dV
M
·
∫∫∫
ρ( , , )
S
Exercises
15. Find the volume of the tetrahedron having vertices (0,0,0), (a,0,0),(0,b,0), and (0,0,c).
16. Find the centroid of the tetrahedron in the previous exercise.
17. Evaluate ( ) xy z dV
S
+
∫∫∫
2
, where S is the set S x y z z x y · ≤ ≤ − − {( , , ):    } 0 1 .
13.15
18. Find the volume of the region in the first octant bounded by the coordinate planes
and the surface z x y · − − 4
2
.
19. Write six different iterated integrals for the volume of the tetrahedron cut from the
first octant by the plane 12 4 3 12 x y z + + · .
20. A solid is bounded below by the surface z y · 4
2
, above by the surface z · 4 , and on
the ends by the surfaces x · 1 and x · −1. Find the centroid.
21. Find the volume of the region common to the interiors of the cylinders x y
2 2
1 + ·
and x z
2 2
1 + · .
14.1
Chapter Fourteen
One Dimension Again
14.1 Scalar Line Integrals
Now we again consider the idea of the integral in one dimension. When we were
introduced to the integral back in elementary school, we considered only functions defined
on nice subsets of the real line. The notion of an integral of a function f : D R → in
which D is a nice one dimensional set, but is not a subset of the reals is our next object of
study. To get some idea of why one might care about such a thing, consider the simple
problem of finding the mass of a piece of wire having the shape of an arc of a space curve
C and having a given density ρ( ) r . How might we approach such a problem? Simple
enough! We subdivide, or partition, the curve with a finite set of points, say
{ , , , } r r r
0 1
K
n
. On the subarc joining r
i −1
to r
i
, we choose a point, say r
i
*
, and evaluate
the function ρ( )
*
r
i
. Now we multiply this times the length of the line segment joining the
points r
i −1
and r
i
for an approximation to the mass of this arc of our curve. Then sum
these to obtain an approximation for the total mass:
S
i i i
i
n
· −
−
·
∑
ρ( ) 
*
r r r
1
1
.
Then we all believe that the "limit" of these sums as we choose finer and finer partitions
of the curve should be the actual, honesttogoodness mass of the wire.
Let's abstract the essence of the discussion. Suppose f :C R → is a function
whose domain C is a curve (in R R
2 3
or , or wherever). We subdivide the curve as in the
preceding discussion and choose a point r
i
*
on the subarc joining r
i −1
to r
i
. The sum
14.2
S f
i i i
i
n
· −
−
·
∑
( ) 
*
r r r
1
1
again is called a Riemann sum. If there is a number L such that all Riemann sums are
arbitrarily close to L for sufficiently fine partitions, then we say f is integrable on C, and
the number L is called the integral of f on C and is denoted f d ( ) r r
C
∫
. This integral is
also frequently referred to as a line integral.
This is wonderful, but how do find such an integral? It is remarkably simple and
easy. Suppose we have a vector description of the curve C; say r( ) t a t b , for ≤ ≤ . We
partition the curve by partitioning the interval [ , ] a b : If { , , , } a t t t b
n
· ·
0 1
K is a partition
of the interval, then the points { ( ), ( ), , ( )} r r r t t t
n 0 1
K partition the curve C. We obtain
the point r
i
*
on the subarc joining r( ) t
i −1
to r( ) t
i
by choosing t t t
i i i
*
[ , ] ∈
−1
and letting
r r
i i
t
* *
( ) · . Our Riemann sum now looks like
S f t t t
i i i
i
n
· −
−
·
∑
( ( ) ( ) ( )
*
r r r
1
1
.
14.3
Next, multiply the terms on the right by one, but one disguised as
∆
∆
t
t
i
i
, where, of course,
∆t t t
i i i
· −
−1
. Then we see
S f t
t t
t
t
i
i i
i i
n
i
·
−
−
·
∑
( ( )
( ) ( )
*
r
r r
1
1
∆
∆ .
We know that lim
( ) ( )
∆
∆
t
i i
i
t t
t
d
dt
→
−
−
·
0
1
r r r
, and so it is not hard to convince oneself that the
"limiting" value of the Riemann sums is
f t
d t
dt
dt
a
b
( ( ))
( )
r
r
∫
.
We have thus turned the problem into one we know how to solve—a plain old everyday
elementary calculus integral. Hence,
f d f t
d t
dt
dt
a
b
( ) ( ( ))
( )
r r r
r
C
∫ ∫
· .
Example
Suppose we have a wire in the shape of a quarter circle of radius 2, and the
density of the wire is given by ρ( , ) x y y · . What is the mass of the wire? Well, we
know the mass is simply the integral ydr
C
∫
, where C is the quarter circle:
14.4
A vector description of the curve is r i j ( ) cos sin t t t · + 2 2 , for 0
4
≤ ≤ t
π
. Thus we have
d
dt
t t
r
i j ·− + ·  sin cos  2 2 2 , and the integral becomes simply
yd t dt r
C
∫ ∫
· · 4 4
0
4
sin
/
π
.
Let's see what happens if we use a different vector description of the curve, say
r i j ( ) t t t · + − 4
2
for 0 2 ≤ ≤ t . We have
d
dt
t
t t
r
i j · −
−
·
− 4
2
4
2 2
. Hence
yd t
t
dt dt r
C
∫ ∫ ∫
· −
−

.
`
,
· · 4
2
4
2 4
2
2
0
2
0
2
.
We get, as we must, the same answer.
Exercises
1. Evaluate the integral ( ) x y z d − + +
∫
2 r
C
, where C is the curve r i j k ( ) ( ) t t t · + − + 1 ,
0 1 ≤ ≤ t .
14.5
2. Evaluate the integral x y d
2 2
+
∫
r
C
, where C is the curve r i j k ( ) cos sin t t t t · + + 4 4 3 ,
− ≤ ≤ 2 2 π π t .
3. Find the centroid of a semicircle of radius a.
4. Find the mass of a wire having the shape of the curve r j k ( ) ( ) t t t t · − + ≤ ≤
2
1 2 0 1 ,
if the density is ρ( ) t t ·
3
2
.
5. Find the center of mass of a wire having the shape of the curve
r i j k ( )
/
t t t
t
t · + + ≤ ≤
2 2
3 2
0 2
3 2
2
, ,
if the density is ρ( ) t
t
·
+
1
1
.
6. What is dr
C
∫
?
14.2 Vector Line Integrals
Now we are introduce something perhaps a little different from what we have seen
to now—integrals with vector valued integrands. Specifically, suppose C is a space curve
and f C R
3
: → is a function from C into the Euclidean space R
3
. We are going to define
an integral f r r
C
( ) ⋅
∫
d . Why should we care about such a thing? Again, let's think about a
physical model. You learned in fifth grade physics that the work done by a force F acting
through a distance d is simply the product Fd. The force F and the displacement d are, of
course, really vectors, and we saw earlier in life that the "product" of the two is actually
14.6
the scalar, or dot, product of the two vectors. Now, in general, neither of these quantities
will be constant, and we will have a variable force F(r) acting along a curve C in space.
How do we compute the work done in this situation? Let's see. Once more, we partition
the curve by choosing a sequence of points { , , , } r r r
0 1
K
n
on the curve, with r
0
being the
initial point and r
n
being the final point. Now, of course, there is an orientation, or
direction, specified on the curve. One may think of specifying an orientation by simply
putting an arrow on the curve—it thus makes sense to speak of the initial point and the
terminal point of the curve. Exactly as in the scalar integrand case, we choose a point r
i
*
on the subarc joining r
i −1
to r
i
, and evaluate F r ( )
*
i
. Now then, the work done in going
from r
i −1
to r
i
is approximately the scalar product F r r r ( ) ( )
*
i i i
⋅ −
−1
. Add all these up for
an approximation to the total work done:
S
i i i
i
n
· ⋅ −
−
·
∑
F r r r ( ) ( )
*
1
1
.
The course should be obvious now; we take finer and finer partitions, and the limiting
value of the sums is the integral
F r r
C
( ) ⋅
∫
d .
This integral too is called a line integral. To prevent confusion, we sometimes
speak of scalar line integrals and vector line integrals. How to find such a vector integral
should be clear from the discussion of scalar line integrals. We let r( ), t a t b ≤ ≤ , be a
vector description of C. (Here r( ) a is the initial point and r( ) b is the terminal point.)
The discussion proceeds almost exactly as it did in the previous section and we get
14.7
F r r F r
r
C
( ) ( ( )) ⋅ · ⋅
∫ ∫
d t
d
dt
dt
a
b
.
Example
Find [( ) ( ) ] xy z x z yz d + + + + ⋅
∫
2
2 i j k r
C
, where C is the straight line from the
origin to the point (1,2,3). The line C has a vector description r i j k ( ) t t t t · + + 2 3 . Thus,
d
dt
r
i j k · + + 2 3 , and so
[( ) ( ) ] [( ) ( ) ] ( )
( ) ( )
xy z x z yz d t t t t t dt
t t t dt t t dt
+ + + + ⋅ · + + + + ⋅ + +
· + + · +
∫ ∫
∫ ∫
2 2 2 2
0
1
2 2
0
1
2
0
1
2 2 9 3 12 2 3
2 8 36 38 8
i j k r i j k i j k
C
· + ·
38
3
4
50
3
3 2
0
1
t t .
Nothing to it.
Another Example
Now let's integrate the same function from (0,0,0) t0 (1,2,3), but this time along
the path P in the picture:
14.8
Here the path P is the union of the three nice curves, P P
1 2
, , and P
3
, so our integral is the
sum of three integrals:
F r F r F r F r ( , , ) ( , , ) ( , , ) ( , , ) x y x d x y x d x y x d x y x d
P P P P
⋅ · ⋅ + ⋅ + ⋅
∫ ∫ ∫ ∫
3 2 1
,
where
F i j k ( , , ) ( ) ( ) x y z xy z x z yz · + + + +
2
2 .
A vector description of P
1
is simply r i ( ) t t t · ≤ ≤ , 0 1. Thus
F r F i j i ( , , ) ( , , ) x y z d t dt t dt
P
⋅ · ⋅ · ⋅ ·
∫ ∫ ∫
00 0
0
1
0
1
1
.
For P
2
, we have r i j ( ) , t t · + ≤ ≤ 0 t 2 . This gives us
F r F j i j j ( , , ) ( , , ) ( ) x y z d t dt t dt dt
P
⋅ · ⋅ · + ⋅ · ·
∫ ∫ ∫ ∫
1 0 2
0
2
0
2
0
2
2
.
14.9
Finally, for P
3
, there is r i j k ( ) t t t · + + ≤ ≤ 2 0 3 , ; and so
F r F k i j k k ( , , ) ( , , ) [( ) ( ) ] x y z d t dt t t t dt
P
⋅ · ⋅ · + + + + ⋅
∫ ∫ ∫
3
12 2 1 4
2
0
3
0
3
· ·
∫
4 18
0
3
t dt .
At last, we have then F r ( , , ) x y z d
P
⋅ · + + ·
∫
0 2 18 20 .
Exercises
7. Evaluate [ ] xy x d
C
i j r + ⋅
∫
2
, where C is the arc of the curve y x ·
2
from (0,0) to (1,1).
8. Evaluate (cos ) x y d
C
i j r − ⋅
∫
where C the part of the curve y x · sin from (0,0) to
(π,0).
9. Evaluate the line integral of F i j k ( , , ) ( ) x y z xyz xy yz z · + + +
2
from (0,0,0) to
(1,1,2) along the line segment joining these two points.
10. Evaluate the line integral of F i j k ( , , ) ( ) ( ) ( ) x y z x z y z x y · − + − − + along the
polygonal path from (0,0,0) to (1,0,0) to (1,1,0) to (1,1,1).
11. Integrate F i j ( , ) ( ) x y
x y
y x ·
+
− +
1
2 2
one time around the circle x y a
2 2 2
+ · in the
counterclockwise direction.
14.10
14.3 Path Independence
Suppose we evaluate the vector line integral F r r ( ) ⋅
∫
d
C
, where C is a curve from
the point p to the point q. Let r( ), t a t b ≤ ≤ , be a vector description of C. Then, of
course, we have r p ( ) a · and r q ( ) b · . As we have already seen,
F r r F r
r
( ) ( ( )) ⋅ · ⋅
∫ ∫
d t
d
dt
dt
a
b
C
.
Now let us make the very special assumption that there exists a realvalued (or scalar)
function g: R R
3
→ such that the derivative, or gradient, of g is the integrand F :
∇ · g F .
Next let's use the Chain Rule to compute the derivative of the composition
h t g t ( ) ( ( )) · r :
h t g
d
dt
t
d
dt
'( ) ( ( )) · ∇ ⋅ · ⋅
r
F r
r
.
This is, mirabile dictu, precisely the integrand in our line integral:
F r r F r
r
p q ( ) ( ( )) ' ( ) ( ) ( ) ( ) ( ) ⋅ · ⋅ · · − · −
∫ ∫ ∫
d t
d
dt
dt h t dt h b h a g g
a
b
C a
b
.
This is a very exciting result and calls for some meditation. Note that the curve C
has completely disappeared from the answer. The value of the integral depends only on
the values of the function g at the endpoints; the path from p to q does not affect the
answer. The line integral is path independent. The result is esthetically pleasing and is
clearly the lineal descendant of the fundamental theorem of calculus we learned so many
years ago.
14.11
A moment's reflection on the examples we have seen should convince us that a lot
of integrals are not path independent, thus many very nice functions F (or vector fields )
are not the gradient of any function. A function F that is the gradient of a function g is
said to be conservative and the function g is said to be a potential function for F.
Let's suppose the domain D of the function F D R
3
: → is open and connected
(Thus any two points in D may be joined by a nice path.) We have just seen that if there
exists a function g: D R → such that F · ∇g , then the integral of F between any two
points of D does not depend on the path between the two points. It turns out, as we
shall see, that the converse of this is true. Specifically, if every integral of F in D is path
independent, then there is a function g such that F · ∇g . Let's see why this is so.
Choose a point p D · ∈ ( , , ) x y z
0 0 0
. Now define g g x y z ( ) ( , , ) s · to be the
integral from p to s along any curve joining these points. We are assuming path
independence of the integral, so it matters not what curve we choose. Okay, now we
compute the partial derivative
∂
∂
g
x
. The domain D is open and hence includes an open
ball centered at s D · ∈ ( , , ) x y z . Choose a point q · ( , , ) x y z
1
in such an open ball, and
let L be the straight line segment from s to q . Then, of course, L lies in D. Now let's
integrate F from p to s by going along any curve C from p to q and then along L from q to
s :
g g x y z d d ( ) ( , , ) ( ) ( ) s F r r F r r
C L
· · ⋅ + ⋅
∫ ∫
.
The first integral on the right does not depend on x, and so
∂
∂ x
d F r r
C
( ) ⋅ ·
∫
0 . Thus
∂
∂
∂
∂
g
x x
d · ⋅
∫
F r r
L
( ) .
14.12
We clearly need to find F r r
L
( ) ⋅
∫
d . This is easy. Suppose
F r r i r j r k ( ) ( ) ( ) ( ) · + + f f f
1 2 3
.
A vector description of L is simply r i j k ( ) , t t y z x t x · + + ≤ ≤
1
. Thus
d
dt
r
i · , and our
line integral becomes simply F r r
L
( ) ( , , ) ⋅ ·
∫ ∫
d f t y z dt
x
x
1
1
. We are almost done, for note
that now
∂
∂
∂
∂ x
d
x
f t y z dt f x y z
x
x
F r r
L
( ) ( , , ) ( , , ) ⋅ · ·
∫ ∫ 1 1
1
.
Hence
∂
∂
g
x
f ·
1
.
It should be clear to one and all how to show that
∂
∂
g
y
f ·
2
and
∂
∂
g
z
f ·
3
, thus
giving us the desired result: F · ∇g .
Exercises
12. Prove that
∂
∂
g
y
f ·
2
, where g and f
2
are as in the preceding discussion.
13. Prove that if F D R
3
: → , where D is open and connected, and every F r r
C
( ) ⋅
∫
d is
path independent, then F r r
P
( ) ⋅ ·
∫
d 0 for every closed path in D.( A closed path, or
14.13
curve, is one with no endpoints.) [Physicists and others like to use a snake sign with a
little circle superimposed on it
∫
to indicate that the path of integration is closed.]
14. Prove that if F D R
3
: → , where D is open and connected, and F r r
P
( ) ⋅ ·
∫
d 0 for
every closed path in D, then every F r r
C
( ) ⋅
∫
d is path independent .
15. a)Find a potential function g for the function F r i j k ( ) · + + yz xz xy .
b)Evaluate the line integral F r r
C
( ) ⋅
∫
d , where C is the curve
r i j k ( ) ( sin ) cos , t e t t e t t
t t
· + + ≤ ≤
2 3 3
0 1.
16. a)Find a potential function g for the function F r i j k ( ) ( ) · + +
+
e x x
y z 2
2 .
b)Find another potential function for F in part a).
b)Evaluate the line integral F r r
C
( ) ⋅
∫
d , where C is the curve
r i j k ( ) cos t t t t e t
t
· + + ≤ ≤ 2 4 0
2 2
, π .
17. Evaluate [( sin ) ( cos ) ] e y y e y x y d
x x
+ + + − ⋅
∫
3 2 2 i j r
E
where E is the ellipse
4 4
2 2
x y + · oriented clockwise.
[Really good hint: Find the gradient of g x y z e y xy y
x
( , , ) sin · + − 2
2
.]
15.1
Chapter Fifiteen
Surfaces Revisited
15.1 Vector Description of Surfaces
We look now at the very special case of functions r D R
3
: → , where D R
2
⊂ is a
nice subset of the plane. We suppose r is a nice function. As the point ( , ) s t ∈D moves
around in D, if we place the tail of the vector r( , ) s t at the origin, the nose of this vector
will trace out a surface in threespace. Look, for example at the function r D R
3
: → ,
where r i j k ( , ) ( ) s t s t s t · + + +
2 2
, and D R
2
· ∈ − ≤ ≤ {( , ) : , } s t s t 1 1 . It shouldn't be
difficult to convince yourself that if the tail of r( , ) s t is at the origin, then the nose will be
on the paraboloid z x y · +
2 2
, and for all ( , ) s t ∈D, we get the part of the paraboloid
above the square − ≤ ≤ 1 1 x y , . It is sometimes helpful to think of the function r as
providing a map from the region D to the surface.
The vector function r is called a vector description of the surface. This is, of course,
exactly the two dimensional analogue of the vector description of a curve.
15.2
For a curve, r is a function from a nice piece of the real line into three space; and for a
surface, r is a function from a nice piece of the plane into three space.
Let's look at another example. Here, let
r i j k ( , ) cos sin sin sin cos s t s t s t t · + + ,
for 0 ≤ ≤ t π and 0 2 ≤ ≤ s π . What have we here? First, notice that
 ( , ) (cos sin ) (sin sin ) (cos )
sin (cos sin ) cos
sin cos
r s t s t s t t
t s s t
t t
2 2 2 2
2 2 2 2
2 2
1
· + +
· + +
· + ·
Thus the nose of r is always on the sphere of radius one and centered at the origin.
Notice next, that the variable, or parameter, s is the longitude of r( , ) s t ; and the variable t
is the latitude of r( , ) s t . (More precisely, t is colatitude.) A moment's reflection on this
will convince you that as r is a description of the entire sphere. We have a map of the
sphere on the rectangle
15.3
Observe that the entire lower edge of the rectangle (the line from ( , ) 0 0 to ( , ) 2 0 π ) is
mapped by r onto the North Pole, while the upper edge is mapped onto the South Pole.
Let r D ( , ), ( , ) s t s t ∈ be a vector description of a surface S, and let p r · ( , ) s t be
a point on S. Now, c r ( ) ( , ) s s t · is a curve on the surface that passes through he point p.
Thus the vector
d
ds s
s t
c r
·
∂
∂
( , ) is tangent to this curve at the point p. We see in the same
way that the vector
∂
∂
r
t
s t ( , ) is tangent to the curve r( , ) s t at p.
15.4
At the point p r · ( , ) s t on the surface S, the vectors
∂
∂
r
s
and
∂
∂
r
t
are thus tangent to S.
Hence the vector
∂
∂
∂
∂
r r
s t
× is normal to S.
Example
Let's find a vector normal to the surface given by the vector description
r i j k ( , ) ( ) s t s t s t · + + +
2 2
at a point. We need to find the partial derivatives
∂
∂
r
s
and
∂
∂
r
s
:
∂
∂
r
i k
s
s · + 2 , and
∂
∂
r
j k
t
t · + 2 .
The normal N is
N
r r
i j k
i j k · × · · − − +
∂
∂
∂
∂
s t
s
t
s t 1 0 2
0 1 2
2 2 .
Meditate on the geometry here and convince yourself that this result is at least
reasonable.
Exercises
1. Give a vector description for the surface z x y · + 2
2
, x y , ≥ 0.
2. Give a vector description for the ellipsoid 4 8 16
2 2 2
x y z + + · .
3. Give a vector description for the cylinder x y
2 2
1 + · .
15.5
4. Describe the surface given by r i j k ( , ) cos sin s t s t s t s · + + , 0 2 ≤ ≤ t π , − ≤ ≤ 1 1 s .
5. Describe the surface given by r i j k ( , ) cos sin s t s t s t s · + +
2
, 0 2 ≤ ≤ t π , 1 2 ≤ ≤ s .
6. Give a vector description for the sphere having radius 3 and centered at the point
(1,2,3).
7. Find an equation (I.e., a vector description) of the line normal to the sphere
x y z a
2 2 2 2
+ + · at the point ( , , )
a a a
3 3 3
− .
8. Find a scalar equation (I.e., of the form f x y z ( , , ) · 0 ) of the plane tangent to the
sphere x y z a
2 2 2 2
+ + · at the point ( , , )
a a a
3 3 3
− .
9. Find all points on the surface r i j k ( , ) ( ) ( ) s t s t s t st · + + + −
2 2
3 at which the tangent
plane is parallel to the plane 5 6 2 7 x y z − + · , or show there are no such points.
10. Find an equation of the plane that contains the point (1,2,3) and is parallel to the
plane tangent to the surface r i j k ( , ) ( ) s t s t s t · + + −
2 2
2 at the point (1, 4,18).
15.2 Integration
Suppose we have a nice surface S and a function f S : → R defined on the surface.
We want to define an integral of f on S as the limit of some sort of Riemann sum in the
way in which we have already defined various integrals. Here we have a slight problem in
that we really are not sure at this point exactly what we might mean by the area of a
15.6
small piece of surface. We assume the surface is sufficiently smooth to allow us to
approximate the area of a small piece of it by a small planar region, and then add up these
approximations to get a Riemann sum, etc., etc. Let's be specific.
We subdivide S into a number of small pieces S S S
n 1 2
, , , K each having area ∆A
i
,
select points r
i i i i i
x y z S
* * * *
( , , ) · ∈ , and form the Riemann sum
R f A
i i
i
n
·
·
∑
( )
*
r ∆
1
.
Then, of course, we take finer and finer subdivisions, and if the corresponding Riemann
sums have a limit, this limit is the thing we call the integral of f on S: f dS
S
( ) r
∫∫
.
Now, how do find such a thing. We need a vector description of S , say
r D r D : ( ) → · S . The surface S is subdivided by subdividing the region D R
2
⊂ into
rectangles in the usual way:
15.7
The images of the vertical lines, s = constant, form a family of "parallel" curves on the
surface, and the images of the horizontal lines t = constant, also form a family of such
curves:
Let's look closely at one of the subdivisions:
15.8
We paste a parallelogram tangent to the surface at the point r( , ) s t
i i
as shown. The
lengths of the sides of this parallelogram are
∂
∂
r
s
s t s
i i i
( , )∆ and
∂
∂
r
t
s t t
i i i
( , )∆ . The area
is then
∂
∂
∂
∂
r r
s
s t s
t
s t t
i i i i i i
( , ) ( , ) ∆ ∆

.
`
,
×

.
`
,
, and we use the approximation
∆ ∆ ∆ A
s
s t
t
s t s t
i i i i i i i
≈

.
`
,
×

.
`
,
∂
∂
∂
∂
r r
( , ) ( , )
in the Riemann sums:
R f s t
s
s t
t
s t s t
i i
i
n
i i i i i i
·

.
`
,
×

.
`
,
·
∑
( ( , )) ( , ) ( , ) r
r r
1
∂
∂
∂
∂
∆ ∆ .
These are just the Riemann sums for the usual old time double integral of the function
F s t f s t
s
s t
t
s t
i i
i
n
i i i i
( , ) ( ( , )) ( , ) ( , ) ·

.
`
,
×

.
`
,
·
∑
r
r r
1
∂
∂
∂
∂
over the plane region D. Thus,
f dS f s t
s
s t
t
s t dA
S
( ) ( ( , )) ( , ) ( , ) r r
r r
D
· ×
∫∫ ∫∫
∂
∂
∂
∂
.
Example
Let's use our newfound knowledge to find the area of a sphere of radius a .
Observe that the area of a surface S is simply the integral dS
S
∫∫
. In the previous section,
we found a vector description of the sphere:
15.9
r i j k ( , ) cos sin sin sin cos s t a s t a s t a t · + + ,
0 ≤ ≤ t π and 0 2 ≤ ≤ s π . Compute the partial derivatives:
∂
∂
∂
∂
, and
r
i j
r
i j k
s
a s t a s t
t
a s t a s t a t
· − +
· + −
sin sin cos sin
cos cos sin cos sin
Then
∂
∂
∂
∂
cost sint
r r
i j k
i j k
s t
a s t s t
s t s
a s t s t t t
× · −
· − − −
2
2 2 2
0 sin sin cos sin
cos cos sin
[ cos sin sin sin sin cos ]
Next we need to find the length of this vector:
∂
∂
∂
∂
r r
s t
a s t s t t t
a t t t a t t t
a t
× · + +
· + · +
·
2 2 4 2 4 2 2 1 2
2 4 2 2 1 2 2 2 2 2 1 2
2
[cos sin sin sin sin cos ]
[sin sin cos ] [sin (sin cos )]
sin 
/
/ /
Hence,
Area = dS
s t
dA a t dA
S
· × ·
∫∫ ∫∫ ∫∫
∂
∂
∂
∂
r r
D D
2
sin 
15.10
·
· ·
∫ ∫
∫
a t dsdt
a tdt a
2
0
2
0
2
0
2
2 4
sin 
sin
π π
π
π π
Another Example
Let's find the centroid of a hemispherical shell H of radius a. Choose our
coordinate system so that the shell is the surface x y z a z
2 2 2 2
0 + + · ≥ , . The centroid
( , , ) x y z is given by
x
xdS
dS
H
H
·
∫∫
∫∫
y
ydS
dS
H
H
·
∫∫
∫∫
and z
zds
dS
H
H
·
∫∫
∫∫
.
First, note from the symmetry of the shell that x y · · 0 . Second, it should be clear
from the precious example that dS a
H
∫∫
· 2
2
π . This leaves us with just integral to
evaluate:
zdS
H
∫∫
. Most of the work was done in the example before this one. This hemisphere has
the same vector description as the sphere, except for the fact that the domain of r is the
rectangle 0 2 0
2
≤ ≤ ≤ ≤ s t π
π
, . Thus
15.11
zdS a a t
s t
dsdt
a t t dsdt a t t dt
a t a
H
∫∫ ∫ ∫
∫ ∫ ∫
· ×
· ·
· ·
2
0
2
0
2
3
0
2
0
2
3
0
2
3 2
0
2
3
2
cos
cos sin cos sin
sin
/
/ /
/
∂
∂
∂
∂
π
π π
π π
π π π
π
r r
And so we have z
a
a
a
· ·
π
π
3
2
2 2
. Is this the result you expected?
Yet One More Example
Our new definition of a surface integral certainly includes the old one for plane
surfaces. Look at the "surface" described by the vector function
r i j ( , ) cos sin θ θ θ r r r · + ,
with r defined on some subset D of the θ − r plane. For what we hope will be obvious
reasons, we are using the letters θ and r instead of s and t . Now consider an integral
f x y dS
S
( , )
∫∫
over the surface S described by r. We know this integral to be given by
f x y dS f r r
r
dA
D S
( , ) ( cos , sin ) · ×
∫∫ ∫∫
θ θ
∂
∂ θ
∂
∂
r r
.
Let's find the partial derivatives:
15.12
∂
∂ θ
θ θ
r
i j · − + r r sin cos , and
∂
∂
θ θ
r
i j
r
· + cos sin .
Thus,
∂
∂ θ
∂
∂
θ θ
θ θ
r r
i j k
k × · − · −
r
r r r sin cos
cos sin
0
0
,
and we have
∂
∂ θ
∂
∂
r r
× ·
r
r . Hence,
f x y dS f r r
r
dA f r r rdA
D S D
( , ) ( cos , sin ) ( cos , sin ) · × ·
∫∫ ∫∫ ∫∫
θ θ
∂
∂ θ
∂
∂
θ θ
r r
.
This should look familiar!
Exercises
11. Find the area of that part of the surface z x y · +
2 2
that lies between the planes z = 1
and z = 2.
12. Find the centroid of the surface given in Problem 11.
13. Find the area of that part of the Earth that lies North of latitude 45°. (Assume the
surface of the Earth is a sphere.)
14. A spherical shell of radius a is centered at the origin. Find the centroid of that part of
it which is in the first octant.
15.13
15. a)Find the centroid of the solid right circular cone having base radius a and altitude h.
b)Find the centroid of the lateral surface of the cone in part a).
16. Find the area of the ellipse cut from the plane z = 2x by the cylinder x y
2 2
1 + · .
17. Evaluate ( ) x y z dS
S
+ +
∫∫
, where S is the surface of the cube cut from the first octant
by the planes x = a, y = a , and z = a.
18. Evaluate x y dS
S
2
1 +
∫∫
, where S is the surface cut from the paraboloid
y z
2
4 16 + · by the planes x = 0, x = 1, and z = 0.
16. 1
Chapter Sixteen
Integrating Vector Functions
16.1 Introduction
Suppose water (or some other incompressible fluid ) flows at a constant velocity v
in space (through a pipe, for instance), and we wish to know the rate at which the water
flows across a rectangular surface S that is normal to the stream lines:
What is the rate at which the fluid flows through S? Let M t ( ) denote the total volume of
fluid that has passed through the surface at time t. The amount of fluid that flows through
during the time between t t t and +∆ is simply
M t t M t a t ( ) ( )   + − · ∆ ∆ v ,
where a is the area of S. Thus, the rate of flow through S is
dM
dt
a ·  v .
The result is slightly more complicated when various exciting changes are made.
Clearly there is nothing special about the surface's being a rectangle. But suppose that S
is placed at an angle to the stream lines instead of being placed normal to the them. Then
we have
dM
dt
a · ⋅ v n , where n is a unit normal to the surface S.
16. 2
Observe that matters which unit normal to the plane surface we choose. If we
choose the other normal ( n ), then our rate will be the negative of this one. We must
thus specify an orientation of the surface. We are computing the rate of flow from one
side of the surface to the other, and so we have to specify the "sides", so to speak.
16.2 Flux
Now, let's look at the general situation. The surface is not restricted to being a
plane surface, and the velocity of the flow is not restricted to being constant in space; it
may vary with position as well as time. Specifically, suppose S is a surface, together
with an orientation—that is, some means of specifying two "sides"—and suppose F r ( )
is a function F R R
3 3
: → , which is the velocity of the incompressible fluid. How do we
find the rate of flow through the surface S from one side to the other?
First, let's come to grips with the problem of specifying an orientation for S. We
say that an orientation for S is a continuous function n R
3
:S → such that n r ( ) is normal
to S and  ( ) n r · 1 for all r ∈S . A surface together with an orientation is called an
oriented surface. At first blush this looks simple enough, and the unsophisticated might
guess that every surface has an orientation (or may be oriented, as we sometimes say).
But this is not so! There are many surfaces for which an orientation does not exist. You
may recall from grammar school a simple example of such a surface, the socalled Möbius
band, or strip. Here is my feeble attempt to draw one:
16. 3
Now we see about finding the rate of flow through the oriented surface S. The
strategy should be oldhat by now. We subdivide S and look at "small" parallelograms
tangent to the surface:
As we have done so often, we suppose the subdivisions are small and approximate the
rate of flow, or flux, through the subdivision by the rate of flow through the tangent
parallelogram.
∆ ∆ S A
i i i
· ⋅ F r n ( )
*
,
and then add them to obtain yet another type of Riemann sum R A
i i
i
n
· ⋅
·
∑
F r n ( )
*
∆
1
. If
these sums have a limiting value as the size of the subdivisions go to zero, this is what we
call the integral of F over the oriented surface S:
F r S ( ) ⋅
∫∫
d
S
.
It should be clear now what we do to evaluate such an integral. As usual, we
consider a vector description of the surface S: r D R
3
: → ∈ S , where D R
2
⊂ . We
subdivide S by subdividing the region D into rectangles formed by lines s = constant and
t= constant, and looking at the curves r( , ) s t and r( , ) s t on the surface, exactly as we did
in integrating a scalar function over a surface S. Most conveniently now, the vector
16. 4
product
∂
∂
∂
∂
s
t
r r
× gives us not only a vector such that
∂
∂
∂
∂
s
t
r r
× ∆ ∆ s t is the area of the
approximating parallelogram, but also one which is normal to the surface. There is the
slight problem of the orientation of S. Thus
∂
∂
∂
∂
s
t
r r
× may not point in the direction of
the specified orientation, in which case, of course, we simply replace
∂
∂
∂
∂
s
t
r r
× by its
negative,
∂
∂
∂
∂
t
s
r r
× . (We may think of just reversing the roles of s and t.) We have in the
Riemann sums,
R s t
i
i
n
i i
· ⋅ ×

.
`
,
·
∑
F r
r r
( )
*
1
∂
∂
∂
∂
s
t
∆ ∆ ,
and, as before, we obtain
F r S F r
r r
D
( ) ( ( , )) ⋅ · ⋅ ×

.
`
,
∫∫ ∫∫
d s t dA
S
∂
∂
∂
∂
s
t
.
The concept we have developed here is purely mathematical and is done
independent of any physical interpretation, such as our fluid flow interpretation. What
we have is just an integral of a vector function F (or field) over an oriented surface S.
This is generally called the flux of F over S. There are many physical interpretations of
this concept; you have perhaps seen some of them in elementary school physics. There is
electric flux, the flux of an electric field; magnetic flux; gravitational flux, etc., etc.
Example
Let S be the sphere of radius a oriented so that the normal points "out" of the
sphere, and let F r
r
r ( )
 
·
c
3
, where c is a constant. Let's find F r S ( ) ⋅
∫∫
d
S
. Use the
vector description of S we used in the first Example of the previous section:
16. 5
r i j k ( , ) cos sin sin sin cos , s t a s t a s t a t · + +
0 2 0 ≤ ≤ ≤ ≤ s t π π , . We have already found that
∂
∂
∂
∂
r r
i j k
s t
a t s t s t t × · − − −
2
sin [ cos sin sin sin cos ] .
Modest meditation should convince you that this normal points into the sphere, and is
thus the negative of the one we need for the specified orientation of S.
Next, the integrand is given by
F r
r
r i j k ( )
 
[cos sin sin sin cos ] · · + +
c c
a
a s t s t t
3 3
,
and our integral becomes
F r S i j k i j k ( ) [cos sin sin sin cos ] sin [cos sin sin sin cos ] ⋅ · + + ⋅ + +
∫∫ ∫ ∫
d
c
a
s t s t t a t s t s t t dsdt
S
2
0
2
0
2
π π
· + +
∫ ∫
c t s t s t t dsdt sin [cos sin sin sin cos ]
2 2 2 2 2
0
2
0
π π
· c sint dsdt ·
0
2π
∫
0
π
∫
2πc sin t dt
0
π
∫
· 4π c .
Note that the radius a of the sphere has disappeared—the value of the integral is
independent of the radius of the sphere.
Exercises
16. 6
1. Find [ ] z x d
S
i k S + ⋅
∫∫
2
, where S is that part of the surface z x y · +
2 2
that lies above
the square {( , ): , } x y x y − ≤ ≤ − ≤ ≤ 1 1 1 1 and , oriented so that the normal points
upward.
2. Find the flux of F i j ( , , ) x y z x z · + out of the tetrahedron bounded by the coordinate
planes and the plane x y z + + · 2 3 6 .
3. Find the flux of F r
r
r ( )
 
·
c
3
out of the surface of the cube − ≤ ≤ a x y z a , , , where c
and a are positive constants.
4. Find the flux of the function F i j k ( , , ) x y z x y · + + 4 4 2 outward through the surface
cut from the bottom of the paraboloid z x y · +
2 2
by the plane z · 1.
5. Find the flux of the function F i j k ( , , ) x y z z x z · + −
2
3 upward throught the surface
cut from the cylinder z y · − 4
2
by the planes x x z · · · 0 1 0 , , and .
6. Let S be the surface defined by
y x · log , 1 ≤ ≤ x e , 0 1 ≤ ≤ z ,
and let n be the orientation of S such that n r j ( ) ⋅ > 0 for all r ∈S . Find the flux
[ ] 2 y z d
S
j k S + ⋅
∫∫
.
16. 7
17.1
Chapter Seventeen
Gauss and Green
17.1 Gauss's Theorem
Let B be the box, or rectangular parallelepiped, given by
B · ≤ ≤ ≤ ≤ ≤ ≤ {( , , ): , , } x y z x x x y y y z z z
0 1 0 1 0 1
;
and let S be the surface of B with the orientation that points out of B. Let F B R
3
: → be
a nice function, or field. For reasons that will become apparent as the drama unfolds, let's
compute the flux
F r S ( ) ⋅
∫∫
d
S
.
We shall do this by computing the surface integral over each of the six sides of B
and adding the results. Let S
1
be the side in the plane x x ·
1
; let S
2
be the side in the
plane x x ·
0
; let S
3
be the side in the plane y y ·
1
; let S
4
be the side in the plane
y y ·
0
; and let S
5
and S
6
be the obvious things. We begin by computing the integral
F r S ( ) ⋅
∫∫
d
S
1
.
A vector description of S
1
is quite easy to come by; it is, of course, simply
r i j k ( , ) s t x s t · + +
1
,
where y s y
0 1
≤ ≤ and z t z
0 1
≤ ≤ . (Obviously, s is simply y, and t is z .) Then
∂
∂
∂
∂
s
r r
j k i × · × ·
t
.
It is clear this is the specified orientation. If F r i j k ( ) ( , , ) ( , , ) ( , , ) · + + p x y z q x y z r x y z ,
then
F r S F i ( ) ( , , )
( , , )
⋅ · ⋅
·
∫ ∫ ∫∫
∫ ∫
d x s t dtds
p x s t dtds
z
z
y
y
S
z
z
y
y
1
1
0
1
0
1
1
0
1
0
1
17.2
A vector description for the opposite side, x x ·
0
, is just
r i j k ( , ) s t x s t · + +
0
,
and we have
F r S F i ( ) ( , , ) ( )
( , , )
⋅ · ⋅ −
· −
∫ ∫ ∫∫
∫ ∫
d x s t dtds
p x s t dtds
z
z
y
y
S
z
z
y
y
0
0
0
1
0
1
2
0
1
0
1
The sum of these two is then
F r S F r S ( ) ( ) ⋅ + ⋅ ·
∫∫ ∫ ∫ ∫∫
d d dtds
S z
z
y
y
S
1 0
1
0
1
2
[p(x , s, t)  p(x , s, t)]
1 0
.
Observe that
p x s t p x s t
p
x
s t d
x
x
( , , ) ( , , ) ( , , )
1 0
0
1
− ·
∫
∂
∂
ξ ξ
.
Substitution of this into the previous equation gives us
F r S F r S ( ) ( ) ( , , ) ⋅ + ⋅ ·
·
∫∫ ∫ ∫ ∫ ∫∫
∫∫∫
d d
p
x
s t d dtds
p
x
dV
S x
x
z
z
y
y
S
B
10 0
1
0
1
0
1
21
∂
∂
ξ ξ
∂
∂
and we have turned the sum of the two surface integrals into a plain ol' volume integral .
It should be clear how we also obtain
F r S F r S ( ) ( ) ⋅ + ⋅ ·
∫∫ ∫∫ ∫∫∫
d d
q
y
dV
S S B
3 4
∂
∂
, and
F r S F r S ( ) ( ) ⋅ + ⋅ ·
∫∫ ∫∫ ∫∫∫
d d
r
z
dV
S S B
5 6
∂
∂
.
17.3
The flux over the entire surface S is thus the sum of these:
(z)
F r S ( ) ⋅ · + +
· + +
]
]
]
∫∫∫ ∫∫ ∫∫∫ ∫∫∫
∫∫∫
d
p
dV
q
dV
r
dV
p q r
dV
B S B B
B
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
x
y
z
x
y
z
We have now found the surface integral, or flux, in terms of an ordinary volume integral.
Now, suppose we have an "arbitrary" solid region B bounded by a surface S,
together with a function F r i j k ( ) ( , , ) ( , , ) ( , , ) · + + p x y z q x y z r x y z defined on B. Trap
B in a box and subdivide the box into parallelepipeds. Consider those parallelepipeds
{ : , , , } B i n
i
· 1 2 K that meet B. The surface that bounds B
i
will be called S
i
, and oriented
so that the normal points out. The union P
n i
B · ∪{ }of all the B
i
is thus an
approximation to the original solid B.
Apply the equation (z) to each of these and sum the equations:
F r S ( ) ⋅ · + +
]
]
] ∫∫
∑
∫∫∫
∑
d
p q r
dV
S
i
B
i
i i
∂
∂
∂
∂
∂
∂
x
y
z
.
The sum on the right hand side is just the integral over P
n
:
F r S ( ) ⋅ · + +
]
]
] ∫∫
∑
∫∫∫
d
p q r
dV
S
i
P
i ni
∂
∂
∂
∂
∂
∂
x
y
z
.
Take a closer look at the sum of the surface integrals on the left hand side of this
equation. Suppose parallelepipeds B
j
and B
k
are adjacent, and call the common side T :
17.4
In the sum of surface integrals, the integral over the common side T appears twice, once
from the integral over S
j
, the surface of B
j
and once from the integral over S
k
, the
surface of B
k
. These integrals, will, however, have opposite signs because the
orientation of T has one direction as a part of the surface of B
j
and the opposite direction
as a part of the surface of B
k
. These two terms thus sum to zero and cancel each other.
In the sum of all the surface integrals, we are therefore left with only the integrals over
sides that are not adjacent to another box. A moments reflection, and you see that what is
left is precisely the integral over the boundary S
n
of P
n
with the outward pointing
orientation. Mirabile dictu, this is precisely the equation (z):
F r S ( ) ⋅ · + +
]
]
] ∫∫∫ ∫∫
d
p q
y
r
z
dV
P S
n n
∂
∂
∂
∂
∂
∂
x
.
Now, as everyone can see coming, we look at the limit of this equation as we take smaller
and smaller subdivisions. Then P B
n
→ and S S
n
→ , giving us precisely the same result
for the arbitrary region B:
F r S ( ) ⋅ · + +
]
]
] ∫∫∫ ∫∫
d
p q
y
r
z
dV
B S
∂
∂
∂
∂
∂
∂
x
.
17.5
This is really a big deal—such a big deal that it has its own name. This is called Gauss's
Theorem, or the Divergence Theorem.
The integrand in the volume integral also has a name; it is called the divergence of
the function F. It is usually designated either div F , or ∇⋅ F . Thus,
div
p
x
q
y
r
z
F F · ∇⋅ · + +
∂
∂
∂
∂
∂
∂
.
With this new definition, Gauss’s Theorem looks like
dV d
S
∫∫ ∫∫∫
⋅ ∇ · ⋅ ) ( ) ( r F S r F
Example
Let's find the divergence of F r
r
r ( )
 
·
c
3
. First we need to see F in the form
F i j k ( , , ) ( , , ) ( , , ) ( , , ) x y z p x y z q x y z r x y z · + + .
That's easy:
F i j k ·
+ +
+ +
c
x y z
x y z
( )
[ ]
/ 2 2 2 3 2
,
and so
p
cx
x y z
·
+ + ( )
/ 2 2 2 3 2
,
q
cy
x y z
·
+ + ( )
/ 2 2 2 3 2
,
r
cz
x y z
·
+ + ( )
/ 2 2 2 3 2
.
A bit of elementary school calculus (remember Mrs. Turner!), and we have
17.6
∂
∂
p
x
c
x y z x
x y z
·
+ + −
+ +
2 2 2 2
2 2 2 5 2
3
( )
/
,
∂
∂
q
y
c
x y z y
x y z
·
+ + −
+ +
2 2 2 2
2 2 2 5 2
3
( )
/
,
∂
∂
r
z
c
x y z z
x y z
·
+ + −
+ +
2 2 2 2
2 2 2 5 2
3
( )
/
.
Hence, ∇⋅ · F 0 everywhere (except, of course, for r = 0, where F is not defined.).
Gauss's Theorem now tells us that the integral of F over any closed surface that
does not enclose r = 0 must be zero. This might be the hohum of the week save for the
fact that the function F is a common one. It is the gravitational field of a point mass fixed
at the origin, or the electric intensity field for a point charge fixed at the origin, or any field
in which the magnitude is inversely proportional to the distance from the origin and which
points in the direction of the origin.
Exercises
1. Find the outward flux of the function F i j k · − + − + − ( ) ( ) ( ) y x z y y x across the
boundary of the cube bounded by the planes x · t4 , y · t4, and z · t4 .
2. Find [ ] y xy z d
S
i j k S + − ⋅
∫∫
, where S is the boundary of the solid inside the cylinder
x y
2 2
1 + ≤ between z = 0 and z x y · +
2 2
, with the outward pointing orientation.
3. Find [log( ) tan ] x y
z
x
y
x
z x y d
S
2 2 1 2 2
2
+ +

.
`
,
+ + ⋅
−
∫∫
i j k S , where S is the boundary of
the solid {( , , ): , } x y z x y z 1 2 1 2
2 2
≤ + ≤ − ≤ ≤ .
17.7
4. Let B a region in R
3
, and let f : B R → be a function such that
∂
∂
∂
∂
∂
∂
2
2
2
2
2
2
0
f
x
f
y
f
z
+ + · in B (Such a function f is said to be harmonic in B.). Let S
be the boundary of B. Show that ∇ ⋅ ·
∫∫
f d
S
S 0.
17.2 Green's Theorem
Let R be the rectangular region in the plane bounded by the rectangle with vertices
( , ), ( , ), ( , ), x y x y x y
0 0 1 0 1 1
and ( , ) x y
0 1
.
( , ) x y
0 1
) , (
1 1
y x
( , ) x y
0 0
( , ) x y
1 0
Suppose F R
2
: R → is a vector function given by F i j ( , ) ( , ) ( , ) x y p x y q x y · + . Now,
let's compute the vector line integral of F around the rectangular boundary C in the
counterclockwise direction. We shall compute the integral in four parts: the integrals
along each of the straight line segments making up the boundary.
C
3
←
C
4
↓ ↑C
2
→
C
1
Thus,
17.8
F r F r F r F r F r
C
2
⋅ · ⋅ + ⋅ + ⋅ + ⋅
∫ ∫ ∫ ∫ ∫
d d d d d
C C C C
1 3 4
.
We shall work out the evaluation of one of these in some painful detail; it should then be
rather obvious how to do the others. Start with a vector description of C
1
:
r i j ( ) t t y · +
0
, x t x
0 1
≤ ≤ .
Then, of course,
d
dt
r
i · , and our line integral becomes
F r i j i ⋅ · + ⋅ ·
∫ ∫ ∫
d p t y q t y dt p t y dt
x
x
C x
x
[ ( , ) ( , ) ] ( , )
0 0 0
0
1
1 0
1
.
In a similar fashion, we get
F r ⋅ · −
∫ ∫
d p t y dt
C x
x
3 0
1
1
( , ) .
Thus,
F r F r ⋅ + ⋅ · − −
· −
· −
∫ ∫ ∫
∫ ∫
∫∫
d d p t y p t y dt
p
y
t s dsdt
p
y
dA
C C x
x
y
y
x
x
R
3 1 0
1
0
1
0
1
1 0
[ ( , ) ( , )]
( , )
∂
∂
∂
∂
In essentially the same manner, we find that
F r F r ⋅ + ⋅ ·
∫ ∫ ∫∫
d d
q
x
dA
C C R
2 4
∂
∂
.
17.9
Thus
F r F r F r F r F r
C
2
⋅ · ⋅ + ⋅ + ⋅ + ⋅
· −
]
]
]
∫ ∫ ∫ ∫ ∫
∫∫
d d d d d
q
x
p
y
dA
C C C C
R
1 3 4
∂
∂
∂
∂
We have turned a one dimensional vector integral into a double integral, similar to
the way in which in the previous section we turned a two dimensional vector integral into
a triple integral.
Now suppose we have a reasonable region R bounded by a reasonable curve C
with a counterclockwise orientation:
Now cover this region with rectangles, and apply the above recipe to each rectangle, and
add all the equations, etc., etc., just as we did with the parallelepipeds in deriving Gauss's
Theorem. When the dust settles, we have the same result:
F r ⋅ · −
]
]
] ∫ ∫∫
d
q
x
p
y
dA
C R
∂
∂
∂
∂
.
This is called Green's Theorem. You should note that the same equation is valid even if
the region R is bounded by more than one closed curve.
17.10
Here the boundary C consists of three curves with the orientation indicated by the arrows
in the fine picture—meditate on the covering by approximating rectangles and you will see
why the orientation of the "inside" curves is clockwise. The line integral on the left side is
simply the sum of the integrals over the pieces of the boundary curve.
Example
Let's evaluate the line integral [ ( ) ] 5 3 1 y x d
C
i j r + + ⋅
∫
, where C is the circle of
radius 2 centered at the origin, oriented counterclockwise. First, note that
∂
∂
∂
∂
x
, and
q p
y
· · 3 5 .
Thus,
[ ( ) ] 5 3 1
2 8
y x d
q
x
p
y
dA
dA
C R
R
i j r + + ⋅ · −
]
]
]
· − · −
∫ ∫∫
∫∫
∂
∂
∂
∂
π
Exercises
5. Evaluate [(sin ) ( ) ] x y x e d
y
C
+ + − ⋅
−
∫
3 2
2
2
i j r , where C is the boundary of the halfdisc
x y y
2 2
9 0 + ≤ ≥ , oriented counterclockwise.
17.11
6. Evaluate (tan ) log( )
−
+ +
]
]
]
⋅
∫
1 2 2
y
x
x y d
C
i j r , where C is the boundary of the region
1 2 0 ≤ ≤ ≤ ≤ r , θ π , oriented clockwise. (These are the usual polar coordinates.)
7. Evaluate the line integral [ ] ye x e d
x y
C
2
3
i j r + ⋅
∫
, where C is the curve given by
r i j ( ) sin sin , t t t t · + ≤ ≤ 2 0 2π by using Green's Theorem.
17.3 A Pleasing Application
Here we shall use Green’s Theorem to find the area of a region R bound by a
polygon P with vertices ). , ( , ), , ( ), , (
2 2 1 1 n n
y x y x y x K How do we do this? We simply
apply Greens’ Theorem to the function
j j i F x y x q y x p y x · + · ) , ( ) , ( ) , ( .
Then Green’s Theorem tells us that
∫ ∫∫
⋅ ·
,
`
.

∂
∂
−
∂
∂
P R
d y x dA
x
p
x
q
r F ) , ( ,
which becomes
r j d x dA
R P
∫∫ ∫
⋅ · .
We thus find the area by evaluating the line integral on the right side. This is easy. We
simply integrate over each line segment of the polygon and add up the integrals.
Let’s integrate along the line segment L
k
from ) , (
k k
y x to ) , (
1 1 + + k k
y x . A vector
description of this segment is
) ( ) )( 1 ( ) (
1 1
j i j i r
+ +
+ + + − ·
k k k k
y x t y x t t , 1 0 ≤ ≤ t .
Thus j i r ) ( ) ( ) ( '
1 1 k k k x
y y x x t − + − ·
+ +
, and we have
17.12
2
) )( (
] ) 1 [( ) (
) ( ' ] ) 1 [(
1 1
1
0
1 1
1
0
1
k k k k
k k k k
L
k k
x x y y
t d tx x t y y
dt t tx x t d x
k
+ −
·
+ − − ·
⋅ + − · ⋅
+ +
+ +
+
∫
∫ ∫
r j r j
Thus, Area =
2
) )( (
2
) )( (
1 1
1
1
1 1 n n
n
k
k k k k
R
x x y y x x y y
dA
+ −
+
+ −
·
∑
∫∫
−
·
+ +
.
Meditate on this result. It is really a very simple formula for the area enclosed by a
polygon.
Example. We shall find the area of the quadrilateral with vertices (0, 0), (2, 4), (1, 7),
and (1, 9):
0
2
4
6
8
1 0.5 0.5 1 1.5 2
x
Area = [ ] 4 )] 0 1 )( 0 9 ( ) 1 1 )( 7 9 ( ) 1 2 )( 4 7 ( ) 0 2 )( 0 4 (
2
1
· + − − + + − − + + − + + −
Exercises
8. Find the area enclosed by the octagon with vertices (0, 0), (1, 0), (2, 3), (0, 5), (2, 2),
(1, 1), (2, 2), (1, 3).
9. By means of a clever choice of the function ) , ( y x F , use Green’s Theorem and derive a
recipe for the integral
∫∫
R
xdA, where R is the region enclosed by the polygon with
vertices ). , ( , ), , ( ), , (
2 2 1 1 n n
y x y x y x K
17.13
10. By means of a clever choice of the function ) , ( y x F , use Green’s Theorem and derive
a recipe for the integral
∫∫
R
ydA, where R is the region enclosed by the polygon with
vertices ). , ( , ), , ( ), , (
2 2 1 1 n n
y x y x y x K
11. Find the centroid of the region enclosed by the triangle with vertices (1, 1), (2, 8), and
(5, 5).
18.1
Chapter Eighteen
Stokes
18.1 Stokes's Theorem
Let F R
3
: D → be a nice vector function. If
F i j k ( , , ) ( , , ) ( , , ) ( , , ) x y z p x y z q x y z r x y z · + + ,
the curl of F is defined by
curl
r
y
q
z
p
z
r
x
q
x
p
y
F i j k · −

.
`
,
+ −

.
`
,
+ −

.
`
,
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
∂
.
Here also the socalled del operator ∇ · + + i j k
∂
∂
∂
∂
∂
∂ x y z
provides a nice
memory device:
curl
x y z
p q r
F F
i j k
· ∇ × ·
∂
∂
∂
∂
∂
∂
.
This definition allows us to look at Green's Theorem from a new perspective by
observing that in case F i j ( , ) ( , ) ( , ) x y p x y q x y · + , Green's Theorem becomes
(♥) F r F S ⋅ · ⋅
∫∫ ∫
d curl d
R C
,
where we are thinking of the region R as an oriented surface with its orientation pointing
in the direction of k.
We want to look at this formula in case the region R is not necessarily in the ij
plane, in which case, the word "clockwise" doesn't help in deciding on the orientation of
the boundary C. Once again, we orient things according to our familiar "righthand" rule.
18.2
Here's the way it goes. Suppose now S is any surface bounded by a finite number of
disjoint curves C C C
n 1 2
, ,K . We say simply that C C C C
n
· ∪ ∪ ∪
1 2
K is the boundary of
S. Now choose an orientation for the surface S. Look at one of these normal vectors
"close" to a curve C
j
and imagine a little circle around the base of the normal oriented so
that the normal vector points in the righthand direction with respect to the direction of
the circle. Then the orientation, or direction, of C
j
that is consistent with the given
orientation of the surface S is the one that "lines up" with the direction on this little circle.
Look at this picture:
The surface and its boundary in this case are said the be consistently oriented.
Now we do what we have done so many times in the past. Look at a surface S in
three space bounded by C. (Here neither S nor C are assumed to lie in a plane.)
Approximate the surface by a bunch of plane regions tangent to S , apply the equation (♥)
to each of these approximating plane regions, and then sum these equations. The sum of
the surface integrals is just the surface integral over the union of the approximating pieces,
and the sum of the line integrals is just the line integral around the boundary of the union
of the pieces—as in the plane case, the line integrals over the boundaries of adjacent
regions cancel. Then, of course, we think of looking at the limit as we take more and more
approximating regions, etc., and we obtain the equation
F r F S ⋅ · ⋅
∫ ∫∫
d curl d
C S
,
18.3
where S and C are oriented consistently. This result is the celebrated Stokes's Theorem.
Example
Let's use Stokes's Theorem to evaluate the line integral
[ ] − + − ⋅
∫
y x z d
C
3 3 3
i j k r ,
where C is the intersection of the cylinder x y
2 2
1 + · and the plane x y z + + · 1
oriented in the clockwise direction when viewed from above (i.e., looking in the direction
of k.). The curve C bounds the part of the plane x y z + + · 1 that lies above the set of
( , ) x y such that x y
2 2
1 + ≤ . A vector description is thus given by
r i j j ( , ) cos sin ( cos sin ) s t s t s t s t s t · + + − − 1 , 0 1 0 2 ≤ ≤ ≤ ≤ s t , π .
Hence,
∂
∂
∂
∂
t
r r
i j k
i j k
s
t t t t
s t s t s t t
s s s
× · − +
− −
· + +
cos sin (cos sin )
sin cos (sin cos )
I hope this result is no surprise. Notice that this is the opposite of the orientation
consistent with that specified for the curve C, and so we must use
∂
∂
∂
∂
s
r r
i j k
t
s × · − + + ( )
in our surface integral. The surface integral curl d
S
F S ⋅
∫∫
looks like
18.4
curl d curl
t
dsdt
S
F S F
r r
⋅ · ⋅ ×

.
`
,
∫ ∫ ∫∫
∂
∂
∂
∂
π
s
0
1
0
2
.
We must find curl F:
curl
x y z
y x z
x y F F
i j k
k · ∇ × ·
− −
· +
∂
∂
∂
∂
∂
∂
3 3 3
2 2
3( ) .
Hence,
curl d curl
t
dsdt
s s dsdt
S
F S F
r r
⋅ · ⋅ ×

.
`
,
· − · − · −
∫ ∫ ∫∫
∫ ∫
∂
∂
∂
∂
π π
π
π
s
0
1
0
2
2
0
1
0
2
3 2
3
4
3
2
( )( )
Exercises
1. Let S be the surface S S S · ∪
1 2
, where S x y z x y z
1
2 2
1 0 1 · + · ≤ ≤ {( , , ): , }, and
S x y z x y z z
2
2 2 2
1 1 1 · + + − · ≥ {( , , ): ( ) , }. Let the function F be given by
k j i F ) 5 ( ) ( ) ( ) , , (
4 3 2
y z x z xy y z x z y x + + + + + · .
Compute the flux integral
∫∫
⋅ × ∇
S
dS F ,
where S has the orientation pointing away from the z axis.
18.5
2. Let S be the hemisphere 0 , 1
2 2 2
≤ · + + z z y x with the orientation pointing toward
the origin.
a)Describe the boundary of S and its orientation that is consistent with the orientation
of S.
b)Evaluate the flux
∫∫
⋅ × ∇
S
dS F , where k j i F z x y z y x + + · 2 ) , , (
3. Let S
1
and S
2
be two surfaces with a common boundary C. Draw a picture indicating
the orientations these surfaces must have to insure that
∇ × ⋅ · ∇ × ⋅
∫∫ ∫∫
F S F S d d
S S
1 21
.
4. Let S be a surface with boundary C . Suppose they are consistently oriented. Suppose
a is a constant vector. Prove that
( ) a r r a S × ⋅ · ⋅
∫ ∫∫
d d
C S
2 .
[Remember, r i j k · + + x y z .]
5. Suppose S is a surface with boundary C and F is a vector function such that F × ∇ is
tangent to S at each point of S. Prove that
∫
· ⋅
C
d 0 r F .
6. Let
3
R r → B : be a vector description of the surface S with boundary C. Let F be a
vector function such that
18.6
,
`
.

∂
∂
×
∂
∂
∂
∂
×
∂
∂
· × ∇
t s
t s
r r
r r
F
1
.
Show that · ⋅
∫
C
dc F area of S.
7. Suppose the vector function F on a domain D is conservative. Prove that 0 · × ∇ F
everywhere in D.
8. Let j i F
2 2 2 2
) , , (
y x
x
y x
y
z y x
+
+
+
−
· , 0
2 2
≠ + y x .
a)Compute F × ∇ .
b)Prove that F is not conservative. [Hint: Evaluate the line integral
∫
⋅
C
dr F , where C
is the circle 0 , 1
2 2
· · + z y x , with the usual counterclockwise orientation.]
18.2 Path Independence Revisited
Problem 7 at the end of the previous section perhaps raised our hopes that an easy
test for a function F to be conservative in a domain D is simply to see if F × ∇ =0. If so,
these hopes were quickly dashed by Problem 8. In this section, we shall see just what we
can do along this line. The concept introduced next provides the key to understanding and
enlightenment.
An open subset D of
3
R is called simply connected if every simple closed curve
in D is the boundary of some surface contained entirely in D. Thus for instance the region
} 1 : ) , , {(
2 2 2
< + + · z y x z y x D
is simply connected, while the region
} 1 : ) , , {(
2 2
> + · y x z y x R
is not.
18.7
Now it easy to see that if F has as domain a simply connected region D, then
0 · × ∇ F everywhere in D implies that F is indeed conservative. We show that F is
conservative by showing that the integral of F around any closed curve is 0. This is easy
to do. Let C be any closed curve in D. Then D is simply connected, so there is a surface S
the boundary of which is C. Now unleash Stokes’s Theorem:
. 0 · ⋅ × ∇ · ⋅
∫ ∫∫
C S
d d S F r F
How about that!
Exercises
9. Explain how you know that j i F
2 2 2 2
) , , (
y x
x
y x
y
z y x
+
+
+
−
· , x > 0. is conservative.
10. Find a potential function for the vector function F given in Problem 9.
19.1
Chapter Nineteen
Some Physics
19.1 Fluid Mechanics
Suppose ) , , , ( t z y x v is the velocity at k j i r z y x z y x + + · · ) , , ( of a fluid flowing
smoothly through a region in space, and suppose ) , , , ( t z y x ρ is the density at r at time t. If
S is an oriented surface, it is not hard to convince yourself that the flux integral
∫∫
⋅
S
dr v ρ
is the rate at which mass flows through the surface S. Now, if S is a closed surface, then
the mass in the region B bounded by S is, of course
∫∫∫
B
dV ρ .
The rate at which this mass is changing is simply
∫∫∫ ∫∫∫
∂
∂
·
∂
∂
B B
dV
t
dV
t
ρ
ρ .
This is the same as the rate at which mass is flowing across S into B:
∫∫
⋅ −
S
dr v ρ , where S
is given the outward pointing orientation. Thus,
∫∫∫ ∫∫
⋅ − ·
∂
∂
B S
d dV
t
r v ρ
ρ
.
We now apply Gauss’s Theorem and get
. ) ( dV d dV
t
B B S
v r v ρ ρ
ρ
∫∫∫ ∫∫∫ ∫∫
⋅ ∇ − · ⋅ − ·
∂
∂
Thus,
dV
t
B
∫∫∫
,
`
.

⋅ ∇ +
∂
∂
) ( v ρ
ρ
.
Meditate on this result. The region B is any region, and so it must be true that the
integrand itself is everywhere 0:
19.2
0 ) ( · ⋅ ∇ +
∂
∂
v ρ
ρ
t
.
This is one of the fundamental equations of fluid dynamics. It is called the equation of
continuity.
In case the fluid is incompressible, the continuity equation becomes quite simple.
Incompressible means simply that the density ρ is constant. Thus 0 ·
∂
∂
t
ρ
and so we have
, 0 ) ( ) ( · ⋅ ∇ · ⋅ ∇ · ⋅ ∇ +
∂
∂
v v v ρ ρ ρ
ρ
t
or
0 · ⋅ ∇ v .
Exercise
1. Consider a one dimensional flow in which the velocity of the fluid is given by
) (x f · v , where 0 ) ( > x f . Suppose further that the density ρ of the fluid does not vary
with time t. Show that
) (
) (
x f
k
x · ρ ,
where k is a constant.
19.2 Electrostatics
Suppose there is a point charge q fixed at the point s. Then the electric field
) (r E
q
due to q is given by
3
) (
s r
s r
r E
−
−
· kq
q
.
It is easy to verify, as we have done in a previous chapter, that this field, or function, is
conservative, with a potential function
 
) (
s r
r
−
−
·
kq
P
q
;
so that
q q
P ∇ · E . Physicists do not like to be bothered with the minus sign in
q
P , so they
define the electric potential
q
V to be 
q
P . Thus,
19.3
 
) (
s r
r
−
·
kq
V
q
,
and
) ( ) ( r r E
q q
V −∇ · .
We have already seen that the flux out of a closed surface S is
¹
'
¹
· ⋅
∫∫
origin the enclose does if 4
origin the enclose not does if 0
S kq
S
d
S
q
π
S E
Some meditation will convince you there is nothing special here about the origin; that is, if
the point charge is at s, then
¹
'
¹
· ⋅
∫∫
s
s
S E
enclose does if 4
enclose not does if 0
S kq
S
d
S
q
π
Next, suppose there are a finite number of point charges , , at , at
2 2 1 1
K s s q q and
n n
q s at . Suppose
j
E is the electric intensity due to
j
q . Then it should be clear that the
electric field due to these charges is simply the sum
∑ ∑
· ·
−
−
· ·
n
j
j
j
j
n
j
j
q k
1
3
1
 
) (
s r
s r
E r E .
Also,
∑
·
−
·
n
j j
j
q
k V
1
 
) (
s r
r ; and
) ( ) ( r r E V −∇ · .
Finally,
∫∫
∑
· ⋅
S
j
q k d π 4 S E
where the sum is over those charges enclosed by S.
19.4
Things become more exciting if instead of point charges, we have a charge
distribution in space with charge density ρ . To find the electric field ) (r E produced by
this distribution of charge in space, we need to integrate:
s
U
dV k
∫∫∫
−
−
·
3
 
) (
) ( ) (
s r
s r
s r E ρ .
But this appears to be a serious breach of decorum. We are integrating over everything,
and at r s · we have the dreaded 0 in the denominator. Thus what we see above is an
improper integral—that is, it is actually a limit of integrals. Specifically, we integrate not
over everything but over everything outside a spherical solid region of radius a centered at
r. We then look at the limit as 0 → a of this integral. With the integral for the electric
field, this limit exists, and so there is no problem with 0 on the bottom of the integrand. In
the same way, we are safe in writing for the potential
s
U
dV k V
∫∫∫
−
·
 
) (
) (
s r
s
r
ρ
.
Everything works nicely so that we also have
) ( ) ( r r E V −∇ · .
If R is a solid region bounded by a closed surface S, then we can also integrate to get
(*)
∫∫ ∫∫∫
· ⋅
S R
dV k d . ) ( 4 s S E ρ π
The divergence of E is the troublesome item in extending matters to distributed
charge. If we simply try to calculate the divergence by
∫∫∫ ∫∫∫
·
U U
dV div dV div stuff) ( stuff ,
then things go wrong because the improper integral of the divergence does not exist.
Gauss saves the day. Let R be any region and let S be the closed surface bounding R.
Then
∫∫ ∫∫∫
⋅ ∇ · ⋅
S R
dV d E S E .
But from equation (*) we have
∫∫∫ ∫∫ ∫∫∫
⋅ ∇ · · ⋅
R S R
dV dV k d ) ( 4 E s S E ρ π .
This gives us
19.5
∫∫∫ ∫∫∫
⋅ ∇ ·
R R
dV dV k 4 E ρ π , or
( )dV k
R
∫∫∫
− ⋅ ∇ ρ π 4 E .
But R is any region, and so it must be true that
ρ πk 4 · ⋅ ∇ E
for all r.
Finally, remembering that V −∇ · E , we get
ρ πk V 4 ) ( · ∇ ⋅ −∇ · ⋅ ∇ E ;
ρ πk V 4
2
− · ∇ , or
ρ πk
z
V
y
V
x
V
4
2
2
2
2
2
2
− ·
∂
∂
+
∂
∂
+
∂
∂
.
This is the celebrated Poisson’s Equation, a justly famous partial differential equation, the
study of which is beyond the scope of this course.
.
Table of Contents
Chapter One  Euclidean Three Space 1.1 Introduction 1.2 Coordinates in ThreeSpace 1.3 Some Geometry 1.4 Some More GeometryLevel Sets Chapter Two  VectorsAlgebra and Geometry 2.1 Vectors 2.2 Scalar Product 2.3 Vector Product Chapter Three  Vector Functions 3.1 Relations and Functions 3.2 Vector Functions 3.3 Limits and Continuity Chapter Four  Derivatives 4.1 Derivatives 4.2 Geometry of Space CurvesCurvature 4.3 Geometry of Space CurvesTorsion 4.4 Motion Chapter Five  More Dimensions 5.1 The space Rn 5.2 Functions Chapter Six  Linear Functions and Matrices 6.1 Matrices 6.2 Matrix Algebra Chapter Seven  Continuity, Derivatives, and All That 7.1 Limits and Continuity 7.2 Derivatives 7.3 The Chain Rule Chapter Eight  f:Rn› R 8.1 Introduction 8.2 The Directional Derivative 8.3 Surface Normals 8.4 Maxima and Minima 8.5 Least Squares 8.6 More Maxima and Minima 8.7 Even More Maxima and Minima Chapter Nine  The Taylor Polynomial 9.1 Introduction 9.2 The Taylor Polynomial 9.3 Error Supplementary material for Taylor polynomial in several variables.
Chapter Ten  Sequences, Series, and All That 10.1 Introduction 10.2 Sequences 10.3 Series 10.4 More Series 10.5 Even More Series 10.6 A Final Remark Chapter Eleven  Taylor Series 11.1 Power Series 11.2 Limit of a Power Series 11.3 Taylor Series Chapter Twelve  Integration 12.1 Introduction 12.2 Two Dimensions Chapter Thirteen  More Integration 13.1 Some Applications 13.2 Polar Coordinates 13.3 Three Dimensions Chapter Fourteen  One Dimension Again 14.1 Scalar Line Integrals 14.2 Vector Line Integrals 14.3 Path Independence Chapter Fifteen  Surfaces Revisited 15.1 Vector Description of Surfaces 15.2 Integration Chapter Sixteen  Integrating Vector Functions 16.1 Introduction 16.2 Flux Chapter Seventeen  Gauss and Green 17.1 Gauss's Theorem 17.2 Green's Theorem 17.3 A Pleasing Application Chapter Eighteen  Stokes 18.1 Stokes's Theorem 18.2 Path Independence Revisited Chapter Ninteen  Some Physics 19.1 Fluid Mechanics 19.2 Electrostatics
George Cain & James Herod School of Mathematics Georgia Institute of Technology Atlanta, Georgia 303320160
Chapter One
Euclidean ThreeSpace
1.1 Introduction. Let us briefly review the way in which we established a correspondence between the real numbers and the points on a line, and between ordered pairs of real numbers and the points in a plane. First, the line. We choose a point on a line and call it the origin. We choose one direction from the origin and call it the positive direction. The opposite direction, not surprisingly, is called the negative direction. In a picture, we generally indicate the positive direction with an arrow or a plus sign:
Now we associate with each real number r a point on the line. First choose some unit of measurement on the line. For r > 0, associate with r the point on the line that is a distance r units from the origin in the positive direction. For r < 0, associate with r the point on the line that is a distance r units from the origin in the negative direction. The number 0 is associated with the origin. A moments reflection should convince you that this procedure establishes a socalled onetoone correspondence between the real numbers and the points on a line. In other words, a real number determines exactly one point on a line, and, conversely, a point on the line determines exactly one real number. This line is called a real line. Next we establish a onetoone correspondence between ordered pairs of real numbers and points in a plane. Take a real line, called the first axis, and construct another real line, called the second axis, perpendicular to it and passing through the origin of the first axis. Choose this point as the origin for the second axis. Now suppose we have an ordered pair ( x1 , x2 ) of reals. The point in the plane associated with this ordered pair is
1.1
in which the first axis is horizontal. A moments reflection here will convince you that there is exactly one point in the plane thus associated with an ordered pair (a. of course.2
. and constructing a line parallel to the first axis through the point on the second axis corresponding to the real number x 2 . x2 ) . b). We thus usually speak of the horizontal axis and the vertical axis.x 1 is the first coordinate and
x 2 is the second coordinate. Suppose we have an equation
1. rather than the first axis and the second axis.
Given any collection of ordered pairs( A collection of ordered pairs is called a relation. x2 ) when. we actually mean the point associated with the ordered pair ( x1 . x2 ) . the second axis is vertical. We also frequently abuse the language by speaking of a point ( x1 . a picture of the collection is obtained by simply looking at the set of points in the plane corresponding to the pairs in the given collection. The
numbers x 1 and x 2 are called the coordinates of the point.). The point at which these two lines intersect is the point associated with the ordered pair ( x1 . and the positive direction on the second axis is up. and each point in the plane is the point associated with some ordered pair (a. the positive direction on the first axis is to the right.found by constructing a line parallel to the second axis through the point on the first axis corresponding to the real number x 1 . b):
It is traditional to assume the point of view we have taken in this picture.
involving two variables, say x and y. Then this equation defines a collection of ordered pairs of numbers, namely all ( x , y ) that satisfy the equation. The corresponding picture in the plane is called the graph of the equation. For example, consider the equation y 2 = x 4 . Let’s take a look at the graph of this equation. A little algebra (very little, actually), convinces us that
{( x , y ): y 2 = x 4 } = {( x, y): y = x 2 } ∪ {(x , y ): y = − x 2 } ,
and we remember from the sixth grade that each of the sets on the right hand side of this equation is a parabola:
What do we do with all this? These constructions are, of course, the bases of analytic geometry, in which we join the subjects of algebra and geometry, to the benefit of both. A geometric figure (a subset of the plane ) corresponds to a collection of ordered pairs of real numbers. Algebraic facts about the collection of ordered pairs of real are reflected by geometric facts about the subset of the plane, and, conversely, geometric
1.3
facts about the plane subset are reflected by algebraic facts about the collection of pairs of reals.
Exercises
Draw a picture of the given relation:
1. R = {( x , y ):0 ≤ x ≤ 1, and 1 ≤ y ≤ 4}
2. R = {(x , y ):− 4 ≤ x ≤ 4, and − x 2 ≤ y ≤ x 2 }
3. R = {(x , y ):1 ≤ y ≤ 2} ∩ {( x , y ): y ≥ x 2 }
4. S = {( x , y ): x 2 + y 2 = 1, and x ≥ 0}
5. S = {( x , y ): x 2 + y 2 ≤ 1} ∩ {( x, y): y ≤ x 2 }
6. E = {(r , s): r =  s}
7. T = {(u, v ): u+  v  = 1}
8. R = {( u, v ):u+  v ≤ 1}
9. T = {( x, y ): x 2 = y 2 }
10. A = {( x, y): x 2 ≤ y 2 }
1.4
11. G = {( s, t ): max { s, t } = 1}
12. B = {( s, t ): max { s, t } ≤ 1}
1.2 Coordinates in ThreeSpace Now let’s see what’s doing in three dimensions. We shall associate with each ordered triple of real numbers a point in three space. We continue from where we left off in the previous section. Start with the plane constructed in the previous section, and construct a line perpendicular to both the first and second axes, and passing through the origin. This is the third, axis. Now we must be careful about which direction on this third axis is chosen as the positive direction; it makes a difference. The positive direction is chosen to be the direction in which a righthand threaded bolt would advance if the positive first axis is rotated to the positive second axis:
We now see how to define a onetoone correspondence between ordered triples of real numbers ( x1 , x2 , x3 ) and the points in space. The association is a simple extension of the way in which we established a correspondence between ordered pair and points in
1.5
a plane. Here’s what we do. Construct a plane perpendicular to the first axis through the point x 1 , a plane perpendicular to the second axis through x 2 , and a plane perpendicular to the third axis through x 3 . The point at which these three planes intersect is the point associated with the ordered triple ( x1 , x2 , x3 ) . Some meditation on this construction should convince you that this procedure establishes a onetoone correspondence between ordered triples of reals and points in space. As in the two dimensional, or plane, case, x 1 is called the first coordinate of the point, x 2 is called the second coordinate of the point, and x 3 is called the third coordinate of the point. Again, the point corresponding to (0,00 ) is called the origin, and we speak of the point ( x1 , x2 , x3 ) , when we actually mean , the point which corresponds to this ordered triple.
The three axes so defined is called a coordinate system for three space, and the three numbers x, y, and z , where ( x , y , z) is the triple corresponding to the point P, are called the coordinates of P. The coordinate axes are sometimes given labelsmost
commonly, perhaps, the first axis is called the x axis, the second axis is called the y axis, and the third axis is called the z axis.
1.6
and suppose space is endowed with a coordinate system such that P = ( x .3 Some Geometry Suppose P and Q are two points.
We saw that in the plane an equation in two variables defines in a natural way a collection of ordered pairs of numbers. look at the picture:
We can see that d 2 = h 2 + (z − w) 2 and h 2 = ( x − u) 2 + ( y − v) 2 . How do we find the distance between P and Q ? This simple enough.7
.1. The analogous situation obtains in threespace: an equation in three variables defines a collection of ordered triples. y . We thus speak of the collection of triples ( x . w) . z) which satisfy the equation
x 2 + y2 + z 2 = 1
1. or d = ( x − u) 2 + ( y − v ) 2 + ( z − w) 2 . v . Thus we have d 2 = ( x − u) 2 + ( y − v ) 2 + ( z − w) 2 . y. z) and Q = ( u.
An example:
This picture was drawn using Maple. When the axes are labeled x.8
. First. What does the graph of x 2 + y2 − z 2 = 1 look like? We’ll go after a picture of this one by slicing the graph with the coordinate planes. this is known as the yz plane. or the y axis and the z axis. . the plane y = 0 is the xz plane. it is difficult to see exactly what a graph of equation looks like.The collection of all such points is the graph of the equation. and z = 0 is the xy plane. y. it is easy to see that the graph is precisely the set of all points at a distance of 1 from the origina sphere of radius 1 and center at the origin. Similarly. and this is clearly the plane determined by the second axis and the third axis. let’s slice through it with the plane z = 0 . then we see
1. Let’s look at a more complicated example. but they usually draw rather poor pictures whose main application is in stimulating your own imagination sufficiently to allow you to see the graph in your mind’s eye. Computers can help. and even more difficult for most of us to draw it. In this example. These special planes are also called the coordinate planes. More often than not. and z. The graph of the equation x = 0 is simply the set of all points with first coordinate 0.
of course.x 2 + y2 = 1 . What the graph looks like should be fairly clear by now:
1. Here we see x 2 − z 2 = 1 .9
. a circle of radius 1 centered at the origin. a hyperbola:
We. let’s slice with the plane y = 0 . see the same hyperbola when we slice the graph with the plane x = 0 . Next.
z): x 2 + y 2 ≤ 1} b) S = {( x . (This graph is called an ellipsoid.10
. b)Sketch G. Describe the following sets a) S = {( x . and z = 0 . z): x ≥ 0. Describe the set of points S = {( x .4 Some More GeometryLevel Sets The curves that result from slicing the graphs with the coordinate planes are special cases of what are called level sets of a set.)
16. Specifically.
14. it is called a hyperboloid. z): x ≥ 5} d) T = {(r. y . t ): r 2 + s2 + t 2 ≤ 4}
15. and z = 0 . the
intersection of S with a plane z = constant is called a level set.
Exercises
13. z): z ≥ 0} c) R = {(x .This graph has a name. In case the level set is a
1. y . if S is a set. (Does this set look at all familiar to you?.)
1. s. y . Let G be the graph of the equation x 2 + 4 y 2 + 9 z 2 = 36 . b)Sketch G. y = 0 . y ≥ 0. a)Sketch the graphs of the curves sliced from G by the coordinate planes x = 0 . y . and z ≥ 0} . a)Sketch the graphs of the curves sliced from G by the coordinate planes x = 0 . Let G be the graph of the equation x 2 − 3y 2 + 4 z 2 = 12 . y = 0 .
in which the contours are.) A family of level sets can provide a nice stimulant to your powers of visualization. the isoclines on a 500mb chart are simply level curves for the 500mb surface. Everyday examples of the use of level sets to describe a set are contour maps. for instance. it is frequently called a level curve. It is clear that these curves are concentric circles of radius c 2 − 1 centered at the
origin. the origin. of course. Let’s illustrate with an example. (The slices by planes x = constant. Notice first that we have the same curve for z = c and z = c. There are no level sets for  c < 1.
1. We shall thus look at just that part of the graph that is above the xy plane. the level set is a single point. or y = constant are also level sets. and weather maps. just level curves .11
.curve. or x 2 + y2 = c2 − 1 . in which. and for c = 1 or 1. The graph is symmetric about the plane z = 0. Let S be the graph of z 2 − y2 − x2 = 1 Now we look at the level set z = c : c 2 − y 2 − x 2 = 1 .
Here is a Maple drawn picture:
1. of course. slice with the planes x = 0 and y = 0 to get a better idea of what this thing looks like.Next. is the same. It is rather easy to visualize this graph. a hyperbola:
The slice by y = 0. For x = 0. we see z 2 − y2 = 1 .12
.
1.This also is called a hyperboloid. x 3 ): x 1 + x 2 +  x 3  = 1} .
Exercises
17. x 2 .). it is a paraboloid. b)Sketch the slices by the planes x = 0 and y = 0. c)Sketch C. a)Sketch some level sets z = c.
This is a hyperboloid of two sheets. a)Sketch the coordinate plane slices of S. Let S = {( x 1 .13
.
18. while the
previously described hyperboloid is a hyperboloid of one sheet. b)Sketch the set S. coordinate plane slices. describe the graph of the equation z = x 2 + y 2 . (This one has a name also. Let C be the graph of the equation z 2 = 4(x 2 + y 2 ) . What does the man on the street call this set?
19. Using level sets. and whatever.
20.14
.
1. coordinate plane slices. Using level sets. and whatever. describe the graph of the equation z = x 2 − y 2 .
2. Convince yourself every segment in an equivalence class is equivalent with every other segment in that class. Given a directed segment u. These equivalence classes of directed line segments are called vectors. The
members of a vector v are called representatives of v.1 Vectors A directed line segment in space is a line segment together with a direction. We frequently denote the direction of a segment by drawing an arrow head on it pointing in its direction and thus think of a directed segment as a spear. An equivalence class containing a segment L is the set of all directed segments equivalent with L. The length.1
. of a vector v is defined to be the common length of the representatives of v.Chapter Two
VectorsAlgebra and Geometry
2. the vector which contains u is called the vector determined by u. We say that two segments have the same direction if they are parallel and their directions are the same:
Here the segments L1 and L2 have the same direction. or magnitude. and two different equivalence classes must be disjoint. Thus the directed line segment from the point P to the point Q is different from the directed line segment from Q to P. We define two directed segments L and M to be equivalent ( L ≅ M ) if they have the same direction and have the same length. It is generally designated by v. The angle between two vectors u and v is simply the angle between the directions of representatives of u and v.
if u is a representative of u and v is a representative of v. Such physical concepts abound: force. Thus the difference uv of two vectors is defined to be the vector you add to v to get u. acceleration. it is traditional to omit them and write simply u + v + w. of vectors. we define what we mean by the sum of two vectors u and v. First.2
. and that u + v = v + u :
Convince yourself also that addition is associative: u + (v + w) = (u + v) + w. Subtraction is defined as the inverse operation of addition. We now introduce an arithmetic. the directed segment from the nose of v to the nose of u is a representative of u . Velocity provides a ready example. The real numbers (or sometimes. Choose a spear u from u and a spear v from v. The vector which contains the directed segment from the tail of u to the nose of v is defined to be u + v . In pictures. Look at the picture and convince yourself that the it does not matter which u spear or v spear you choose. the complex numbers) are frequently called scalars in order to distinguish them from vectors. An easy consequence of elementary geometry is the fact that  u + v  <  u  +  v . Saying the car is traveling 50 miles/hour doesn’t tell the whole story. Thus velocity is a vectorit has both magnitude and direction. the sum of u and v. Place the tail of v at the nose of u. Since it does not matter where the parentheses occur.v:
2. or algebra. etc. you must specify in what direction the car is moving. and we put the tails of u and v together.Vectors are just the right mathematical objects to describe certain concepts in physics. displacement.
e. what are we to make of u . of course.. Our next move is to define a onetoone correspondence between vectors and points in space (This will. real number) with a vector u. also establish a onetoone correspondence between vectors and ordered triples of real numbers.v. or points. we make the zero vector behave as it should: u . and u + (1)v = u .u to stand for (1)u.3
. and direction opposite the direction of u if r < 0. Convince yourself that all the following nice properties of this multiplication hold:
(r + s)u = ru + su .u = 0 and u + 0 = u for all vectors u. Next we define the product of a scalar r (i. 0u = 0.
It is then perfectly safe to write .u ? We define a special vector with 0 length. the direction is not important!). simply take a representative of the vector u and place its tail at the origin. r ( u + v ) = ru + rv. We may think of 0 as the collection of all degenerate line segments.).Now. The point at which is
2. called the zero vector and denoted 0. The correspondence is quite easy. To make our algebra of vectors nice . Note that the zero vector is special in that it has no direction (If you are going 0 miles/hour. The product ru is defined to be the vector with length ru and direction the same as the direction of u if r > 0.
found the nose of this representative is the point associated with u. 0). but this shorthand is usually clear and saves a lot of verbiage (The numbers a. Suppose u = (a. z) = xi + y j + zk .b. y . and ru = (ra. c). 1). rc).4
.x. c + d).
u + v = (a + x.c) or of a point u. let j be the vector corresponding to (0. b. z). y.y. and c are called the coordinates. Strictly speaking this equation makes no sense. 1. and let k be the vector corresponding to (0. b . an equivalence class of directed line segments cannot possible be the same as a triple of real numbers. b. Unleash your vast knowledge of
elementary geometry and convince yourself of the truth of the following statements: u = a 2 + b 2 + c2 . c) is associated with the vector u in this manner is shorthandedly indicated by writing u = (a. Look at the picture:
2. Thus we frequently do not distinguish between points and vectors and indiscriminately speak of a vector (a.). b + y. Any vector u can now be expressed as a linear combination of these special socalled coordinate vectors: u = ( x . We handle the vector with no representatives by associating the origin with the zero vector. 0). b. b. c . c) and v = (x. rb.d).
Let i be the vector corresponding to the point (1. of u.v = (a . or components. 0. 0. u .
Example
Let’s use our newfound knowledge of vectors to find where the medians of a triangle intersect. The fact that the point with coordinates (a.
5
. the result you remember from Mrs.
Exercises
2. and 2 2 s 1 − t − = 0. no doubt. It follows that s=t= 2 . 2 2 2 This means that we must have s t − = 0. 2 2 Tidying this up gives us s s t (1 − t − ) a = ( − ) b . 3
This is. 2
Otherwise.We shall find scalars s and t so that b b−a a + t ( − a ) = s (a + ). a and b would be nonzero scalar multiples of one another. Turner’s high school geometry class. which would mean they have the same direction.
3).
7.6
. z 2 ) . Let u = (2. Find the midpoint of the line segment joining the points (1. y1 . y 2 .
2. z1 ) its nose will be at the point ( x 2 .4. What is the distance between the points ( x1 . 3).
3.1) and v = (1.1). z 2 ) ?
4. Find the vector such that if its tail is at the point ( x1 .8). and w = (3.
2. Describe the set of points L = {t i: − ∞ < t < ∞} . z1 ) and ( x 2 .3).1. and − ∞ < s < ∞} . Describe the set C = {cost i + sin t j: 0 ≤ t ≤ 2π } . Describe the set P = {5k + ti + sj: − ∞ < t < ∞ .and
v
=
(1.
9. Describe the set P = {w + tu + sv: − ∞ < t < ∞ . and − ∞ < s < ∞} .2.
10. y 2 .
11. y1 .3.6.
8. Describe the set P = {tu + sv: − ∞ < t < ∞ .1).3. and − ∞ < s < ∞} .
Describe
the
set
of
points
M = {v + t u: − ∞ < t < ∞ } .
6. and − ∞ < s < ∞ } .7). Let
u
=
(2. 2. Describe the set of points M = {3k + t i: − ∞ < t < ∞} .
5. Let u = (2. Describe the set of points L = {t u: .
12. Describe the set P = {ti + sj: − ∞ < t < ∞ .2.5. Let u = (2. 9) and (3. 5.4. v = (1.∞ < t < ∞} .8).
2. The scalar product is frequently also called the dot product. the angle between vectors.
14. then u ⋅ v is the length of the projection of v onto u. etc. 7). 6). Let T be the triangle with vertices (2. (1.13. We now introduce this product. 2. Describe the set P = {ti + t 2 j : 1 ≤ t ≤ 2} . we omit mention of this. Generally.7
. 2. (More precisely. Find the point at which the medians intersect. Describe the set E = {4 cost i + 3 sin t j: 0 ≤ t ≤ 2π } . where there is no danger of confusion. where θ is the angle between u and v. of course. 4).2 Scalar Product You were perhaps puzzled when in grammar school you were first told that the work done by a force is the product of the force and the displacement since both force and displacement are.)
2. We avoid having to use the latter weasel words by defining the zero vector to be perpendicular to every vector. It is a scalar and hence is called the scalar product. just as we speak of the length of vectors. then we can say u ⋅ v = 0 if and only if u and v are perpendicular. Study the following picture to see that if u = 1. 5. and that u ⋅ v = 0 if and only if u and v are
perpendicular (or orthogonal ). vectors. and (4. or one or the other of the two is the zero vector. Observe that u ⋅ u = u2 . This scalar product u ⋅ v is defined by u ⋅ v = uv cosθ .
15. the length of the projection of a representative of v onto a representative of u.
It is clear that (au) ⋅ (bv ) = (ab )u ⋅ v . since i ⋅ i = j ⋅ j = k ⋅ k = 1 and i ⋅ j = i ⋅ k = j ⋅ k = 0.8
.
Now let’s get a recipe for the scalar product of u = (a .
Example
2. and w. y . b. v. c) and v = ( x .
We thus see that it is remarkably simple to compute the scalar product of two vectors when we know their coordinates. Study the following picture until you believe that u ⋅ ( v + w ) = u ⋅ v + u ⋅ w for any three vectors u. z) : u ⋅ v = (ai + bj + ck ) ⋅ ( xi + y j + zk ) = ax i ⋅ i + ayi ⋅ j + azi ⋅ k + bxj ⋅ i + byj ⋅ j + bzj ⋅ k + cxk ⋅ i + cyk ⋅ j + czk ⋅ k = ax + by + cz.
2. Find an angle between the lines 3x + 2 y = 1 and x − 2 y = 3. 2) to the point (3. Let L be the line passing through the origin and the point (2. 3
Exercises
16. and let M be the line passing through the points (3. let’s see how vectors can make geometry easy by using them to find the angle between a diagonal of a cube and the diagonal of a face of the cube. 2) and (5. s. 2 θ = Cos −1 . s).
2. 5). where θ is the angle we seek. 3 2 s2 =  d  f  2 s2 3s2 2s 2 = 2 . The vector determined by this diagonal is thus d = si + sj + sk and the vector determined by the diagonal of the face in the horizontal coordinate plane is f = si + sj . 3). one vertex is the origin and the vertex at the other end of the diagonal from the origin is (s.
18. Find the work done by the force F = 6i − 3 j + 2 k in moving an object from the point (1. 4. Find the smaller angle between L and M.Again. Introduce a coordinate system so that the faces are parallel to the coordinate planes. This gives us cosθ = Or. Thus d ⋅ f = d  f cosθ = s 2 + s2 . Suppose the cube has edge length s.
17. 5).9
.
and r is the position vector of the point in question. What is the graph of the equation v ⋅ ( x . the celebrated Law of Cosines. 5). of course.3 Vector Product Hark back to grammar school physics once again and recall what you were taught about the velocity of a point at a distance r from the axis of rotation. Prove that
c 2 = a 2 + b 2 − 2ab cosγ . 2. . Find an angle between the diagonal and an adjoining edge of a cube. 2) having slope 2.
21. you were likely told that the velocity is rω . The vector product of vectors u and v is defined by u × v = u vsin θ  n .10
.19.
2. and c. Suppose the lengths of the sides of a triangle are a. and suppose M is the line tangent to the curve y = x 3 at the point (1. We now know that these quantities are actually vectorsω is the angular velocity.
20. where ω is the rate at which the turntable is rotatingthe socalled angular velocity. Suppose L is the line passing through (1. and suppose γ is the angle opposite the side having length c. Find the smaller angle between L and M. The velocity of the point is the socalled vector product of these two vectors. Let v = (1. ).)
22. y .
(This is. The grammar school quantities are the magnitudes of ω (the angular speed) and of r . z) = 0 ?
2. 1). b.
Note first that this is a somewhat more exciting product than you might be used to: the order of the factors makes a difference. Thus u × v = − v × u . ) This gives a vector v ** having the same length as v * and having the direction of u × v .11
. Now project v onto this plane. Let P be a plane perpendicular to u. Thus u × v = u v** :
2.where θ is the angle between u and v and n is a vector of length 1 (such vectors are called unit vectors) which is orthogonal to both u and v and which points in the direction a righthand threaded bolt would advance if u were rotated into the direction of v.” (By the positive direction of rotation about a vector a. we mean the diction that would cause a righthand threaded bolt to advance in the direction of a. Now let’s find a geometric construction of the vector product u × v . Now rotate this vector v * 90 degrees around u in the “positive direction. Proceed as follows. giving us a vector v * perpendicular to u and having length  v sin θ  .
we have u × (v + w) = u × v + u × w . c) = ai + bj + ck and v = ( x . Making these substitutions in the above equation for u × v gives us
2. j × k = −( k × j ) = i . v. y .12
. until we note that i× i = j× j = k× k =0. b. why did we go to all this trouble to construct u × v in this fashion? Simple. and w. i × j = −( j × i ) = k .Now. and k × i = −( i × k ) = j . z) = xi + y j + zk . We have u × v = (ai + bj + ck ) × ( xi + y j + zk ) = ax ( i × i) + ay ( i × j ) + az( i × k ) + bx ( j × i ) + by ( j × j ) + bz( j × k) + cx( k × i ) + cy( k × j ) + cz( k × k )
This looks like a terrible mess.
(Draw a picture!) We shall see how to compute this vector product u × v for u = (a . It makes it much easier to see that for any three vectors u.
For our point on the Earth. In other words. The Earth makes one complete revolution every 24 hours. choose Room 254. Our velocity is thus . .
2. the origin of our coordinate system is the center of the Earth. etc. As we said.). Skiles Classroom Building at Georgia Tech. and it is about 3960 miles from the center of the Earth. is of course. The position vector r of our room is .
This is not particularly hard to remember. the coordinate vector k points through the North Pole. . but changes as the Earth rotates.u × v = (bz − cy )i + (cx − az) j + ( ay − bx ) k . The latitude of the room is about 33. and neglect its motion about the sun. of course. x y z
Example
Let’s find the velocity of a point on the surface of the Earth relative to a coordinate system whose origin is fixed at its centerwe thus shall consider only motion due to the Earth’s rotation. not a constant. We choose the third axis to point through the North Pole. . The velocity of our room. and so its angular velocity ω is ω = 2π k ≈ 02618 k radians/hour.13
. We find the velocity at the instant our room is in the coordinate plane determined by the vectors i and k. 24
r = 3960(cos(3375 ) i + sin(3375) k ) ≈ 32926 i + 22001k miles.75 degrees (North. but there is a nice memory device using determinants: i j k u× v = a b c .
Exercises
2. 10.i j k ω ×r = 0 0 02618 ≈ 862 j miles/hour. 20). the nonparallel sides of which are representatives of the vectors a and b:
The area A is clearly A = a bsin θ =  a × b . and (6. 8).
Suppose we want to find the area of a parallelogram. 5 6 22 and so.14
. . .
Example Find the are of the parallelogram with a vertex (1. 32926 0 22001 . 7. This is easy. So we have
i j k a × b = 3 3 10 = 6i − 16 j + 3k .2) and the vertices at the other ends of the sides adjoining this vertex are (4. This is just as in the above picture with a = (4 − 1) i + (7 − 4) j + (8 − (− 2)) k = 3i + 3 j + 10k and b = (6 − 1) i + (10 − 4) j + (20 − ( −2 ))k = 5i + 6 j + 22 k . Area = a × b = 6 2 + 16 2 + 32 = 301 .4.
6. Find the area of the triangle whose vertices are (0.2. Find the volume of the parallelepiped
2. 20) collinear? Explain how you know?
25.3).7. and (3.2. 6. 8). Find the torque created by the force a = i − 2 j − 7k .
24.0).23. 10. Are the points (0. Find a a vector perpendicular to the plane containing the points (1. (1.
f = 3i + 2 j − 3k acting at the point
26.15
.10).12).0. and (4.6).
27. (2.7).4.7).4. and (5. (1.
we call R a relation on A. c) ∈ R only if b = c is called a function.Chapter Three
Vector Functions
3. and a ∈ dom R .1 Relations and Functions We begin with a review of the idea of a function. If R ⊂ A × B is a relation. The Cartesian product A × B of these sets is the collection of all ordered pairs (a . and is. a function. A relation R ⊂ A × B such that (a . b): b is the mother of a} .b) ∈ R . if R is a function. x 2 ) in which x ∈[01] . there is exactly one ordered pair (a . a ):( a. b) ∈ R} . Thus we may define the function f from the interval [0. In case A = B and the domain of R is all of A. and domS −1 ≠ A .1
. This says that f is the collection of all ordered pairs ( x .b) such that a ∈ A and b ∈ B .1] to the real numbers by f ( x) = x 2 . . in fact. The S is a relation on A. Example Let A be the set of all people who have ever lived and let S ⊂ A × A be the relation defined by S = {( a. Very often a function f is defined by specifying the domain. and we say f is a function from A to B. b) ∈ R} . The relation S −1 is not a function. The domain of R is the set dom R = {a ∈ A:( a . In other words. The second “coordinate” b is thus uniquely determined by a. The fact that f ⊂ A × B is a function with dom f = A is frequently indicated by writing f : A → B . the inverse of R is the relation R −1 ⊂ B × A defined by R −1 = {(b. A relation R is simply a subset of A × B . Suppose A and B are sets. and giving a recipe for finding f(a). It is usually denoted R (a ) .b) ∈ R and (a .
Exercises
3.
is it necessarily true that f ∩ g is a function? Prove your answer. If f and g are both functions. Find dom f.
With X . municipalities. and let Y be the set of all U.
Suppose f ⊂ X × Y and g ⊂ X × Y . Is W a function? Is W −1 a function? Explain. y ): y is the largest city in x} . y ): y is the capital of x} . Explain why c is a function. a)What is b(South Carolina)? b)What is b(California)? c)Let f = c ∩ b . These are called vector functions.
3. If f is a function. As the drama unfolds. b): b is the weight (in pounds) of a} .
2.
−1
b)If y ∈ dom f
. Let X be the entire real line and let the function f be defined by f (t ) = ti + t 2 j . and find c(Nevada).
5. Y as in Exercise 2. It should be reasonably clear that if we place the tail of
3. we shall be solely interested in the special case in which X is a “nice” set of real numbers. Define the relation W ⊂ A × B by W = {( a. S. such as an interval.1. Let’s begin with a simple example.
is also a function.
Let A be the set of all Georgia Tech students. Suppose f : X → Y is a function and the inverse f a)What is f
−1 −1
6.
Suppose f ⊂ X × Y and g ⊂ X × Y . Define the relation c ⊂ X × Y by c = {(x . we shall see that such functions provide just the right tool for describing curves in space. vectorvalued functions.2
. and let B be the set of real numbers. and c(Kentucky).2 Vector Functions Our interest now will be focused on functions f : X → Y in which Y is a set of vectors. Initially.
4. what is f ( f
−1
( y )) ? Explain.
Let X be set of all states of the U. is it necessarily true that f ∪ g is a function? Prove your answer. where c is the function defined in Exercise 2.
3. or sometimes. let b be the function b = {( x ..
( f ( x )) ? Explain. S. c(Missouri).
Let f (t ) = cos ti + sin tj + t k . It’s not difficult to see that. Now. The function f is called a vector description of the curve. a right circular helix. The real usefulness of vector descriptions is most evident when we consider curves in space. for all t ≥ 0 . The nose of g thus always lies on the circle of radius one centered at the origin. the nose traces out this curve. we run around this circle and the third component of our position increases linearly.. a representative of f (t ) ) at the origin. This time.3
. what curve is followed by the nose of f(t)? Notice first that if we look down on this curve from someplace up the positive third axis (In other words. we see the circle described by cos ti + sin tj . let g (t ) = cos t i + sin t j for 0 ≤ t ≤ 4π . as t varies from 0 to 2π. Let’s look at another example. Here’s a computer drawn picture:
3. the nose traces out the circle again.). How about the curve described by the vector function g (t ) = cos t i + sin t j + sin(2 t ) k ? This one is just a bit more exciting.f (t ) (actually. The picture was drawn by Maple. the nose will lie on the curve y = x 2 . Let’s draw another. In fact. What is the curve described by this function? First. k is pointing directly at us. Convince yourself now that this curve looks like this:
This curve is called a helix. we have  g(t ) = 1. in fact. as t varies over the reals. and as t varies on from 2π to 4π. or more precisely. note that for all t. As t increases. the nose moves around the circle once.
[Hint: Find an equation in x and y. y. Exercises 7 .
8 . Sketch or otherwise describe the curve given by f (t ) = ti + t 3 k for −1 ≤ t ≤ 3 . Assuming the axes are labeled x.). the curve described by the vector function r(t ) = f (t ) i + g (t ) j + h( t ) k is equivalently described by the equations x = f (t ) y = g (t ) z = h(t ) These are called parametric equations of the curve (The variable t is called the parameter. and z. the graph of which is the given curve. Sketch or otherwise describe the curve given by f (t ) = (2t − 3) i + ( 3t + 1) j .4
.”) Observe that in giving a vector description.]
3.(This time we put the axes where they are “supposed to be. we are in effect specifying the three coordinates of points on the curves as ordinary real valued functions defined on a subset of the reals.
3.
18. a)Sketch or otherwise describe the curve given by the function r(t ) = a + tb .4.
15.
a)Find a vector function for the straight line passing through the points (1.2.
a)Find a vector equation for the graph of y = x 3 + 2 x 2 + x + 5. 14. b)Find a vector function for the line segment joining the points (1.
the graph of
which is the curve
12.
and y.1.4).1. where a = 2 i − j + 3k and b = i + 3 j − 5k . and let M be the line through the points (2.6) and (3.5).2).4.2) in the direction of the vector v = i − j + 2k .5
.
11.4) and (3.
Find an equation in x g (t ) = 3 cos ti + 4 sin tj .1. Sketch or otherwise describe the curve given by c( t ) = cos ti + sin t j + 7 k . 16.
Describe the curve given by L (t ) = (3t + 1)i + (1 − t ) j + 2tk .2) and is perpendicular to both L and M.4) and (3.9 . Find a vector description of the line which passes through the point (1. b)Find a vector equation for the graph of x = y 3 + 2 y 2 + y + 5.3 Limits and Continuity
3.
Find a vector equation for the graph of x 3 + y 3 = 1 . Let L be the line through the points (1. Sketch or otherwise describe the curve given by
c( t ) = cos(t 2 ) i + sin(t 2 ) j + 7 k .2.
17.1.5. Find a vector function for the straight line passing through the point (1. 10.5). b)Express r(t) in the form r(t ) = f (t ) i + g (t ) j + h( t ) k .2) and (2.
2 2
13.2.
3 ε for 0 <  t − t 0  < δ 3 . Suppose α( t ) is a scalar function for which lim α (t ) = a . Choose δ 1 . Then  α (t ) f (t ) − aL =  a ( f (t ) − L ) + L (α (t ) − a) + (α (t ) − a )( f (t ) − L ) ≤  a( f (t ) − L )+ L (α ( t ) − a )+ (α (t ) − a )( f (t ) − L ) < a  L ε ε ε ε ε ε + ε + < + + =ε 3(1+a) 3(1+ L) 3 3 3 3 3
3. δ 4 } suppose t is such that 0 <  t − t 0  < δ .δ 3 . or scalar. Let ε > 0 be given. This is traditionally written lim f (t ) = L . and L is a vector such that for every real number ε > 0. and f is a vector function
t → t0
for which lim f (t ) = L . δ 2 . δ 3 .
t → t0
To see this. and 3(1+  L ) ε for 0 <  t − t 0  < δ 4 . t 0 is a real number. and δ 4 so that  f ( t) − L<  f ( t) − L<  α (t ) − a <  α (t ) − a < ε for 0 <  t − t 0  < δ 1 . Specifically. there is a δ > 0 such that  f ( t ) − L  < ε whenever 0 <  t − t 0  < δ and t is in the domain of f. The definition for vector functions is essentially the same. suppose f is a vector valued function. 3(1+ a ) ε for 0 <  t − t 0  < δ 2 . It is but a modest exercise to show that
t → t0
lim(α (t ) f ( t )) = aL . we use the “behold!” method. function f is L. 3
Now let δ = min{δ 1 .6
.
t → t0
The vector L is called a limit of f at a.Recall from grammar school what we mean when we say the limit at t 0 of a realvalued.δ 2 .
If f is continuous at each point of its domain. Exercises
3. and lim z( t ) = c. and z(t ) is continuous at t 0 .7
. then f is continuous at t 0 if and only if each of the everyday scalar functions x ( t ).
It is now easy to show that all the usual nice properties of limits are valid for vector functions: lim( f (t ) + g( t )) = lim f ( t ) + lim g(t ).
t → t0 t →t 0 t → t0
We are now ready to say what we mean by a vector function’s being continuous at a point of its domain.
t → t0 t → t0 t → t0
lim( f (t ) ⋅ g(t )) = (lim f (t )) ⋅ (lim g(t )) . It is easy to see that if
t→ t0
f (t ) = x ( t ) i + y (t ) j + z (t ) k . Then we say f is continuous at t 0 if it is true that lim f ( t ) = f ( t 0 ) . in other words. that lim f (t ) = L if and only if
t → t0
Then we see
lim x( t ) = a .
t → t0
which is what we set out to show.
t → t0 t →t 0 t → t0
lim( f (t ) × g(t )) = (lim f (t )) × (lim g(t )) . Now suppose f (t ) = x ( t ) i + y (t ) j + z(t ) k and L = ai + bj + c k . lim(α (t ) f ( t )) = aL . y (t ). This shows there is nothing particularly mysterious or exotic about continuity of vector functions.Or. Suppose t 0 is in the domain of the vector function f.
t → t0 t → t0 t → t0
lim y (t ) = b. then we say simply that f is continuous.
8
. How about the function r(t ) = f (t ) ⋅ g( t ) ? 1 Let r(t ) = ti + t 2 j + k .
Is the function h defined by
h( t ) = f ( t ) × g( t ) continuous? Explain. Is r a continuous function? Explain. Is the vector function h defined by h(t ) = m(t ) f (t ) also continuous?
21.
Is it possible for a function f to have more than one limit at t = t 0 ? Prove your answer.
Suppose m is a continuous realvalued function and f is a continuous vectorvalued function.19. function n(t ) = r(t ) is continuous. t Suppose r is a continuous function.
Let f and g be continuous at t = t 0 .
3.
23. Explain how you know that the length
22.
20. Explain.
and c(t ) is.Chapter Four
Derivatives
4. The derivative is defined just as it is for a plain old everyday real valued function. Then
1 a (t + h) − a( t 0 ) b(t 0 + h) − b(t 0 ) c(t + h) − c(t 0 ) [ f (t 0 + h) − f (t 0 )] = 0 i + j + 0 k .1 Derivatives Suppose f is a vector function and t 0 is a point in the interior of the domain of f ( t 0 in the interior of a set S of real numbers means there is an interval centered at t 0 that is a subset of S. we say that f is differentiable at t 0 if there is a vector v such that
1 lim [ f (t 0 + h) − f (t 0 )] = v . Now we “know” what the derivative of a vector function is. Moreover. really? Let’s see. except. but what is it. the derivative is a vector. Specifically. b(t ). of course. Let f (t ) = ti + t 3 j . Let’s look at the
4. of course.).1
. t → t0 h
The vector v is called the derivative of f at t 0 .
Now. the vector derivative v is v = a '( t ) i + b'( t ) j + c'( t ) k . a vector function which describes the graph of the function y = x 3 . and we know how to compute it. This is. h h h h
It should now be clear that the vector function f is differentiable at t 0 if and only if each of the coordinate functions a( t ). how would we find such a thing? Suppose f (t ) = a( t ) i + b (t ) j + c (t ) k .
If f (t ) specifies the position of a particle at time t. Then the derivative. This means we must not allow the velocity to be zero for any t between a and b.) at r(t ) . vectors come to our rescue.
Example Consider the function r(t ) = cos ti + sin tj . is r'( t ) = − sin t i + cos tj . Note that the scalar product r(t ) ⋅ r'( t ) = − sin t cost + sin t cos t = 0 . of course. t 0 ) . the derivative is the velocity of the particle. It should be clear what we mean by “behaves nicely”. This vector is indeed tangent to the curve described by r (which we already know to be a circle of radius 1 centered at the origin.2
. Turner in 4th grade. . the direction of the derivative f '( t ) vector is the tangent direction at the point f (t ) the derivative f '( t ) is said to be tangent to the curve at f (t ) . and so the tangent vector and the vector from the center of the circle to the point on the circle are perpendiculara wellknown fact you learned from Mrs. Note that the derivative is never
4. It is not so clear what
we should define to be the tangent to a curve other than a plane curve. Again. this distance is precisely the length of the arc of the curve from f (a ) to f (b) . Convince yourself that the direction of the vector v is 3 the direction tangent to the graph of y = x 3 at the point (t 0 . there must be no “backtracking”. then. and its length  f '( t ) is the speed. or reversing direction. If f is a vector description of a space curve. or velocity.
a
b
If the particle behaves nicely. For the distance traveled by the particle to be the same as the length of its path.2 derivative of f at t 0 : v = i + 3t 0 j . Thus the distance the particle travels from time t = a to time t = b is given by the integral of the speed:
d = ∫  f '( t ) dt .
0. a)Find the velocity of the particle. 1.3
. a)Find a vector tangent to the curve f (t ) = t 2 i + t 3 j + (1 − t )k at the point (1.4π ) . 6) in the direction of b = i − 3 j + 2k . and let M be the straight line passing through the point (0. The length of a piece of the curve can thus be found by integrating the speed:
p=
2π
∫  r'( t ) dt =
0
2π
∫
0
sin 2 t + cos 2 tdt =
2π
∫ dt = 2π . The position of a particle is given by r(t ) = cos(t 3 ) i + sin(t 3 ) j . c)Describe the path of the particle.
3. Let L be the straight line passing through the point (5. Let L be the line tangent to the curve g (t ) = 10 cos ti + 10 sin t j + 16tk at the point ( 10 10 . 0.zerothere is no value of t for which both cost and sint vanish. .
0
No surprise here. b)Find the speed of the particle. 1.
4. 2 2
4. b)Find a vector equation for the line tangent to this same curve at the point (1. b)Do L and M intersect? Explain. 3) in the direction of the vector a = i + 2 j − k . Find the point at which L intersects the ij plane. 0).
Exercises
1. and find its length. 0).
2. a)Are L and M parallel? Explain.
simply s(t ) = ∫  R '( ξ ) dξ . 1) and (1. 5) in the direction of b = 3i − j + 2k . as we have seen. Find an integral the value of which is the length of the curve y = x 2 between the points (1. Find the length of the arc of the curve R (t ) = 3cos t i + 3 sin t j + 4tk between the points (3. 16π). that R'( t ) ≠ 0 . 0. The speed is ds = R'( t ) .4
. 1. Let L be the straight line passing through the point (1. 1. and let M be the straight line passing through the point (0.
4. 0) and (3. 3) in the direction of the vector a = 2 i + j − k . of course.5.2 Geometry of Space CurvesCurvature Let R(t ) be a vector description of a curve. It is called the unit tangent vector.
7. dt Now then the vector T= R '( t ) R'( t ) dt dR = = R'( t ) =  R'( t ) ds / dt ds ds
is tangent to R and has length one.
t0 t
assuming.
6. 0. Then the distance s(t ) along the curve from the point R(t 0 ) to the point R(t ) is. 1).
Consider next the derivative
4. Find the distance between L and M.
and its direction is sometimes called the principal normal direction. Thus
κ= The unit vector N=
dT . ds dt ds a dt a
4. or orthogonal. which means that the vector is ds ds
perpendicular. dt
T=
1 R' (t ) = − sin ti + cost j . dT dT dt 1 dT 1 = = = ( − cos t i − sin t j ) . or normal. Then R '( t ) = − a sin ti + a cos tj . a
Let’s not stop now. Thus T ⋅
dT dT = 0.5
. to the tangent vector T. ds ds ds ds
But we know that T ⋅ T = T  2 = 1.d dT dT dT T ⋅T = T ⋅ + ⋅ T = 2T ⋅ . The length of this vector is called the curvature and is usually denoted by the letter κ.
Example Consider the circle of radius a and center at the origin: R (t ) = a cost i + a sin t j . ds
1 dT κ ds
is called the principal unit normal vector. and Thus ds =  R'( t ) = a 2 sin 2 t + a 2 cos 2 t = a 2 =  a = a .
viz. Now differentiate:
A(t). and N = − (cos t i + sin t j ) .
Fine.
Another Example This time let R = (t + 1) i + 2t j + t 2 k . The unit tangent is then dt T= 1 5 + 4t 2 ( i + 2 j + 2tk ) .Thus κ =
dT 1 = . so let’s use a computer algebra system.. and so
It’s a bit of a chore now to find the curvature and the principal normal.
We need to tidy this up:
4. Maple: First.6
. So the curvature is the reciprocal of the ds a
radius and the principal normal vector points back toward the center of the circle. ds =  R'( t ) = 5 + 4t 2 . R '( t ) = i + 2 j + 2tk . First. let’s enter the unit tangent vector T:
See if we got it right: T(t).
N(t).
4. 15). Find a line tangent to the curve R (t ) = (t 2 + t ) i + (t + 1) j − ( t 3 + 5) k and passing through the point (5. the unit tangent T. and the dt
Exercises
8..7
.B(t)))). the curvature κ. of course.
This vector is.
So there we have at last the speed principal normal N. We continue and find the curvature κ and the ds
kappa:=t>simplify(sqrt(dotprod(B(t). 2. kappa(t). the normal principal normal N.
ds . or show there is no such line.
dT .B(t).
9. Find the unit tangent T, the principal normal N, and the curvature κ, for the curves: a) R (t ) = 5 cos(t )i + 5 sin(t ) j + 2 tk b) R (t ) = (2t + 3) i + (5 − t 2 ) j c) R (t ) = e t cos t i + e t sin tj + 6k
10. Find the curvature of the curve y = f ( x ) at ( x0 , f ( x0 )) .
11. Find the curvature of R (t ) = ti + t 2 j . At what point on the curve is the curvature the largest? smallest?
12. Find the curvature of R (t ) = ti + t 3 j . At what point on the curve is the curvature the largest? smallest?
4.3 Geometry of Space CurvesTorsion
Let R(t ) be a vector description of a curve. If T is the unit tangent and N is the principal unit normal, the unit vector B = T × N is called the binormal. Note that the binormal is orthogonal to both T and N. Let’s see about its derivative to arclength s. First, note that B ⋅ B =  B2 = 1, and so B ⋅ being orthogonal to B, the derivative dB with respect ds
dB = 0 , which means that ds
dB is in the plane of T and N. Next, note that B is ds dB ⋅ T = 0 . So what have ds
perpendicular to the tangent vector T, and so B ⋅ T = 0 . Thus
4.8
we here? The vector
dB is perpendicular to both B and T, and so must have the ds
direction of N (or, of course,  N). This means
dB = −τ N . ds
The scalar τ is called the torsion.
Example
Let’s find the torsion of the helix R (t ) = a cos ti + a sin tj + btk . Here we go! R '( t ) = − a sin ti + a cos tj + bk . Thus ds =  R'( t ) = a 2 + b 2 , and we have dt
T=
1 a + b2
2
(− a sin t i + a cost j + bk ) .
Now then dT dT dt −a = = 2 (cos ti + sin tj ) . ds dt ds (a + b 2 )
Therefore, κ= a and N = − (cos t i + sin t j ) . (a + b 2 )
2
Let’s don’t stop now:
4.9
B =T ×N =
1 a2 + b2
i − a sin t − cos t
j a cos t − sin t
k b = 0
1 a2 + b2
(b sin ti − b cost j + ak ) ;
and dB dB dt b −b = = 2 (cos ti + sin tj ) = 2 N. 2 ds dt ds ( a + b ) (a + b2 )
The torsion, at last: τ = b . a + b2
2
Suppose the curve R(t ) is such that the torsion is zero for all values of t. In other words, dB ≡ 0 . Look at ds d dR dB [( R( t ) − R (t 0 )) ⋅ B] = ⋅ B + ( R( t ) − R (t 0 )) ⋅ = 0. ds ds ds
Thus the scalar product ( R(t ) − R (t 0 )) ⋅ B is constant. It is 0 at t 0 , and hence it is 0 for all values of t. This means that R (t ) − R( t 0 ) and B are perpendicular for all t, and so R (t ) − R( t 0 ) lies in a plane perpendicular to B. In other words, the curve described by R(t ) is a plane curve.
Exercises
13. Find the binormal and torsion for the curve R (t ) = 4 cos ti + 3sin tk .
14. Find the binormal and torsion for the curve R (t ) =
sin t sin t i + cos t j + k. 2 2
4.10
15. Find the curvature and torsion for R (t ) = ti + t 2 j + t 3 k.
16. Show that the curve R (t ) = ti +
1+ t 1− t 2 j+ k lies in a plane. t t
17. What is the vector B × T ? How about N × T ?
4.4 Motion Suppose t is time and R(t ) is the position vector of a body. Then the curve described by R(t ) is the path, or trajectory, of the body, v( t ) = a(t ) = dR is the velocity, and dt
dv ds is the acceleration. We know that v( t ) = T , and so the direction of the dt dt
velocity is the unit tangent T. Let’s see about the direction of the acceleration:
a( t ) =
dv d 2 s ds dT = 2T+ dt dt dt dt , 2 2 d s ds = 2 T + κN dt dt
since
dT ds = κ N . This tells us that the acceleration is always in the plane of the dt dt d 2s is the tangential component of the dt 2
vectors T and N. The derivative of the speed
2
ds acceleration, and κ is the normal component of the acceleration. dt
Example
4.11
Find its position at time t. A force f (t ) = t 2 i + (t − 1) j + k newtons is applied to an object of mass 2 kilograms. dt dt
ds ds  f  = mκ N  = mκ . 20
Exercises
18.
4. dt dt
2
2
The mass m = thus  f =
160 1 = 5 slugs. The position of an object at time t is given by r(t ) = ti + (t 3 − 2) j + 2t k . The magnitude of the force is 32 20
5 ⋅882 = 1936 pounds. the object is at the origin. in other words. the speed. we know the force f is the mass times the acceleration: f (t ) = ma (t ) . and the tangential and normal components of the acceleration.12
. At time t = 0. or 88 feet/second.Suppose a person who weighs 160 pounds moves around a circle having radius 20 feet at a constant speed of 60 miles/hour. and the curvature κ = . Find the velocity.
19. Thus
f =m
d 2s ds T + mκ N 2 dt dt
2
also have The speed is a constant 60 miles/hour. Hence. ds d 2s = 88 and = 0 . What is the magnitude of the force on this person at any time? First.
and the only force acting on the projectile is f = −wj . bowling ball is rolled along a track with a circular vertical loop of radius a feet.13
. ball be in order for it not to fall?
4.20. A projectile of weight w is fired from the origin with an initial speed v0 in the direction of the vector cosθ i + sin θ j .
21. a)Find a vector description of the trajectory of the projectile. What must the speed of the ball be in order for it not to fall from the track? What must the speed of an 8 lb. b)Find an equation the graph of which is the trajectory. A 16 lb.
R 2 is the plane.K . ax n ) . x 2 .K . x n ) and y = ( y1 . x3 ) and spoke of the triple as being a vector. The ndimensional Euclidean space R n is simply the set of all ordered ntuples of real numbers x = ( x1 . Again we see that these definitions are entirely consistent with what we have done in dimensions 1. x 2 . x2 .Chapter Five
More Dimensions
5. ax2 . Continuing. we define the length. This definition does not contradict our previous definition of a vector in case n =3 in that we identified each vector with an ordered triple ( x1 . y 2 . It is easy to verify that a( x + y) = ax + ay and (a + b) x = ax + bx . These ordered ntuples are called points. and R 3 is Euclidean threespace.1 The Space R n We are now prepared to move on to spaces of dimension greater than three. K.
5. We now define various arithmetic operations on R n in the obvious way. Let n be a positive integer. x n ) .K . x n + yn ) . Thus R 1 is simply the real
numbers. If we have vectors x = ( x1 . 2. These spaces are a straightforward generalization of our Euclidean space of three dimensions.1
. or vectors. y n ) in R n .K. x 2 + y 2 . the sum x + y is defined by x + y = ( x 1 + y1 . and 3there is nothing to unlearn. or norm of a vector x in the obvious manner  x = The scalar product of x and y is
2 2 x12 + x 2 +K+ xn . and multiplication of the vector x by a scalar a is defined by ax = (ax 1 .
This is a quadratic function in a and is never negative.2
. This inequality remains true in higher dimensions. This last inequality is the celebrated CauchySchwarzBuniakowsky inequality. We can. x n ) and y = ( y1 .
5. in fact. we also speak of d( x . It is exactly the ingredient we need to prove the triangle inequality. no longer rely on our vast knowledge of Euclidean geometry in our reasoning about R n when n > 3.x ⋅ y = x1 y1 + x 2 y 2 +K+ x n y n = ∑ x i y i . or  x ⋅ y  ≤ x y . is called the triangle inequality. y n ) . x ⋅ y = y⋅x . but requires an essentially algebraic proof. we have  ax + y2 = (ax + y ) ⋅ (ax + y) ≥ 0 . of course. In fact. Let x = ( x1 . and (ax ) ⋅ y = a ( x ⋅ y ) . thus we speak of the “length” of an ntuple of numbers.
i =1
n
It is again easy to verify the nice properties:  x 2 = x ⋅ x ≥ 0. y ) =  x − y as the distance between x and y . and. x 2 . Thus for n ≤ 3 .K . (ax + y ) ⋅ ( ax + y ) = a 2 x ⋅ x + 2ax ⋅ y + y ⋅ y ≥ 0. The geometric language of the three dimensional setting is retained in higher dimensions. x ⋅ ( y + z) = x ⋅ y + x ⋅ z . the fact that  x + y  ≤  x  +  y for any vectors x and y was a simple consequence of the fact that the sum of the lengths of two sides of a triangle is at least as big as the length of the third side. y 2 . Thus. Then if a is a scalar.K .  ax = a  x. Let’s see if we can prove it. it must therefore be true that 4( x ⋅ y ) 2 − 4( x ⋅ x )( y ⋅ y ) ≤ 0 .
.3
. we have  x + y 2 ≤ x 2 + 2 x  y+  y 2 = ( x+  y ) 2 .0 . = (0. Let x and y be two vectors in R n . e n are defined in R n by e1 e2 e3 M en = (10 .1. e 2 . Prove that  x + y 2 =  x2 + y 2 if and only if
x ⋅ y = 0. Let x and y be two vectors in R n . x + y 2 = ( x + y ) ⋅ ( x + y) = x ⋅ x + 2 x ⋅ y + y ⋅ y .0) 0 = ( 0.1.K .K . or  x + y  ≤  x  +  y . .K .00 .) 2 .00 . e1 . and k. we say that x and y are perpendicular or orthogonal if x ⋅ y = 0. the coordinate vectors
Thus each vector x = ( x1 .K.
i= 1 n
Exercises 1 .0) . Applying the CSB inequality. = (0.0 . x 2 . j .0) . b) x + y 2 + x − y 2 = 2[ x2 +  y 2 ] .0 .K . Corresponding to the coordinate vectors i . (Adopting more geometric language from three space. . x n ) may be written in terms of these coordinate vectors: x = ∑ x i ei . Prove a) x + y 2 −  x − y 2 = 4 x ⋅ y .K .
5.01) .
x n ) ∈ P if and only if the coordinates of x satisfy the equation. F ( x)) resides in R n+ p since x ∈ R n and F ( x) ∈ R p . This F is a linear function. Find vectors n and a such that P = {x ∈ R 4 : n ⋅ ( x − a) = 0} . and x n such that x = ( x1 . and ii) F (ax ) = aF ( x ) for all scalars a and x ∈ R n . Prove that   x −  y   ≤  x + y .3 . x 2 . A function F: R n → R p is said to be a linear function if i) F ( x + y ) = F ( x) + F ( y ) for all x . K. Describe it in case n =2. we have the vector valued functions of the previous chapter. x 2 . y ∈ R n . The set of points ( x. and let P = {x ∈ R n: n ⋅ ( x − a) = 0 . this is difficult to “see” unless n + p ≤ 3 . We begin with a very special kind of functions. x 4 ) such that 3x1 + 5x 2 − 2x 3 + x 4 = 15 . 4 . Let n and a be vectors in R n . Then F ( x + y ) = 3( x + y ) = 3x + 3 y = F ( x ) + F ( y ) and F (ax ) = 3(ax ) = a3 x = aF ( x ) . the socalled linear functions. Note also that except for very special circumstances.K . Example Let n = p = 1. graphs of functions will not play a big role in our understanding. x 3 . b)Describe the set P be in case n = 3.] 5. a)Find an equation in x 1 .2 Functions We now consider functions F: R n → R p . and define F by F ( x ) = 3x .
5. we have the usual grammar school calculus functions. x 2 . Note that when n = p = 1. 5 . Let P ⊂ R 4 be the set of all vectors x = ( x1 .4
. Let x and y be two vectors in R n . and when n = 1 and p = 2 or 3. [The set P is called a hyperplane through a.
0. F (at ) = ati + 2 atj − 7at k = a[ ti + 2t j − 7t k ] = aF (t ) We see yet another linear function. x 2 .Another Example Let F: R → R 3 be defined by F (t ) = t i + 2tj − 7tk = (t . where L is a linear function. x 2 . x 3 )) = (2 x1 − x 2 + 3 x3 .−7t ) . It is easy to verify that F is indeed a linear function. 4t − 3. x 3 ) = x 1 x 2 + x3 f) h( x1 . x 1 + x 3 ) . 4t . 0) b) g ( x ) = 2 x − 5 e) F (t ) = (2t . Clearly F (t ) = a + L(t ) . 0. 3 x1 . a) f ( x) = −7 x d) G( x1 . x 4 ) = (1. x1 )
5. Which of the following functions are linear? Explain your answers. 5 x1 − 2 x 2 .4 . t ) . t . An affine function F thus has the form F ( x) = a + L( x) .2t . Then F is affine. Exercises 6 . t ) . One More Example Let F: R 3 → R 4 be defined by F (( x1 . Let a = (2 . A function that is the composition of a linear function followed by a translation is called an affine function.5
. Then F (t + s) = (t + s) i + 2( t + s) j − 7 (t + s)k = [ti + 2tj − 7tk ] + [ si + 2sj − 7 sk ] = F (t ) + F ( s ) Also. where a is a fixed vector in R n . x 2 ) = (2 x 1 + x2 . x1 + 4 x 2 − 5x 3 . − x 1 + 2 x2 + x 3 . x 3 .0) and L(t ) = (t . x 1 − x 2 . x 2 . Example Let F: R → R 3 be defined by F (t ) = (2 + t . − 2t ) g) f ( x) = sin x c) F ( x1 . A translation is a function T:R p → R p such that T( x ) = a + x .
5. a)Describe the graph of a linear function from R to R.6
. b)Describe the graph of an affine function from R to R.7 .
a 22 .Chapter Six
Linear Functions and Matrices
6. we arrange them in a rectangular array. Specifically. a p 1 x 1 + a p 2 x 2 +K+ a pn x n ). Let e1 .K . f ( e 2 ) = (a12 . a pn ). a 21 x1 + a 22 x 2 +K+ a 2 n x n . a p1 ). it says that a linear function is entirely determined by its values f ( e1 ). a 2 n . we have x = x 1e 1 + x 2 e 2 +K+ x n e n . e 2 . suppose f ( e1 ) = (a11 . a p 2 ).
The numbers aij thus tell us everything about the linear function f.K . . K. Thus f ( x) = f ( x1 e 1 + x 2 e2 +K+ x n e n ) = x1 f ( e1 ) + x 2 f ( e2 )+K+ xn f (e n ) .
Then f ( x) = ( a11 x1 + a12 x 2 +K+ a1n x n . To avoid labeling these numbers.K .K . M f ( e n ) = (a1 n . f (e n ) . called a matrix:
6. f (e 2 ).1 Matrices Suppose f : R n → R p be a linear function.K . e n be the coordinate vectors for R n .1
. For any x ∈ R n .
Meditate on this. a21 .
x 2 ) = (2 x1 − x 2 . This calculation is systematized by introducing an arithmetic of matrices. and f ( e 2 ) = f (01) = ( −15.
Then f ( e1 ) = f (10 ) = ( 2. is thus 2 −1 1 5 3 − 2
Given the matrix of a linear function. The matrix representing f .3) .− 2) . x1 + 5x 2 . a11 a12 K a1 n a 21 a 22 K a2 n M M a p 1 a p 2 K a pn
The matrix is said to represent the linear function f. . we can use the matrix to compute f ( x) for any x. For the matrix
a11 a 21 A= M a p1
a12 a22 a p2
K a1n K a 2n . For example. 3x 1 − 2 x2 ) . . suppose f : R 2 → R 3 is given by the receipt
f ( x1 .1. we need some jargon. First.2
. M K a pn
6.
and f ( e 3 ) = f ( 0. Thus A has p rows and n columns. ain ]
a1 j a 2j are called rows of A.
Example Suppose f : R 3 → R 2 is defined by
f ( x 1 .
So f ( e1 ) = f (10 . called a column vector.− 5) .K .0) = (2.1.0 .the matrices [ai 1 . f (e n ) . A vector in R n can be displayed as a matrix in the obvious way. x 3 ) = (2 x1 − 3 x2 + x 3 .3
. either as a 1× n matrix. in which case the matrix is called a row vector. Suppose A is a p × n matrix and x is a n × 1 column
6. or as a n × 1 matrix.K .− 1) . . − x 1 + 2 x2 − 5x 3 ) . and the matrices are called M a pj
columns of A.1) = (1. f (e1 ). Thus the matrix representation of f is simply the matrix whose columns are the column vectors f ( e1 ). .0 ) = ( −32 ) . a i2 . The matrix which represents f is thus 2 −3 1 −1 2 − 5
Now the recipe for computing f(x) can be systematized by defining the product of a matrix A and a column vector x. x 2 . f ( e 2 ) = f (0. the size of A is said to be p × n .
x 3 ) = (2 x1 − 3 x2 + x 3 . −1 2 −5 Then x1 2 − 3 1 2x 1 − 3x 2 + x 3 Ax = x2 = = f ( x) − 1 2 − 5 x − x1 + 2 x 2 − 5x 3 3
Exercises
6. where. then a linear function f : R n → R p is given by f ( x) = Ax . We define the product Ax . A is the matrix representation of f. We found the matrix representing f to be 2 −3 1 A= . For each i = 12 . K. p. to be the p × 1column vector given by r1 ⋅ x r ⋅ x 2 .4
.
Example Consider the preceding example: f ( x 1 . − x 1 + 2 x2 − 5x 3 ) . x 2 . of course. let ri denote the i th row of A .vector. Ax = M rp ⋅ x
If we consider all vectors to be represented by column vectors.
Find the matrix representation for f. b) R (t ) = 4t i − 5tj − 2t k . Let g be define by g( x ) = Ax . c) L( x ) = 6x . d)Find f(4.1) = (2. −3 5
3.
4. Let f : R 2 → R 2 be the function in which f(x) is the vector that results from rotating the vector x about the origin π in the counterclockwise direction. where A = 0 3
−1 1 . Let f : R 2 → R 2 be the function in which f(x) is the vector that results from rotating the vector x about the origin θ in the counterclockwise direction.−9 ) .
5. x1 + 4 x 2 .1. Find g(3.
2 −2 2. x 2 ) = (2 x1 − x 2 . 7x 1 . Find the matrix representation of each of the following linear functions: a) f ( x1 .2).2) = (4.5
. 3x1 + 5x 2 ) . Suppose g: R 2 → R 2 is a linear function such that g(1.9).7) and g(2.
6. b)Find the matrix representation for f. 4
a)Explain why f is a linear function.
n .K.K. where.K .K . j = 1. Suppose f : R n → R p and g: R n → R p are linear functions.
6.6
.
Prove that the
7. k 2 . C = [ Bk1 . We know the columns of C are the vectors g ( f (e j )).2. . so . Prove that the function f + g: R n → R p defined by ( f + g)( x ) = f ( x ) + g ( x) is a linear function. j = 12 . then the columns of C are Bk1 .K. n . Now the columns of A are just the vectors f ( e j ). composition g o f : R n → R q is a linear function. of course. Suppose f : R n → R p and g: R p → R q are linear functions.Find the matrix representation of g.2 Matrix Algebra Let us consider the composition h = g o f of two linear functions f : R n → R p and g: R p → R q . k n ] . the vectors e j are the coordinate vectors for R n . Bk2 . that A = [k1 . In other words. Thus the vectors g ( f (e j )) are simply the products Bf (e j ) .K. if we denote the columns of A by ki . i = 12 . Bkn ] . or in other words. Let’s see about the matrix C of h. Bkn . n .
6. Bk2 . Suppose A is the matrix of f and B is the matrix of g.
Example
6.
−3
Thus
f : R 2 → R 3 and g: R 3 → R 4 (Note that for the
composition h = g o f to be defined.2 1 0 −1 −5 8 and let the matrix A of f be Let the matrix B of g be given by B = 2 7 −3 2 −2 1 3 given by A = 1 −4 1 2. The matrix C of the composition is thus 25 25 −3 0 −5 −5 −40 −35 C= . if B is an n × p matrix. it must be true that the number of columns of B be 3 1 1 and k = 2 .). Now we can simply say that the matrix representation of the composition of two linear functions is the product of the matrices representing the two functions. and A is a p × q matrix.7
. Now. the product BA of these matrices is defined to be the
n × q matrix whose columns are the column vectors Bk j . where k j is the j th column of
A. Thus. 25 25 0 −3
These results inspire us to define a product of matrices. k1 = 2 − 4 −3 −5 −5 −40 −35 Bk1 = and Bk2 = . and so the same as the number of rows of A.
6.
note that it follows directly from the fact that h o ( f o g ) = (h o f ) o g that for C(BA) = (CB)A. we usually omit them entirely. K a pn + b pn
where a11 a 21 A= M a p1 a12 a22 a p2 L a1n L a 2n L a pn
is the matrix of f. and
6. Next.There are several interesting and important things to note regarding matrix products.8
.). then the matrix representation for the sum f + g: R n → R p is simply the matrix a11 + b11 a + b 21 21 A+ B = M a p 1 + bp 1 a12 + b12 a22 + b22 a p2 + b p2 K a1 n + b1 n K a 2 n + b2 n . It should be clear that if f and g are both functions from R n to R p . First and foremost is the fact that in general BA ≠ AB . even when both products are defined (The product BA obviously defined only when the number of columns of B is the same as the number of rows of A. Since it does not matter where we insert the parentheses in a product of three or more matrices.
9
. we have ( A + B)C = AC + BC . B.
Exercises 8.
Similarly. Find the products: 2 1 −2 a) 0 3 1 2 1 1 b) 0 3 3 1 5 −2 3 d) [ −3 2 − 1] 1 0 2 −3 4
2 1 −2 1 c) 0 3 1 3
1 0 0 a 11 9. Find a) 0 1 0 a 21 0 0 1 a 31
a12 a 22 a 32
a13 a 23 a 33
0 0 0 a 11 b) 0 0 0 a 21 0 0 0 a 31
a12 a 22 a 32
a13 a 23 a 33
6. it is true that
A( B + C ) = AB + AC . b11 b12 L b1n b 21 b22 L b2 n B= M b p 1 b p 2 L bpn
is the matrix of g. for appropriately sized matrices. and C. Meditating on the properties of linear functions should convince you that for any three matrices (of the appropriate sizes) A.
Let C be the circle of radius 1 and center at the origin. and use the result to give identities for cos(θ + η ) and sin(θ + η ) in terms of cosθ . sinθ . 4
b)Find a vector description for the curve that results from applying the linear transformation in a) to the curve R (t ) = cos ti + sin tj + t k .). Find a vector description for the curve f(C). Let A(θ ) be the 2 × 2 matrix for the linear function that rotates the plane θ counterclockwise.
11. Suppose g: R 2 → R n is linear.
6. Compute the product A(θ ) A(η ) .3) .
12. and sinη .
13. Suppose f : R 2 → R 2 is linear. a)Find the matrix for the linear function that rotates R 3 about the coordinate vector j by π (In the positive direction.− 5) . according to the usual “right hand rule” for rotation. .10.
and
g( −11) = ( 4.10
. cosη . Find the matrix of g.
Suppose moreover that
g(11) = (2.
x →a x →a
Now we are ready to say what we mean by a continuous function f : D → R p .
x →a
and y is called the limit of f at a. r ) = {x ∈ R n : x − a ≤ r} . The usual properties of limits are relatively easy to establish:
lim( f ( x ) + g ( x)) = lim f ( x ) + lim g ( x) .1 Limits and Continuity Let x 0 ∈ R n and r > 0. r) ⊂ D . The closed ball of radius r centered at x 0 is the set B (a . then we say that y is the limit of f at a. A point a ∈ D is called an
interior point of D if there is an open ball B (a. Suppose f : D → R p . The collection of all interior points of D is called the interior of D.2.1
. The set B (a . where D ⊂ R n . or 3. If y ∈ R p is such that for every ε > 0. where D ⊂ R n and suppose a ∈ R n is such that every open ball centered at a meets the domain D. and is usually denoted int D. and All That
7. Again this definition will not contradict our previous lower dimensional
7. This is written
lim f ( x) = y . Notice that this agrees with our previous definitions in case n = 1 and p =1. Derivatives. and
x →a x →a x →a
lim af ( x ) = a lim f ( x ) . there is a δ > 0 so that f ( x ) − y < ε whenever 0 <  x − a  < δ . r ) = {x ∈ R n : x − a < r} is called the open ball of radius r centered at x 0 . A set U is said to be open if U = int U.Chapter Seven
Continuity. Now suppose D ⊂ R n .
suppose f : R n → R p is linear and a ∈ R n .2
.α ) = 2 = .definitions. Then for all α ≠ 0. f ( e2 ). we say simply that f is continuous. To see this. Let x = α (11) . f (e n )} and let δ = Then for x such that 0 <  x − a  < δ . we say that f is continuous at a ∈ D if lim f ( x) = f ( a) . 2 α +α 2
7. nM
 f ( x ) − f (a ) =  f ( x1 e1 + x 2 e 2 +K+ xn en ) − f (a1 e 1 + a2 e 2 +K+ a n e n ) =( x 1 − a1 ) f ( e1 ) + ( x2 − a2 ) f (e 2 ) +K+ ( x n − an ) f (e n ) ≤  x1 − a1  f (e1 )+  x 2 − a2  f ( e 2 )+K+  xn − an  f (e n ) ≤ ( x1 − a1 +  x2 − a 2 + K+  x n − an ) M ≤ n x − a  M <ε
Thus lim f ( x) = f ( a) and so f is continuous. Now let M = max{ f (e 1 ). for x1 + x 2 ≠ 0 2 2 f ( x ) = f ( x1 . x2 ) = x1 + x2 .K. we have .
x →( 0 .
Example Every linear function is continuous. otherwise
Let’s see about lim f ( x). Specifically.0 )
α2 1 f ( x) = f (α . If f is
x →a
continuous at each point of its domain D. Let ε > 0. we have ε .
x →a
Another Example Let f : R → R be defined by
2
x1 x 2 2 2 . 0.
x j + t . Then f is differentiable at x 0 if there is a linear function L such that
lim
h→0
1 [ f ( x0 + h) − f ( x 0 ) − L( h)] = 0 . let x = α (10) = (α .2 Derivatives Let f : D → R p . xn ) f ( x . f ( x) = exist.K. 1 and such that f ( x) = 0 .Now.K . xn )
and
7. f 1 ( x1 . there are vectors x with 0 <  x − ( 00 ) < δ such that .0. x j + t . x ) 2 1 2 j n f ( x + h) = f ( x1 . First.K . and let x 0 ∈int D . x + t .K. x j + t .0) . It is usual to identify the linear function L with its matrix representation and think of the derivative at a p × n matrix.K .K . t .K. x . Then . x 2 . x 2 .0) . how do find the derivative of f ? Suppose f has a derivative at x 0 .3
. M f p ( x1 . x 2 .K . What does this tell . the matrix L is simply the 1 × 1 matrix whose sole entry is the every day grammar school derivative of f . x n ) = . It follows that all α ≠ 0.K. us? It tells us that for any δ > 0 . let h = te j = ( 00 . means that 2
x →( 0 . Note that in case n = p = 1. Now. f ( x) = 0 .0. This. where D ⊂ R n .K.  h
The linear function L is called the derivative of f at x 0 .0 )
lim f ( x) does not
7. of course.
xn ) s = xj ds i 1 2
This derivative has a name.x 2. xj + t. xn ) t = d f (x . x 2. x2 . It is called the partial derivative of f i with respect to the j th variable. x . xn ) − m pj t
Meditate on this vector.x n ) − m1 j t f2(x1 .K. x2 . x 2 . xn ) . x2. x j + t. x .K.
1 [ f (x 0 + h) − f (x0 ) − L(h)]  h f1(x1.K.x j + t. M mpj t
where x 0 = ( x 1 . xn ) − m2 j = t M f p (x1 . x ) − m t j n p 1 2 n pj p 1 2 f1(x1.. xn ) − fi (x1. x2 .x2 . x n ) − m1 j t 1 f2 (x1 .K. etc.K.K.K. lim
t→0
fi (x1 .K. x . xn ) − m2 j t = M t f (x .K. xn ) − f p (x1 .K. For each component.x ) − f (x . x n ) .K.4
.x 2.K. x j + t. x + t.. Now then. x 2 . xn ) − f2 (x1 .K.x 2.x2 .K.K . xn ) − f1(x1.K. The two most common are:
7.K.K.x2 .K .K.x n ) − f1(x1.K.K..K.K. There are many different notations for the partial derivatives of a function g ( x1 . x j + t. x 2 . x j + t.K. s. xn ) − f2 (x1 . 0 m11 m12 L m1n 0 m M 21 m22 L m2 n = Lh = M t mp 1 m p 2 L m pn M 0
m1 j t m t 2j .
x 2 . x 2 ) = 3x1 sin x 2 . the matrix of the derivative. xn ) ∂x j
The requirement that lim
h→0
1 [ f ( x0 + h) − f ( x 0 ) − L( h)] = 0 now translates into  h
mij =
∂f i . Now m22 f 1 ( x1 . be 2 × 2 : L = 11 m21 m12 . x2 ) = x 1 + x1 x2
Compute the partial derivatives:
∂f 1 = 3 sin x 2 ∂x 1 . ∂x j
and. we have found the matrix L !
Example 3x sin x 2 Let f : R 2 → R 2 be given by f ( x1 . m of course. mirabile dictu. Assume f is differentiable 2 x1 + x 1 x 2 and let’s find the derivative (more precisely. j ( x 1 .K . This matrix will.K . x2 ) = 3 1 . ∂f 2 2 2 = 3x 1 + x 2 ∂x 1
7. and
3 2 f 2 ( x1 . x n ) ∂ g ( x1 . x 2 .g .5
.
−2 7 0
3.
Exercises
1. x 3 . x 2 . but how do we know when there is a derivative? The function f is differentiable at x if the partial derivatives exist and are continuous. Find all partial derivatives of the given functions: a) f ( x.6
. What is the derivative a linear function whose matrix is A ?
7.and ∂f 1 = 3x 1 cos x 2 ∂x 2 . x 2 . z) = x 2 yz + z cos( xy) x 3 sin( e x1 ) d) h( x1 . y) = x 2 y 3 c) g ( x1 . 2 x 1x 2
We now know how to find the derivative of f at x if we know the derivative exists. ∂f 2 = 2 x1 x 2 ∂x 2
The derivative is thus
3 sin x L = 2 22 3x1 + x 2
3x 1 cos x 2 . x 3 ) = x1 x 2 x 3 + x 2 b) f ( x. x 4 ) = x2 + x 4
1 3 2 2. y. It should be noted that it is not sufficient just for the partial derivatives to exist. Find the derivative of the linear function whose matrix is .
x3 ) = .
5. it is also continuous there. Find the derivative of R (t ) = cos ti + sin tj + t k .
x →a h→0
f (a + h) − f (a ) − L (h) f ( a + h) − f (a ) =  h − L (h)  h
Now look at the limit of this as  h → 0 :
f ( a + h) − f (a ) − L (h) lim =0 h→0  h
7. Let h = x − a . Find the derivative of
x1 x3 + e x 2 2 x3 log(x1 + x 2 ) f (x1 . Find the derivative of f ( x.
6. Now. The same is true here in the more general setting of functions f : R n → R p .4. x2. x2 2 x1 x 3 + 5
7.7
. Let’s see why this is so. Then lim f ( x) = lim f ( a + h) . Suppose f is differentiable at a with derivative L. y) = x 2 y 3 .3 The Chain Rule Recall from elementary one dimensional calculus that if a function is differentiable at a point.
r(a + h) − r(a ) − ML (h) g( f ( a) + k ) − g( f ( a)) − M( k ) k − L (h ) = + M( )  h  h  h
Now we are ready to see what happens as  h → 0 . Let r = g o f . and hence continuous. Let f : R n → R p and g: R p → R q . let’s see what the celebrated chain rule looks like in higher dimensions. and look at r(a + h) − r(a ) = g( f (a + h)) − g( f ( a)) .8
. and lim L( h) = L (0) = 0 because the linear function L is
h→0
continuous. Thus lim( f (a + h) − f (a )) = 0 . h→0  h
7.because f is differentiable at a. look at the second term first:
lim M(
h→0
f ( a + h) − f (a ) − L (h) f (a + h) − f (a ) − L ( h) k − L (h ) ) = lim M = M (lim ) h→0 h→0  h  h  h = M (0) = 0
since L is the derivative of f at a and M is linear. Suppose the derivative of f at a is L and the derivative of g at f ( a) is M. Next. or lim f (a + h) = f ( a) . which means f is
h→0 h→0
continuous at a. Next. Then we may write
r(a + h) − r(a ) − ML (h) = g ( f (a + h)) − g ( f (a )) − ML (h) = g ( f (a ) + k ) − g ( f (a )) − M ( k ) + M ( k) − ML( h) . let k = f (a + h) − f (a ) . = g ( f (a ) + k ) − g ( f (a )) − M ( k ) + M ( k − L( h))
Thus. We go on a quest for the derivative of the composition g o f : R n → R q at a . Now we need to see what happens to the term
g ( f (a) + k ) − g( f ( a)) − M ( k ) lim .
1 + t 3 ) and g ( x1 . or matrix product. we know that
 k  f ( a + h) − f (a) =  h  h
behaves nicely as  h → 0 . and let r = g o f . What could be more pleasing from an esthetic point of view!
Example Let f ( t ) = (t 2 . h→0  h
which means the derivative of the composition r = g o f is simply the composition. we have  h
r (a + h) − r (a) − ML( h) lim = 0. Finally at last. x 2 ) = ( 2x 1 − x2 ) 3 .
g ( f (a ) + k ) − g ( f (a )) − M ( k )  k  lim ⋅ h→0  h  k  g ( f (a ) + k ) − g ( f (a )) − M ( k )  k = lim =0 h→0  k  h
since the derivative of g at f ( a) is M. Note first that because f is differentiable at a . The derivative of f is
7. we shall find the derivative of r at t = 2 using the Chain Rule. Next. First.9
. and shown that
 k is wellbehaved. of the derivatives.This is a bit tricky.
x 2 ) = ( x1 x2 sin x 1 . e ( x− y ) ) and g ( x . let’s find an explicit recipe for r and differentiate: r(t ) = g( f ( t )) = g( t 2 . L = .
8.4). and at g ( f (2 )) = g (4 . It is.
7. Let f ( x. x 2 − 2 x 1 ) . Thus the derivative of the 12 4 composition is ML = [6 − 3] = [ − 12] = − 12 . very comforting to get the same answer as before. s.
Exercises
2 7. x 3 ) = ( x1 x3 . z) = ( x + y 2 .1 + t 3 ) = (2t 2 − 1 − t 3 ) 3 . y . x 2 x 3 + 1) and f ( x1 . t . b)Find the derivative of your answer to part a).
Find the derivative of g o f at (2. M = [6 −3]. x 2 .
and so r'( 2) = 3(1)(8 − 12 ) = − 12. of the composition functions u o v or v o u is defined? Explain. There are several different notations for the matrix of the derivative of f : R n → R p at x ∈ R n The most usual is simply f '( x ) . Let g ( x1 . Let u( x . x sin y .( r − t )e s ) .
]
4 At t = 2 . x 3 y 2 ) and v( r.9 ) .
9. y ) = ( x − y 3 . q ) = (r + s − q 3 .2 xy. a)Which. Thus r'( t ) = 3( 2t 2 − 1 − t 3 ) 2 (4 t − 3t 2 ) . of course. 2t L = 2. if either. 12 Now for fun. 3t
and the derivative of g is M = 6( 2 x1 − x2 ) 2
[
−3(2 x1 − x 2 ) 2 . x 2 + y ) .10
. x 1 + 3x 2 . y) = (e ( x + y ) .
we have r = g o f : R n → R q . Suppose r = t 2 cos t and t = x 2 − 3y 2 . ∂ r1 ∂ x 1 ∂ r2 r'( x) = r '( x1 . f : Rn → R p . Find the partial derivatives
7. as we now know. x n ) = ∂ x1 M ∂ r p ∂ x1 ∂ r1 ∂ x2 ∂ r2 ∂ x2 ∂ rp ∂ x2 ∂ r1 ∂ xn ∂ r2 L ∂ xn . The the derivative is. ∂r ∂r and . Suppose. ∂x ∂y
10. d) Find the derivative of g o g at the point (1. b)Find the derivative of g o f at the point (1.a)Find the derivative of f o g at the point (1. we seldom compute the derivative of the composition of two functions by explicitly multiplying the two derivative matrices. c) Find the derivative of f o f at the point (1.2). x 2 . x n ) . Let r = go f .
7. x 2 .11
. ∂ rp L ∂ xn L
We can thus find the derivative using the Chain Rule only in the very special case in which the compsite function is real valued. Specifically.2).4 More Chain Rule Stuff In the everyday cruel world. Let’s use the Chain Rule to find the partial derivatives.2).2). as usual.K .K . suppose g: R p → R and
Then r is simply a realvalued function of
x = ( x1 .
xn ) . and z = s 2 + 3t 2 . Since y j = f j ( x1 . ∂ x j ∂ y1 ∂ x j ∂ y 2 ∂ x j ∂ yp ∂ xj
Frequently. z) = x 2 y + yez and x = s + t . engineers and other malefactors do not use a different name for the composition g o f . K. x 2 .K .K. The Chain Rule given above then looks even nicer:
∂g ∂ g ∂ y1 ∂ g ∂ y 2 ∂ g ∂ yp = + +L+ .K. y 2 . Let us find the partial derivatives ∂g ∂g and . y . x 2 . f p ( x1 .∂r r'( x) = ∂ x1
∂r ∂ x2
L
∂r ∂ g = ∂ xn ∂ y1
∂g ∂ y2
∂ f1 ∂ x 1 ∂f ∂ g 2 L ∂ x ∂ y p M 1 ∂ f p ∂ x1
∂ f1 ∂ x2 ∂ f2 ∂ x2 ∂ fp ∂ x2
∂ f1 ∂ xn ∂ f2 L ∂ xn ∂ fp L ∂ xn L
Thus makes it clear that
∂r ∂ g ∂ f1 ∂ g ∂ f 2 ∂ g ∂ fp = + + L+ . xn ). xn ).K . y = st 3 .K . and simply use the name g to denote both the composition g o f ( x 1 . x 2 . x n ) = g ( f 1 ( x1 .12
. x 2 .K . x 2 . ∂ x j ∂ y1 ∂ x j ∂ y 2 ∂ x j ∂ yp ∂ xj
Example Suppose g ( x . these same folks also frequently just use y j to denote the function f j . We know that ∂r ∂t
7. x n )) and the function g given by g ( y ) = g( y1 . f 2 ( x1 . y p ) .
for instance. y . and x = t + z . and z = t 3 . y . y . y = t 2 + 2 z . ∂z
Exercises
11. Suppose g ( x . Suppose. however. ∂x ∂y
12. y − s) . be aware that it is a practice sometimes fraught with peril.13
. Suppose the temperature T at the point ( x . Now it is not at all clear what is meant by the symbol ∂g .
7. you should. Find the derivative with respect to t of a particle moving along the curve described by r(t ) = cos ti + sin tj + 3tk . z) in space is given by the function T( x .∂ g ∂ g ∂ x ∂ g ∂ y ∂ g ∂z = + + ∂ s ∂x ∂ s ∂y ∂ s ∂z ∂ s = 2xy(1) +(x 2 + ez )t3 + ye z (2s) = 2xy +(x 2 + e z )t 3 + 2syez
Similarly. z) = x 2 + xyz − zy 2 . z) = x 2 + y 2 + z 2 . Find
∂g ∂g + . Meditate on this. ∂ g ∂g ∂ x ∂ g ∂y ∂ g ∂ z = + + ∂ t ∂x ∂ t ∂y ∂ t ∂z ∂ t = 2xy(1) + (x 2 + ez )3st 2 + yez (6t) = 2xy + 3(x 2 + ez )st 2 + 6tyez
These notational shortcuts are fine and everyone uses them. y ) = f ( x − y . you have g ( x .
z) = x 2 + y 2 + z 2 . y . y . y ) = f ( x ) g ( y ) . A particle moves along the curve described by
r(t ) = sin π ti + cos πt j + (t 2 − 2t + 2) k . z) in space is given by the function T( x . Find the coldest point on the trajectory. Suppose the temperature T at the point ( x . Let r( x . dt
7.13.
14. and suppose x = t and y = t .14
. Use the Chain Rule to find dr .
Then the directional derivative of f at a in the direction of the vector u is defined to be
Du f ( a) =
d f (a + tu) t =0 . valued.1
. x 3 ) ∈ R 3 : f ( x1 . In the very. or scalar. then S is what we call a surface. y . but important. and let u ∈ R n be a vector such that  u = 1. Suppose. Let us now return to the general case of a function f : R n → R .Chapter Eight
f : Rn → R
8. z) in space. Specifically. for instance. Let f : R n → R .2 The Directional Derivative In the applications of scalar fields it is of interest to talk of the rate of change of the function in a specified direction. The derivative of f is just ∂f a row vector f '( x) = ∂ x 1 and denoted grad f or ∇f . z) gives the temperature at points ( x . The
collection S = {( x 1 . suppose f : R n → R . It is frequently called the gradient of f ∂ xn
8. and we might want to know the rate at which the temperature changes as we move in a specified direction. y . dt
8.1 Introduction We shall now turn our attention to the very important special case of functions that are real.
If f is a
reasonably nice function. x 2 . special subcase in which the dimension of the domain space is 2. x 2 ) = x 3 } is called the graph of f. the function T( x . ∂f ∂ x2 L ∂f . These are sometimes called scalar fields. We shall see more of this later. let a ∈ R n . we can actually look at the graph of a function.
z) = x 2 y + yz 3 . we know at once how to compute such a thing. At 2 this becomes ∇f ( 2.1). then the positive xaxis points East. This gives us the slope of the path. but we need a unit vector u in the direction we are walking. it is positive so we are going 2 2
uphill.Now that we are experts on the Chain Rule.− 10 y] . dt
Example The surface of a mountain is the graph of f ( x. just u = the point a 1 (1. The answers to these questions call for the directional derivative. the height is f (x. of course. y). where θ is the angle between ∇T and u.4) = [ −2. This is. and at last we have
∇f ⋅ u =
−2 + 40 38 = . y) = 700 − x 2 − 5 y 2 . It is simply
D u f (a ) =
d f ( a + t u) t =0 = ∇f ⋅ u . In other words. What is the slope of the path at the starting point? Are you going uphill or downhill? (Which!?). at the point (x.− 1) . Thus T in creases most rapidly in
8. of course. We know we are at the point a = (2 . and. Can you tell in which direction the path will be level?
Another Example The temperature in space is given by T( x . Anyone can see that this will be largest when θ = 0. y). The
directional derivative is simply ∇T ⋅ u =∇ Tcosθ .1. 4) and begin to walk Southeast. You are on the mountain side above the point (2. y ) = [ −2 x .− 40] . The positive yaxis points North. y .4 ) .2
. From the point (1. Next we compute the gradient ∇f ( x . in which direction does the temperature increase most rapidly? We clearly need the direction in which the directional derivative is largest.
1). y. this becomes [2. x 2 + z 3 . The position of a particle at time t is given by r(t ) = 3( t 2 − sin t ) i + t j − cos tk . Find the derivative of f ( x. You stand on the hill above the point (5. 2 .3 Surface Normals
8.the direction of the gradient of T.2] . At time t = π.2 . and the position of another particle is R (t ) = t 2 i + (t 3 + t ) j + sin t k . Find the derivative of f ( x. 3]. Here that direction is [2 xy. 1) in the direction of the vector [1. . At (1.3 yz 2 ] .3) and pour out a glass of water. The surface of a hill is the graph of the equation z = 1000 + x 2 − x 4 − y 2 . 2. y ) = x 2 y + e xy sin y increases and decreases most rapidly from the point (1.
2. 2.
3.3
. Find the directions in which g ( x . z) = x log z + 2 xy at (1. what is the rate of change of the distance between the two particles? Are they getting closer to one another.1.2] .
Exercises
1. 1) in the direction of the vector [3. or are they getting farther apart? (Which!) Explain. y. 0). z) = x cos y + 3z 3 − xz at (1.
4. π. In which direct will it begin to run? Explain.
8.
5.
2).
8. y . In the case at hand. or level surface. where x = ( x . it is any curve on the surface. y .4 z] . z) . y ( t ). z) = c} is called a level set.
∇f ( x . 1. that f ( r( t )) = f ( x (t ). z( t )) = c . y . For an equation of a plane. z) ∈ R 3 : f ( x . we need a point a on the plane and a vector N normal to the plane. y. Let’s find a normal vector N. dt
In other words. y. z) = [2 x . The gradient ∇f is thus perpendicular. the gradient of f and the tangent to the curve are perpendicular. 2). z) = x 2 + 3 y 2 + 2 z2 does the job.6 y . we have a point on the plane: a = (1. or normal to the surface f ( x. Recall that the set S = {( x .Let f : R 3 → R be a function and let c be some constant. of the function f . We have just learned that the gradient of
f ( x.4
. y . This means.
Example Suppose we want to find an equation of the plane tangent to the surface
x 2 + 3y 2 + 2z 2 = 12
at the point (1. Then the equation we seek is simply N ⋅ ( x − a) = 0 . Now look at the derivative with respect to t of this equation:
d f ( r(t )) = ∇f ⋅ r'( t ) = 0 . of course. Suppose r(t ) = x (t )i + y( t ) j + z(t )k describes a curve in R 3 that lies on the surface S. 1. Note there was nothing special about our choice of r(t). z) = c .
− 12 ) = [2 . Find an equation for the plane tangent to the surface z = log( x + y ) at the point
2 2
(10 . the steepest path is in the direction normal to the contour lines. Combining these results with what we know about the directional derivative.
Exercises
6.0) .
8.3). for example. and the gradient of f is normal to such a curve. Find an equation for the plane tangent to the surface cosπ x − x 2 y + e xz + yz = 4 at the point (0.5
.and so N = ∇ f (1. we see that at a point the value of a function increases most rapidly in a direction normal to the level set passing through that point. On a contour map of a portion of the Earth’s surface.1. 1. The tangent plane is thus given by the equation . . y ) = c} is a level curve of f.1. y ) ∈ R 2 : f ( x . 3).−68 ] . N ⋅ ( x − a) = 0 . Thus if f : R 2 → R .
7.
You should note that the discussion here didn’t depend on the dimension of the domain. .
9.
8.2). Find an equation of the straight line tangent to the curve of intersection of the surfaces x 3 + 3x 2 y 2 + y 3 + 4 xy − z 2 = 0 and x 2 + y 2 + z 2 = 11 at the point (1. which in this case is
2( x − 1) − 6( y + 1) + 8( z − 2) = 0 . Find an equation for the plane tangent to the surface z = x 2 + 2 y 2 at the point (1. then the set {( x .
or gradient. A point a in the domain of f is called a local minimum if there is an open ball B (a . If f is a nice function. then this means the directional derivative Du f ( a) ≥ 0 for all unit vectors u. then the derivative ∇f (a ) = 0. Then it must be true that both ∇f (a ) ⋅ u ≥ 0 and −∇f (a) ⋅ u = ∇ f ( a) ⋅ ( u) ≥ 0 .4 Maxima and Minima Let f : R n → R . So let’s begin by finding all points at which ∇f (a ) = 0 .
Example Let us find all local maxima and local minima of the function
f ( x . You should be aware of the fact that here. Thus f has − a local minimum at a point at which it has a derivative only if the derivative is zero there. In other words.6
. ∇f (a ) ⋅ u ≥ 0 . just as in Mrs. Turner’s elementary calculus class. and if f has a derivative at a. y) at which ∂f ∂f = 0 and = 0: ∂x ∂y
8. the converse is not necessarily true. This can be so for every u only of ∇f (a ) = 0 . r ) centered at a such that f ( x) − f (a) ≥ 0 for all x ∈ B( a. You should guess the definition of a local maximum and see why it must be true that the gradient is zero at such a point. r ) . In other words.
Meditate on just how should proceed. y ) = x 2 + xy + y 2 + 3x − 3y + 4 . this derivative. We may have ∇f (a ) = 0 without a being either a local minimum or a local maximum. we want (x. so at any local maximum or minimum. Thus if a is a local minimum or a local maximum of f. must be zero.8. This function clearly has a derivative everywhere.
3 + y ) − f ( − 3. These are all points except (3.
Exercises
In each of the following.3 ) = f ( − 3 + x . 3) .. What we have actually found is all the points that cannot possibly be local minima or maxima.3 ) :
f ( − 3 + x .7
. Now let us reflect on what we have here. 3 + y ) − ( − 5) = ( − 3 + x ) 2 + ( − 3 + x)( 3 + y) + ( 3 + y) 2 + 3( − 3 + x) − 3( 3 + y) + 9 = x 2 + xy + y 2 = x + y 3y 2 + 2 4
2
It is therefore clear that f (− 3 + x . 3). which means that (3. 3 + y ) − f ( − 3. find all local maxima and minima:
8. 3 + y ) − f ( − 3. y ) = (−3.∂f = 2x + y + 3 = 0 ∂x ∂f = x + 2y − 3 = 0 ∂y
We are thus faced with the borderline trivial problem of solving the system of equations
2 x + y = −3 .3 ) ≥ 0 . x + 2y = 3
There is just one solution: ( x . All we know right now is that this point is the only possible candidate. Let’s find out what we have by the hammer and tongs method of examining the quantity f (− 3 + x . 3) is a local minimum.
8
. We first decide what we mean by "best". applications in which the location of a minimum value of a function is sought. Let's say we mean the line that minimizes the sum of the squares of the vertical distances from the points to the line. Suppose we have a set of n points in the plane. y) = x 2 + xy + 3x + 2 y + 5
12. f ( x . Knowing these values will give us our line. y) = y − x 2
8.( x 2 . yet important. We simply apply our vast and growing knowledge of calculus and find where the gradient of f is 0:
8. f ( x. y n ) . Thus every such line has the form y = mx + b . y ) = x 2 + 3xy + 3y 2 − 6 x + 3 y − 6
11. f ( x. Our quest is thus for the values of m and b at which the function f ( m. f ( x.( x n .10. y) = x 2 + 2 xy
14. f ( x. We can describe all
nonvertical lines in the world by means of two variables. traditionally called m and b. y2 ). y) = 2 xy − 5x 2 − 2 y 2 + 4x − 4
13.K. say ( x1 . y1 ).5 Least Squares We shall next look at some very simple. and we seek the straight line that "best" fits this collection of points. b) = ∑ (mxi + b − y i ) 2
i =1 n
has its minimum value.
Suppose we have the following table of values:
x 0 1 2 3 4
y 1 2 4 3. ∂ b i =1 i =1 i =1
We are thus faced with solving the 2 x 2 linear system
m∑ x i2 + b ∑ xi = ∑ x i y i
i= 1 n i =1 i =1
n
n
n
m∑ x i + bn = ∑ yi
i= 1 i= 1
n
Meditate sufficiently to convince yourself
that there is always exactly one
solution to this system. and ∂ m i =1 i =1 i= 1 i =1
n n n ∂f = ∑ 2( mxi + b − yi ) = 2[m∑ xi + nb − ∑ yi ].
n n n n ∂f = ∑ 2 x i (mxi + b − yi ) = 2[m∑ x i2 + b∑ x i − ∑ xi y i ].9
.5 5
8. Let's have a go at an example. and continue meditating sufficiently to convince yourself that there must be an honesttogoodness minimum of the original function at this solution.
Now. ) =0 ∂m ∂ b
.∇f = (
∂f ∂f .
In other words. the line that best 142 568 fits the data in the “sense of least squares” is Solving this system gives us m = y= 255 993 x− 142 568
Here’s a picture of this line together with the data points:
8.5 76m +12b = 115.5 255 993 and b = − .5 7 8 9 10 12 15
4 7 9 12 18 21 29
The linear system for m and b is 718m + 76b = 1156.10
.
Mozart. Here is a table of Köchel numbers versus year of composition for the compositions of W. Find the "least squares" straight line approximation to this table and use it to estimate the year in which Mozart's Sinfonia Concertante in Eflat major was composed.]
8.11
. by Thomas & Finney.
Köchel Number 1 75 155 219 271 351 425 503 575 626
Year composed 1761 1771 1772 1775 1777 1780 1783 1786 1789 1791
[This problem is taken from Calculus and Analytic Geometry (8th Edition).Looks pretty good! Exercises
15. A.
Let's think a moment about what the candidates for such points are. The domain R is certainly closed. First. then there must necessarily be a maximum and a minimum value for f on D. if the domain D is both closed and bounded. what properties of D will insure the existence of a biggest and smallest value of f ? The answer is fairly simple. however. then the gradient must be 0 . Consider. we suppose that f is continuous—otherwise.16.). but f has neither a . If the biggest or smallest value of f occurs in the interior of D. Having the domain be closed. How did you find the maximum value of a function f whose domain D is a closed interval [a . maximum nor a minimum. (Here D is a subset of R n . It turns out that for continuous f . is not sufficient to guarantee the . one is most likely interested in finding the places at which the largest and smallest values of a function f : D → R occur.). b] ⊂ R ? Recall
8. for example f : R → R again with f :( 01) → R given by f ( x) = x . let's think a moment about how we can tell if there is a maximum or minimum value of f on D. We need also to have the domain be bounded. Find some data somewhere (The Statistical Abstract of the United States is a good source of interesting data. The points at which the largest or smallest values occur must therefore be either i)points in the interior of D at which the gradient of f vanishes. Turner's third grade calculus class. Hark back to Mrs. D = (01) . which has neither a maximum nor a minimum on .12
. for example the function f :( 01) → R given simply by f ( x) = x . or iii)points in D but not in the interior of D (that is. then surely the point at which it occurs is a local maximum (or minimum). find the least squares linear approximation to the data.
8. Certainly D must be a closed subset of R n . To begin. consider. existence of a maximum and minimum. ii)points in the interior at which the gradient of f does not exist. rather than in simply finding local maxima and minima. and say something intelligent about your results.6 More Maxima and Minima In real life. If f has a gradient there. points on the boundary of D). anything can happen! Next.
found all points in the interior (that is. and t = 2π .13
. y ) = x 2 + 2 y 2 − x . it .
Example A flat circular plate has the shape of the region {( x . candidates for the hottest and coldest points are all points inside the circular boundary at which the gradient of T is 0 and all points on the boundary. mostly because the boundary of even a nice domain D is not a nice finite set as in the case of an interval. by using the same name for the function T( x . but is an infinite set. We have at this point added one more point to our list
8. Note that the boundary may be described by the vector equation r(t ) = cos t i + sin t j .) First. First.0) . in the open interval (a. (Note that T has a gradient at all points inside the circle. the boundary of D is particularly simple. This is easy. 0 ≤ t ≤ 2π [Here we are abusing the notation. let's find where among all points ( x . 1 should be clear there is just one such point: ( . You then simply evaluated f at these points. 2 finding the candidates on the boundary. evaluated f at any points in (a. evaluated f at the two end points of the interval (in this one dimensional case. The temperature on this set is then given by T( t ) = T ( r(t )). Our
assignment is to find the hottest and coldest points on the plate. the ones at which ∇T = (2 x − 14 y) = 0 . where 0 ≤ t ≤ 2π . y ) ∈ R 2 : x 2 + y 2 ≤ 1} .] We are now faced with the one dimensional problem of finding the maximum and minimum values of a nice differentiable function of one variable on a closed interval. as we have done before. The temperature at the point ( x . y ) on the plate is given by T( x . we know the endpoints of the interval are candidates: t = 0.). Now for the more difficult part. y ) such that x 2 + y 2 < 1 . Let's look at an example. The situation in higher dimensions is a bit more complicated.b) at which there is no derivative. and then picked out the biggest and smallest numbers you computed. According to our previous discussion. y ) and the composition T( r(t )) .b)) at which the derivative vanishes.
0) = − = − .of candidates: r(0) = r(2π ) = (10) . ( . . 1 3 1 3 1 3 1 9 T( − . ).− ). From the Chain Rule. y = 0 . When 2
1 3 3 ..0). 2 2 2 2
8. places at which the derivative dT = 0 . and (− . 2 4 2 4 T(10 ) = 1 − 1 = 0 .14
. or 2 cos t + 1 = 0. then y = or y = − . we know dt
dT = ∇ T( r(t )) ⋅ r'( t ) = (2 cos t − 14 sin t ) ⋅ (− sin t . or x = − y = 0 .0) . T( −10 ) = 1 + 1 = 2 . ) and (− . We have.0) . Thus our 2 2 2
1 3 1 3 new candidates are (1. we have our answer. . make up our entire list of possibilities for the hottest and 2 coldest points on the plate. (− 10 ). ) = T( − . or sin t ( 2 cos t + 1) = 0 Thus sint = 0 . then x = 1 or x = −1.− ) = + + = 2 2 2 2 4 2 2 4
1 Finally. and the hottest points are 2 1 3 1 3 (− .− ) .cos t ) = 2 cos t sin t + sin t . Now for candidates inside the interval. (. The coldest point is ( . we seek . These together with the one 2 2 2 2 1 we have already found. All we need do now is to compute the temperature at each of these points: 1 1 1 1 T( . dt The equation dT = 0 now becomes dt 2 cos t sin t + sin t = 0. and when x = − 1 . in other words.
15
. 4 4
8. and so there are no endpoints to worry about. In the problems of the previous section. y ) = 0 . Let's see what we might do about this difficulty. we were fortunate in that it was easy to find parametric representations of the these sets. (Here g is dt
sufficiently nice to insure that g ( x . Thus at a maximum or dt
minimum. Suppose we are faced with the problem of finding the maximum or minimum value of the function f : D → R . y ) = 0} .) Suppose r(t ) is a vector description of the curve D.7 Even More Maxima and Minima It should be clear now that the really troublesome part of finding maxima and minima is in dealing with the constrained problem. y = 4 . then it must have the
8.
18. and y = x . in general. where g is a nice function. At a maximum or minimum. y ) = 0 . But if ∇f is perpendicular to the tangent to the level curve g ( x . of course. we must have dF = 0 .) The Chain Rule tells us that dF = ∇f ⋅ r' = 0 . that is. the problem of finding the maxima and minima of a given function on a set of lower dimension than the domain of the function. we are seeking a maximum or minimum of the function F (t ) = f ( r( t )) . D is a level curve of g . Find the maximum and minimum values of f ( x . Find the maximum and minimum value of f ( x. where D = {( x. y) = x 2 − xy + y 2 + 4 on the closed area in the first quadrant bounded by the triangle formed by the lines x = 0 . this.Exercises
17. y ) = (4 y − y 2 )cos x on the closed area bounded by the rectangle 1 ≤ y ≤ 3 . Now then. (In other words. y) ∈ R 2 : g ( x . − π π ≤x≤ . the gradient of f must be perpendicular to the tangent to g ( x . could be quite difficult. y ) = 0 is a closed curve.
What do you make of the solution λ = 0 ? These values of λ give us two candidates for places at which extrema occur: x = 0 and y = 0 . This is just what we need to know.16
. ∇f and ∇g must "line up". and our equation ∇f = λ ∇ g becomes 2x = λ (2 x − 2) 2 y = λ ( 2 y − 4) We obtain a third equation from the requirement that the point ( x .same direction as the normal to this curve. 2 2 (λ − 1) λ − 1 (λ − 1) λ −1 λ2 − 2λ (λ − 1) = 0. y ) = 0 . λ2 − 2λ = 0 We have two solutions: λ = 0 and λ = 2 . and f ( 2. Thus at a maximum or minimum. y ) = 0 . or. There . The last λ −1 λ −1 λ2 2λ 4λ 2 8λ − + − = 0 .4) = 4 + 16 = 20 . In other words.1) = λ y (λ − 1) = 2λ Thus x = λ 2λ and y = . Now then f ( 00 ) = 0 . y ) be on the curve g ( x . Let's see this idea in action.
equation then becomes
8. and x = 2 and y = 4 . Then ∇f = 2x i + 2 y j . y ) = x 2 − 2 x + y 2 − 4 y . we may take g ( x . (What about the possibility that λ − 1 = 0 ?). y) = x 2 + y 2 on the curve x 2 − 2 x + y 2 − 4 y = 0 . and
∇g = (2 x − 2) i + (2 y − 4) j . and there is no need actually to know a vector representation r for g ( x . we need to find all solutions to the system of equations 2 x = λ (2 x − 2) 2 y = λ (2 y − 4) x 2 − 2 x + y2 − 4 y = 0 The first two equations become x ( λ . Suppose we wish to find the largest and smallest values of f ( x. for the gradient of g is normal to this curve. Here. Thus ∇f = λ ∇ g .
y) = xy has its extreme values. This method for finding "constrained" extrema is generally called the method of Lagrange Multipliers. Find the points on the ellipse x 2 + 2 y 2 = 1 at which f ( x. and the maximum value is 20. and it occurs at (2. y) = 4 x + 3y on the circle x 2 + y 2 = 1 .4).
21.
8. Use the method of Lagrange multipliers to find the largest and smallest values of f ( x. (The variable λ is called a Lagrange multiplier.)
Exercises
19.we have them—the minimum value is 0 and it occurs at (0.
20. Find the points on the curve x 2 + xy + y 2 = 1 that are nearest to and farthest from the origin.0).17
.
Chapter Nine
The Taylor Polynomial
9.1 Introduction Let f be a function and let F be a collection of "nice" functions. The approximation problem is simply to find a function g ∈ F that is "close" to the given function f . There are two issues immediately. How is the collection F selected, and what do we mean by "close"? The answers depend on the problem at hand. Presumably we want to do something to f that is difficult or impossible (This might be something as simple as finding
f ( x ) for some x.). The collection F would thus consist of functions to which it is easy to
do that which we wish to do to f . Our measure of how close one function is to another would try to reflect the closeness of the results of our operations. Now, what are we talking about here. Suppose, for example, we wish to find
f ( x ) . Our collection F of
functions should include functions that are easy to evaluate at x , and two function would be "close" simply if there values are close. We might, for instance, want to evaluate sin x for all x is some interval I. The collection F could be a collection of second degree polynomials. The approximation problem is then to find elements of F that make the "distance" max{sin x − p (x ): x ∈ I } as small as possible. Similarly, we might want to find the integral of some function f over an interval I . Here we would want F to consist of functions easily integrated and measure the distance between functions by the difference of their integrals over I . In the previous chapter, we found the "best" straight line approximation to a set of data points. In that case, the collection F consisted of all nonvertical straight lines, and we measured the distance between functions by the sum of the squares of their differences on a specified set of points { x1 , x2 ,K , x n } . imagine many other examples. 9.2 The Taylor Polynomial We look first at a simple but useful problem: Given a nice function f : D ⊂ R → R , a point a in the interior of the domain D , and an integer n , find a polynomial p of degree ≤ n such that You can
9.1
p (a ) = f (a ) p '( a ) = f '( a ) p ''( a ) = f ' ' ( a ) M p ( n ) (a ) = f
( n)
(a )
We solve the problem by the Behold Method. Simply verify that
f ''( a ) f '''( a ) f ( n ) (a ) ( x − a) 2 + ( x − a ) 3 +K+ (x − a ) n 2! 3! n!
p ( x ) = f ( a ) + f '( a )( x − a ) +
does the job! It is also fairly easy to see that this polynomial is the only polynomial of degree ≤ n that does the job. Suppose q is also a polynomial with degree g ≤ n such that
p (a ) = f (a ) p '( a ) = f '( a ) p ''( a ) = f ' ' ( a ) M p ( n ) (a ) = f
( n)
(a )
and consider the function r = p − q . Note that r is also a polynomial of degree ≤ n . But
r ( a ) = r '( a ) = r ''( a ) =K = r ( n ) ( a ) = 0 .
Or, in other words, r has a zero of order n + 1, and the only way this can happen is if
r ( x ) ≡ 0 for all x . That is, p ( x ) ≡ q ( x ) identically.
Example Let f (x ) = sin x and let a = 0 . Let's find the Taylor polynomial for a few different values of n. For n = 1, we have simply p1 ( x) = f (a ) + f '( a )( x − a ) = sin 0 + cos 0( x) = x . Note that for n = 2, we have p 2 ( x) = sin 0 + cos0 (x ) − sin 0( x 2 ) = x , also. Let's take a look at the next Taylor polynomial. Here p 3 (x ) = x − at the graph of p3 and f . We shall use Maple.
x3 . Let's draw some pictures; we'll look 6
9.2
What we see is that the Taylor polynomial looks like a pretty good approximation as long as we don't get too far away from a = 0. Let us continue. Convince yourself that p 4 = p 3 , and p 5 ( x) = x −
x3 x5 + . Another picture: 6 120
9.3
Exercises 1 . Find the Taylor polynomial of degree n for f (x ) = e x at a = 0. 2 . Find the Taylor polynomial of degree n for f (x ) = x 3 at a = 1. 3 . Find the Taylor polynomial of degree 3 for f (x ) = log x at a = 1. 4 . Find the Taylor polynomial of degree n for f (x ) = sin x at a = 0. 5 . Find the Taylor polynomial of degree 3 for f (x ) = x at a = 4. 9.3 Error Let's see how close the Taylor polynomial is to the function f . To do this, suppose p is the Taylor polynomial of degree ≤ n for the function f at a , and consider the function
( t − a ) n +1 ( f ( x ) − p ( x )) . ( x − a ) n +1
g ( t ) = f ( t ) − p( t ) −
(We assume x ≠ a .) Note that g (a ) = g ( x) = 0 . Now, from the Mean Value Theorem (or Rolle's Theorem, or whatever.) we know that g' (ξ 1 ) = 0 for some ξ 1 between a and x . But note also that g ' ( a ) = f '( a ) − p '( a ) −
( n + 1)(a − a ) n ( f ( x ) − p ( x )) = 0 . It thus follows ( x − a ) n +1
from the Mean Value Theorem that the derivative of g' is zero at some ξ 2 between a and
ξ 1 . Also, g ' ' ( a ) = f ' ' (a ) − p' ' ( a ) − ( n + 1) n ( a − a ) n −1 ( f ( x ) − p ( x )) = 0 . Once again, from ( x − a ) n +1
the celebrated Mean Value Theorem, we conclude that g'''( ξ 3 ) = 0 for some ξ 3 between a and ξ 2 . Continuing in this fashion, we are finally able to conclude that g ( n +1) (ξ ) = 0 for some ξ . Let's see what this looks like.
(n + 1)! ( f ( x) − p( x)) ( x − a ) n +1
g ( n +1) (t ) = f
( n +1)
(t ) − p ( n +1 ) (t ) −
9.4
and so p ( n +1) (t ) ≡ 0 . we don't know what ξ is. θ = we use
π π instead of sin is estimated by 18 18 π π 1θ sin − ≤ ≤ 0. thus
sin θ − θ ≤ θ3 .5
. Observe that p(θ ) = θ is simply the Taylor polynomial of degree ≤ 2 for f (θ ) = sinθ at
a = 0 . but we do know that sin ξ } ≤ 1 . what was going on here? Let's see if our newfound knowledge of Taylor polynomials will help. Crews replaced the sine of a "small" angle θ by θ itself ? He assured us that for small angles this was just fine." Exercises
9. ( x − a ) n +1
(Remember. p is a polynomial of degree ≤ n . I guess that's "not too bad.008862. ( n + 1)!
f (x ) − p (x ) =
Example Remember when in 7th grade physics class. that θ = 10 o . 6
and we have a precise estimate of the error incurred by substituting θ for sinθ . 6
Now. we have that sinθ − θ = − sin ξ 3 θ . then what? Well. Using the result just derived. 18 18 6 18
3
10 π 2π = .and so g ( n +1) (ξ ) = 0 becomes
f
(n +1)
(ξ ) −
(n + 1)! ( f ( x) − p (x )) = 0 . Then the error we get when 360 18
Now we know exactly what "pretty close" means. for example. From this we obtain an expression for the difference between f and the Taylor polynomial g :
f ( n +1) (ξ ) ( x − a ) n +1 . Mr. Well. Suppose. For 10 degrees.
6 . ) and find an 11 upper bound for the magnitude of the difference between your approximation and
log(. ) . Use the Taylor polynomial found in Exercise 3 to approximate log(. c)Find as small an upper bound as you can for the difference between your approximation found in part b) and
e. Calculate e with an error of less than 106 .
7 .
x3 with an error of magnitude no 6
9. For what values of x can you replace sin x by x − greater than 3 × 10 −4 ? 9 . a)Find the Taylor polynomial of degree ≤ 2 for f (x ) = e x at a =0.6
. b)Use the result of part a) to find an approximation for e . 11
8 .
u. i 2 . p ÝnÞ ÝaÞ = f ÝnÞ ÝaÞ. in any number of variables. Then recall from Ms. Ýx. This polynomial is given by ÝnÞ f vv ÝaÞ Ýx ? aÞ 2 + u + f Ýx ? aÞ n pÝxÞ = fÝaÞ + f v ÝaÞÝx ? aÞ + n! 2! We also know the difference between fÝxÞ and pÝxÞ: fÝxÞ ? pÝxÞ = where Y is somewhere between a and x. yÞ at a becomes the polynomial such that
1
. yÞ =
> a ij x i y j. how might we extend the idea of the Taylor polynomial of degree ² n for a function f at a point a ? Simple enough. This looks pretty ferocious in general.
i. a polynomial pÝx. or indeed. Turner’s class the single variable version of Taylor’s Theorem tells us that there is exactly one polynomial p of degree ² n such that pÝaÞ = fÝaÞ. u. Before we worry about what the Taylor polynomial might be in higher dimensions. yÞ of degree ² n is a function of the form
i+j=n
f Ýn+1Þ ÝYÞ Ýx ? aÞ n+1 . Suppose f is a onevariable function that has n + 1 derivatives on an interval about the point x = a. iq i i1 i2 /x 1 /x 2 u/x q /x i11 /x i22 u/x qq for all i 1 .j=0
Thus a polynomial of degree ² 2 (perhaps more commonly known as a quadratic) looks like pÝx. yÞ = a 00 + a 10 x + a 01 y + +a 11 xy + a 20 x 2 + a 02 y 2 . In this case. In two dimensions. Ýn + 1Þ!
pÝx. so let’s see what it says for just two variables. Now. p v ÝaÞ = f v ÝaÞ. bÞ and the Taylor polynomial pÝx. i q such that i 1 + i 2 + u + i q ² n. It’s a polynomial pÝxÞ of degree ² n so that / i 1 +u+i q pÝaÞ / i 1 +u+i q fÝaÞ = . we have a =Ýa. we need to be sure we understand what is a polynomial in more than one dimension. Introduction. The polynomial p is called the Taylor Polynomial of degree ² n for f at a. I hope it easy to guess what one means by a polynomial in three variables. y.Taylor’s Theorem 1. p vv ÝaÞ = f vv ÝaÞ. zÞ.
0Þ = ?1 = / p Ý0. 0Þ = 1.0)? Explain. /f Ý0. /x 2 /x 2 2 / 2 f = ? cosÝx + yÞ. and / p = ?1. /x /x /f = ? sinÝx + yÞ. 0Þ. a) pÝx. /x 2 /x 2 2 / 2 f Ý0. and /y 2 /y 2 2 / 2 f Ý0.
2
y2 2
. /x/y /x/y Exercises 1. Example Let fÝx. yÞ = xy
2
. Let fÝx. 0Þ. and /p = ?x ? y./ i+j fÝaÞ / i+j pÝaÞ = i j . 2. /f = ? sinÝx + yÞ. yÞ = 1 ? x2 ? xy ? polynomial of degree ² 2 for f at Ý0. He we go. /y /y 2 / 2 f = ? cosÝx + yÞ.0) of degree ² 3. /x/y /x/y Now it’s easy to see that fÝ0. Let’s verify that p is the Taylor
fÝ0. 0Þ = 0 = pÝ0. yÞ = sinÝx + yÞ. /y 2 /y 2 2 / 2 f = ? cosÝx + yÞ. 0Þ. yÞ = cosÝx + yÞ. /x /x /f Ý0. 0Þ = ?1 = / p Ý0. 0Þ = 1. and pÝ0. Verify that the polynomial in the Example is also the Taylor polynomial for f at (0.Which Which of the following is the Taylor polynomial of degree ² 2 for f at (0. 0Þ. and /p = ?x ? y. 0Þ = ?1 = / p Ý0. and / p = ?1. 0Þ = 0 = /p Ý0. 0Þ. 0Þ. and let pÝx. and / p = ?1. yÞ = 1 + x 2 + y 2 b) pÝx. 0Þ = 0 = /p Ý0. /y /y 2 / 2 f Ý0. / i x/ j y / x/ y for all i + j ² n. 0Þ.
We know from the chain rule that g v ÝtÞ is given by g v ÝtÞ = 4fÝa + thÞ 6 h = = /f . Let f be a realvalued function defined on a subset of R q . h q Þ f
Ýa+thÞ
h1 / + h2 / + u + hq / /x 1 /x 2 /x q
In keeping with our general practice of restricting ourselves to dimensions one. we’ll write x =Ýx. Prior to finding a general recipe for the Taylor polynomial. Derivatives. u. As usual. two. g vv ÝtÞ = = h / +k / /x /y
2
f f
2 2 2 h 2 / 2 + 2hk / + k 2 / 2 /x/y /x /y 2 2 2 f f = h 2 / 2 + 2hk / f + k 2 / 2 /x/y /x /y
Example Suppose fÝx. kÞ. yÞ = x 2 y 3 + y 2 . h 2 . yÞ and h = Ýh. Suppose that in a neighborhood of the point x. ?2 + tÞ
3
. The expression for g v ÝtÞ now looks like: g v ÝtÞ = h / +k / f /x /y
Ýx+thÞ
We are now in business. /f . a 2 . or three. Define the function g by gÝtÞ = fÝa + thÞ. u. we need look at finding higher order derivatives of certain composite functions. Let’s find the second derivative of the function gÝtÞ = fÝ1 + 3t. h 2 . yÞ = x + y
2.c) pÝx. let’s look first at the case q = 2. a q Þ and h = Ýh 1 . where a = Ýa 1 . yÞ = x 2 + xy + 2y
d) pÝx. h q Þ. the function f has a lot of continuous partial derivatives. u. for we have a nice recipe for higher order derivatives of g : g ÝmÞ ÝtÞ = h / +k / /x /y
m
f
Ýx+thÞ
For example. u. /f /x 1 /x 2 /x q 6 Ýh 1 .
Thus. 3 ? 4tÞ. Next. Find g vvv ÝtÞ. Find the second derivative of the function g defined in Problem 3. The Taylor polynomial. gÝtÞ = Ý0Þ > g m! t m +
m=0 n ÝmÞ
g Ýn+1Þ ÝYÞ n+1 t . where g is the function defined in the Example. To find the Taylor polynomial for a function f of several variables at a point a. vÞ = u 3 v + v 2 . Let fÝx.First. Let FÝu.
g vv ÝtÞ = 18Ý?2 + tÞ 3 + 36Ý?2 + tÞ 2 + 6Ý1 + 3tÞ 2 Ý?2 + tÞ + 2 Exercises 3. and
/f /y
= 3x 2 y 2 + 2y. and so
= 2y 3 . 4. Ýn + 1Þ!
where Y is a number between 0 and t. Thus.
/2f /y/x
= 6y 2 . g vv ÝtÞ = 3 / + / f /x /y 2 2 2 f f = 9 / 2 +6 / f + / 2 /x/y /x /y
/2f /x 2 2
Now. 3. substitute t = 1 into the above: gÝ1Þ = fÝaÞ = >
m=0 n
g Ýn+1Þ ÝYÞ g ÝmÞ Ý0Þ + m! Ýn + 1Þ!
We know the value of g ÝkÞ from Section 2: 1 > m!
m=0 n
fÝa + hÞ =
h1 / + h2 / + u + hq / /x 1 /x 2 /x q
m
fÝaÞ
4
. and
/2f /y 2
= 6x 2 y + 2. Find the second derivative of RÝzÞ = FÝz. Find the derivative of gÝtÞ = fÝ1 + t. 5.
/f /x
= 2xy 3 . 3zÞ. 6. we shall simply apply the onedimensional results to the function gÝtÞ = fÝa + thÞ. yÞ = xe y .
Taylor’s polynomial becomes
2 2 2 2 2 f f pÝh.
/f /y
= sin x cos y. WÞ
where ÝY. This is pretty fierce looking. WÞ is one the segment joining Ýx. h 2 . Let’s get an estimate for how well this approximates sin x sin y near Ý0. 0Þ 2 /x 2 /y /x /y /x/y
We have /f = cos x sin y.
/2f /x 2
= ? sin x sin y.+
1 h1 / + h2 / + u + hq / /x 1 /x 2 /x q Ýn + 1!
n+1
fÝcÞ
The point c lies somewhere on the line segment joining a and a + h. The polynomial pÝhÞ =pÝh 1 . We know that sin x sin y ? xy = 1 x / +y / 3! /x /y
3
fÝY. if we let h = x ? a. 0Þ + k / 2 Ý0. For n = 2 and a = Ý0. a 2 Þ
n+1
+
1 h / +k / /x /y Ýn + 1!Þ
fÝc 1 . then qÝxÞ =pÝx ? aÞ is the thing we called the Taylor polynomial in the first section. the last term is traditionally called the error term or sometimes. /x Thus. 0Þ. Example Let fÝx. Let’s look at the two variable case:
fÝa 1 + h. Now. a 2 + kÞ. a 2 Þ and Ýa 1 + h. 0Þ + k /f Ý0.
pÝh. yÞ = sin x sin y.
/2f /y 2
= ? sin x sin y. kÞ = fÝ0.
/2f /x/y
= cos xcosy. a 2 + kÞ =
>
m=0
n
1 h / +k / m! /x /y
m
fÝa 1 . the remainder term. 0Þ + h / 2 Ý0. 0Þ + h /f Ý0. yÞ and the origin. u. c 2 Þ
where Ýc 1 . 0Þ. h q Þ = 1 > m!
m=0 n
h1 / + h2 / + u + hq / /x 1 /x 2 /x q
m
fÝaÞ
is the Taylor polynomial of degree ² n for f at a.
5
. c 2 Þ is on the line joining Ýa 1 . Actually. 0Þ + hk / f Ý0. kÞ = hk.
Exercises 7. Estimate the error in the approximation found in Problem 11 if x ² 0. 0Þ.1.
6
. 11.
3 sin x sin y ? xy ² 8c = 4 c 3 6 3 3
f ² 8c 3 . let’s suppose that x ² c and y ² c for some constant c. Find the Taylor polynomial of degree ² 3 for fÝx. 8. Find the Taylor polynomial of degree ² 1 for fÝx.1 and y ² 0. Noting that all the partial derivatives in the above expression are simply products of sine and cosines. yÞ = e xy at Ý0.x / +y / /x /y
3
3 3 3 3 f f f = x 3 / 3 + 3x 2 y /2 f + 3xy 2 / f 2 + y 3 / 3 . yÞ = e xy at Ý0. we can estimate x / +y / /x /y and so. /x /x /y /x/y /x
Next. 0Þ. 0Þ. yÞ = e x cos y at Ý0. Use Taylor’s Theorem to find a quadratic approximation of e x cos y at the origin. Find the Taylor polynomial of degree ² 2 for fÝx. 12. Find the Taylor polynomial of degree ² 1 for fÝx. 10. 0Þ. 9. yÞ = e xy at Ý0. at last.
1 Introduction Suppose we want to compute an approximation of the number e by using the Taylor polynomial pn for f (x ) = e x at a =0. If the domain is infinite. and the sequence itself by ( f n ) . 2 6 n!
We could now use p n (1) as an approximation to e .) We shall be concerned only with infinite sequences. and All That
10.Chapter Ten
Sequences. It tells us that we can make this error as small as we like by choosing n sufficiently large. and so the modifier will usually be omitted.1
. ( n + 1)!
where 0 < ξ < 1 . We know from the previous chapter that the error is given by
e − p n (1) = eξ 1 n +1 . (Here Z + denotes the positive integers. This polynomial is easily seen to be
p n ( x) = 1 + x + x2 x3 xn + +K+ . is the sequence n
1
10. Assume we know that e <3. it is customary to denote f (n ) by f n . for example. and we have the estimate
3 . If f : Z + → R . 2 Sequences A sequence of real numbers is simply a function from a subset of the nonnegative integers into the reals.) Thus.
1 0 . There are several notational conventions involved in writing and talking about sequences. This is the idea we shall study in this chapter. We shall also almost always consider sequences in which the domain is either the entire set of nonnegative or positive integers. (Guess what a finite sequence is. we say the sequence is an infinite sequence. Series. ( n + 1)!
0 ≤ e − pn ( 1) ≤
Meditate on this error estimate. This is expressed formally by saying that the limit of
p n (1) as n becomes infinite is e .
This is usually written lim a n = L . there is an N so that to the right of N . Let (a n ) be a sequence and suppose there is a number L such that for any ε >0.f defined by f (n) = . . Exercises
10. the terms of the sequence get close to L .
n →∞
what does this really mean? It says simply that as n gets big. all the spots are in the strip of width 2ε centered at L.
1 n
The function values f n are called terms of the sequence.9 . it should be reasonably clear that a limit of the sequence
The graph of a sequence is pretty dreary compared with the graph of a function whose domain is an interval of reals. a look at some pictures can help understand some of these definitions. there is an integer N such that  a n − L < ε for all n > N .2
. Look at the graph of (a n ) :
The fact that L is a limit of the sequence means that for any ε >0. but nevertheless. n
1 1 1 1 + + +K+ is e . Then L is said to be a limit of the sequence. n 2 . Now. I hope it is clear that 0 is a limit of the sequence . Suppose the sequence (a n ) converges to L .
Frequently one sees a sequence described by writing something like
14 .K . 2 6 n! 1
From the discussion
in the Introduction to this chapter.K.
This is simply another way of describing the sequence (n 2 ). and (a n ) is said to converge to L .
n
10. one would. 3. It is a little neater to write if we use the usual summation notation: ∑ a k . Thus. for example. a 6 .
1
k =0
∞
10. We have seen an
k =0
n
example of such a thing previously.3 Series Suppose (a n ) is a sequence.
4. Suppose the sequence (a n ) = a 0 . or explain why it does not converge.
Find the limit of the sequence converge. viz.
Find the limit of the sequence
6.. n2
2. n2
3n 2 + 2n − 7 . 1 + + +K+ = 2 6 n ! k =0 k !
∑
It is usual to replace lim ∑ a k by
n →∞ k=0
n
∑a
k =0
∞
k
. Give an example of a sequence that does not have a limit.
Find the limit of the sequence
5. or explain why it does not converge.
5n 3 − n 2 + 7 n + 2 . a 2 .1. Explain how you know that the sequence (a n+5 ) = a5 . write
e=
∑k!.
Find the limit of the sequence
logn .
Prove that a sequence can have at most one limit (We may thus speak of the limit of a sequence.). or explain why it does not converge. The sequence (a 0 + a1 +K+ a n ) is called a series.K also converges to L.K converges to L.3
. a 1 . or explain why it does not 3n 3 + n 2 − n + 10
7. a 7 . Explain.
3 .
1 1 1 n 1 .
k =0 k =0 k =0
∞
n
n −1
n lim a n = lim a k − n →∞ n →∞ k=0
∑
∑a
k =0
n −1
k
= lim
n →∞
∑a
k =0
n
k
− lim
n→∞
∑a
k =0
n −1
k
= L − L = 0. It is usually clear whether the series or the limit of the series is meant. Some poor misguided souls also use
∑a
k =0
∞
k
to stand simply for the series
n a k . Let k 2 4 2 k =0 2
n
Let's consider the series ∑
Sn = 1 +
1 1 1 1 + + +K+ n . And one more word
of warning. suppose L = ∑ a k . In other words.4
.
10. Example
1 1 1 1 = 1 + + +K+ n . To see
k =0
n →∞
n
this.One also frequently sees the limit
∑a
k =0
∞
k
written as a 0 + a 1 +K+a n +K . it must be true that lim a n = 0 . n→∞ 2
∞
∑2
k =0
1
k
=2.
Observe that for series ∑ a k to converge. but it is k =0
∑
nevertheless an offensive practice that should be ruthlessly and brutally suppressed. Then 2 4 8 2
1 1 1 1 1 1 S n = + + +K+ n + n +1 . 2 2 2
This makes it quite easy to see that lim
n→∞
Sn = 1 . or lim S n = 2 . and observe that a n = ∑ a k − ∑ a k . Thus. 2 2 4 8 2 2
Thus
Sn 1 1 = S n − S n = 1 − n +1 .
In other words. In other words.3 0. we still don't know anything.
1 . if lim a n ≠ 0 .7 0. note that lim
k =1
n
1 k
1 = 0 . x
10.5
. Observe that the area under the "stairs" is simply x
n
∑k.8 0. Look at the following picture:
1 0.9 0. that is.5 0. First.6 0. then the series ∑ a k does not have a limit.
1
k =1
11
∑k
k =1
1
is larger than the area under the curve y =
1 from x =1 x
∑ k >∫
1
k =1
n
n +1
1
1 dx = log(n + 1) .
n →∞
n
k =0
Another Example Consider the series ∑ .2 0.4 0.1 0 0 2 4 6 8 10 12
The curve is the graph of y = Now convince yourself that to x = n+1. Thus we do not know that n →∞ k
the series does not converge.
Find a value of n that will insure that 1 +
1 1 1 + +K+ > 10 6 .We know that log(n + 1) can be made as large as we wish by choosing n sufficiently large. 2 3 n θ 2 k +1 . or explain why it does not converge. the series is divergent.
Find the limit of the series ∑
k =0
n
. Let 0 ≤ θ ≤ 1 . Suppose we have a series ∑ a k such that a k > 0 for all k . ) The method we used to show that the harmonic series does not converge can be used on many other series. Thus
∑k
k =1
n
1
can be made as large as we wish by choosing n sufficiently large. Suppose f is a decreasing function such that
k =1 f ( k ) = a k for all k . We simply consider a picture like the one above. or explain why it does not converge. n + 3 5
1 0 . (This series has a name. It is called
k =1
1 k
the harmonic series.]
10. Prove that sinθ = ∑ (−1)k
k =0
∞
[Hint: p 2 n +1 (θ ) = ∑ (−1) k
k =0
n
θ 2 k +1 is the Taylor polynomial of degree < 2n+1 for ( 2 k + 1)!
the function f (θ ) = sinθ at a = 0.
Exercises
1 . Then if the limit lim
n
R →∞
∫ f ( x)dx
1
R
does not exist.
Find the limit of the series ∑
9. ( 2 k + 1)!
1 1 . n k =0 3
n
8.6
. From this it
n
follows that the series ∑ does not have a limit.
kp
b) Find all p for which the series in a) diverges. First note the obvious: ∑ b k converges if and only if
k =0 n b k converges. This is called a positive
k =0
n
series.
k =0
n
This tells us all about the series ∑ b k . Suppose that b k ≤ a k for all k > N . and suppose f is a
k =1
n
decreasing function such that f (k ) = a k for all k . Specifically. observe that for all n . Suppose we have a series ∑ a k such that a k > 0 for all k . Now suppose further that we know that ∑ a k converges.7
. Next.4 More Series Consider a series ∑ a k in which a k ≥ 0 for all k .
1 3 .
n →∞ k=N
n
Example
10. we have k= N
n
∑
k=N
∑
n
bk ≤
k=N
∑a
n
k
. Show that if the limit
R →∞
lim
∫ f ( x)dx
1
R
exists. from which it
follows at once that lim ∑ bk exists.
10. Let ∑ b k be another positive series. Let's see why that is.1 2 . then the series is convergent. it tells us that this series also
k =0
n
converges. a) Find all p for which the series ∑
k =1
n
1 converges. where
k =0
n
N is simply some integer.
suppose we know that ∑ b k is divergent. as before. Thus ∑ 3 converges. where N is some number. 2 3 R→∞ 3 R 3 3 x k =1 n + 3n + n + 4
Suppose that. and b k ≤ a k for all
k =0 k =0
n
n
k > N .
1 k =0 2 e + k
n k
1 4 . ∑
1 k =2 log k
n
1 7 .What about the convergence of the series ∑
1 n + 3n + n + 4
3 2
? Observe first that k =1 n + 3n + n + 4
n
1
2
3
<
1 . This time. ∑
k =0
n
1 k + k − 1
2
10. ∑ a k and ∑ b k are positive series. ∑
1 5 . also. ∑
1 k =0 2 k + 1
n
1 6 .
k =0
n
Then it should not be too hard for you to convince yourself that ∑ a k must be
k =0
n
divergent.8
.
Exercises Which of the following series are convergent and which are divergent? Explain your answers. n3
Then observe that the series ∑
1 converges because 3 k =1 n
n
R →∞ 1
lim
∫
R
n 1 1 −1 1 1 dx = lim 2 + = .
). Let's
k =0
n
see about it.10. Then the number ρ = r + For all sufficiently large k. suppose that r < 1. there is an N so that ak
Look now at the series
n a N ρ k − N = a N (1 + ρ + ρ 2 +K ρ n −N ) . A similar argument should convince you that if r > 1. and let ak
r = lim
k →∞
a k +1 .5 Even More Series We look at one more very nice way to help us determine if a positive series has a limit. First.
k =0
n
Next suppose the
sequence
a k +1 is convergent. and suppose a k > 0 for all k.9
. It now follows from the previous section that our original series a k has a k =0
n
∑
n
limit. Consider a series ∑ a k . The "method" of the previous section is usually called the Comparison Test.
10. 2
a k +1 ≤ ρ . we know that
a k +1 ≤ a k ρ for all k ≥ N . k= N
∑
(
)
This one converges because the Geometric series ∑ ρ k converges (Recall that
k =0 n 0 < ρ < 1. ak
The number r tells us almost everything about the convergence of the series ∑ a k . In other words. Thus a k +1 ≤ a k ρ ≤ a k −1 ρ 2 ≤ a k −2 ρ 3 ≤K≤ a N ρ k +1− N . while that of this section is usually called the Ratio Test. 1− r is positive and less than 1. then the series ∑ a k does
k =0
not have a limit.
6 A Final Remark The "tests" for convergence of series that we have seen so far all depended on the series having positive terms. Thus. First. we
k =0 k =0
n
n
10. The second result is a bit harder to see. then lim a k = 0 is sufficient to insure convergence of the
k →∞
series. This is not too hard to see—meditate on it for a while. It says simply that if the series ∑  a k  converges. ∑
3k k 4 k =1 5 ( k + k + 1)
n
2 2 . and we'll just put out the result as the word. if the terms of a series ∑ a k alternate in sign.
k =0
n
then so also does the series ∑ a k . and if it is
k =0
n
true that  a k +1  ≤  a k  for all k . faced with an arbitrary series ∑ a k .
10 k k =0 k !
n
1 8 . ∑
3 2 k +1 k k =0 5
n
2 0 . asking that you accept it on faith. ∑
3 k ( k 4 + k + 1) 5k k =1
n
10. We need to say a word about the situations in which this is not necessarily the case. ∑
1 9 .10
.Exercises Which of the following series are convergent and which are divergent? Explain your answers. ∑
3 2 k +1 k k =0 10
n
2 1 .
then we still do not know about the original series.11
. of course.
10. this series turns out not to be convergent.
k =0
n
If we find this one to be
convergent. then the original series is also convergent.may unleash out arsenal of tests on the series ∑  a k  . If.
Chapter Eleven
Taylor Series
11.1 Power Series Now that we are knowledgeable about series. The number k →∞ c k +1
k →∞
ck is called the radius of convergence. or
11. We begin with the idea of a socalled power series. we ck and divergence for
have
absolute
convergence
for
 x − a  < lim R = lim
ck k →∞ c k +1
 x − a  > lim
ck . Thus this series converges absolutely for all values of x if the number lim
k →∞
ck +1 = 0 . The k th term of the series is ck (x − a)k so the Ratio Test calculation looks like r(x) = lim c k +1 (x − a)k +1 c = x − a lim k +1 . k k →∞ k →∞ c ck (x − a) k
Recall that our series converges for r(x) < 1 and diverges for r(x) > 1 . ck +1
and the interval  x − a < R is called the interval of convergence. we can return to the problem of investigating the approximation of functions by Taylor polynomials of higher and higher degree. Clearly the question of convergence will depend on x . Otherwise. as will the limit where there is one. k= 0 k A power series is thus a sequence of special polynomials: each term is obtained from the previous one by adding a constant multiple of the next higher power of (x − a). There are thus exactly n three possibilities for the convergence of our power series ∑ ck (x − a) k : k= 0 (i)The series converges for no value of x except x = a .1
. A power series is a series of the form n ∑ c (x − a) k .
Here R = lim k = lim k +1 = lim = . ∑ (x + 5)k k=0
n 1 2. Then R = lim k = lim = lim = 0. k →∞ c k →∞ 3 k →∞ 3 k=0 3 k +1 Thus. or (iii)There is a positive number R so that the series converges for  x − a < R and diverges for  x − a  > R . k →∞ c k →∞ (k +1)! k →∞ k +1 k=0 k +1 Thus this series converges only when x = 0. 3 3
Exercises
Find the interval of convergence for each of the following power series:
n 1.(ii)The series converges for all values of x . this one converges for  x − 1< 1 1 and diverges for  x − 1> . ∑ (x −1)k k=0 k
11.
Another Example n c 3k 1 1 Now look at the series ∑ 3k (x −1)k .
Example n c k! 1 Consider the series ∑ k!x k .
Note that the Ratio Test tells us nothing about the convergence or divergence of the series at the two points where  x − a = R .2
.
we have f '( x) =
∑ kc (x − a)
k=1 k
∞
k −1
. ∑ (x − 9) k 2 k = 0 7(k +1)
11. n k 3. the limit of the series defines a function f : f (x) =
k=0
∑ c (x − a)
k
∞
k
. ∑ (x +1) k k = 0 k!
n k! 5. ∑ (x − 4)k k = 0 3k + 1
n 3k 4. Thus for all x in the interval of convergence. k= 0 of course.2 Limit of a Power Series n If the interval of convergence of the power series ∑ ck (x − a) k is  x − a < R . and they are given by f (p ) (x) =
k= p
∑ k(k −1)K(k − p + 1)c (x − a)
k
∞
k− p
.
It is known that this function has a derivative. for  x − a < R . and this derivative is the limit of the derivative of the series. then.
We can now apply this result to the power series for the derivative and conclude that f has all derivatives. Moreover.3
.
Example
11. the differentiated series has the same interval of convergence as that of the series defining f .
for  x < r .
11. and the integral of the limit is the limit of the integral. k=0 k + 1
∞
It is also valid to perform all the usual arithmetic operations on power series. Thus if f (x) =
k=0
∑c x
k
∞
k
and g(x) =
k=0
∑d x
k
∞
k
for  x < r . also true that the limit of a power series can be integrated.
k =0
Also. or 0 < 1 − x < 2. miraculously enough.We know that
∞ 1 = ∑ x k for  x < 1 .
∞ k f (x)g(x) = ∑ ∑ ci dk −i c kx k . for −1 < x < 1. Once again.4
. for  x < r . then
∞
f (x) ± g(x) = ∑ (c k ± dk )x k . i= 0 k= 0
The essence of the story is that power series behave as if they were “infinite degree” polynomials—the limits of power series are just about the nicest functions in the world.
Example We may simply integrate the Geometric series to get log(1− x) = − ∑ xk +1 .
It is. the interval of convergence of the integrated series remains the same as that of the original series:
∫
a
x
f (t)dt =
k=0
∑ k +1(x − a)
∞
ck
k +1
. It follows that 1 − x k=0 ∞ 1 = ∑ kx k −1 = 1+ 2x + 3x 2 + 4x3 +K (1 − x) 2 k =1
for  x < 1 .
9. k = 0 k! The Taylor Series is thus an “infinite degree” Taylor Polynomial> In general. the Taylor Series for f at a is the series n f (k ) (a) ∑ (x − a)k . for many sufficiently nice functions it does. Thus if f is a given function. What is the limit of the series ∑ 2(−1) k kx 2k −1 ? What is its interval of convergence? k =1
8.5
. but. What is the limit of the series ∑ x 2k ? What is its interval of convergence? k=0
n 7.3 Taylor Series Our major interest in finding a power series that converges to a given function. and a is a point in the interior of the domain of f. mercifully.Exercises
n 6. Find a power series that converges to tan −1 x on some nontrivial interval. The obvious candidate for such a series is simply the sequence of Taylor polynomials of increasing degree. the Taylor series for a function may not converge on any nontrivial interval to f . Suppose f (x) =
k=0
∑ c (x − a)
k
∞
k
. What is f (p ) (a) ?
11. we are provided with the nice answer to the question proposed back in Chapter Nine: Can we approximate the function f as well as we like by a Taylor Polynomial for sufficiently large degree?
11. In such cases.
Find the interval of convergence and show that the series converges to f on this interval.
11. this power series converges for all x . Find the derivative of the cosine function by differentiating the Taylor Series you found in Problem #11. Find the interval of convergence and show that the series converges to f on this interval. Find the Taylor Series at a = 0 for f (x) = cos x . From Section 9.
12. or in other words. (2k +1)! (2n + 3)!
and we know that  x 2 n + 3 = 0. Find the Taylor Series at a = 1 for f (x) = logx .
13. Find the Taylor Series at a = 0 for f (x) = ex .
11. (2k +1)! k=0
k ∞
Exercises
10. Thus we have x 2 k+ 1 sin x = ∑(−1) . for all x .6
.Example n x 2k + 1 The Taylor series for f (x) = sin x at x = a is simply ∑ (−1)k . But is the limit sin x ? That’s easy to decide. we know that sin x − ∑ (−1)k
k =0 n
x 2 k+ 1  x 2 n+ 3 ≤ . An easy k=0 (2k +1)! calculation shows us that the radius of convergence is infinite.3. n→ ∞ (2n + 3)! lim no mater what x is. Find the interval of convergence and show that the series converges to f on this interval.
for x ≠ 0 e
Find the Taylor Series at a = 0 for f. Let the function f be defined by 0. Find the interval of convergence and the limit of the series. for x = 0 f (x) = −1/ x 2 .7
.
11.14. .
x n } of [a . the only nice closed pieces of R are the closed intervals. the sum obviously also depends on the choices of the points xi* . S ( P ) for this partition to be a sum S ( P ) = ∑ f ( xi* )∆ xi . x1 . b] with a = x 0 < x1 < x 2 <K < x n = b . xi ] and xi* is any point in this subinterval.1 Introduction We now turn our attention to the idea of an integral in dimensions higher than one. in this context. Recall that we defined a partition P of the interval to be simply a finite subset {x 0 . Mrs. where the domain D is a nice closed subset of Euclidean nspace R n . then later we shall see just what such an abstract thing might be good for in real life. there will be nothing to unlearn.K . n} . Consider a realvalued function f : D → R . our definition of the integral in higher dimensions will include the definition for dimension 1 we learned in grammar school—as always.1
. Let us again hark back to our youth and review what we know about the integral of f : D → R in case D is a nice connected piece of the real line R. (Thus there is not just one Riemann sum for a partition P. The mesh of a partition is max{ x i − xi −1 :i = 12K . As it was in extending the definition of a derivative to higher dimensions. b] .) Now. Turner taught us all about the case n = 1 . This is not reflected in the notation.Chapter Twelve
Integration
12.
i =1 n
where ∆x i = x i − x i−1 is simply the length of the subinterval [ xi −1 . We then defined a Riemann sum . First. we thus have D is a set [a . where b > a . if there is a number L such that we can make all Riemann sums as close as we like to L by choosing the mesh of the partition sufficiently small. then f is said to be
12. We shall begin by seeing what we mean by the integral of f over the set D.
we say that L is the
integral of f over [a . in extending calculus ideas to higher dimensions. On the real line. We shall thus begin by looking at the integral of f : D → R for the case in which D is a nice closed subset of the plane.
12. b]. and the number L is called the integral of f over [a.2 Two Dimensions Let us begin our study of higher dimensional integrals with the two dimensional case. You no doubt remember from your first encounter with this integral that it initially seemed like an impossible thing to compute in any reasonable situation. Complications appear at once.integrable over the interval.2
. b] if for every ε > 0 . nice closed sets are considerably more interesting:
12. As we have seen so often in the past. This number L is almost always denoted
∫ f ( x )dx . there is a δ so that  S ( P ) − L < ε for every partition P having mesh < δ. nice closed sets are simply closed intervals. most of the excitement occurs in taking the step from one dimension to two dimensions—seldom is the step from 97 to 98 dimensions very interesting. but then some version of the Fundamental Theorem of Calculus came to the rescue. in the plane.
a
b
More formally.
D
Such integrals over two dimensional domains are frequently referred to as double integrals. label the subrectangles that meet D. capture D inside a rectangle with sides parallel to the coordinate axes.A moment's reflection convinces us that the domain D can. The largest .
i= 1 n
where ∆Ai is the area of the rectangle from which ( xi* .
12. area of all such rectangles is called the mesh of the subdivision. The number L is usually written with two snake signs:
∫∫ f ( x. and the number L is the integral of f over D. say with subscripts i = 12 . n . First. Now if there is a i number L such that we can get as close to L as we like by choosing the mesh of the subdivision sufficiently small. and then divide this rectangle into subrectangles by partitioning each of its sides:
Now. choose a point ( xi* . K. In each such rectangle. then f is said to be integrable over D. y * ) is chosen. y * ) in D. y) dA . A Riemann sum S now looks like i S = ∑ f ( x i* .3
. yi* ) ∆Ai . even in just two dimensions. be considerably more complicated than it is in one dimension.
and on the sides by the cylinder determined by the region D. above by the xy plane. etc. let us suppose that f ( x. y) ≥ 0 everywhere on D. y ) . etc. . y ) . We capture D inside a box. we get the negative of the volume bounded below by the surface z = f ( x . Suppose a and b are constants.4
. For the purpose of drawing a reasonable picture. and D = E ∪ F . The top of the box thus meets the surface z = f ( x . above by the surface z = f ( x . y) ≤ 0 . where E and F are nice domains whose interiors do not meet. The Riemann sum is i thus the total volume of all such boxes.
Each term f ( xi* . Convince yourself that as the size of the bases of the boxes goes to 0. Let's look a bit at some geometry. The integral
∫∫ f ( x. and subdivide the box into boxes. etc. The following important properties of the double integral should be evident:
12. y) dA is thus equal to the volume of this solid. y * ) ∆Ai is the volume of a box with base the rectangle Ai and height i f ( xi* .
of course. then. y ) . There will be more of the higher dimensional stuff later. the boxes "fill up" the solid bounded below by the xy plane.I hope the definition of the integral in case D is a nice subset of R 3 is evident. y * ) .
D
If f ( x.
∫∫ [af ( x . y) dA = ∫∫ f ( x . y ) + bg ( x . Again. y )dA . y) dA as the volume of the blob
D
bounded below by D in the xy plane and above by the surface z = f ( x . and
D D D
∫∫ f ( x. and again we shall draw our picture as if f ( x. we suppose it to be bounded above by the curve y = h ( x ) . we
shall look at a picture. y )]dA = a ∫∫ f ( x . Think of finding this volume by dividing the blob into slices parallel to the yaxis and adding up the volumes of the slices. and on the right by x = b :
It is convenient for us to think of the integral
∫∫ f ( x. It should be clear what happens if this is not the case. y) dA ?
D
Let's see. how on Earth do we ever find an integral
∫∫ f ( x. y )dA + b ∫∫ g ( x . y ) . on the left by x = a . To approximate the volumes of these slices. y )dA . We assume our domain D has a special form.
D E F
Now. y )dA + ∫∫ f ( x .5
. specifically. we use slabs:
12. y) ≥ 0 . below by y = g ( x ) .
The sum of the volumes of the approximating slabs is thus
S = ∑ [∫
i= 1
n
h ( xi* ) g ( xi* )
f ( x * . In each subinterval [ xi −1 . or [ ∫
h ( x* ) i g ( x* ) i
f ( x i* .We partition the x interval [a. y) dy ]∆ xi . y )dy ]∆x i . b ]: a = x0 < x1 <K < x n −1 < x n = b . The volume i of the slab is the cross section area times the thickness. i
The double integral we seek is just the "limit" of these as we take thinner and thinner slabs. But Lo! The above sums are
12. Our approximating slab has as its base the rectangle of "width" ∆x i = x i − x i−1 and height h( xi* ) − g( x * ) . y ) .6
. or finer and finer partitions of the interval [a. b]. xi ] choose a point xi* . the roof is z = f ( xi* .
where D is the area enclosed by the
lines y = x . and so the double integral is given by
∫∫ f ( x.):
12. x = 0 . y )dydx . The first item of business here is to draw a picture of D (We always need a picture of the domain of integration. It is traditional to omit the brackets and write the iterated integral simply as
∫ ∫ f ( x. y )dy ]dx
The double integral is thus equal to an integral of an integral.
a g( x )
b h( x )
Example Let's find the double integral
∫∫ [ x
D
2
+ y 2 ]dA . usually called an iterated integral.Riemann sums F ( x) = ∫
h( x ) g( x )
for
the
ordinary
one dimensional integral of
the
function
f ( x . and x + y = 2 . y) dA = ∫ F ( x) dx
D a
b
= ∫[
b h( x )
a g ( x)
∫ f ( x . y )dy .7
.
and b = 1. Thus
2 2 2 ∫∫ [ x + y ]dA = ∫ [2 x + D 0
1
( 2 − x) 3 7 3 − x ]dx 3 3
1
2 x 3 (2 − x ) 4 7 x 4 = − − 3 12 12 = 16 4 = 12 3
0
Exercises
1.It should be clear from the picture that in the language of our discussion. h( x ) = 2 − x . a = 0 .8
. 3 3 3 3
and we have such a slice for all x from x = 0 to x = 1. The "volume" is thus
2 −x
y3 [ x + y ]dy = x y + ∫ 3 x
2 2 2
y= 2 −x
= x 2 (2 − x) +
y= x
(2 − x ) 3 x3 (2 − x ) 3 7 3 − x3 − = 2x 2 + − x . g ( x ) = x . Find
∫∫ x
D
2
dA . So slice parallel to the y axis:
The lower end of the slice is at y = x and the upper end is at y = 2 − x .
12. where D is the domain bounded by the curves y = 4 − x 2 and y = 3 x .
Find
∫∫ (x
D
2
− y )dA . Suppose you have a double integral
∫∫ f ( x.
6.2. Use double integration to find the area of the region enclosed by the curves x − y = 2 and y = − x 2 . Find the volume of the wedge cut from the first octant by the cylinder z = 12 − 3y 2 and the plane x + y = 2 . Sketch the domain of integration and evaluate the iterated integral:
∫∫ y
0 x
1 1
2
e xy dydx . y) dA in which the domain D is bounded
D
on the left by the curve x = g( y ) . Sketch the domain of integration and evaluate the iterated integral:
log 8log x
∫ ∫e
1 0
x+ y
dydx .9
.
7.
12.
8. and above by y = b .
4. on the right by x = h( y ) .
3. where D is the area in the first quadrant enclosed by the
coordinate axes and the line 2 x + y = 4 . Find the volume of the solid cut from the first octant by the surface z = 4 − x 2 − y .
5. below by y = a .
12.Give an iterated integral for the double integral in which the first integration is with respect to x . and explain what's going on.10
. Give a double integral for the area of the region bounded by x = y 2 and x = 2 y − y 2 . and evaluate the integral.
9.
13. then the equation becomes F = M . In case the total mass is continuously distributed in space.Chapter Thirteen
More Integration
13. This point R is called the center of mass of the system. we have f i = mi for each i.. Now sum these equations to get F = ∑ fi = ∑ mi
i =1 i =1 n n
d 2 ri dt 2
d 2 ri . the vector ri . Specifically. K. Thus the sum of the external dt 2
i
forces on the system of masses is the total mass times the acceleration of the mystical point R. Thus according to Sir Isaac Newton. If we define R by
i =1 n
R=
∑mr
i =1 n
n
i i
∑m
i =1
d 2R . Assume moreover that there is a force f i acting on this mass. or dt 2
n mr 2 ∑ i i d i =1 . Let's look at what this means in two dimensions. Suppose we have a system of point masses and forces acting on the masses.1
.1 Some Applications Think now for a moment back to elementary school physics. n we have a point mass mi whose position in space at time t is given by . F=M dt n ∑ mi i =1 where M = ∑ mi . Reflect for a moment on this equation. suppose that for each i = 12 . the "sum" in the equation for R becomes an integral.
yi* ) ri ∆ Ai . we obtain for R.Suppose we have a plate and the mass density of the plate at (x. we approximate its location by chopping it into a bunch of small pieces and treating each of these pieces as a point mass. The equation for the center of mass of this system of rectangles is then
~ R=
∑ m r ∑ ρ( x
i i i =1 n
n
n
* i
.
Now choose a point ri = xi* i + y i* j in each rectangle. yi* ) x * ∆Ai i + ∑ ρ ( xi* .y) is given by ρ( x .. The mass of this rectangle will be approximately ρ( x i* . the location of the center of mass
13. where ∆Ai is the area of the rectangle. y * ) ∆Ai i
∑m
i =1
=
i =1 n
i
∑ρ (x
i =1
* i
=
n n ρ ( x i* . yi* ) ∆Ai . y * ) yi* ∆ Ai j ∑ i i n i =1 ∑ ρ( xi* . y ) . y *i ) ∆Ai i =1 1
i =1
The three sums in the previous line are Riemann sums for two dimensional integrals! Thus as we take smaller and smaller rectangles.2
. etc. To find the center of mass of the plate.
y )dA ∫∫ yρ ( x . y ) = k . we say the mass is uniformly distributed over the region. y )dA = k ∫
T 0 T
a b (1− x /a )
∫ xdydx = k ∫ xb(1 − x / a )dx = k
0 0 a b (1 −x / a ) 0
a
a2 b . y ) dA = k ∫
kb 2 a ab 2 ydydx = (1 − x / a ) 2 dx = k .3
. ∫ 2 ∫ 6 0 0
13. y )dA . y ) of the center of mass of P are given by
∫∫ x ρ ( x . y) dA i + ∫∫ yρ ( x .
P
Example Let's find the center of mass of a plate having the shape of the plane region enclosed by the triangle
and having constant density (In this case. and then 6
∫∫ yρ( x . the coordinates ( x .
∫∫ x ρ( x . y) dA j P ∫∫ ρ( x . Suppose ρ( x .R=
∫∫ xρ ( x. and y = P . y ) dA P 1
P
In other words. First. y )dA is the total mass of the plate. x = P M M
where M = ∫∫ ρ( x .
y ) . comforting to see this fact come out in the mathematical wash. it is thus a geometric property of the region. then the constant slips out through the integral signs and cancels top and bottom in the recipe for the coordinates ( x . y ) = 5x . Find the center of mass of the smaller of the two regions cut from the elliptical region x 2 + 4 y 2 = 12 by the parabola x = 4 y 2 if the density ρ( x .
Exercises
1. and y ≥ 0} . y ) = y + 1 having the shape of the area bounded by the line y = 1 and the parabola y = x 2 . 2 x= a b .Also. In this case of constant density. intimately related though they be.
3.
2. Thus. the center of mass thus depends only on the geometry of the plate. while the centroid is an abstract mathematical something. One must never confuse the two concepts. Find the center of mass of a plate of density ρ( x . It is. nevertheless. Find the centroid of the semicircular region {( x . they are different.
13.4
. 3 3
Meditate on the fact that the location of the center of mass does not depend on the value of the constant k. M = ∫∫ kdA = k ∫∫ dA = k
T T
ab . Note that in general. I believe. It is called the centroid of the region. This is what most of our intuitions tell us. if the density is constant. y ) ∈ R 2 : x 2 + y 2 ≤ a 2 . and y = . The center of mass is something a physical body has.
The area of a region A is
∫ ∫ dydx + ∫ ∫ dydx .
6. Find the centroid of the region bounded by the curves y 2 = 2x . y )dA .5
. b)Find the point on the semicircular boundar of D that is closest to the centroid. The average value A of f on D is defined to be A = 1 f ( x .
13. Let D be the region inside the circle x 2 + (y − a) 2 = a 2 that lies below the line y = a .
b)Find the average depth of a bowl having the shape of the part of the paraboloid z = x 2 + y 2 − 1 below the xy plane.
13. (That is.
8. x + y = 4 .
0 x2 −4 0 0
2
0
4
x
Draw a picture of the region. and y =0.4. area of D ∫∫ D
a)Find the average depth of a bowl having the shape of the bottom half of the sphere x 2 + y2 + z 2 = 1 . a)Find the centroid of D.
7. the region between x = 0 and x = π bounded above by y = sin x and below by y = 0. Let f :D → R be a function defined on a nice subset D ⊂ R 2 .2 Polar Coordinates Now we shall see what happens when we express a double integral as an iterated integral in some coordinate system other than the usual rectangular. Find the centroid of the region bounded by the horizontal axis and one arch of the sine curve. or Cartesian.)
5.
Now we divide the plane into regions formed by the curves r = constant and θ = constant . we got rectangles. we know that
Now. and y = r sin θ . a typical region looks like
13. etc. Suppose we have the integral we must substitute x = r cosθ .coordinate system.. etc. There is.6
. This results in funny shaped regions:
∫∫ f ( x. y) dA . let's look at what happens in polar coordinates. When we divided the plane into regions formed by the curves x = constant and y = constant. We shall see more of this later. where r and θ are the usual polar coordinates. right now. however. more to it than this.
D
In polar coordinates.
and our iterated integral looks like
∫∫ f ( x. y)dA = ∫∫ f (r cosθ . Here is a picture drawn by Maple:
13.
Example Let's find the centroid of the region enclosed by the curve whose equation in polar coordinates is r = 1 + cosθ .) We desperately need to see an example. of course. r sin θ )rdrdθ
D
together with the appropriate limits of integration. integrate first with respect to θ and then with respect to r if this is convenient.7
.The area of this region is thus something like ∆A ≈ r∆r∆θ . (We may.
Finally.The centroid ( x . y ) is given by
x=
∫∫ xdA
D
∫∫ dA
D
. It should be clear that
∫∫ ydA = ∫ ∫ r
D 0 0
2 π 1+cos θ 2
sin θ drdθ = 0 . let's find the integral
∫∫ xdA . and so our iterated integral looks like
∫∫ xdA =
D
2 π1 +cos θ
∫
0
∫ r cosθ rdrdθ =
0
2π 1+ cosθ
∫ ∫r
0 0
2
cosθ drdθ . Now. and y =
∫∫ ydA
D
∫∫ dA
D
. when we hold θ fixed and integrate first with
D
respect to r. the lower limit is independent of θ and is always r = 0 . We have a slice for each value of θ from θ = 0 to θ = 2π . of course on θ and is r = 1 + cosθ . while the upper limit depends.
13.8
.
It is downhill all the way now:
2 π1+ cosθ
∫ ∫
0 0
1 r cosθ drdθ = ∫ (1 + cosθ ) 3 cosθ dθ 30
2
2π
1 2π = ∫ [cosθ + 3 cos2 θ + 3 cos3 θ + cos 4 θ ]dθ 3 0 1 3 1 = [ 0 + ∫ (1 + cos 2θ )dθ + 0 + ∫ (1 + cos 2θ ) 2 dθ ] 3 2 0 40 π 1 2π 2 π π 15π 5 =π + + ∫ cos 2θ dθ = π + 6 + 12 = 12 = 4 π 6 12 0
2π 2π
Now for the other integrals.
First.
Find the area of the region enclosed by the curve with polar equation r = sin2θ .
5 π 4 = 5 . where 0 ≤ α ≤ centroid as α → 0 ? π .∫∫ dA = ∫
D 0
2 π1 +cos θ
1 2π 2 ∫ rdrd θ = 2 ∫ (1 + cosθ ) dθ 0 0 1 2π (1 + cos 2θ )dθ 4∫ 0 π 3 = π 2 2
=π + =π + We are. x= 3 π 6 2
Exercises 9. done.
10. where R is the semicircular region bounded above by
y = 1 − x 2 and below by the x axis. Find the centroid of the region in the first quadrant inside the circle r = a and between the rays θ = 0 and θ = α .
11. where D is the region in the first quadrant inside
D
the circle x 2 + y 2 = a 2 and below the line y = x 3 . Evaluate
∫∫ e
R
x 2 + y2
dA .
13. What is the limiting position of the 2
12. Evaluate the integral
∫∫ ( x + y )dA .9
. at last. and y = 0.
where D is a nice subset of R 3 . i i
(The
summation is over all boxes that meet D.)
13. 14.13. y . y * . then we say that f is integrable over D. (We choose the xy plane as an example. choose a point ( xi* . A Riemann sum S now looks i i like S = ∑ f ( xi* . First.e. project D onto a coordinate plane. z* ) in D. z)dV . Find the area enclosed by one leaf of the rose r = cos3θ . z* ) was chosen.
13. y * . a rectangular parallelepiped). 3 Three Dimensions We move along to integrals in three dimensions. Find the area of the region inside r = 1 + cosθ and outside r = 1. The idea is quite simple. and the number L is called the integral of f over D. Capture D inside a big box (i.10
. This integral is usually written with three snake signs:
∫∫∫ f ( x .
D
Let's see how to evaluate such a thing by considering iterated integrals. zi* )∆Vi . Now subdivide this box by partitioning each of its sides. The volume of the largest such box is called the mesh of the subdivision. Here's what we do.
Suppose we have a function f : D → R . y i* .) If there is a number L such that  S − L can be made arbitrarily small by choosing a subdivision of sufficiently small mesh.
i =1 n
where ∆Vi is the volume of the box from which ( xi* .. In each box that meets D.
z)dz dA . A g ( x .Let A be the region in the xy plane onto which D projects.
h( x . (0. In other words.11
.0).0. y )
Now we simply integrate the integral
g ( x .0). y ) ∈ A enters D through the surface z = g( x . z)dV = ∫∫ ∫ f ( x.1). y ) and z = h( x . y . the blob D is the solid above the region A between the surfaces z = g( x .
13. Assume that a vertical line through a point ( x . z)dz over the region A:
h ( x. and (0.0. y ) .2. (1. where D is the tetrahedron with vertices
D
(0. y. y ) .0).0. y ) f ( x .y )
∫∫∫
D
Example Let's find the integral
∫∫∫ ( x + 2 y + z)dV . y .y )
∫ f ( x . y ) and exits through the surface z = h( x .
12
.
0
13. We now write the double A 0 integral over A as an iterated integral. and we have
1 − x − y /2 ( x + 2 y + z)dV = ∫∫ ∫ ( x + 2 y + z )dz dA ∫∫∫ D A 0
=∫
0
1 2 ( 1− x) 1− x − y/ 2
∫
0
∫ ( x + 2 y + z)dzdydx . the bottom of D is the surface z = 0 and the top of D is x + y y + z = 1. or z = 1 − x − . The projection is simply the triangle 2 2
1− x − y /2 Our iterated integral is thus simply ∫∫ ∫ ( x + 2 y + z) dz dA .When we project D onto the xy plane.
but look at the intermediate steps.
∫ ( x + 2 y + z )dz :
0
Thus. we want int(x+2*y+z. 2 2 8 2
and our next integral is
2 (1 −x ) 1− x −y / 2
∫
0
∫
0
(1 + 2 y + z)dzdy =
2 (1 −x )
∫
0
1 3 7 1 ( − x 2 − 2 xy + y − y 2 + )dy .
2 (1 −x )
∫
0
1 3 7 1 2 5 ( − x 2 − 2 xy + y − y 2 + )dy = − 4 x − x 3 + 3x 2 + . 2 2 8 2
Maple again:
int((x^2)/22*x*y+(3/2)*y(7/8)*y^2+1/2.
1 − x − y /2
For the first integration. it is traditional to omit the parentheses in the iterated integral.
Thus. All we need do now is integrate three times.1).
1 − x − y /2
∫
0
1 3 7 1 ( x + 2 y + z)dz = − x 2 − 2 xy + y − y 2 + .z=0. int(4*x(2/3)*x^3+3*x^2+(5/3).2*(1x)).
13.13
. 2 2 8 2 3 3
and finally.x=0. rather than just use one statement. Here we go.Again...y=0.(1xy/2)). Let's use Maple for the calculations..
y.
13. z)dV
S
M
y=
∫∫∫ yρ( x .0. the volume V of the solid is simply V = ∫∫∫ dV . y .
0
1
We make a few obvious observations. z) dV . Find the volume of the tetrahedron having vertices (0. then the mass M of the blob is M = ∫∫∫ ρ( x . Evaluate
∫∫∫ ( xy + z
S
2
)dV .0).
17. If the mass density of a blob having the shape of S is
S
ρ( x .14
. (a.c).0). and (0. y . and the location
S
( x . where S is the set S = {( x . First.(0. y . z ) of the center of mass is given by
x=
∫∫∫ xρ ( x.b. y .0).At last!
∫ ∫
0 0
1 2 ( 1− x) 1 −x − y /2
∫ ( x + 2 y + z) dzdydx = 2 .0. z):0 ≤ z ≤ 1−  x−  y} . y .0. y . Find the centroid of the tetrahedron in the previous exercise. z) dV
S
M
Exercises
15. z) . if S is a solid.
16. z)dV
S
M
z=
∫∫∫ zρ( x .
13. A solid is bounded below by the surface z = 4 y 2 .18. Find the centroid. Find the volume of the region in the first octant bounded by the coordinate planes and the surface z = 4 − x 2 − y . Write six different iterated integrals for the volume of the tetrahedron cut from the first octant by the plane 12 x + 4 y + 3z = 12 . Find the volume of the region common to the interiors of the cylinders x 2 + y 2 = 1 and x 2 + z 2 = 1 .
19. above by the surface z = 4 . and on the ends by the surfaces x = 1 and x = −1.
20.15
.
21.
or partition.
i= 1
n
Then we all believe that the "limit" of these sums as we choose finer and finer partitions of the curve should be the actual. How might we approach such a problem? Simple enough! We subdivide. rn } . The notion of an integral of a function f : D → R in which D is a nice one dimensional set.1 Scalar Line Integrals Now we again consider the idea of the integral in one dimension. On the subarc joining ri −1 to ri .1
. We subdivide the curve as in the preceding discussion and choose a point ri* on the subarc joining ri −1 to ri . To get some idea of why one might care about such a thing. we considered only functions defined on nice subsets of the real line. say
{r0 . consider the simple problem of finding the mass of a piece of wire having the shape of an arc of a space curve C and having a given density ρ(r) . we choose a point. Suppose f :C → R is a function whose domain C is a curve (in R 2 or R 3 . Let's abstract the essence of the discussion. Now we multiply this times the length of the line segment joining the points ri −1 and ri for an approximation to the mass of this arc of our curve. say ri* . but is not a subset of the reals is our next object of study. The sum
14. Then sum these to obtain an approximation for the total mass:
S = ∑ ρ( ri* ) ri − ri −1  . the curve with a finite set of points.Chapter Fourteen
One Dimension Again
14. r1 . and evaluate the function ρ( ri* ) . When we were introduced to the integral back in elementary school. or wherever). K. honesttogoodness mass of the wire.
then the points {r(t 0 ). for a ≤ t ≤ b . t n = b} is a partition of the interval. t 1 . b] : If {a = t 0 . but how do find such an integral? It is remarkably simple and easy.
∫ f ( r)dr . K. say r(t ).S = ∑ f (ri* ) ri − ri−1 
i =1
n
again is called a Riemann sum. and the number L is called the integral of f on C and is denoted also frequently referred to as a line integral. We partition the curve by partitioning the interval [a .2
. If there is a number L such that all Riemann sums are arbitrarily close to L for sufficiently fine partitions.K .
C
This integral is
This is wonderful. r( t n )} partition the curve C. t i ] and letting i ri* = r(t i* ) . Our Riemann sum now looks like
S = ∑ f ( r(t i* ) r(t i ) − r( ti −1 ) . We obtain the point ri* on the subarc joining r(t i −1 ) to r(t i ) by choosing t * ∈ [t i −1 . Suppose we have a vector description of the curve C. r(t 1 ). then we say f is integrable on C.
i= 1
n
14.
multiply the terms on the right by one. dt
Example Suppose we have a wire in the shape of a quarter circle of radius 2. and so it is not hard to convince oneself that the ∆t →0 ∆t i dt
"limiting" value of the Riemann sums is
∫ f ( r(t ))
a
b
dr(t ) dt . where C is the quarter circle:
C
14. y ) = y .Next. where. ∆ti
We know that lim
r (t i ) − r (t i − 1 ) dr = . but one disguised as ∆t i = t i − t i −1 . Then we see
∆t i . we know the mass is simply the integral
∫ ydr . Hence. ∆t i
S = ∑ f ( r(t i* )
i= 1
n
r(t i ) − r( ti −1 ) ∆t i . What is the mass of the wire? Well. dt
We have thus turned the problem into one we know how to solve—a plain old everyday elementary calculus integral. of course.
∫
C
f ( r)dr = ∫ f ( r(t ))
a
b
dr(t ) dt .3
. and the density of the wire is given by ρ( x .
where C is the curve
C
r(t ) = ti + (1 − t ) j + k . and the integral becomes simply dt
π .
Exercises
1. Evaluate the integral 0 ≤ t ≤ 1. Hence
∫ ydr = ∫
C 0
2
2 2 4−t dt = ∫ 2dt = 4 . for 0 ≤ t ≤ dr = − 2 sin t i + 2 cos t j = 2 .
0
Let's see what happens if we use a different vector description of the curve. 4− t2 0 2
We get. Thus we have 4
∫ ydr =
C
π/4
∫ 4 sin t dt = 4 . as we must.
14. say r(t ) = ti + 4 − t 2 j for 0 ≤ t ≤ 2 .4
. the same answer.A vector description of the curve is r(t ) = 2 cos ti + 2 sin tj .
∫ ( x − y + z + 2 )dr . We have dr t = i− j= dt 4 − t2 2 4 − t2 .
Find the mass of a wire having the shape of the curve r(t ) = (t 2 − 1) j + 2 tk . where C is the curve r(t ) = 4 cos ti + 4 sin tj + 3tk . You learned in fifth grade physics that the work done by a force F acting through a distance d is simply the product Fd. of course. suppose C is a space curve and f : C → R 3 is a function from C into the Euclidean space R 3 .2 Vector Line Integrals Now we are introduce something perhaps a little different from what we have seen to now—integrals with vector valued integrands. 3 2
if the density is ρ(t ) =
1 . really vectors.2. Specifically. let's think about a
physical model.
3. 0 ≤ t ≤ 2 . The force F and the displacement d are.
∫
C
2 2 x + y dr . t +1
6. What is
∫ dr ?
C
14.
4.
C
Why should we care about such a thing? Again. and we saw earlier in life that the "product" of the two is actually
14. Find the center of mass of a wire having the shape of the curve r(t ) = ti + 2 2 3 /2 t2 t j + k. 2
5. 0 ≤ t ≤ 1 if the density is ρ(t ) = 3 t.5
. Find the centroid of a semicircle of radius a. Evaluate the integral −2π ≤ t ≤ 2π . We are going to define an integral
∫ f ( r) ⋅ dr .
i= 1
n
The course should be obvious now. of course.the scalar. Now then. specified on the curve. we partition the curve by choosing a sequence of points {r0 . Exactly as in the scalar integrand case. How do we compute the work done in this situation? Let's see. we choose a point ri* on the subarc joining ri −1 to ri .
C
This integral too is called a line integral. We let r(t ). One may think of specifying an orientation by simply putting an arrow on the curve—it thus makes sense to speak of the initial point and the terminal point of the curve. and the limiting value of the sums is the integral
∫ F ( r) ⋅ dr . rn } on the curve. the work done in going from ri −1 to ri is approximately the scalar product F ( ri* ) ⋅ (ri − ri−1 ) . a ≤ t ≤ b . product of the two vectors. Now.) The discussion proceeds almost exactly as it did in the previous section and we get
14.6
. K. in general. or direction. be a vector description of C. and evaluate F( ri* ) . there is an orientation. Now. r1 . we sometimes speak of scalar line integrals and vector line integrals. To prevent confusion. Add all these up for an approximation to the total work done:
S = ∑ F (ri* ) ⋅ ( ri − ri −1 ) . (Here r(a ) is the initial point and r(b) is the terminal point. Once more. and we will have a variable force F(r) acting along a curve C in space. we take finer and finer partitions. How to find such a vector integral should be clear from the discussion of scalar line integrals. with r0 being the initial point and rn being the final point. neither of these quantities will be constant. or dot.
0) t0 (1. dt
Example Find
∫ [( xy + z
C
2
) i + ( x + z) j + 2 yzk ] ⋅ dr .7
.3).2. where C is the straight line from the
origin to the point (1.2. 3
Nothing to it. The line C has a vector description r(t ) = ti + 2tj + 3tk . and so dt
∫ [( xy + z
C
2
) i + ( x + z) j + 2 yzk ] ⋅ dr = ∫ [( 2t 2 + 9 t 2 ) i + (t + 3t ) j + 12t 2 k ] ⋅ ( i + 2 j + 3k )dt
0 1 1
1
= ∫ ( 2t + 8t + 36t )dt = ∫ (38t 2 + 8t ) dt
2 2 0 0
=
38 3 t + 4t 2 3
1
=
0
50 . Thus.∫ F ( r) ⋅ dr = ∫ F (r( t )) ⋅
C a
b
dr dt .
Another Example Now let's integrate the same function from (0. but this time along the path P in the picture:
14.0. dr = i + 2 j + 3k .3).
y . 0 ≤ t ≤ 1. so our integral is the 1 2 sum of three integrals:
∫ F ( x . t .
P1 0 0
1
1
For P2 . z) ⋅ dr = ∫ F ( t . y . z ) ⋅ dr = ∫ F (1. we have r(t ) = i + tj .Here the path P is the union of the three nice curves. z) = ( xy + z 2 ) i + ( x + z) j + 2 yzk .
A vector description of P1 is simply r(t ) = ti . Thus
. This gives us
P2
∫ F ( x . y. 0 ≤ t ≤ 2 . x ) ⋅ dr . and P3 .
0 0 0
2
2
2
14.8
. y.
P P1 P2 P3
where F ( x . P . x ) ⋅ dr + ∫ F ( x. P .00 ) ⋅ i dt = ∫ tj ⋅ i dt = 0 . y .0) ⋅ j dt = ∫ (t i + j ) ⋅ j dt = ∫ dt = 2 . y . y . ∫ F ( x . x ) ⋅ dr + ∫ F ( x . x ) ⋅ dr = ∫ F ( x .
y .Finally.0) to (1. ∫ F ( x .0) to (1. z) ⋅ dr = ∫ F (12 .
11.1.
∫ (cos x i − y j) ⋅ dr
C
where C the part of the curve y = sin x from (0. for P3 . y . y . 14. there is r(t ) = i + 2 j + tk .1.
10. and so
P3
. y ) =
1 (− yi + x j) one time around the circle x 2 + y 2 = a 2 in the 2 x +y
2
counterclockwise direction.0) to (1. t ) ⋅ k dt = ∫ [( 2 + t 2 )i + (1 + t ) j + 4t k] ⋅ k dt
0 0
3
3
= ∫ 4t dt = 18 . Integrate F ( x . y . Evaluate (π. Evaluate ∫ [ xyi + x 2 j ] ⋅ dr . z) = ( x − z) i + ( y − z) j − ( x + y ) k along the polygonal path from (0.0. we have then
∫ F ( x . where C is the arc of the curve y = x 2 from (0.0) to (1.9
.
P
Exercises
7. Evaluate the line integral of F ( x .0.0).
C
8. z) ⋅ dr = 0 + 2 + 18 = 20 .2) along the line segment joining these two points. z) = xyzi + ( xy + yz ) j + z 2 k from (0.1.0. 0 ≤ t ≤ 3 .
0
3
At last.0) to
9.0) to (1. Evaluate the line integral of F ( x .1).1).
14.
∫ F ( r) ⋅ dr = ∫ F (r( t )) ⋅
C a
b
dr dt . As we have already seen. of g is the integrand F : ∇g = F . Then. Let r(t ).3 Path Independence Suppose we evaluate the vector line integral
∫ F ( r) ⋅ dr . The line integral is path independent. precisely the integrand in our line integral:
b dr ∫ F ( r) ⋅ dr = ∫ F ( r(t )) ⋅ dt dt = ∫ h'( t ) dt = h(b) − h(a ) = g( p) − g (q ) .14. The result is esthetically pleasing and is clearly the lineal descendant of the fundamental theorem of calculus we learned so many years ago. we have r(a ) = p and r(b) = q . The value of the integral depends only on the values of the function g at the endpoints. a ≤ t ≤ b . Next let's use the Chain Rule to compute the derivative of the composition h(t ) = g ( r(t )) : h'( t ) = ∇g ⋅ dr dr = F ( r(t )) ⋅ . or gradient. of course. C a a b
This is a very exciting result and calls for some meditation. mirabile dictu.10
. Note that the curve C has completely disappeared from the answer. dt dt
This is. be a vector description of C. where C is a curve from
C
the point p to the point q. dt
Now let us make the very special assumption that there exists a realvalued (or scalar) function g: R 3 → R such that the derivative. the path from p to q does not affect the answer.
Specifically. Choose a point p = ( x 0 . y. of course. y 0 . Thus ∂x∫ C
14. z) in such an open ball. It turns out. y . now we compute the partial derivative ∂g . Let's suppose the domain D of the function F: D → R 3 is open and connected (Thus any two points in D may be joined by a nice path. that the converse of this is true. Okay. then there is a function g such that F = ∇g . and let L be the straight line segment from s to q .A moment's reflection on the examples we have seen should convince us that a lot of integrals are not path independent. z) ∈ D . thus many very nice functions F (or vector fields ) are not the gradient of any function.
independence of the integral. Let's see why this is so. Now let's integrate F from p to s by going along any curve C from p to q and then along L from q to s:
g ( s) = g( x . y . The domain D is open and hence includes an open ∂x
ball centered at s = ( x. then the integral of F between any two points of D does not depend on the path between the two points. z) to be the We are assuming path
integral from p to s along any curve joining these points. y .11
. A function F that is the gradient of a function g is said to be conservative and the function g is said to be a potential function for F. so it matters not what curve we choose. z) = ∫ F ( r) ⋅ dr + ∫ F ( r) ⋅ dr . Then. L lies in D.
C L
The first integral on the right does not depend on x. ∂x ∂x∫ L
∂ F ( r) ⋅ dr = 0 . Now define g ( s) = g( x .) We have just seen that if there exists a function g: D → R such that F = ∇g . and so ∂g ∂ = F ( r) ⋅ dr . if every integral of F in D is path independent. Choose a point q = ( x1 . z0 ) ∈ D . as we shall see.
and our dt
∫ F ( r) ⋅ dr = ∫ f
L x1
x
1
( t .( A closed path. z)dt . We are almost done. z) . y. and every path independent. y . ∂x∫ ∂ x x∫ 1 L 1 Hence ∂g = f1 . for note
It should be clear to one and all how to show that giving us the desired result: F = ∇g . ∂y
13. This is easy. then
∫ F ( r) ⋅ dr
C
is
∫ F ( r) ⋅ dr = 0 for every closed path in D.12
. Prove that
∂g = f 2 . Prove that if F: D → R 3 . where D is open and connected. y . Suppose
L
F (r ) = f 1 (r ) i + f 2 ( r ) j + f 3 ( r ) k .
∂g ∂g = f 2 and = f 3 .We clearly need to find ∫ F ( r) ⋅ dr . x1 ≤ t ≤ x . Thus line integral becomes simply that now ∂ ∂ x F ( r) ⋅ dr = f (t . thus ∂y ∂z
Exercises
12.
A vector description of L is simply r(t ) = ti + yj + zk . ∂x
dr = i . where g and f 2 are as in the preceding discussion. or
P
14. z)dt = f 1 ( x.
0 ≤ t ≤ 1 . a)Find a potential function g for the function F ( r) = yzi + xzj + xyk .]
14.13
.
15. b)Evaluate the line integral ∫ F ( r) ⋅ dr . a)Find a potential function g for the function F ( r) = e y+ 2 z (i + x j + 2 xk ) . Prove that if F: D → R 3 . then every
∫ F ( r) ⋅ dr = 0 for
P
∫ F ( r) ⋅ dr
C
is path independent . where D is open and connected. 0 ≤ t ≤ π . z) = e x sin y + 2xy − y 2 . where C is the curve
C
r(t ) = t cos 2t 2 i + 4t j + e 2 t k . and every closed path in D. [Really good hint: Find the gradient of g ( x .
17. b)Evaluate the line integral ∫ F ( r) ⋅ dr . b)Find another potential function for F in part a).curve. y . where C is the curve
C
r(t ) = (e t sin t )i + t 2 e 3 t j + cos3 t k .) [Physicists and others like to use a snake sign with a little circle superimposed on it
∫
to indicate that the path of integration is closed. Evaluate
∫ [( e
E
x
sin y + 3y ) i + (e x cos y + 2x − 2 y ) j ] ⋅ dr
where E is the ellipse
4 x 2 + y 2 = 4 oriented clockwise.]
14. is one with no endpoints.
16.
y ≤ 1 . and for all (s.Chapter Fifiteen
Surfaces Revisited
15. As the point (s. t ) ∈ D . We suppose r is a nice function.
15. t ) ∈ D moves around in D.1 Vector Description of Surfaces We look now at the very special case of functions r: D → R 3 . t ) = si + t j + ( s2 + t 2 ) k .1
. and D = {(s. the nose of this vector will trace out a surface in threespace. Look. It is sometimes helpful to think of the function r as providing a map from the region D to the surface. This is. where D ⊂ R 2 is a nice subset of the plane. for example at the function r: D → R 3 . we get the part of the paraboloid above the square −1 ≤ x . t ≤ 1} . of course. exactly the two dimensional analogue of the vector description of a curve. if we place the tail of the vector r( s.
The vector function r is called a vector description of the surface. It shouldn't be
difficult to convince yourself that if the tail of r( s. t ) is at the origin. where r( s. t ) at the origin. t ) ∈ R 2 :− 1 ≤ s. then the nose will be on the paraboloid z = x 2 + y 2 .
What have we here? First. We have a map of the sphere on the rectangle
15.2
. Let's look at another example. let
r( s. and for a surface. t ) 2 = (cos s sin t ) 2 + (sin s sin t ) 2 + (cost ) 2 = sin 2 t (cos2 s + sin 2 s) + cos2 t = sin 2 t + cos2 t = 1
Thus the nose of r is always on the sphere of radius one and centered at the origin. Notice next. s is the longitude of r( s.) A moment's reflection on this will convince you that as r is a description of the entire sphere. t ) = cos s sin t i + sin s sin t j + cos t k . r is a function from a nice piece of the real line into three space. notice that
 r(s.
for 0 ≤ t ≤ π and 0 ≤ s ≤ 2π . (More precisely. t ) . t ) . Here. and the variable t is the latitude of r( s. t is colatitude. or parameter.For a curve. r is a function from a nice piece of the plane into three space. that the variable.
∂t
way that the vector
15. t ). t ) is a curve on the surface that passes through he point p. ( s. t ) is tangent to the curve r(s . We see in the same ds ∂ s ∂r (s . Now. t ) be a point on S. t ) ∈ D be a vector description of a surface S.0) ) is mapped by r onto the North Pole.0) to (2π . Thus the vector dc ∂ r = ( s . Let r( s. t ) at p. t ) is tangent to this curve at the point p.3
.Observe that the entire lower edge of the rectangle (the line from (0. and let p = r( s. c( s) = r( s. while the upper edge is mapped onto the South Pole.
x . ∂s ∂t 0 1 2t
Meditate on the geometry here and convince yourself that this result is at least reasonable. ∂s ∂t
∂r ∂r and are thus tangent to S.
3.
2. ∂s ∂t
Example Let's find a vector normal to the surface given by the vector description r( s. ∂s ∂t ∂r and ∂s
The normal N is i j k ∂r ∂r N= × = 1 0 2 s = − 2 si − 2tj + k . t ) = si + t j + ( s2 + t 2 ) k at a point. the vectors Hence the vector ∂r ∂r × is normal to S. y ≥ 0.
15. Give a vector description for the ellipsoid 4 x 2 + y 2 + 8z 2 = 16 . and = j + 2tk . t ) on the surface S. We need to find the partial derivatives ∂r : ∂s ∂r ∂r = i + 2sk .
Exercises 1. Give a vector description for the cylinder x 2 + y 2 = 1 .4
.At the point p = r( s. Give a vector description for the surface z = x + 2 y 2 .
5. 0 ≤ t ≤ 2π . Find an equation of the plane that contains the point (1. t ) = s cos t i + s sin tj + sk .− ).
7. z) = 0 ) of the plane tangent to the sphere x 2 + y 2 + z 2 = a 2 at the point ( a a a . Describe the surface given by r( s.4. 3 3 3
8. Find a scalar equation (I.3) and is parallel to the plane tangent to the surface r( s. .2. or show there are no such points. . t ) = (s 2 + t 2 )i + (s + 3t ) j − stk at which the tangent plane is parallel to the plane 5x − 6 y + 2z = 7 . 4.2 Integration Suppose we have a nice surface S and a function f : S → R defined on the surface. 0 ≤ t ≤ 2π . t ) = (s + t )i + s2 j − 2t 2 k at the point (1.e.
15. We want to define an integral of f on S as the limit of some sort of Riemann sum in the way in which we have already defined various integrals. a vector description) of the line normal to the sphere x 2 + y 2 + z 2 = a 2 at the point ( a a a .− ). Here we have a slight problem in that we really are not sure at this point exactly what we might mean by the area of a
15. Describe the surface given by r( s. of the form f ( x.2. Find all points on the surface r( s.e.18).. 1 ≤ s ≤ 2 ..
10. y. Give a vector description for the sphere having radius 3 and centered at the point (1. 3 3 3
9. t ) = s cos t i + s sin tj + s2 k .
6.5
. −1 ≤ s ≤ 1 .3). Find an equation (I.
. and then add up these approximations to get a Riemann sum. zi* ) ∈ Si . We need a vector description of S .
S
Now. how do find such a thing. We assume the surface is sufficiently smooth to allow us to approximate the area of a small piece of it by a small planar region. We subdivide S into a number of small pieces S1 .small piece of surface.6
. The surface S is subdivided by subdividing the region D ⊂ R 2 into rectangles in the usual way:
15. etc. and form the Riemann sum
R = ∑ f (ri* )∆Ai . we take finer and finer subdivisions. this limit is the thing we call the integral of f on S:
∫∫ f (r) dS . select points ri * = ( x i* . yi* . of course. say r: D → r( D) = S . S 2 .K . and if the corresponding Riemann sums have a limit. S n each having area ∆Ai .
i =1
n
Then. etc. Let's be specific.
and the images of the horizontal lines t = constant.7
. s = constant. form a family of "parallel" curves on the surface.The images of the vertical lines. also form a family of such curves:
Let's look closely at one of the subdivisions:
15.
t ) = ∑ f ( r( si . ti ) ∆ si ∆ t i ∂s ∂t
in the Riemann sums:
∂r ∂r R = ∑ f ( r( si . and we use the approximation ∂s ∂t
∂r ∂ r ∆Ai ≈ (si . t i ) × ( si .
S D
∂r
∂r
Example Let's use our newfound knowledge to find the area of a sphere of radius a . Observe that the area of a surface S is simply the integral we found a vector description of the sphere:
∫∫ dS . t )) ∂ s (s. t i ) as shown. t i ) ∆t i . t i ) × ( si .
15.
∫∫ f (r) dS = ∫∫ f ( r( s. t ) dA . t i )) (s i . t )∆t i . The lengths of the sides of this parallelogram are ∂r ∂r ( si . The area ∂s ∂t i i
∂r ∂r is then ( si .8
. ∂s ∂t i= 1
n
These are just the Riemann sums for the usual old time double integral of the function
n ∂ r ∂ r F ( s. t i ) ∆si × ( si . ti )∆si and ( s .We paste a parallelogram tangent to the surface at the point r( si . t ) × ∂ t ( s. t i ) ∂s ∂t i= 1
over the plane region D. t i ) × (s i . Thus. t i ) ∆si ∆t i . t i )) (s i .
S
In the previous section.
r( s. Compute the partial derivatives:
∂r = − a sin s sin t i + a cos s sin t j . t ) = a cos s sin t i + a sin s sin t j + a cos t k .9
.
Area = ∫∫ dS = ∫∫
S D
∂r ∂r × dA = ∫∫ a 2 sin t dA ∂s ∂t D
15. and ∂s ∂r = a cos s cos t i + a sin s cos t j − a sin t k ∂t
Then
i j ∂r ∂r 2 × = a − sin s sin t cos s sin t ∂s ∂t cos s cos t sin s cost
k 0 sint
= a 2 [ − cos s sin 2 t i − sin s sin2 t j − sin t cos t k ]
Next we need to find the length of this vector:
∂r ∂r × = a 2 [cos 2 s sin 4 t + sin 2 s sin 4 t + sin 2 t cos2 t ]1 /2 ∂s ∂t = a 2 [sin 4 t + sin2 t cos2 t ]1/2 = a 2 [sin 2 t (sin 2 t + cos2 t )]1/2 = a 2 sin t 
Hence.
0 ≤ t ≤ π and 0 ≤ s ≤ 2π .
This leaves us with just integral to
∫∫ zdS . z ≥ 0 . y . Second. Thus 2
15.=a
2
∫ ∫ sin t  dsdt
0 0 π 2
π 2π
= 2π a
∫ sin tdt = 4π a
0
2
Another Example Let's find the centroid of a hemispherical shell H of radius a.10
. The centroid ( x . This hemisphere has
the same vector description as the sphere. it should be clear from the precious example that evaluate:
∫∫ dS = 2π a
H
2
. except for the fact that the domain of r is the rectangle 0 ≤ s ≤ 2π .
H
Most of the work was done in the example before this one. Choose our
coordinate system so that the shell is the surface x 2 + y 2 + z 2 = a 2 . z ) is given by
x=
∫∫ xdS
H
∫∫ dS
H
y=
∫∫ ydS
H
∫∫ dS
H
and z =
∫∫ zds
H
∫∫ dS
H
. 0 ≤ t ≤ π . note from the symmetry of the shell that x = y = 0 .
First.
We know this integral to be given by
∫∫ f ( x. r ) = r cosθ i + r sin θ j . r sin θ ) ∂ θ × ∂ r dA .2 ∫∫ zdS = a H
π / 2 2π
∫ ∫ a cost ∂ s × ∂ t
0 0 π /2 2 π
∂r
∂r
dsdt
=a3
∫
0
∫ cos t sin t dsdt = 2π a 3 ∫ cost sin t dt
0 0 π /2 0
π /2
= π a 3 sin 2 t
= π a3
And so we have z =
πa3 a = . y) dS
S
over the surface S described by r. we are using the letters θ and r instead of s and t . Is this the result you expected? 2 2π a 2
Yet One More Example Our new definition of a surface integral certainly includes the old one for plane surfaces.
S D
∂r
∂r
Let's find the partial derivatives:
15.11
. Now consider an integral
∫∫ f ( x. Look at the "surface" described by the vector function
r(θ .
with r defined on some subset D of the θ − r plane. y) dS = ∫∫ f (r cosθ . For what we hope will be obvious reasons.
)
14.
15.12
. ∂r Thus. Find the area of that part of the Earth that lies North of latitude 45°. and ∂θ ∂r = cosθ i + sin θ j . r sinθ )rdA
S D D
∂r
∂r
. Find the centroid of that part of it which is in the first octant. 0
and we have
∂r ∂r × = r . y) dS = ∫∫ f (r cosθ .∂r = − r sin θ i + r cos θ j .
This should look familiar!
Exercises
11. Hence.
12. ∂θ ∂r
∫∫ f ( x. r sin θ ) ∂ θ × ∂ r dA = ∫∫ f (r cosθ .
13. Find the centroid of the surface given in Problem 11. i ∂r ∂ r × = − r sin θ ∂ θ ∂r cosθ j r cosθ sinθ k 0 = −rk . A spherical shell of radius a is centered at the origin. Find the area of that part of the surface z = x 2 + y 2 that lies between the planes z = 1 and z = 2. (Assume the surface of the Earth is a sphere.
Evaluate
∫∫ ( x + y + z)dS .
15. and z = a.15. where S is the surface cut from the paraboloid
y 2 + 4 z = 16 by the planes x = 0.13
. Find the area of the ellipse cut from the plane z = 2x by the cylinder x 2 + y 2 = 1 . Evaluate
∫∫ x
S
2 y + 1 dS . x = 1. and z = 0.
18. y = a .
16. where S is the surface of the cube cut from the first octant
S
by the planes x = a. a)Find the centroid of the solid right circular cone having base radius a and altitude h. b)Find the centroid of the lateral surface of the cone in part a).
17.
But suppose that S is placed at an angle to the stream lines instead of being placed normal to the them. dt
The result is slightly more complicated when various exciting changes are made. the rate of flow through S is dM = v a . The amount of fluid that flows through during the time between t and t + ∆ t is simply M (t + ∆t ) − M (t ) =  v a∆t . and we wish to know the rate at which the water flows across a rectangular surface S that is normal to the stream lines:
What is the rate at which the fluid flows through S? Let M (t ) denote the total volume of fluid that has passed through the surface at time t. dt
16.Chapter Sixteen
Integrating Vector Functions
16. Then we have dM = v ⋅ n a . Thus. where a is the area of S. for instance). 1
. Clearly there is nothing special about the surface's being a rectangle. where n is a unit normal to the surface S.1 Introduction Suppose water (or some other incompressible fluid ) flows at a constant velocity v in space (through a pipe.
then our rate will be the negative of this one. it may vary with position as well as time. 2
. suppose S is a surface. We are computing the rate of flow from one side of the surface to the other.n ).Observe that matters which unit normal to the plane surface we choose. as we sometimes say). We must thus specify an orientation of the surface. The surface is not restricted to being a plane surface. and so we have to specify the "sides". But this is not so! There are many surfaces for which an orientation does not exist. We say that an orientation for S is a continuous function n:S → R 3 such that n( r) is normal to S and  n( r) = 1 for all r ∈ S . or strip. which is the velocity of the incompressible fluid. If we choose the other normal (. Specifically. At first blush this looks simple enough. let's come to grips with the problem of specifying an orientation for S. How do we find the rate of flow through the surface S from one side to the other? First. together with an orientation—that is.
16. You may recall from grammar school a simple example of such a surface. A surface together with an orientation is called an oriented surface. and the velocity of the flow is not restricted to being constant in space. some means of specifying two "sides"—and suppose F ( r) is a function F: R 3 → R 3 . and the unsophisticated might guess that every surface has an orientation (or may be oriented.2 Flux Now. so to speak. Here is my feeble attempt to draw one:
16. let's look at the general situation. the socalled Möbius band.
Most conveniently now. and then add them to obtain yet another type of Riemann sum R = ∑ F (ri* ) ⋅ n∆ Ai . where D ⊂ R 2 . and looking at the curves r(s. we consider a vector description of the surface S: r: D → S ∈ R 3 . If
i= 1 n
these sums have a limiting value as the size of the subdivisions go to zero. t ) and r(s .Now we see about finding the rate of flow through the oriented surface S.
S
It should be clear now what we do to evaluate such an integral. 3
. We
subdivide S by subdividing the region D into rectangles formed by lines s = constant and t= constant. we suppose the subdivisions are small and approximate the rate of flow. t ) on the surface. the vector
16. ∆S i = F (ri* ) ⋅ n∆ Ai . The strategy should be oldhat by now. or flux. As usual. We subdivide S and look at "small" parallelograms tangent to the surface:
As we have done so often. this is what we call the integral of F over the oriented surface S:
∫∫ F ( r) ⋅ dS . through the subdivision by the rate of flow through the tangent parallelogram. exactly as we did in integrating a scalar function over a surface S.
product
∂r ∂r ∂r ∂r × gives us not only a vector such that × ∆s∆t is the area of the ∂s ∂t ∂s ∂t
approximating parallelogram. we simply replace negative. we obtain
∫∫ F ( r) ⋅ dS = ∫∫ F ( r( s. There is the slight problem of the orientation of S. ∂ s ∂ t i i i= 1
and. such as our fluid flow interpretation. t )) ⋅ ∂ s × ∂ t dA . This is generally called the flux of F over S.. as before.
S D
∂r
∂ r
The concept we have developed here is purely mathematical and is done independent of any physical interpretation.) We have in the ∂ t ∂s
Riemann sums.
∂r ∂r × . Thus ∂r ∂r × may not point in the direction of ∂s ∂t ∂r ∂r × by its ∂s ∂t
the specified orientation. Use the
vector description of S we used in the first Example of the previous section:
16.
n ∂ r ∂ r R = ∑ F (ri* ) ⋅ × ∆ s ∆t . the flux of an electric field. and let F ( r) = c r . (We may think of just reversing the roles of s and t. etc. Let's find  r 3
∫∫ F ( r) ⋅ dS
S
. 4
. but also one which is normal to the surface. There are many physical interpretations of this concept. What we have is just an integral of a vector function F (or field) over an oriented surface S. gravitational flux. in which case. etc. you have perhaps seen some of them in elementary school physics. magnetic flux.
Example Let S be the sphere of radius a oriented so that the normal points "out" of the sphere. There is electric flux. of course. where c is a constant.
r( s. ∂s ∂t
Modest meditation should convince you that this normal points into the sphere.
Exercises
16. Next. the integrand is given by F ( r) = and our integral becomes c c r = 3 a[cos s sin t i + sin s sin t j + cos tk ] . We have already found that
∂r ∂r × = a 2 sin t [ − cos s sin ti − sin s sin tj − cos tk ] .
0 0 0
π 2π
π
Note that the radius a of the sphere has disappeared—the value of the integral is independent of the radius of the sphere. 5
. 0 ≤ t ≤ π .
0 ≤ s ≤ 2π . t ) = a cos s sin ti + a sin s sin tj + a cos tk . and is thus the negative of the one we need for the specified orientation of S. 3  r a
∫∫ F ( r) ⋅ dS = ∫
S
π 2π
0 0
∫a
c
2
[cos s sin ti + sin s sin tj + cos tk ] ⋅a 2 sin t [cos s sin ti + sin s sin tj + cos tk ]dsdt
= c∫ ∫ sin t[cos2 s sin 2 t + sin2 s sin 2 t + cos 2 t ]dsdt
0 0
π 2π
= c ∫ ∫ sint dsdt =2πc∫ sin t dt = 4π c .
0 ≤ z ≤ 1 . Find the flux of the function F ( x . x = 1. oriented so that the normal points upward.
4. and let n be the orientation of S such that n(r) ⋅ j > 0 for all r ∈ S . Find the flux
∫∫ [2 yj + zk ] ⋅ dS . Find
∫∫ [ zi + x
S
2
k ] ⋅ dS . 6
. y ):− 1 ≤ x ≤ 1.
5.1. Find the flux of F ( r) =
c r out of the surface of the cube − a ≤ x.
6. where c  r 3
and a are positive constants.
3.
2. y . z) = 4x i + 4 y j + 2 k outward through the surface cut from the bottom of the paraboloid z = x 2 + y 2 by the plane z = 1. y . and − 1 ≤ y ≤ 1} . z) = x i + zj out of the tetrahedron bounded by the coordinate planes and the plane x + 2 y + 3z = 6 . 1 ≤ x ≤ e . where S is that part of the surface z = x 2 + y 2 that lies above
the square {( x . z ≤ a . y .
S
16. and z = 0 . z) = z 2 i + xj − 3zk upward throught the surface cut from the cylinder z = 4 − y 2 by the planes x = 0. y . Let S be the surface defined by y = log x . Find the flux of F ( x . Find the flux of the function F ( x .
16. 7
.
where y 0 ≤ s ≤ y1 and z 0 ≤ t ≤ z1 . and let S5 and S 6 be the obvious things. of course. let's compute the flux
∫∫ F ( r) ⋅ dS . or field. let S 2 be the side in the plane x = x 0 . s is simply y. (Obviously. If F ( r ) = p( x . then
∫∫ F ( r) ⋅ dS = ∫ ∫ F ( x . t ) dtds
1
y1 z1
17. z 0 ≤ z ≤ z1 } .Chapter Seventeen
Gauss and Green
17. let S 3 be the side in the plane y = y1 . y . z ) k . Let S1 be the side in the plane x = x1 . and t is z .) Then ∂r ∂r × = j×k = i. For reasons that will become apparent as the drama unfolds. Let F : B → R 3 be a nice function. t ) = x1 i + sj + tk . s. z ): x 0 ≤ x ≤ x1 . z ) j + r ( x .1
. y . or rectangular parallelepiped. t ) ⋅ i dtds
1 S1 y 0 z0
y1 z1
=
y 0 z0
∫ ∫ p( x . simply r ( s. let S 4 be the side in the plane y = y 0 . ∂s ∂t It is clear this is the specified orientation. y .
S1
A vector description of S1 is quite easy to come by. given by B = {( x .
S
We shall do this by computing the surface integral over each of the six sides of B and adding the results. s. it is. y 0 ≤ y ≤ y1 . We begin by computing the integral
∫∫ F ( r) ⋅ dS .1 Gauss's Theorem Let B be the box. y . and let S be the surface of B with the orientation that points out of B. z ) i + q ( x .
t ) =
x1
x0
∫ ∂ x ( ξ . t ) ⋅ ( − i ) dtds . s.
Observe that p( x1 . s. x = x 0 . s. t )dξdtds
S10 S21 y0 z0 x0
y1 z1 x1
∂p
=
∫∫∫ ∂ x dV
B
∂p
and we have turned the sum of the two surface integrals into a plain ol' volume integral .s. is just r ( s .
∂p
Substitution of this into the previous equation gives us
∫∫ F ( r) ⋅ dS + ∫∫ F ( r ) ⋅ dS = ∫ ∫ ∫ ∂ x (ξ . and
S3 S4 B
∫∫ F ( r) ⋅ dS + ∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ z dV .
S5 S6 B
∂r
17. and we have
∫∫ F ( r) ⋅ dS = ∫ ∫ F ( x
S2 y 0 z0
y1 z1
0
. It should be clear how we also obtain ∂q
∫∫ F ( r) ⋅ dS + ∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ y dV . s.2
. s. t ) = x 0 i + s j + tk . s. t ) − p( x 0 .p(x
1 S1 S2 y 0 z0
y1 z1
0
. t)] dtds . t ) dξ . t) . t ) dtds
=
y 0 z0
∫ ∫ − p( x
y1 z1
0
The sum of these two is then
∫∫ F ( r) ⋅ dS + ∫∫ F ( r) ⋅ dS = ∫ ∫ [p(x . s .A vector description for the opposite side.
3
.
i Si i Bi
∂ p
∂q
∂ r
The sum on the right hand side is just the integral over Pn :
∑ ∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ x + ∂ y + ∂ z dV . together with a function F ( r ) = p( x . n} that meet B. The surface that bounds Bi will be called S i . in terms of an ordinary volume integral. z ) i + q ( x .2. and oriented so that the normal points out.K .The flux over the entire surface S is thus the sum of these: ∂p ∂q ∂r
∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ xdV + ∫∫∫ ∂ ydV + ∫∫∫ ∂ zdV
S B B B
(z) ∂ p ∂ q ∂ r dV = ∫∫∫ + + ∂ x ∂ y ∂ z B
We have now found the surface integral. y . Suppose parallelepipeds B j and Bk are adjacent. The union Pn = ∪{Bi }of all the Bi is thus an
approximation to the original solid B. suppose we have an "arbitrary" solid region B bounded by a surface S. z ) k defined on B. Consider those parallelepipeds {Bi : i = 1. y . or flux. y .
i Si Pni
∂ p
∂q
∂ r
Take a closer look at the sum of the surface integrals on the left hand side of this equation. Apply the equation (z) to each of these and sum the equations:
∑ ∫∫ F ( r ) ⋅ dS = ∑ ∫∫∫ ∂ x + ∂ y + ∂ z dV . and call the common side T :
17. z ) j + r ( x . Trap B in a box and subdivide the box into parallelepipeds. Now.
In the sum of surface integrals. once from the integral over S j .
S B
∂ p
∂q
∂ r
17. will. Then Pn → B and S n → S . have opposite signs because the
orientation of T has one direction as a part of the surface of B j and the opposite direction as a part of the surface of Bk . the integral over the common side T appears twice. and you see that what is left is precisely the integral over the boundary S n of Pn orientation. giving us precisely the same result for the arbitrary region B:
∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ x + ∂ y + ∂ z dV . as everyone can see coming.
Sn Pn
∂ p
∂q
∂ r
Now. These two terms thus sum to zero and cancel each other. we look at the limit of this equation as we take smaller and smaller subdivisions. however.4
. Mirabile dictu. These integrals. the surface of B j and once from the integral over S k . this is precisely the equation (z): with the outward pointing
∫∫ F ( r) ⋅ dS = ∫∫∫ ∂ x + ∂ y + ∂ z dV . In the sum of all the surface integrals. the surface of Bk . we are therefore left with only the integrals over sides that are not adjacent to another box. A moments reflection.
5
. y . y .
That's easy: F= and so p= q= r= cx . z ) k . First we need to see F in the form  r3
F ( x .This is really a big deal—such a big deal that it has its own name. z ) i + q( x . ( x + y 2 + z 2 ) 3/ 2
2
A bit of elementary school calculus (remember Mrs. Turner!). ( x + y 2 + z 2 ) 3/ 2
2
cz . Gauss’s Theorem looks like
∫∫ F(r) ⋅ dS = ∫∫∫ ∇ ⋅ F(r) dV
S
Example Let's find the divergence of F ( r ) = c r . ∂p ∂q ∂r . or ∇ ⋅ F . z ) = p( x . ( x + y 2 + z 2 ) 3/ 2
2
c ( x + y + z 2 ) 3/ 2
2 2
[ xi + yj + zk ] . or the Divergence Theorem. This is called Gauss's Theorem.
cy . and we have
17. It is usually designated either div F . The integrand in the volume integral also has a name. z ) j + r ( x . y . Thus. it is called the divergence of the function F. y . + + ∂x ∂y ∂z
div F = ∇ ⋅ F =
With this new definition.
Exercises 1. where S is the boundary of the solid inside the cylinder
S
x 2 + y 2 ≤ 1 between z = 0 and z = x 2 + y 2 . =c 2 ∂z ( x + y 2 + z 2 ) 5/ 2
Hence. of course.
3. =c ∂y ( x 2 + y 2 + z 2 ) 5/ 2 ∂r x 2 + y 2 + z 2 − 3z 2 . y . It is the gravitational field of a point mass fixed at the origin. y = ±4 . =c ∂x ( x 2 + y 2 + z 2 ) 5/ 2 x 2 + y 2 + z 2 − 3y 2 ∂q .6
. Find
∫∫ [ yi + xyj − zk ] ⋅ dS . Find
∫∫ [log( x
S
2
y 2z + y 2 ) i + tan −1 j + z x 2 + y 2 k ] ⋅ dS .
17.). ∇ ⋅ F = 0 everywhere (except. or the electric intensity field for a point charge fixed at the origin. with the outward pointing orientation. and z = ±4 . for r = 0. − 1 ≤ z ≤ 2} . Find the outward flux of the function F = ( y − x ) i + ( z − y ) j + ( y − x ) k across the boundary of the cube bounded by the planes x = ±4 . where F is not defined.
2. Gauss's Theorem now tells us that the integral of F over any closed surface that does not enclose r = 0 must be zero. This might be the hohum of the week save for the fact that the function F is a common one. or any field in which the magnitude is inversely proportional to the distance from the origin and which points in the direction of the origin. z ):1 ≤ x 2 + y 2 ≤ 2.x 2 + y 2 + z 2 − 3x 2 ∂p . where S is the boundary of x x
the solid {( x .
y1 ) ( x1 . Now.( x1 . C3 ←
C4 ↓
↑C2
→ C1 Thus.2 Green's Theorem Let R be the rectangular region in the plane bounded by the rectangle with vertices ( x 0 . y1 ).). Show that
∫∫ ∇f ⋅ dS = 0 . Let B a region in
R 3 .
17. y ) j . y 0 ).
S
17.7
. y1 )
( x0 .4. y0 )
Suppose F : R → R 2 is a vector function given by F ( x . We shall compute the integral in four parts: the integrals along each of the straight line segments making up the boundary. ( x 0 . y0 )
( x1 . and ( x 0 . let's compute the vector line integral of F around the rectangular boundary C in the counterclockwise direction. and let
f :B → R
be a function such that
∂2 f ∂2 f ∂2 f + + = 0 in B (Such a function f is said to be harmonic in B. y1 ) . y 0 ).( x1 . Let S ∂ x2 ∂ y2 ∂ z2 be the boundary of B. y ) = p( x . y ) i + q ( x .
1 x0
x1
Thus. Then. y )dt .
C C1 C2 C3 C4
We shall work out the evaluation of one of these in some painful detail. we find that ∂q
C2
∫ F ⋅ dr + ∫ F ⋅ dr = ∫∫ ∂ x dA . y 0 )]dt
C3 x0
x1
=
x0 y 0
∫ ∫ − ∂ y (t .
C1
∫ F ⋅ dr + ∫ F ⋅ dr = ∫ −[ p(t . y
0
)dt .
In a similar fashion.8
. y0 ) j] ⋅ i dt =
x0
x1
x1
x0
∫ p( t . x 0 ≤ t ≤ x1 . and our line integral becomes dt
C1
∫ F ⋅ dr = ∫ [ p(t . of course. we get
C3
∫ F ⋅ dr = ∫ − p(t . s)dsdt
R
x1 y1
∂p
= ∫∫ −
∂p dA ∂y
In essentially the same manner. y1 ) − p(t .∫ F ⋅ dr = ∫ F ⋅ dr + ∫ F ⋅ d r + ∫ F ⋅ d r + ∫ F ⋅ d r . it should then be rather obvious how to do the others. y0 ) i + q(t .
C4 R
17. dr = i . Start with a vector description of C1 : r (t ) = ti + y 0 j .
etc. When the dust settles.. just as we did with the parallelepipeds in deriving Gauss's Theorem. and add all the equations.
C R
∂ q
∂ p
This is called Green's Theorem. etc.9
. and apply the above recipe to each rectangle.Thus
∫ F ⋅ dr = ∫ F ⋅ dr + ∫ F ⋅ dr + ∫ F ⋅ dr + ∫ F ⋅ dr
C C1 C2 C3 C4
=
∂ q ∂ p ∫∫ ∂ x − ∂ y dA R
We have turned a one dimensional vector integral into a double integral.
17. similar to the way in which in the previous section we turned a two dimensional vector integral into a triple integral. we have the same result:
∫ F ⋅ dr = ∫∫ ∂ x − ∂ y dA . Now suppose we have a reasonable region R bounded by a reasonable curve C with a counterclockwise orientation:
Now cover this region with rectangles.. You should note that the same equation is valid even if the region R is bounded by more than one closed curve.
where C is the boundary of the halfdisc
2
C
x 2 + y 2 ≤ 9.10
.
∫ [5yi + 3( x + 1) j] ⋅ dr = ∫∫ ∂ x − ∂ y dA
C R
∂ q
∂ p
= −2 ∫∫ dA = −8π
R
Exercises 5. The line integral on the left side is simply the sum of the integrals over the pieces of the boundary curve.
17. note that ∂q ∂p = 3. oriented counterclockwise. where C is the circle of
C
radius 2 centered at the origin. y ≥ 0 oriented counterclockwise. ∂x ∂y Thus. and =5 . Evaluate ∫ [(sin x + 3 y 2 ) i + (2 x − e − y ) j ] ⋅ dr . First.Here the boundary C consists of three curves with the orientation indicated by the arrows in the fine picture—meditate on the covering by approximating rectangles and you will see why the orientation of the "inside" curves is clockwise.
Example Let's evaluate the line integral
∫ [5yi + 3( x + 1) j ] ⋅ dr .
where C is the curve given by
r (t ) = sin t i + sin 2t j . (These are the usual polar coordinates.11
. where C is the boundary of the region x
1 ≤ r ≤ 2. y k +1 ) . ( x 2 . How do we do this? We simply apply Greens’ Theorem to the function F( x.
R P
∂q
∂p
which becomes
∫∫ dA = ∫ xj ⋅dr . y n ). and we have
17. Evaluate the line integral
∫ [ ye
C
x2
i + x 3 e y j ] ⋅ dr . Then Green’s Theorem tells us that
∫∫ ∂x − ∂x dA = ∫ F( x. Let’s integrate along the line segment Lk from ( x k . y 2 ). 0 ≤ t ≤ 2π by using Green's Theorem. oriented clockwise. Thus r ' (t ) = ( x x +1 − x k )i + ( y k +1 − y k ) j .
R P
We thus find the area by evaluating the line integral on the right side.3 A Pleasing Application Here we shall use Green’s Theorem to find the area of a region R bound by a polygon P with vertices ( x1 . We simply integrate over each line segment of the polygon and add up the integrals.
17. Evaluate
∫ (tan
C
−1
y ) i + log( x 2 + y 2 ) j ⋅ dr . y1 ).K.6. This is easy. y )i + q( x.)
7. 0 ≤ t ≤ 1 . A vector description of this segment is r (t ) = (1 − t )( x k i + y k j) + t ( x k +1 i + y k +1 j) . y ) = p( x. y) ⋅ dr . ( x n . y ) j = xj . y k ) to ( x k +1 . 0 ≤ θ ≤ π .
17. y ) . 0).
9. 1). (2. 4). 3).K. (1. y 2 ). y1 ).Lk
∫ xj ⋅ dr = ∫ [(1 − t ) xk + txk +1 ]j ⋅ r' (t )dt
0
1
= ( y k +1 − y k ) ∫ [(1 − t ) x k + tx k +1 ]dt
0
1
=
( y k +1 − y k )( x k +1 + x k ) 2
Thus. 2). (2. 2).5
2
Area =
1 [(4 − 0)(2 + 0) + (7 − 4)(2 + 1) + (9 − 7)(−1 + 1) + (9 − 0)(−1 + 0)] = 4] 2
Exercises 8.
Example.5
0
0. 0). (1. 9):
8
6
4
2
1
0. By means of a clever choice of the function F( x. (1. It is really a very simple formula for the area enclosed by a polygon. and (1. y n ). ( x 2 . 7). 3). where R is the region enclosed by the polygon with
R
vertices ( x1 . Area =
∫∫ dA = ∑
R
( y k +1 − y k )( x k +1 + x k ) ( y1 − y n )( x1 + x n ) . + 2 2 k =1
n −1
Meditate on this result.12
. 0). (2. We shall find the area of the quadrilateral with vertices (0. Find the area enclosed by the octagon with vertices (0. (2. use Green’s Theorem and derive a recipe for the integral
∫∫ xdA . (0. 5). (1. ( x n .5
1 x
1.
By means of a clever choice of the function F( x. (2.
17. y n ). use Green’s Theorem and derive a recipe for the integral
∫∫ ydA . y 2 ).13
.10. Find the centroid of the region enclosed by the triangle with vertices (1. where R is the region enclosed by the polygon with
R
vertices ( x1 . ( x n . y ) . and (5.
11. 5). y1 ).K. 1). 8). ( x 2 .
z ) j + r ( x .1 Stokes's Theorem Let F : D → R 3 be a nice vector function. in which case. If F ( x . We want to look at this formula in case the region R is not necessarily in the ij plane. z ) = p( x . we orient things according to our familiar "righthand" rule. y .1
. z ) i + q ( x . ∂z r
This definition allows us to look at Green's Theorem from a new perspective by observing that in case F ( x .
C R
where we are thinking of the region R as an oriented surface with its orientation pointing in the direction of k. y . z ) k . the word "clockwise" doesn't help in deciding on the orientation of the boundary C. y ) i + q ( x .Chapter Eighteen
Stokes
18. Once again. y . Green's Theorem becomes
(♥)
∫ F ⋅ dr = ∫∫ curlF ⋅ dS .
18. y ) = p( x . i + ∂ x ∂ y ∂ y ∂ z ∂ z ∂ x ∂ ∂ ∂ +j +k provides a nice ∂x ∂y ∂z
Here also the socalled del operator ∇ = i memory device:
i ∂ curlF = ∇ × F = ∂x p
j ∂ ∂y q
k ∂ . y ) j . y . the curl of F is defined by ∂ q ∂ p ∂ r ∂ q ∂ p ∂ r − − − curlF = j + k .
and the sum of the line integrals is just the line integral around the boundary of the union of the pieces—as in the plane case. Then the orientation.)
Approximate the surface by a bunch of plane regions tangent to S . or direction. apply the equation (♥) to each of these approximating plane regions. etc. Then.
C S
18. (Here neither S nor C are assumed to lie in a plane. Suppose now S is any surface bounded by a finite number of disjoint curves C1 . of C j that is consistent with the given orientation of the surface S is the one that "lines up" with the direction on this little circle. Look at this picture:
The surface and its boundary in this case are said the be consistently oriented. Look at a surface S in three space bounded by C. and then sum these equations. we think of looking at the limit as we take more and more approximating regions.2
. The sum of the surface integrals is just the surface integral over the union of the approximating pieces. and we obtain the equation
∫ F ⋅ dr = ∫∫ curlF ⋅ dS . We say simply that C = C1 ∪ C2 ∪K∪Cn is the boundary of S. Now we do what we have done so many times in the past.Here's the way it goes. Now choose an orientation for the surface S. the line integrals over the boundaries of adjacent regions cancel. of course.K Cn . C2 . Look at one of these normal vectors "close" to a curve C j and imagine a little circle around the base of the normal oriented so that the normal vector points in the righthand direction with respect to the direction of the circle..
3
. looking in the direction of k. y ) such that x 2 + y 2 ≤ 1 .e. The surface integral
∫∫ curlF ⋅ dS looks like
S
18. The curve C bounds the part of the plane x + y + z = 1 that lies above the set of ( x . t ) = s cos ti + s sin tj + (1 − s cos t − s sin t ) j . Notice that this is the opposite of the orientation consistent with that specified for the curve C. This result is the celebrated Stokes's Theorem.
i ∂r ∂r × = cos t ∂s ∂t − s sin t
j sin t s cos t
k − (cos t + sin t ) s(sin t − cos t )
= si + sj + sk
I hope this result is no surprise..
Example Let's use Stokes's Theorem to evaluate the line integral
∫ [− y
C
3
i + x 3 j − z 3 k ] ⋅ dr . 0 ≤ s ≤ 1. and so we must use ∂r ∂r × = − s( i + j + k ) ∂t ∂s in our surface integral.
Hence. 0 ≤ t ≤ 2π .where S and C are oriented consistently.). A vector description is thus given by r ( s.
where C is the intersection of the cylinder x 2 + y 2 = 1 and the plane x + y + z = 1 oriented in the clockwise direction when viewed from above (i.
0 0
∂ r
∂ r
We must find curl F:
i ∂ curlF = ∇ × F = ∂x − y3
j ∂ ∂y x3
k ∂ = 3( x 2 + y 2 ) k . y . and S 2 = {( x . Compute the flux integral
∫∫ ∇ × F ⋅ dS . 0 ≤ z ≤ 1} . y .axis. z ): x 2 + y 2 = 1. where S1 = {( x . Let S be the surface S = S1 ∪ S 2 .∫∫ curlF ⋅ dS =
S
2π 1
∫ ∫ curlF ⋅ ∂ t × ∂ s dsdt .
S
where S has the orientation pointing away from the z. z ): x 2 + y 2 + ( z − 1) 2 = 1.4
. Let the function F be given by F( x. ∂z −z3
Hence.
∫∫ curlF ⋅ dS =
S
2π 1
∫ ∫ curlF ⋅ ∂ t × ∂ s dsdt
0 0 2π 1
∂ r
∂ r
=
∫ ∫ 3( s
0 0
2
)( − s)dsdt = −2π
3 3 =− π 4 2
Exercises 1.
18. y . z ≥ 1} . z ) = ( x 2 z 3 + y )i + ( xy + z ) j + (5 x z + y 4 )k .
Prove that
∫ (a × r ) ⋅ dr = 2∫∫ a ⋅ dS . Let F be a vector function such that
18.2.]
5. Suppose they are consistently oriented. Suppose a is a constant vector. Let S be a surface with boundary C . Suppose S is a surface with boundary C and F is a vector function such that ∇ × F is tangent to S at each point of S.
C S
[Remember. Let S1 and S 2 be two surfaces with a common boundary C.5
. where F( x. Draw a picture indicating the orientations these surfaces must have to insure that
∫∫ ∇ × F ⋅ dS = ∫∫ ∇ × F ⋅ dS
S1 S21
. Prove that ∫ F ⋅ dr = 0 .
C
6. r = xi + yj + zk . z ) = 2 yi + xj + zk
3. y . Let S be the hemisphere x 2 + y 2 + z 2 = 1.
4. Let r : B → R 3 be a vector description of the surface S with boundary C. b)Evaluate the flux
∫∫ ∇ × F ⋅ dS
S
. a)Describe the boundary of S and its orientation that is consistent with the orientation of S. z ≤ 0 with the orientation pointing toward the origin.
where C
C
is the circle x 2 + y 2 = 1 . [Hint: Evaluate the line integral
∫ F ⋅ dr .
18. Let F( x. If so. Prove that ∇ × F = 0 everywhere in D.2 Path Independence Revisited Problem 7 at the end of the previous section perhaps raised our hopes that an easy test for a function F to be conservative in a domain D is simply to see if ∇ × F =0. z ) : x 2 + y 2 > 1} is not. with the usual counterclockwise orientation. z = 0 . Thus for instance the region D = {( x.6
. z ) =
a)Compute ∇ × F . 2 2 x +y x +y
2
8. b)Prove that F is not conservative. these hopes were quickly dashed by Problem 8. y. In this section.∇×F =
∂r ∂r × . while the region R = {( x. y . we shall see just what we can do along this line. Suppose the vector function F on a domain D is conservative. z ) : x 2 + y 2 + z 2 < 1} is simply connected. ∂r ∂r ∂s ∂t × ∂s ∂t 1
Show that
C
∫ F ⋅ dc
= area of S.]
18. y . x2 + y 2 ≠ 0 . The concept introduced next provides the key to understanding and enlightenment. An open subset D of R 3 is called simply connected if every simple closed curve in D is the boundary of some surface contained entirely in D.
7. −y x i+ 2 j.
We show that F is conservative by showing that the integral of F around any closed curve is 0. 2 x +y x + y2
2
10. is conservative. then ∇ × F = 0 everywhere in D implies that F is indeed conservative. y . x > 0.7
.Now it easy to see that if F has as domain a simply connected region D. Find a potential function for the vector function F given in Problem 9. Explain how you know that F( x. Then D is simply connected.
S
How about that!
Exercises 9. Let C be any closed curve in D. This is easy to do. Now unleash Stokes’s Theorem:
C
∫ F ⋅ dr = ∫∫ ∇ × F ⋅ dS = 0.
18. so there is a surface S the boundary of which is C. z ) = −y x i+ 2 j .
of course
∫∫∫ ρdV
B
. then the mass in the region B bounded by S is. The region B is any region.
B S B
∂ρ
Thus. t ) is the density at r at time t.
The rate at which this mass is changing is simply ∂ ∂ρ ∫∫∫ ρdV = ∫∫∫ ∂t dV . z . where S
S
is given the outward pointing orientation. it is not hard to convince yourself that the flux integral
∫∫ ρv ⋅ dr
S
is the rate at which mass flows through the surface S.
Meditate on this result.
∫∫∫ ∂t dV = −∫∫ ρv ⋅ dr .Chapter Nineteen
Some Physics
19. y. y.
∫∫∫ ∂t
B
∂ρ
+ ∇ ⋅ ( ρv ) dV . if S is a closed surface.
B S
∂ρ
We now apply Gauss’s Theorem and get
∫∫∫ ∂t dV = −∫∫ ρv ⋅ dr = ∫∫∫ − ∇ ⋅ (ρv)dV . Now. y . Thus.1
. and suppose ρ ( x. z ) = xi + yj + zk of a fluid flowing smoothly through a region in space. and so it must be true that the integrand itself is everywhere 0:
19. ∂t B B This is the same as the rate at which mass is flowing across S into B: − ∫∫ ρv ⋅ dr . z . t ) is the velocity at r = ( x.1 Fluid Mechanics Suppose v ( x. If S is an oriented surface.
In case the fluid is incompressible.2
. Consider a one dimensional flow in which the velocity of the fluid is given by v = f (x ) . Suppose further that the density ρ of the fluid does not vary with time t. or ∂t ∇⋅v = 0. the continuity equation becomes quite simple. Show that ρ ( x) = where k is a constant. Then the electric field E q (r ) due to q is given by E q (r ) = kq r−s r−s
3
. It is called the equation of continuity. Thus. ∂t This is one of the fundamental equations of fluid dynamics.Pq . with a potential function Pq (r ) = − kq . Incompressible means simply that the density ρ is constant. Thus ∂ρ + ∇ ⋅ ( ρv ) = ∇ ⋅ ( ρv) = ρ∇ ⋅ v = 0. Exercise 1. that this field.2 Electrostatics Suppose there is a point charge q fixed at the point s. f ( x) ∂ρ = 0 and so we have ∂t
19.
It is easy to verify.
19. is conservative. where f ( x ) > 0 . k . r −s 
so that E q = ∇Pq . or function.∂ρ + ∇ ⋅ ( ρv) = 0 . so they define the electric potential Vq to be . Physicists do not like to be bothered with the minus sign in Pq . as we have done in a previous chapter.
Also. r −s
E q (r ) = −∇Vq (r ) . Suppose E j is the electric intensity due to q j . K . if the point charge is at s.
We have already seen that the flux out of a closed surface S is
∫∫ E
S
q
if S does not enclose the origin 0 ⋅ dS = if S does enclose the origin 4πkq
Some meditation will convince you there is nothing special here about the origin.3
.
∫∫ E ⋅ dS = 4πk ∑ q
S
j
where the sum is over those charges enclosed by S. suppose there are a finite number of point charges q1 at s1 .
19. that is. q 2 at s 2 . Finally.Vq (r ) = and
kq . and
E(r ) = −∇V (r ) . V (r ) = k ∑
j =1 n
qj r−sj 
. and q n at s n . Then it should be clear that the electric field due to these charges is simply the sum E(r ) = ∑ E j = k ∑ q j
j =1 j =1 n n
r−sj  r − s j 3
. then if S does not enclose s 0 ⋅ dS = if S does enclose s 4πkq
∫∫ E
S
q
Next.
r−s
Everything works nicely so that we also have E(r ) = −∇V (r ) . With the integral for the electric field.Things become more exciting if instead of point charges. Specifically. Let R be any region and let S be the closed surface bounding R.
U U
then things go wrong because the improper integral of the divergence does not exist. it is actually a limit of integrals.  r − s 3
But this appears to be a serious breach of decorum. If R is a solid region bounded by a closed surface S. and so there is no problem with 0 on the bottom of the integrand. If we simply try to calculate the divergence by div ∫∫∫ stuff dV = ∫∫∫ div (stuff) dV . then we can also integrate to get (*)
∫∫ E ⋅ dS = 4πk ∫∫∫ ρ (s)dV . we integrate not over everything but over everything outside a spherical solid region of radius a centered at r. Then
∫∫ E ⋅ dS = ∫∫∫ ∇ ⋅ E dV . Gauss saves the day.
S R
But from equation (*) we have
∫∫ E ⋅ dS = 4πk ∫∫∫ ρ (s)dV = ∫∫∫ ∇ ⋅ E dV . we are safe in writing for the potential V (r ) = k ∫∫∫
U
ρ (s) dVs . Thus what we see above is an improper integral—that is. In the same way. We then look at the limit as a → 0 of this integral.
S R
The divergence of E is the troublesome item in extending matters to distributed charge. we need to integrate: E(r ) = ∫∫∫ kρ (s)
U
(r − s ) dV s .4
.
S R R
This gives us
19. this limit exists. We are integrating over everything. and at s = r we have the dreaded 0 in the denominator. we have a charge distribution in space with charge density ρ . To find the electric field E(r ) produced by this distribution of charge in space.
or ∂ 2V ∂ 2V ∂ 2V + + = −4πkρ . ∂x 2 ∂y 2 ∂z 2 This is the celebrated Poisson’s Equation. and so it must be true that ∇ ⋅ E = 4πkρ for all r. a justly famous partial differential equation.∫∫∫ 4πkρdV = ∫∫∫ ∇ ⋅ E dV . we get ∇ ⋅ E = −∇ ⋅ (∇V ) = 4πkρ . remembering that E = −∇V . or
R R
∫∫∫ (∇ ⋅ E − 4πkρ )dV . the study of which is beyond the scope of this course.
.5
. ∇ 2V = −4πkρ .
R
But R is any region.
19. Finally.