THE JACOBIAN MATRIX

A Thesis
Presented to
the Department of Mathematics
The Kansas State Teachers College of Emporia

"*'

In Partial Fulfillment
of the Requirements for the Degree
Master of Arts

by

-
Sheryl A. Cole
June 1970
aAo.xddv
 PREFACE

The purpose of this paper is to investigate the

Jacobian matrix in greater depth than this topic is dealt
with in any individual calculus text.
The history of Carl Jacobi, the man who discovered
the determinant, is reviewed in chapter one.
The several illustrations in chapter two demon­
strate tile mechanics of finding the Jacobian matrix.
In chapter three the magnification of area under
a transformation is related to ti1e Jacobian. The amount
of magnification under coordinate changes is also discussed
in this chapter.
In chapter four a definition of smooth surface area
is arrived at by application of the Jacobian.
It is my pleasure to express appreciation to Dr.
John Burger for all of his assistance in preparing this
paper. An extra special thanks to stan for his help and
patience.
C.G.J. JACOBI (1804-1851)
"Man muss immer umkehren"
(Man must always invert.)

,
,I,~'

"1,1'1
 TABLE OF CONTENTS .

CHAPTER PAGE

I. A HISTORY OF JACOBI • • • • • • • • • • • • 1

II. TIlE JACOBIAN .AND ITS RELATED THEOREMS • • • 8

III. TRANSFORMATIONS OF AREA • • • • • • • • • • 14

IV. SURFACE AREA • • • • • • • • • • • • • • • 23

CONCLUSION • • • • • • • • • • • • • • • • • • • • 28

FOOTNOTES • • • • • • • • • • • • • • • • • • • • 30

BIBLIOGRAPHY • • • • • • • • • • • • · .. • • • • • 32
 CHAPTER I

A HISTORY OF JACOBI

Carl Gustav Jacob Jacobi was a prominant mathema­

tician noted chiefly for his pioneering work in the field
of elliptic functions.
Jacobi was born December 10, 1804, in Potsdam,
Prussia. He was the second of four children born to a
very prosperous banker.
Carl Gustav Jacob Jacobi should not be confused with
his equally famous older brother Moritz II. Jacobi. Moritz
H. Jacobi achieved fame, while still living, as the founder
of galvanoplastics. The importance of the ideas and teach­
ings of Carl Gustav was not realized until after his death.
Carl's impatience with his brother's popularity during his
lifetime once broug~out the following statement: "I am
not his brother, he is mine." Today his statement aptly
describes their relative importance.
Jacobi's first mathematics teacher was an uncle.
Under the guidance of his uncle, he was prepared to enter
the Potsdam Gymnasium at the age of twelve. After grad­
uating from the Gymnasi~m, Jacobi was still undecided as
 2

to whether to pursue study in the field of mathematics or

the field of philosophy.
In 1821 he went to the University of Berlin. It was
here that he made his decision to become a mathematician.
He spent much time studying work on the masters of mathe­
matics such as Euler and La Grange, as he felt the Univer­
sity instructors had little to offer.
He received his Ph.D. in mathematics from the
University of .Berlin in 1825 at the age of 21.
Having become "certified", his teaching career began
as a lecturer at his alma mater, the University of Berlin.
Due to a lack of respect for his own University professors,
he attacked his new position with much energy and enthu­
siasm. His basic teaching philosophy is summed up by the
following quotation.

That irrepresible innovator believed the infallible

method to advance mathematics was for domineering
professors in the leading universities to drill their
own ideas and as little else as possible, into as many
advanced students as could be induced to scribble
lecture notes. L!. p. 4417

The greatest portion of his lectures were devoted

to his personal discoveries and ideas which he was pre­
sently investigating.
Some credit the lack of mathematical progress in
the twentieth century to his teaching methods. LT. p. 4267
The logic behind this feeling is that many poor mathema­

.lo '_
 3
maticians do "drill their own ideas, and as little else· as
possible" into their students. Thus, they offered the
student little help and encouragement.
After staying approximately one half year at the
University of Berlin, he took a position as lecturer at
the University of Konigsberg. A year later (1827) he was
promoted to assistant professor largely due to a publica­
tion of his works on cubic reciprocity. His greatest work
FUNDAMENTA NOVA THEORIAE FUNCTIONUM ELLIPTICARUM (New
Foundations of the Theory of Elliptic Functions) was com­
pleted and published in 1829.
The death of his father in 1832 resulted in the
ultimate loss of the family fortune. However, Jacobi
continued to work hard in producing works of mathematics.
In fact, he worked so diligently that it began to affect
his health. Jacobi's physician suggested that he take an
active part in politics to get his mind away from his
mathematical works. This decision nearly cost him his
position of favor with the King of Prussia.
Jacobi soon became the stoolpidgeon for the local
liberals and gained the wrath of his only remaining
monetary benefactor, the King.
At the age of 45, real poverty had finally struck.
He had neither his job nor the family fortune, and a wife
and seven children to support. A generous offer from Vienna

.
"
 4

helped him solve this problem. Germany did not want to,
lose one of their greatest mathematicians, so the King
reconsidered. IIe returned to his position at the University
of Konigsberg. He died shortly thereafter (February 18,
1858) of e1e small pox.

Jacobi, often called 'the great algorist', was second

in his field only to Euler. (An algorist is one who develops
a way to solve a specific type of problem.) A.lthough he
made many contributions to the field of mathematics, his
greatest was in the area of elliptic functions. Later, it
was found that Guass, working independently, had previously
discovered elliptic functions and tileir doUble periodicity.
Had he published his findings, the total contribut.ions of
Jacobi may have been much more advanced.
At approximately the same time Jacobi ~as working
on elliptic functions, so was his elder rival Niels Henrik
Abel. Neither of them realized the existence of the other
when they first started their work. Abel was credited with
first inverting elliptic integrals, thus smoothing the way
for further development in that area.
In 1830 the academy awarded Jacobi and Abel the
Grand Prize in Mathematics for work done with trancen­
dental functions.
While working with rational numbers Jacobi discovered
that an elliptic identity would determine the number of
 5
representations of an integer as the sum of two squares .•
This is found in his "Fundamenta nova", published in 1829.
In 1841, Jacobi first presented the functional deter­
minants which carry his name. DE FOIU-mTIONE ET PROPRIET­
ATIBUS DETERMINANTIUM was a "landmark on the subject of
determinants". L!. p. 4267
The basic works of these functional determinants
are as follows:

If Y , Y , ••• Y , are functions of x , x , ••• x ,

1 2 n . 1 2 n
then the functional determinant, or Jacobian, is the
equations as follows:
I·

~
ax; ~
2
... .~
n

~1 ~
2
.
J I:
a(Yl,Y2, • • • •X,n ) I:
I'M'x
0\ 1
,x 2 , •••• x
n
1
•
•

;~n aYn
~ ••••••••
1 2

This determinant is often more briefly written as:

J(y~!x~). L!.
p. 916-9177

Some properties of the Jacobian are: (a) J(y~!x~)

J(XI.!Yi.)=l. (b) J(y<.!zdJ(Zi,!xi.) = J(y,,!x~). A Definition

for the Jacobian is: A Jacobian is a functional determinant
formed by the n 2 different coefficients of n given functions
of n independent variables. L? p. 340-34!7 Sylvester ts
 6
attributed for having given the name Jacobian to this
functional determinant.
With the possible e~ception of Cauey, Jacobi is gen­
erally, considered to have contributed more to the field
of functional determinants than any other person.
Additional areas in which Jacobi worked would include
continued fractions, Abelian functions, maxima and minima,
and the geometric interpretation of the conjugate point.
His early works led to a field (late 19th century) later
known as elliptic modular functions. He established a theory
of dynamics which was not extended until fifty y~ars later
by Poincare.
Jacobirs famous memoirs appeared in "Crellers' Journal
For Pure and Applied Sciences" in 1841. Many of his
important writings were also published by Crellers.
It is obvious that Jacobi's theory of elliptic
functions was responsible for much of the mathematical
analysis in the 18th and 19th century.
Much of the work Jacobi started was continued by
Weirstrass and Hermite.
Kronecker and Reimann were former students of
Jacobi's who would have to be _considered famous mathema­
ticians in their -own right.

,; .,
 CHAPTER 2

THE JACOBIAN AND ITS RELATED THEOREMS

The purpose of this chapter is to define a Jacobian

and discuss its related theorems. It generalizes the theory
of inverses of one variable to transformations defined by
a system of equations.
The equations u = f(x,y), v = g(x,y), define a
transformation from points in the xy plane to points in
the uv plane. The Jacobian matrix for this transformation
is the matrix I

au au
ax dy

dV av
ax ay

The determinant of the above matrix is the Jacobian deter­

minant of the transformation (noted T) or the Jacobian of
T. This is also denoted by a(u,v). The idea of the .
a(x,y)
Jacobian is easily extended to dimensions greater than two.
The above is the Jacobian of u and v with respect to
x and y. That is, the transformation is from the xy plane
to the uv plane. If one wishes to define the inverse
 a
function from the uv plane to the xy plane, the equations
would be in the form x = F(u,v) and y = G(u,v). Then the
Jacobian matrix of these equations would be:

ax ax
au av
2Z ~
au, av

These are inverse transformations (if inverses exist) and

their JacObians are rec~procals.

Some theorems about J acobians are as follows. In

each one must agree that the functions are continuously
differentiable.

THEOREM 1. A necessary and sufficient condition that the

equations F(x, y, z, u, v) = 0', G(x, y, z, u,
v) = 0 can be solved for x and y '(for example)
is that a(u,v) is not equal to zero in a region
a(x,y)
R. In other words a necessary and sufficient
condition that the equations be a 1-1 trans­
formation is that J ~ 0 in any region. 'In
this case J-1 exists and is also nonzero--JJ-1 • 1.
THEOREM 2. If x and yare functions of u and v, while u
and v are functions of r and s, then,
a(x,y) a(x,y) a(u,v)
:i(r ,s) = a(u,v) • a(r,s)
This is known as' the chain rule for Jacobians.
THEOREM 3. If u = f(x,y), and v = g(x,y), then a necessary
and sufficient condition that a functional
relation of the form ~(u,v) = 0 exists between
u and v is that a(u,v) be identically zero.
a(x,y)

. "
 9

Example ,1. GIVEN: u'= 3x 2 - xy

v = 2xy 2 + y3

au: = 6x-y au = -x
ax ay

av = 2y 2
ax
av
ay .. 4xy + 3y2

J • 6X y • (6x-y) (4xy + 3y) 2 _ (-x) ( 2y 2)

1,2 -2 4xy -x
+ 3y 2 I .
y
= y'(24x 2 + 16xy - 3y 2)

Since the Jacobian ,does not equal zero, there exists

no functional relation of the form ~(u,v) = O. This follows
by theorem 3. But when y = 0, the Jacobian is zero. By
Theorem 1, this is not a 1-1 transformation. There exists
no inverse.
In investigating this particular transformation when

y • o. One obtains (by direct substitution) u • 3x 2

v .. 0 •

Thus if y = 0, so also must v.

u = 3x 2 x = ± {U7!
+ y
x==
Y
±3] MAPS, TO
0
= 27
",) v
[u = 0

These points (3,0) and (-3,0) in

the xy plane both map to (27,0) in
the plane. (See sketch)
I (-3,0)
I k I(3,~)
l I I I I I I I I I .. X

As u cannot be a negative number and

produce a real value for x, it is
obvious that the x.axis maps its
points only to the positive u axis.

I I I I I I "I I .. U
I I 1 1 1 1(27,0)1
 10
Example 2. GIVEN: u - x + 2z
v = 2y + 3z
w = 2x - 2y + Z
au
ax - 1 au
ay = 0 au
az 11:1 2

av • 0 av 2 av 3
ax ay
11:1

az
11:1

aw aw aw
ax 11:1 2 ay = -2 az
11:1 1

.I-~ tI I~ tI j~ ~21
1 o 2
THUS J • 10 2 3 11:1 1 ·-0 + 2 11:1 0
3 -2 1
This transformation has no inverse by theorem ,1, but there is
a functional relationship of the form ~(u,v,w)= o.
By ~e following methods one can arrive at this
relationship.

by using the first and u =x + 2z

third equations to -2w = -4x + 4y - 2z
eliminate z u - 2w = -3x + 4y *

by using the second and v = 2y + 3z

third equations to -3w = -6x + 6~ - 3z
eliminate z v - 3w - -6x + By
v - 3w = -3x + 4y
2 *
By using the two resulting equations (*)
u - 2w c v - 3w or 2u - w - v = 0
2
which is a plane in the uvw space.

Example 3. GIVEN: u = x + 2z
v = 2y + 3z
w 11:1 2x - 2y + Z
1 o 2
"J. 10 2 31 11:1 0
2 -21

. ".
 11

Again this indicates a functional relationship between u

and v.
Some 1-1 transfor.mations and their inverses are
illustrated below.

Example 4. GIVEN: u = 2x + 2y
v = 5x + 7y
au =2 au IOl 2
ax ay a(u,v) _ J
a(x,y)
IOl
1;;1 • 4
av = 5 av =7
ax ay
Since this Jacobian will never equal zero, it is a 1-1
transfor.mation. Thus their exists the following inverse:
By using determinants to solve for x and y:

J; ~I y = I; ~I
x -
I; ~ j I; ~ I
NOTE: Notice that the determinant in the denominator is
actually the Jacobian.
x = -7 -1 y = 5 u +-l
lU+"2v "4 -ZV
a (x,y). _ J-1 =
-7
4"
1
'!
I 1
="4
a(u,v) 5 -1
"4 '!
This inverse could als9 have been found by finding the
inverse matrix associated with J, solving for x and y is
not necess.ary, but is possible.
It is noted that JJ-I = 1.

Example 5: GIVEN: u =
2x + 3y - z
V x -7y + z
I:

w - x + Y + 2z' .;. " .

 12
2. 3-1
J .. 1 -7 1 = -41
112

This is also a 1-1 transformation. The inverse can

be found by solving for x, y, and z in terms of u, v, and w
or by finding the inverse matrix. Since in this case finding
the inverse matrix is easiest, one will proceed in that
manner.

15 7 4
14 14 14
J-1 • I iIT1 -5
4I'
3
4I' = -1
4I'
-8 -1 17
iIT iIT iIT
It is noted again that J and J-I are reciprocals.
Assume that one wishes to evaluate the Jacobian of a trans­
formation at a specific point. Given T(x,y) = (2x + 3y,

5x + 7y) it follows that J =I;~I = 14 _ 15 =-1.

In this particular transformation ,the Jacobian will be -1

regardless of the point chosen, as there are no variables
present in the value of the Jacobian.
Following is an example where the Jacobian chapges
values at different points.

Example 6. GIVEN: T(X,y) = '(X+l

x+y
, x+Y)

J ..
1~2
x+Y
-x-l
(x+Y) 2 • ~
x+y 2 - (x+Y)
(-x-l~ ..
1 1
I
--1:
-rffir
x+Y 2
1:1
x+Y
13
 CHAPTER 3

TRANSFORMATIONS OF AREA

In this chapter the importance of the Jacobian

when transforming one area to another area will be shown.
For most transformations, closed regions in one
space map to closed regions in the image space. If 6Axy
and 6Auv denote respective areas in these regions, then:

lim ~Axy
Auv
1:1

Ia(u,v) I
a (x ,y)

where lim denotes the limit at 6Axy (or 6Auv) approaches

zero. This Jacobian is often called the Jacobian of the
Transformation. lb. p. lO!7 If the Jacobian of the func­
tion is a constant the expression lim is unnecessary as
the value of. the Jacobian varies with position.

Example 1. u =XiY
v = xi Y
Suppose one wishes to transform the area bounded by the
followi~g lines to the uv-plane: x =y
x = 1
Y • 4 •
 y

-"
IC
,,'\
~

15

I ~J!li('..u £
~ y II 4

X I 1.1 I 1 I 1 I - X

Each of these lines map as follows: (These may be found

by substituting the respective lines into the transforma­
tion.)
x =y ( ~ v • 0
x =1 ( ~ u + v = 1.
Y c 4 ( ) u - v c 4
v.

I 'ttL!. nLE"" 1 1 1 1 I - u

Since both triangles are right triangles, the areas of each

may be figured as shown below:
Area of R. 1/2(3) (3) • 9/2, Area of R' D 1/2(3/211) .
, (3/211) D 9/4 •
"
 16
It is interesting to note that both orthogonality and the
type of triangle remain unchanged under the transformation.
The Jacobian of T from the xy-plane to the uv-plane
is 1/2 1/21 = 1/2
1 1/2 ·1/2

This is also the ratio of the area in the uv-plane to the

area in the xy-plane. This might be explained in this
manner. The transformation from the xy-plane to the uv­
plane "magnifys" the area. The Jacobian tells one the
amount of magnification.

Example 2. Given the transformations: x . 2u + 3v

y =u - v

Find and discuss the region bounded by u = 0, u = 1, v - 0,

and v • 1.
v

o
"I "
:I :I

y: I
y=o u
~4f(H::t-=:;;rfi~:>,)';'j;I'4';:~I \ x

The following are the line transformations:

u =0 ( ) 3y + x = 0
v = 0 ( ) x - 2y = 0
u == 1 ( ) 3y + x = 5
v • 1 < )
x - 2y • 5
 17
The unit square is mapped to the parallelogram bounded by
the above lines with vertices (0,0), (2,1), (5,0), (3,-1).
The transformation is from the uv-plane to the xy-plane,
so the area in the uv-plane is magnified by 5 (since the
Jacobianis 5) and placed in the xy-plane. Thus, the area
of the parallelogram is 5 units. All of this can be
verified by analytic methods.
If one returns for a moment to look at some proper­
ties of vectors, these should help relate why the Jacobian
shows magnification.
First one wishes to show that the absolute value of
a 2 x 2 determinant whose rows are components of vectos u
and v is the area of a parallelogram with u and v as adjacent
y
sides.

~ )x

One wishes to prove the area of the parallelogram which is

(u) (v) sin e is
I
U'I
v~v~
u

Proof. The following identities will help in the proof;

 18
(u.v) - lui Ivlcos e
I (uxv) I. ju U
V 1V 2
1 2

(u.v)2+ 1 (uxv) 1 2 - IUl2jvl2

By starting with the last identity, we get

Dull vi cos] 2 + I (uxv) 12 = 1ul 2 1vl 2
I (uxv) 12 = lul 2 1vl 2 - lul21vl2cos2e
= [I u/ 2 1v1 2J[1 - cos 2 e1_
D lul 2 1vl 2 sin 2 e

UiU2
/ V1V2 1
"":' I ull vi sin e

This idea can be extended to dimensions greater than 2.

That is, a 3 x 3 determinant whose rows are components ov
vectors u, v, and w will give the volume of the parallel­
piped formed,by these vectors. This leads to the follow­
ing theorem which associates the Jacobian with this area.

THEOREM 4.

Suppose T is a linear transformation of R

. with matrix A. If T maps an n-dimensional
parallelpiped P onto a parallelpiped P*, then
(volume of P*) =,ldet(A) I (volume of p)
Proof: The volume of P is \det(B) I where B has as its
rows components of the vectors forming P. The
edges of P* are rows of matrix AB and its volume
is Idet (AB) I. Since det (AB) :II (det (A» (det (b», we
get the above statement. L'7. p. 40!7
 19
If a transformation is not linear, then it can be
approximated by a linear ~ansformation which is the
Jacobian of the transformation. If it is linear, this
matrix A (above) is just the Jacobian of the transforma-.
tion. Thus one has the Jacobian as the "magnification
factor".

Example 3.
Find the linear transformation that takes the
following parallogram into the unit square. What is its
Jacobian? Its Area?
y v
( 3,8)

(0,1 ) ( I, I)

r I I I I I I I ) X xx<*t I
(1,0)

~~ -~)
The transformation is

The value of the Jacobian is 1, then the area of the parallelo­

gram is magnified byl to get the area of the unit square
which also is 1. The area of the parallelogram is 1.
 20
THEOREM 5.

Suppose T is a one-to-one transformation whose

domain and range are in R N and that there is an
open set U on which the component functions of
T have continuous first-order partial derivatives
and the Jacobian of T is nonzero. Let R be a
subset of U that is mapped onto a compact set S
that has n-D volume. If f is continuous on S, then
~f(p)dv = ~f(T(q») IJ(T)I dv L"[. p. 41~.7
This theorem generalizes a little more the role that the
Jacobian plays in finding .areas.

Example 4.
Given a parallelogram with vertices (0,0) ,(2,1),
'-­
(5,0),(3,-1) under a transformation defined as x = 2u + ~:
3v and y = u - v. Express ~(x + y) dA in the form 11 du dv.
"

I'
Since this transformation was discussed in example" 2
one knows the limits of integration are 0 and 1.

1,(x+y) c1A a 101

I'
0 (2u +3v+ u-v) (5) du elv
, (,
~Jo (3u +av) (5) duav

.(,'ic: (lSu + lOY) auav

lSu2 + lOUV]'
-r 0
,i~
~(~ + lOy) av 'I'

lSv + lOv 2
~ ... ~
J' 0

12 1/2

The "local magnification factor" when rectangular

coordinates are changed to other coordinate systems follows.
 21
v
POLAR COORDINATES:
The proper coordinate' p(x,y)
changes are: I
I
I
x r cos
Ill: e I
IY
y • r sin e . I
I
I
I
x
x
The corresponding
JaCObian(a(X,~»)iSl
:Hr, )

cos e -r sin e
=r .
sin e r cos e

CYLINDRICAL COORDINATES:
The transformation for rectangular coordinates to
cylindrical coordinates is the same as for polar coordi­
nates except a z coordinate is added as follows:
~

x • p cos e y .p sin e ~. z •

cos e-p sin ea X .

I
.~ ~ v
a(x,y,z) = sin e p cos e a
J • a(p,e,z) I.

o a 1

The magnification factor is p which corresponds to r in

the 2-d figure above.

SPHERICAL COORDINATES:
Spherical coordinates and rectangular coordinates are
related in the following manner:
x • r sin ~ cos e
 !
P(ll,y,Z)
22
y = r sin ~ sin e

z • r cos ~ )"-:: I I » )ly

)I;

The Jacobian of this transformation is

sin ~ cos e .-r sin ~. sin e r cos ~ cos e

a (x,~, z)
a(r, ,4»
.. sin ~ sin e r sin 4> cos e r cos 4> sin e .
cos 4> 0 -r sin 4>

-r 2sin 4> •

The absolute value of the expression r 2 sin 4> is the

magnification.
 CHAPTER 4

SURFACE AREA

Chapter three was concerned mainly with magnification

of areas from one coordinate system to another. This
chapter will cover the i~vestigation of the Jacobian as
it relates to surface areas. A concept of smooth surface
areas will be developed.
,
j:

The relationship of the length of a line segment to

its projection on an axis is basic to the concept of the
relation of an area and its projection. If the line segment
AB (fig. 1) has projection CE on the axis, the length of AB

is 1 (CE) •
Coser

B 1"1'1
I ~ii!

I
I
I
I
I
I
a;'
A lt ' 'F
I
I I
I I
I I
I
I I
I I
I,
C
I
E
»)(

FIGURE'
 24

One can at this time find another angle equal to « and thus
a method of evaluating the angle ex:. V f is the direction
of greatest change at a point and is perpendicular to a
point on the segment. af at this same point would be
ay
perpendicular to thexaxis. (see fig. 2) The angle between
the gradient (Vf) and a unit vector in the direction of af
ay
is equal to ex:. So the cos CIC can be evaluated by the formula
at divided by the length of Vf. It might be well to note
ay
here that it is necessary to know the identity IVfjcosc
• af (directional derivative) •
as
I::'
y af ,
ch

A,ee:
I
- , IF
I I
I I
I I
I I
I I
I I >x
c E

FIGURE 2

One can apply the idea developed above to the area

of a small portion of a surface. If a surface has a tangent
plane at a point, then a reasonable approximation of the
area of the piece of surface is the area of the projection
of the tangent plane divided by the cosex: , where ~ is

relatively in the same position as discussed in the two

 25

dimensional illustration.
One can be a little more explicit in explaining this
in reference to figure 3. If one wishes a good approxima­
tion to the area of !:J.S one can use the area of !:J.D/cos lit •

So the area of the whole surface is given by the following

formula I

area of S -= -b Icos lit I-I dA

Z

) ~~ A ) Y

x FIGURE 3

The difficulty here is in finding the cos lit I but the

method for finding cos lit in 2-space (refer tQ fig. 2) also
applies here.
cos lIll - to fig. 4)

AS
l
I
I
I
J.._
,'"
,." "
",-' AD

FIGURE 4
 26
All of this is related to the Jacobian. Let one presume
figure 5 to be the small piece of the tangent plane in
figure 4 projected on the xy plane.' Although the axis
have been moved this will no~ change the final result.
The area of this projection (~D) is the area of a
parallelogram with (u , u , u ) and (v , v , v ) as its
. 1 2. 3 1 2. 3

adjacent sides. This can be expressed as ux v or

u U
1 2.'"
• (This was determined in chapter 3.)
v V
:, 1 2.

/C~c:::;==t'(u,.u., ,0) / . < :

I'
),y

FIGURE ~,
 27
u may be thought of as au since in the xy plane x is held
1 ay
constant and the movement takes place in the y direction for
this coordinate. Likewise, u as au; v as dV ; and v as
ax 1 ay
2 2

dV. Then the determinant becomes

ax
z
(ul,~,u3) au dU
I
I
I
I
I
I

;
v)
~
dV
~
-ax I
dV
~
- d (u,v)
cHx,y)

I
I "
" ""
...v/
(u/,u ,O)
2
I'
X
FIGURE 6

A surface is considered to be smooth at ~ point P if

a tangent plane exists at that point and at every point in
its neighborhood. This leads to the conclusion that the
area of the projection of the tangent plane in each of the
xy, yz, and xz planes cannot be zero. NoW let one introduce
the following notation for these Jacobians.
If a surface is parametrically defined as:·
x= f(u,v) y g(u,v) z= h(u,v) =
then:
These Jacobians have to be nonzero for the surface to'be
smooth since they represent the area of the projection of
the tangent plane. The, ratio of these special Jacobians
also gives the direction of the normal.
 CONCLUSION

Jacobi made many contributions to the mathematical

world. . Although his work in elliptical functions was
probably most important, the discovery of the Jacobian
matrix is not to be slighted. In analysis, this matrix
is used to evaluate the change in area or volume under a
transformation. This is particularly useful when an area
or volume is difficult to find before transformation.
'. This matrix is used to approximate:'surface area and to
define smooth surface area.

This paper has only discussed a few of the physical

applications of the Jacobian.
 FOOTNOTES

1. Bell, Eric T. Development of Mathematics. Second

edition. New" York: McGraw=ii'il1 BOOk Company, 1945~
p. 441.
2. ~." p. 426.
3. Ibid.
4. "Jacobi, C.G.J.", Van Nostrand's Scientific Encyclopedia.
(3rd ed.), pp. 916=9I7.
5. "Jacobi, C.G.J.", Encyclopedia Britanica. (1968) XII.
pp. 340-341.

6. Spiegel, Murray R., Theory and Problems of Advanced

Calculus. New York: Schaum Publishing. p. 10 a-.
7. ~. p. 401
8. James, Robert C. Advanced Calculus. Belmont, Calif.:
Wadsworth" Publishing Co., Inc., 1966. p. 414.
XHdWDOI'IgIg
 ..,

BIBLIOGRAPHY

A. BOOKS

Bell, Eric T,.' Develo1ment of Mathematics. Second Edition.

New York: McGraw-Hi 1 Book Company, Inc. 1945.
Bell, Eric T. Men of Mathematics. New York: Simon and
Schuster, 1937.----­

Chisholm, J.S.R. and Morris, Rosa M. Mathematical Methods

in Physics. Philadelphia: W. B. Saunders Co. 1965.
Dolciani, Mary P. et ale Modern Introductory Analysis.
Boston: Houghton MiffIin, 1967.
Jamer, Robert C. Advanced Calculus. Belmont, Calif ••
Wadsworth Publishing Co., Inc., 1966.

Kramer, Edna E. The Main Stream of Mathematics. New York.

Oxford University-preSS;-1951. -­

Smith, David E. History of Mathematics. Vol. II of Special

Topics of Elementary Mathematics. Boston. Ginn and
, Company, 1935.

Taylor, Angus E. Advanced Calculus. Waltham, Massachusetts.

Blaisdell Publishing Co. 1955.

B. ENCYCLOPEDIAS

"Jacobi, C.G.J.," Enclyclopedia Britanica (L96B), XII.

"Jacobi, C.G.J,.," ~ Nostrand I s Scientific Encyclopedia
(3rd ed.).