Chapter 6

Differential Calculus of Functions of

Several Variables

Contents
6.1 Notation, examples, graphs and level curves 4
6.1.1 Notation and Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
6.1.2 Level Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

6.2 Limit and Continuity 6

6.2.1 Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
6.2.2 Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

6.3 Partial Derivatives; Higher Order Partial Derivatives 13

6.3.1 Partial Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6.3.2 Higher Order Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6.3.3 Tangent Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

6.4 Directional Derivatives and Gradients 16

6.4.1 Directional Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
6.4.2 Gradients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

6.5 Tangent Plane and Tangent Plane Approximation 19

6.5.1 Total Differential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.5.2 Tangent Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
6.5.3 Tangent Plane Approximation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

6.6 Chain Rule and Implicitly Differentiation 22

6.6.1 Chain Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
6.6.2 Implicitly Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

6.7 Relative Extrema of Function of Two Variables 23

6.8 Largest and Smallest Values of a Function on a Given Set 25

6.9 Extreme Value under Constraint Condition 27

6.9.1 Method of Lagrange Multipliers for One Constraint . . . . . . . . . . . . . . . . 27
6.9.2 Method of Lagrange Multipliers for Two Constraints . . . . . . . . . . . . . . . . 29

6.10 Exercise 31

In this chapter, basic concepts of calculus along with linear algebra and functions of two or more
variables, commonly called functions of several variables are discussed.

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
2 Differential Calculus of Functions of Several Variables

6.1 Notation, examples, graphs and level curves

6.1.1 Notation and Examples
Definition 6.1.1 A function of two variables is a rule that assign a unique real number f (x, y) to each
ordered pair of real numbers (x, y) in the domain of the function; that is, f : D ⊆ R2 → R.

 Example 6.1 f (x, y) = xy2 and g(x, y) = x2 − y2 are both examples of function of two variables. 

Definition 6.1.2 A function of several variables is a rule that assign a unique real number f (x1 , x2 , · · · , xn )
to each ordered n − tuple of real numbers (x1 , x2 , · · · , xn ) in the domain of the function; that is,
f : D ⊆ Rn → R.

The domain of function of several variables is taken to be the set of all values of the variables for which
the function is defined.
 Example 6.2 Find the domain for f (x, y) = x ln y.
Solution: The domain of f is then the set D = {(x, y)|y > 0}, since ln y is defined only for y > 0; that is,
the half-plane lying above the x − axis. 

cos(x+z)
 Example 6.3 Find the domain of a function f (x) = xy .
Solution: f is defined unless there is a division by zero if xy = 0, which occurs if x = 0 or y = 0. The
domain is then {(x, y, z)|x 6= 0 and y 6= 0}, which is all of the three-dimensional space excluding the
yz-plane (x = 0) and the xz-plane (y = 0). 

2x
 Example 6.4 Find the domain for g(x, y) = y−x 2.

Solution: The function g(x, y) is defined for all ordered pair of real numbers except the denominator is
zero which occurs when y − x2 = 0. Then the domain of g is {(x, y)|y 6= x2 }, which is the entire xy-plane
with the parabola y = x2 removed. 

Properties
The sum, product, and quotient of two functions f and g of several variables are defined as
1. ( f + g)(x, y) = f (x, y) + g(x, y)
2. ( f − g)(x, y) = f (x, y) − g(x, y)
3. ( f g)(x, y) = f (x, y)g(x, y)
f (x,y)
4. ( gf )(x, y) = g(x,y) , provided that g(x, y) does not assume the value 0.
The formulas for functions of three variables are analogous.
Definition 6.1.3 If f is a function of two variables and g is a function of a single variable, then the
function go f is defined by
(go f )(x, y) = g( f (x, y))
for all (x, y) in the domain of f such that f (x, y) in the domain of g.

The definition of the composite of a function of three variables and a function of a single variable is
similar.

Graphs
The graph of a function f of two variables is the collection of points (x, y, f (x, y)) for which (x, y) is in
the domain of f . It is customary to let z = f (x, y); then the graph of f consists of all points (x, y, z) such
that z = f (x, y).
 Example 6.5 Sketch the graph of f (x, y) = x2 + y2 .
Solution: The surface z = x2 + y2 is a circular parabolic and the traces in the planes z = k > 0 are circles,
while the traces in the planes x = k and y = k are parabolas. 
p
 Example 6.6 Find and describe in graphical terms the domain of g(x, y, z) =
p 9 − x 2 − y2 − z2 .
Solution: To define g(x, y, z) = 9 − x − y − z it should to have 9 − x − y − z ≥ 0 or x2 + y2 + z2 ≤ 9.
2 2 2 2 2 2

Therefore the domain of g is sphere of radius 3 centered at the origin and its interior. 

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.1 Notation, examples, graphs and level curves 3

Figure 6.1: f (x, y) = x2 + y2

p
Figure 6.2: g(x, y, z) = 9 − x2 − y2 − z2

6.1.2 Level Curves

In sketching the graph of a function f of two variables, it is often helpful to determine the intersections of
the graph of f with planes of the form z = k. We call each such intersection the trace of the graph of f in
the plane z = k. Thus the trace of the graph of f in the plane z = k is the collection of points (x, y, k) such
that f (x, y) = k.
Definition 6.1.4 The level curves of a function f of two variables are the set of points (x, y) in xy
plane such that f (x, y) = k, where k is a constant in the range of f and it shows the height value of f
at (x, y) from the xy plane.

Figure 6.3: Level curves

The level curve f (x, y) = k are just the traces of the graph of f in the horizontal plane z = k projected
down to the xy plane and the trace of f in the plane z = k either lies directly above or directly below the
level curve f (x, y) = k or coincides with it. The idea behind level curves is to provide three dimensional
informatin in a two dimensional setting.
Example 6.7 Sketch the level curves of the function f (x, y) = 6 − 3x − 2y for the values k = −6, 0, 6, 12.
Solution: The level curves are 6 − 3x − 2y = k,this is a family of lines with slope − 23 . The four particular

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
4 Differential Calculus of Functions of Several Variables

Figure 6.4: Level curves

level curves with k = −6, 0, 6 and 12 are

3x + 2y − 12 = 0
3x + 2y − 6 = 0
3x + 2y = 0 and
3x + 2y + 6 = 0

The level curves are equally spaced parallel lines because the graph of f is a plane. The contour map of
f (x, y) = 6x − 3y − 2y is given below.


Figure 6.5: The level curves of f (x, y) = 6x − 3y − 2y

p

p curves of the function f (x, y) =
Example 6.8 Sketch the level 9 − x2 − y2 for k = 0, 1, 2, 3.
2 2 2 2 2
Solution: The level curves are√ 9 − x − y = k or x + y = 9 − k . This is a family of concentric circles
with center (0, 0) and radius 9 − k2 . The cases k = 0, 1, 2, 3 are shown in the figure below.


 Example 6.9 The level curve(contour curve) the graph of f (x, y) = x2 + y2 is the set of circles given by


6.2 Limit and Continuity

6.2.1 Limit
Definition 6.2.1 Let f be defined on the interior of a circle centered at the point (a, b) except possibly
at (a, b) it self. We say that lim f (x, y) = L if for every ε > 0, there is a δ > 0 such that
(x,y)→(a,b)
q
| f (x, y) − L| < ε whenever 0 < (x − a)2 + (y − b)2 < δ

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.2 Limit and Continuity 5

p
Figure 6.6: The level curves of f (x, y) = 9 − x2 − y2

Figure 6.7: The level curves of f (x, y) = x2 + y2

 Example 6.10 Verify that lim x = a and lim y = b.

(x,y)→(a,b) (x,y)→(a,b)
Solution: Certainly, both of these limits are intuitively quite clear, we can use the above definition to
verify them, however. Given any number ε > 0, we must find another number δ > 0 so that
q
|x − a| < ε whenever 0 < (x − a)2 + (y − b)2 < δ

Notice that q q
(x − a)2 + (y − b)2 ≥ (x − a)2 = |x − a|

and so, taking δ = ε, we have that

q q
|x − a| = (x − a)2 ≤ (x − a)2 + (y − b)2 < ε
p
whenever 0 < (x − a)2 + (y − b)2 < δ . Lickwise, we can show that

lim y=b
(x,y)→(a,b)

Definition 6.2.2 The function f has a limit L at the point (a, b), written

lim f (x, y) = L
(x,y)→(a,b)

if the deference | f (x, y) − L| is as small as we wish whenever the distance from the point (x, y) to the
point (a, b) is sufficiently small, but not zero.

The limit of a function of two variables (if exists) is unique.

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6 Differential Calculus of Functions of Several Variables

Operation of Limit
If f (x, y) and g(x, y) have limits at (x, y) approaches to (a, b), respectively; that is,

lim f (x, y) = K and lim g(x, y) = L

(x,y)→(a,b) (x,y)→(a,b)

then
• lim ( f (x, y) ± g(x, y)) = K ± L
(x,y)→(a,b)
• lim f (x, y) · g(x, y) = K · L
(x,y)→(a,b)
f (x,y)
• lim = KL , provided L 6= 0
(x,y)→(a,b) g(x,y)
• lim c · f (x, y) = c · K, where c is some constant number.
(x,y)→(a,b)

x sin y+x +2 2
 Example 6.11 Evaluate the limit of the function f (x, y) = √ 2 2 at (x, y) approaches to (0, 0).
x +y +1
Solution:
x sin y + x2 + 2
lim f (x, y) = lim p
(x,y)→(0,0) (x,y)→(0,0) x 2 + y2 + 1
lim x sin y + x2 + 2
(x,y)→(0,0)
= p
lim x 2 + y2 + 1
(x,y)→(0,0)
= 2


2
 Example 6.12 Evaluate lim 2x y+3xy2 .
(x,y)→(2,1) 5xy +3y
Solution: First, note that this limit of rational function is the quationt of the two polynomials. Since the
limit in the denominator is
lim (5xy2 + 3y) = 10 + 3 = 13 6= 0
(x,y)→(2,1)

we have
lim (2x2 y + 3xy)
2x2 y + 3xy (x,y)→(2,1)
lim =
(x,y)→(2,1) 5xy2 + 3y lim (5xy2 + 3y)
(x,y)→(2,1)
14
=
13


Remark 6.2.1 1. If f (x, y) approaches L1 as (x, y)approaches (a, b) along a path p1 and f (x, y)
approaches L2 6= L1 as (x, y) approaches (a, b) along a path p2 , then lim f (x, y) does not
(x,y)→(a,b)
exist. So, the following paths are the simplest paths to check the existence of limit
(a) Vertical line test (x = a, y → b)
(b) Horizontal line test (y = b, x → a)
(c) y = g(x), x → a where b = g(a) and
(d) x = g(y), y → b where a = g(b)
2. Use polar coordinate to check the existence of limit.
Consider (x, y) approaches to (a, b). We can convert it by polar coordinate in the form x =
a + r cos θ and y = b + r sin θ . So,

lim f (x, y) = lim f (r, θ )

(x,y)→(a,b) r→0

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.2 Limit and Continuity 7

If lim f (r, θ ) exist; that is, constant number and unique for any θ , then the limit exist. Otherwise,
r→0
the limit does not exist.

Figure 6.8: f approaches from (x, y) to (a, b) in different paths

y
 Example 6.13 Evaluate lim x+y−1 .
(x,y)→(1,0)
Solution: First, we consider the vertical line path along the line x = 1 and compute the limit as y approaches
0.
If (x, y) → (1, 0) along the line x = 1, we have
y
lim = lim 1 = 1
(x,y)→(1,0) x + y − 1 y→0
Next consider the horizontal line y = 0 and compute the limit as x approaches 1, then we have
y
lim = lim 0 = 0
(x,y)→(1,0) x + y − 1 x→1
Since the function is approaching two different values along two different paths to the point (1, 0), the
limit does not exist. 

xy
 Example 6.14 Show that lim 2 2 doesn’t exist.
(x,y)→(0,0) x +y
Solution: To show that lim x2xy 2 doesn’t exist, it suffice to show that the limits doesn’t exist at
(x,y)→(0,0) +y
different direction of approaching the point (0, 0).

Figure 6.9: f approaches from (0, 0) along different paths

Suppose, we approach (0, 0) through the parabola y = x2 . Then,

xy x3
lim = lim =0
(x,y)→(0,0) x2 + y2 (x,y)→(0,0) x3 (x + 1 )
x

On the other hand, approaching (0, 0) through the line y = x. we get,

xy x2 1
lim 2 2
= lim 2 2
=
(x,y)→(0,0) x + y (x,y)→(0,0) x + x 2

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
8 Differential Calculus of Functions of Several Variables

Hence, since
xy xy
lim 6= lim
(x,x2 )→(0,0) x2 + y2 (x,x)→(0,0) x2 + y2
xy
We conclude that lim 2 2 doesn’t exist. 
(x,y)→(0,0) x +y

xy2
 Example 6.15 Evaluate lim 2 4.
(x,y)→(0,0) x +y
Solution: First we consider the path x = 0 and get

0y2
lim = lim 0 = 0
(0,y)→(0,0) 02 + y4 y→0

Next, following the path y = 0, we get

x02
lim = lim 0 = 0
(x,0)→(0,0) x2 + 04 x→0

Since the limits along the first two paths are the same, we try another path. The next most obvious choice
of path through (0, 0) is the line y = x. As it turns out, this limit is

xx2 x
lim = lim =0
(x,x)→(0,0) x2 + x4 x→0 1 + x2

also, so we will need to try yet another path. Path y = kx

Finally, notice that along the path x = y2 , the terms x2 and y4 will be equal. We then have

y2 y2 y4 1
lim = lim =
(y2 ,y)→(0,0) (y2 )2 + y4 y→0 2y4 2

Since this limits does not agree with the limits along the earlier paths, the original limit does not exist. 

xy2
Figure 6.10: f (x, y) = x2 +y2

 Example 6.16 Determine whether lim (x2 + y2 ) ln(x2 + y2 ) exist or not by using polar coordinate.
(x,y)→(0,0)
Solution: Let x = r cos θ and y = r sin θ , then

lim (x2 + y2 ) ln(x2 + y2 ) = lim r2 ln r2 , but lim r2 ln r2 = 0 · −∞ type

(x,y)→(0,0) r→0 r→0

ln r2
= lim 1
r→0
r2
2r
r2
= lim −2 by L’Hopital Rule
r→0 3
r
= lim −r2
r→0
= 0

Therefore, lim (x2 + y2 ) ln(x2 + y2 ) exist. 

(x,y)→(0,0)

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.2 Limit and Continuity 9

Theorem 6.2.2 Suppose that | f (x, y) − L| ≤ g(x, y) for all (x, y) in the interior of some circle centered
at (a, b) except possibly at (a, b). If lim g(x, y) = 0, then lim f (x, y) = L
(x,y)→(a,b) (x,y)→(a,b)

x2 y
 Example 6.17 Evaluate lim 2 2.
(x,y)→(0,0) x +y
Solution: As we did in the earlier example, we start by looking at the limit along several paths through
(0, 0).
Along the path x = 0, we have
02 y
lim =0
(0,y)→(0,0) 02 + y2

Along the path y = 0, we have

x2 0
lim =0
(x,0)→(0,0) x2 + 02

Further, along the path y = x, we have

x3 x
lim = lim =0
(x,x)→(0,0) x2 + x2 x→0 2

At this stage, we know that if the limit exists, it must equal 0. We could try other paths, but our last
calculation gives an important clue that the limit does not exist. In that calculation, after simplifying the
expression, there remained an extra power of x in the numerator forcing the limit to 0. To show that the
limit equals L = 0, consider

x2 y
| f (x, y) − L| = | f (x, y) − 0| =
x2 + y2

Notice that if there was no y2 term in the denominator, we could cancel the x2 terms. Since x2 + y2 ≥ x2 ,
we have for x 6= 0.
x2 y x2 y
| f (x, y) − L| = 2 2
≤ 2 = |y|
x +y x
x2 y
certainly, lim |y| = 0 and so the above theorem gives as lim x 2 +y2 = 0. 
(x,y)→(0,0) (x,y)→(0,0)

6.2.2 Continuity
Definition 6.2.3 Suppose that f (x, y) is defined in the interior of a circle centered at the point (a, b).
We say that f is continuous at (a, b) if

lim f (x, y) = f (a, b)

(x,y)→(a,b)

If f (x, y) is not continuous at (a, b), then we call (a, b) a discontinuity of f .

If a function f is continuous at each point in its domain, then we will refer to f as a continous function.
For example, if f (x, y) = x and g(x, y) = y, then f and g are continous functions.

Continuity on a Set
Let R be a set in the plane. Then for each point P in the plane, one of the following conditions hold
1. There is an open disk centered at P and contained in R. In this case P is an interior point of R.
2. There is an open disk centered at P and containing no points of R. In this case P is an exterior point
of R.
3. Every open disk centered at P contains a point in R and a point outside of R. In this case P is an
boundary point of R. If a set R contains its boundary, then R is closed.

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
10 Differential Calculus of Functions of Several Variables

If (a, b) is a boundary point of the domain D of a function f , we say that f is continuous at (a, b) if

lim f (x, y) = f (a, b), where (x, y) ∈ D

(x,y)→(a,b)

This notation indicates that the limit is taken only along paths lying completely inside D.
 Example 6.18 Let R be a rectangular region consisting of all points (x, y) such that 0 ≤ x ≤ 1 and
0 ≤ y ≤ 2. Let

4−x−y for (x, y) in R
f (x, y) =
0 for (x, y) not in R
Show that f is continous on R but f is not a continous function.
Solution: Since the polynomial 4 − x − y is continous, both the polynomial and f are continous on R.
Next, since | f (x, y)| ≥ 1 for all (x, y) in R and f (x, y) = 0 for (x, y) not in R, it follows that f has no limit
at any boundary point of R and thus is not continous. 

Theorem 6.2.3 If f (x, y) and g(x, y) are continuous at (a, b), then f + g, f − g and f · g are all continu-
ous at (a, b). Further gf , provided that g(a, b) 6= 0 is continuous at (a, b).

Theorem 6.2.4 All polynomials are continuous everywhere.

x2 +y3
 Example 6.19 The function defined by f (x, y) = x3 y+4 is continuous at the point (0, 0) because

lim f (x, y) = f (0, 0)

(x,y)→(0,0)

Theorem 6.2.5 Suppose that f (x, y)is continuous at (a, b) and g(x) is continuous at the point f (a, b).
Then h(x, y) = (go f )(x, y) = g( f (x, y)) is continuous at (a, b).

2
 Example 6.20 Show that f (x, y) = ex y is continuous.
Solution: f (x, y) = g(h(x, y)), where g(t) = et and h(x, y) = x2 y. Since g is continuous for all values
of t and h is a polynomial in x and y (and hence continuous for all x and y). It follows from the above
theorem that f is continuous for all x and y. 

 Example 6.21 Find all points where the given function is continuous f (x, y) = x2x−y .
Solution: Since f (x, y) is a quotient of two polynomials and so it is continuous at any point where we
do not divide by 0. Since division by zero occurs only when y = x2 . We have that f is continuous at all
points (x, y) with y 6= x2 . 

 Example 6.22 Find all points where the given function is continuous
(
x4
x(x2 +y2 )
if (x, y) 6= (0, 0)
g(x, y) =
0 if (x, y) = (0, 0)

Solution: Since the function g(x, y) is a quotient of polynomials, except at the origin. Notice that there
is a division by zero whenever x = 0. We must consider the point (0, 0). Separately, however, since the
function is not defined by the rational expression there. We can verify that

lim g(x, y) = 0
(x,y)→(0,0)

Using the following string of inequalities. Notice that for (x, y) 6= (0, 0).

x4 x4
|g(x, y)| = |g(x, y) − 0| = | |≤| | = |x|
x(x2 + y2 ) x(x2 )

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.3 Partial Derivatives; Higher Order Partial Derivatives 11

and |x| → 0 as (x, y) → (0, 0), we have that

lim g(x, y) = 0 = g(0, 0)

(x,y)→(0,0)

So that g is continuous at (0, 0). Putting this all together, we get that g is continuous at the origin and also
at all points (x, y) with x 6= 0. 

Definition 6.2.4 Suppose that f (x, y, z) is defined in the interior of a sphere centered at the point
(a, b, c). We say that f is continuous at (a, b, c) if

lim f (x, y, z) = f (a, b, c)

(x,y,z)→(a,b,c)

If f (x, y, z) is not continuous at (a, b, c), then we call (a, b, c) a discontinuity of f .

x2 +y2 −z2
 Example 6.23 Evaluate lim 2 2 2.
(x,y,z)→(0,0,0) x +y +z
Solution: First, we consider the path x = y = 0 (the z-axis). There, we have

02 + 02 − z2 −z2
lim = lim = −1
(0,0,z)→(0,0,0) 02 + 02 + z2 z→0 z2

Along the path x = z = 0 (the y-axis), we have

02 + y2 − 02 y2
lim 2 2 2
= lim 2 = 1
(0,y,0)→(0,0,0) 0 + y + 0 z→0 y

Since the limits along these two specific paths do not agree, the limits does not exist. 

6.3 Partial Derivatives; Higher Order Partial Derivatives

6.3.1 Partial Derivatives
∂f
Definition 6.3.1 • The partial derivative of f (x, y) with respect to x written ∂x , is defined by

∂f f (x + h, y) − f (x, y)
(x, y) = lim
∂x h→0 h
for any values of x and y for which the limit exists.
∂f
• The partial derivative of f (x, y) with respect to y written ∂y , is defined by

∂f f (x, y + h) − f (x, y)
(x, y) = lim
∂y h→0 h
for any values of x and y for which the limit exists.

Remark 6.3.1 • Prime notation; that is, f 0 (x, y) is not applicable in several variables.
• The notation of partial derivatives is represented by ∂∂ xf (x, y) = fx (x, y) = ∂∂x f (x, y). Similarly,
we have ∂∂ yf (x, y) = fy (x, y) = ∂∂y f (x, y).
• To compute the partial derivative of f (x, y) respect to x; that is, ∂∂ xf simply take an ordinary
derivative with respect to x, while treating y as a constant. Similarly, we can compute ∂∂ yf by
taking an ordinary derivative with respect to y, while treating x as a constant.

 Example 6.24 Let f (x, y) = x2 + y2 . Find the partial derivatives of f at (0, 0).

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
12 Differential Calculus of Functions of Several Variables

Solution: By definition

∂f f (0 + h, 0) − f (0, 0)
(0, 0) = lim
∂x h→0 h
= lim h
h→0
= 0
∂f
Similarly, ∂ y (0, 0) = 0. 

Theorem 6.3.2 If f is differentiable at (x0 , y0 ), then f is continuous at (x0 , y0 ).

2
 Example 6.25 Evaluate the partial derivatives of f (x, y) = exy cos y.
∂f ∂f
Solution: We have to compute ∂x and ∂y respectively as

∂f 2
(x, y) = exy (y2 cos y)
∂x
2
= y2 (cos y)exy
∂f 2 2
(x, y) = (− sin y)exy + 2xy cos yexy
∂y
2
= exy (− sin y + 2xy cos y)

∂f ∂f
 Example 6.26 For f (x, y) = 3x2 + x3 y + 4y2 . Compute ∂ x (x, y), ∂ y (x, y), f x (1, 0) and fy (2, −1).
∂f
Solution: Compute ∂x by treating y as a constant. We have

∂f ∂
(x, y) = (3x2 + x3 y + 4y2 )
∂x ∂x
= 6x + (3x2 )y + 0
= 6x + 3x2 y

Notice here that the partial derivative of 4y2 with respect to x is 0, since 4y2 is treated as if it were a
constant when differentiating with respect to x. Next we compute ∂∂ yf by treating x as a constant. We have

∂f ∂
(x, y) = (3x2 + x3 + 4y2 )
∂y ∂y
= 0 + x3 (1) + 8y
= x3 + 8y

Substituting values in for x and y, we get

∂f
fx (1, 0) = (1, 0) = 6 + 0 = 6
∂x
∂f
fy (2, −1) = (2, −1) = 8 − 8 = 0
∂y


6.3.2 Higher Order Derivative

For functions of two variables, there are four different−second order partial derivatives. Those are
2
• ∂∂x ( ∂∂ xf ) = ∂∂ x2f = fxx
∂ ∂f ∂2 f
• ∂ y ( ∂ y ) = ∂ y2 = fyy

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.3 Partial Derivatives; Higher Order Partial Derivatives 13

∂ ∂f ∂2 f
• ∂x ( ∂y ) = ∂ x∂ y = fyx , mixed 2nd order partial derivatives.
∂ ∂f ∂2 f
• ∂y ( ∂x ) = ∂ y∂ x = fxy , mixed 2nd order partial derivatives.

 Example 6.27 Find all second-order partial derivative of f (x, y) = x2 y − y3 + ln x.

Solution: We start by computing the first−order partial derivatives

∂f 1 ∂f
= 2xy + and = x2 − 3y2
∂x x ∂y

we then have

∂2 f ∂ ∂f
= ( )
∂ x2 ∂x ∂x
∂ 1
= (2xy + )
∂x x
1
= 2y − 2
x

∂2 f ∂ ∂f
= ( )
∂ y∂ x ∂y ∂x
∂ 1
= (2xy + )
∂y x
= 2x

∂2 f ∂ ∂f
= ( )
∂ x∂ y ∂x ∂y
∂ 2
= (x − 3y2 )
∂x
= 2x and finally

∂2 f ∂ ∂f
= ( )
∂ y2 ∂y ∂y
∂ 2
= (x − 3y2 )
∂y
= −6y

Theorem 6.3.3 If fxy (x, y) and fyx (x, y) are continuous on an open set containing (a, b), then

fxy (a, b) = fyx (a, b)

 Example 6.28 fxy (x, y) and fyx (x, y) of f (x, y) = x2 y − y3 + ln x are equal with in the domain of f (x, y).


Partial derivatives of order 3 or higher can also be defined in similar ways as second order. For instance,
3
fxyy = ( fxy )y = ∂∂y2 ∂f x and fxyy = fyxy = fyyx if these functions are continuous.

 Example 6.29 For f (x, y) = cos(xy) − x3 + x4 . Compute fxyy and fxyyy .

Solution: we have
∂
fx (x, y) = (cos(xy) − x3 + x4 )
∂x
= −y sin(xy) − 3x2

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
14 Differential Calculus of Functions of Several Variables

Differentiating fx with respect to y gives us

∂ ∂f
fxy (x, y) =
∂y ∂x
∂
= (−y sin(xy) − 3x2 )
∂y
= − sin(xy) − xy cos(xy) and

∂ ∂ ∂f
fxyy (x, y) = ( ( ))
∂y ∂y ∂x
∂
= (− sin(xy) − xy cos(xy))
∂y
= −2x cos(xy) + x2 y sin(xy) finally, we have

∂ ∂ ∂ ∂f
fxyyy (x, y) = ( ( ( )))
∂y ∂y ∂y ∂x
∂
= (−2x cos(xy) + x2 y sin(xy))
∂y
= 2x2 sin(xy) + x2 sin(xy) + x3 y cos(xy)
= 3x2 sin(xy) + x3 y cos(xy)

6.3.3 Tangent Line

The vertical plane y = y0 ; that is, {(x, y, z)|y = y0 } intersects the graph of f in a curve C which is
essentially the graph of f (x, y0 ) considered as a function of one variable x. Then, fx (x0 , y0 ) is the
derivative of the function k(x) = f (x, y0 ) and hence, fx (x0 , y0 ) is the slope of the tangent line to the curve
C at (x0 , y0 , f (x0 , y0 )). Similarly, fy (x0 , y0 ) is the slope of the tangent line to the graph of f (x0 , y) at
(x0 , y0 , f (x0 , y0 )).
 Example 6.30 A point move along the intersection of the graph of z = x2 + 3y2 and the plane x = 2.
Find the rate at which z changes with y when the point is at (2, 1, 7).
Solution: First we have to find ∂∂ xz

∂z ∂ 2
(x, y) = (x + 3y2 )
∂y ∂y
= 6y
∂z
Now we can evaluate ∂y at a point (2, 1, 7)

∂z
(2, 1) = 6(1) = 6
∂y
Therefore, the rate at which z changes with y when the point is at (2, 1, 7) is 6. 

 Example 6.31 Find the slope of the tangent line at (2, 1, 2) to the curve of intersection of x2 +y2 +z2
=9
and the planes x = 2.
Solution: Consider x2 + y2 + z2 = 9. The intersection of the surface x2 + y2 + z2 = 9 and the plane x = 2
is the curve y2 + z2 = 5. Then, ∂∂ yz (2, 1, 2) = −1
2 . 

6.4 Directional Derivatives and Gradients

6.4.1 Directional Derivatives

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.4 Directional Derivatives and Gradients 15

Definition 6.4.1 The directional derivative of f (x, y) at the point (a, b) in the direction of the unit
vector u = (u1 , u2 ) is given by

f (a + hu1 , b + hu2 ) − f (a, b)

Du f (a, b) = lim
h→0 h
provided the limit exists.

Theorem 6.4.1 Suppose that f is differentiable at (a, b) and u = (u1 , u2 ) is any unit vector. Then, the
directional derivative of f (x, y) at the point (a, b) in the direction of the unit vector u = (u1 , u2 ) can
write as
Du f (a, b) = fx (a, b)u1 + fy (a, b)u2
√
 Example 6.32 For f (x, y) = x2 y − 4y3 , compute Du f (2, 1) for the direction u = ( 23 , 12 ) and v in the
direction from (2, 1) to (4, 0).
Solution: Regardless of the direction, we first need to compute the first partial derivatives

∂f ∂f
= 2xy and = x2 − 12y2 , then
∂x ∂y
fx (2, 1) = 4 and fy (2, 1) = −8

First we
√
have to evaluate the directional derivative in the direction of u. So, the unit vector is given as
u = ( 23 , 21 ) and so, from the above theorem we have

Du f (2, 1) = fx (2, 1)u1 + fy (2, 1)u2

√
3 1
= 4( ) − 8( )
√ 2 2
= 2 3−4

In particular, this says that the function is decreasing in this direction.

Next, we have to compute the directional derivative in the direction of v. We must first find the unit vector
u in the indicated direction. Observe that the vector from (2, 1) to (4, 0) corresponding to the position
−1
vector (2, −1) and so, the unit vector in that direction is u = ( √25 , √ 5
), we then have from the above
theorem that

Du f (2, 1) = fx (2, 1)u1 + fy (2, 1)u2

2 −1
= 4( √ ) − 8( √ )
5 5
16
= √
5
Notice that this says that the function is increasing rapidly in this direction. 

6.4.2 Gradients
Definition 6.4.2 The gradient of f (x, y) is the vector-valued function is given by

∂f ∂f ∂f ∂f
∇ f (x, y) = ( , )= i+ j
∂x ∂y ∂x ∂y
provided both partial derivatives exist.

We denote the gradient of a function f by grad f or ∇ f ( read as del f).

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
16 Differential Calculus of Functions of Several Variables

Theorem 6.4.2 If f is a differentiable function of x and y and u is any unit vector, then

Du f (x, y) = ∇ f (x, y) · u

 Example 6.33 For f (x, y) = x2 + y2 , find Du f (1, −1) for u in the direction of v = (−3, 4).
Solution: First, note that
∂f ∂f
∇ f (x, y) = ( , ) = (2x, 2y)
∂x ∂y
At the point (1, −1), we have ∇ f (1, −1) = (2, −2). A unit vector in the direction of v is u = ( −3 4
5 , 5 ). The
directional derivative in this direction is then
Du f (1, −1) = fx (2, 1)u1 + fy (2, 1)u2
−3 4
= 2( ) − 2( )
5 5
−14
=
5


Definition 6.4.3 The directional derivative of f (x, y, z) at the point (a, b, c) in the direction of the unit
vector u = (u1 , u2 , u3 ) is given by

f (a + hu1 , b + hu2 , c + hu3 ) − f (a, b, c)

Du f (a, b, c) = lim
h→0 h
provided the limit exists.

Definition 6.4.4 The gradient of f (x, y, z) is the vector-valued function

∂f ∂f ∂f ∂f ∂f ∂f
∇ f (x, y, z) = ( , , )= i+ j+ k
∂x ∂y ∂z ∂x ∂y ∂z
provided all partial derivatives exist.

Theorem 6.4.3 If f is a differentiable function of x, y and z, respectively and u is any unit vector, then

Du f (x, y, z) = ∇ f (x, y, z) · u

We have
Du f (x, y, z) = ∇ f (x, y, z) · u
= k∇ f (x, y, z)kkuk cos θ
= k∇ f (x, y, z)k cos θ
Where θ is the angle between the vector ∇ f (x, y, z) and u.

Theorem 6.4.4 Suppose f (x, y) is differentiable at (x0 , y0 ). Then, f (x, y) has directional derivative at
any direction.

Remark 6.4.5 Let f be a function of three variables x, y and z. Then,

• ∇ f (x0 , y0 , z0 ) = fx (x0 , y0 , z0 )i + fy (x0 , y0 , z0 ) j + fz (x0 , y0 , z0 )k.
• Du f (x, y, z) = ∇ f (x, y, z) · u, for any unit vector u = u1 i + u2 j + u3 k.

Theorem 6.4.6 Suppose f is differentiable at (x0 , y0 ). Then

1. ∇ f (x0 , y0 ) = 0 implies Du f (x0 , y0 ) = 0 for any vector u.
2. If ∇ f (x0 , y0 ) 6= 0, then the directional derivative of f in the direction of ∇ f (x0 , y0 ) is the largest

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.5 Tangent Plane and Tangent Plane Approximation 17

and it’s value is k∇ f (x0 , y0 )k.

3. If ∇ f (x0 , y0 ) 6= 0, then Du f (x, y), regarded as a function of u, attains its maximum value when u
points in the same direction as ∇ f (x0 , y0 ).

 Example 6.34 Find the maximum and minimum rates of change of the function f (x, y) = x2 + y2 at the
point (1, 3).
Solution: We first compute the gradient ∇ f (x, y) = (2x, 2y) and evaluate it at the point (1, 3)
∇ f (1, 3) = (2, 6)
From the above theorem, the maximum rate of change of f at (1, 3) is
√
k∇ f (1, 3)k = k(2, 6)k = 40
∇ f (1,3) (2,6) √1 (2, 6).
and occur in the direction u = k∇ f (1,3)k =
√
40
= 40
Similarly, the minimum rate of change is
√
−k∇ f (1, 3)k = −k(2, 6)k = − 40
∇ f (1,3) (2,6) √1
which occurs in the direction u = − k∇ f (1,3)k = − 40 = − 40 (2, 6).
√ 

Theorem 6.4.7 Suppose f is differentiable at (x0 , y0 , z0 ). Then

1. ∇ f (x0 , y0 , z0 ) = 0 implies Du f (x0 , y0 , z0 ) = 0 for any vector u.
2. If ∇ f (x0 , y0 , z0 ) 6= 0, then the directional derivative of f in the direction of ∇ f (x0 , y0 , z0 ) is the
largest and it’s value is k∇ f (x0 , y0 , z0 )k.
3. If ∇ f (x0 , y0 , z0 ) 6= 0, then Du f (x, y, z), regarded as a function of u, attains its maximum value
when u points in the same direction as ∇ f (x0 , y0 , z0 ).
2 2

z
Example 6.35 If the temperature at the point (x, y, z) is given by T (x, y, z) = 85 + (1 − 100 )e−(x +y ) .
Find the direction from the point (2, 0, 99) in which the temperature increases most rapidly.
Solution: We first compute the gradient
∂T ∂T ∂T
∇T (x, y, z) = ( , , )
∂x ∂y ∂z
z 2 2 z 2 2 1 −(x2 +y2 )
= (−2x(1 − )e−(x +y ) , −2y(1 − )e−(x +y ) , − e )
100 100 100
−1 −4 −1 −4
∇T (2, 0, 99) =( e , 0, e )
25 100
To find a unit vector in this direction, you can simplify the algebra by canceling the common factor of
e−4 and multiplying by 100. A unit vector in the direction of (−4, 0, −1) and also in the direction of
∇T (2, 0, 99), is then √117 (−4, 0, −1). 

6.5 Tangent Plane and Tangent Plane Approximation

6.5.1 Total Differential
For a differentials of two variables z = f (x, y), we define the differentials dx and dy to be independent
variables; that is, they can be given any values. Then the differential, also called the total differential, is
defined by
dz = fx (x, y)dx + fy (x, y)dy
∂z ∂z
= dx + dy
∂x ∂y
The sum of the partial differentials with respect to all of the independent variables is the total differential
given by
∂y ∂y
dy = dx1 + · · · + dxn
∂ x1 ∂ xn

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
18 Differential Calculus of Functions of Several Variables

where y = f (x1 , x2 , · · · , xn ).
 Example 6.36 Let z = 4x3 y2 , find the total differential.
Solution: Since f (x, y) = 4x3 y2 , and then fx (x, y) = 12x2 y2 and fy (x, y) = 8x3 y. Therefore

dz = 12x2 y2 dx + 8x3 ydy

6.5.2 Tangent Plane

Definition 6.5.1 Let f be a differentiable function at a point (x0 , y0 , z0 ) on a level surface S of f . If
∇ f (x0 , y0 , z0 ) 6= 0, the plane through (x0 , y0 , z0 ) whose normal is ∇ f (x0 , y0 , z0 ) is the plane tangent
to S at (x0 , y0 , z0 ). Any vector that is perpendicular to this tangent plane is said to be normal to the
level surface S. This means ∇ f (x0 , y0 , z0 ) is normal to the level surface S force f (x0 , y0 , z0 ) = c at
(x0 , y0 , z0 ).

 Example 6.37 Find a normal vector to the surface xy3 z2 = 4 at (−1, −1, 2).
Solution: Let f (x, y, z) = xy3 z2 − 4, so ∇ f (x, y, z) = (y3 z2 , 3xy2 z2 , 2xy3 z). Therefore the normal vector to
the surface at a point (−1, −1, 2) is

∇ f (−1, −1, 2) = (−4, −12, 4)

Suppose f has continuous partial derivatives. An equation of the tangent plane to the surface z = f (x, y)
at the point (x0 , y0 , z0 ) is

z − z0 = fx (x0 , y0 )(x − x0 ) + fy (x0 , y0 )(y − y0 )

Alternatively, let g(x, y, z) = z − f (x, y) and let (x, y, z) be any arbitrary point on a tangent plane. Since
∇g(x0 , y0 , z0 ) is the normal vector to the surface at (x0 , y0 , z0 ), we have

∇g(x0 , y0 , z0 ) · (x − x0 , y − y0 , z − z0 ) = 0
⇒ (− fx (x0 , y0 ), − fy (x0 , y0 ), 1) · (x − x0 , y − y0 , z − z0 ) = 0
⇒ − fx (x0 , y0 )(x − x0 ) − fy (x0 , y0 )y − y0 ) + z − z0 = 0
⇒ z − z0 = fx (x0 , y0 )(x − x0 ) + fy (x0 , y0 )(y − y0 )

is the equation of the tangent plane to the surface z = f (x, y) at the point (x0 , y0 , z0 ).
Definition 6.5.2 Suppose f is differentiable at P0 = (x0 , y0 ). Then, the equation of the tangent plane
to the graph of f at (P0 , f (P0 )) is given by z = f (P0 ) + (P − P0 ) · ∇ f (P0 ) or

f (x0 , y0 ) + fx (x0 , y0 )(x − x0 ) + fy (x0 , y0 )(y − y0 ) − z = 0

Theorem 6.5.1 The equation of the tangent plane to the surface f (x, y, z) = c at the point (x0 , y0 , z0 ) is
given by
∂ f (x, y, z) ∂ f (x, y, z) ∂ f (x, y, z)
(x − x0 ) + (y − y0 ) + (z − z0 ) = 0
∂x ∂y ∂z

 Example 6.38 Find the tangent plane to the elliptic paraboloid z = 3x2 + 2y2 at the point (1, 1, 5).
Solution: Let f (x, y) = 3x2 + 2y2 , then

fx (x, y) = 6x and fy (x, y) = 4y, then

fx (1, 1) = 6 and fy (1, 1) = 4

Therefore, the equation of the tangent plane is

z − 5 = 6(x − 1) + 4(y − 1) ⇒ z = 6x + 4y − 5

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.5 Tangent Plane and Tangent Plane Approximation 19

 Example 6.39 Find the equation of the tangent plane to a surface x2 + 2y2 + 3z2 at (4, −1, 1).
Solution: First we have to find the normal vector to the plane as ∇ f (x, y, z) = (2x, 4y, 6z) and the normal
vector at (4, −1, 1) will be ∇ f (4, −1, 1) = (8, −4, 6). Let (x, y, z) be any arbitrary point on a tangent plane.
So
∇ f (4, −1, 1) · (x − 4, y + 1, z − 1) = 0
(8, −4, 6) · (x − 4, y + 1, z − 1) = 0
8(x − 4) − 4(y + 1) + 6(z − 1) = 0
8x − 4y + 6z = 42
Therefore, 8x − 4y + 6z = 42 is the tangent plane to the surface x2 + 2y2 + 3z2 at (4, −1, 1). 

6.5.3 Tangent Plane Approximation

Definition 6.5.3 If z = f (x, y), f is differentiable at the point (x0 , y0 ) if 4z can be expressed in the
form
4z = fx (x0 , y0 )4x + fy (x0 , y0 )4y + ε1 4x + ε2 4y
where ε1 → 0 and ε2 → 0 as (4x, 4y) → (0, 0).

Replace 4 f by 4z in the tangent plane equation and then we can rewrite this as
f (x0 + 4x, y0 + 4y) = f (x0 , y0 ) + fx (x0 , y0 )4x + fy (x0 , y0 )4y + ε1 4x + ε2 4y
Thus, if 4x ≈ 0 and 4y ≈ 0, the terms involving ε1 and ε2 will be small, and we can approximate
f (x0 + 4x, y0 + 4y) as
f (x0 + 4x, y0 + 4y) ≈ f (x0 , y0 ) + fx (x0 , y0 )4x + fy (x0 , y0 )4y
This formula, which is the two-variables, is called the local tangent planer approximation of f at (x0 , y0 ).
When convenient, we can replace 4x and 4y in this equation by 4x = x − x0 and 4y = y − y0 and then
rewrite the equation as
f (x, y) ≈ f (x0 , y0 ) + fx (x0 , y0 )(x − x0 ) + fy (x0 , y0 )(y − y0 )
It’s sometimes hard to use the above definition directly to check the differentiability of a function, but the
next theorem provides a convenient sufficient condition for differentiability.

Theorem 6.5.2 If the partial derivatives fx and fy exist near (x0 , y0 ) and are continuous at (x0 , y0 ), then
f is differentiable at (x0 , y0 ).

 Example 6.40 Show that f (x, y) = xexy is differentiable at (1, 0) and find its linearization there. Then
use it to approximate f (1, 1, −0.1).
Solution: The partial derivatives are
fx (x, y) = exy + xyexy fy (x, y) = x2 exy
fx (1, 0) = 1 fy (1, 0) = 1
Both fx and fy are continouse functions, so f is differentiable by the above theorem. The linearization is
L(x, y) = f (x, y) + fx (1, 0)(x − 1) + fy (1, 0)(y − 0)
= 1 + 1(x − 1) + 1(y)
= x+y
The corresponding linear approximation is
xexy ≈ x + y
So, f (1.1, −0.1) ≈ 1.1 − 0.1 = 1. Comparare this with the actual value of
f (1.1, −0.1) ≈ 1.1e−0.11 ≈ 0.98542


Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
20 Differential Calculus of Functions of Several Variables

 Example 6.41 An equation of the tangent plane to the graph of the function f (x, y) = 3x2 + 2y2 at the
point (1, 1, 3) is z = 6x + 4y − 5. Therefore,the linear function of two variables L(x, y) = 6x + 4y − 5 is a
good approximation to f (x, y) when (x, y) is near (1, 1).The function L is called the linearizion of f at
(1, 1) and the approximation
f (x, y) ≈ 6x + 4y − 5
is called a tangent plane approximation of f at (1, 1). 

6.6 Chain Rule and Implicitly Differentiation

6.6.1 Chain Rule
If z = f (x(t), y(t)), where x(t) and y(t) are differentiable and f (x, y) is a differentiable function of x and
y, then

dz d
= ( f (x(t), y(t)))
dt dt
∂f dx ∂ f dy
= (x(t), y(t)) + (x(t), y(t))
∂x dt ∂ y dt

For a function z = f (x(t), y(t))

• t is independent variables
• x and y are intermediate variables
• z is dependent variable
∂f
 Example 6.42 Consider f (x, y) = x2 cos y, where x(t) = t 2 + 2t, y(t) = sint. Then evaluate ∂t .
Solution:
∂f ∂ f dx ∂ f dy
= · + ·
∂t ∂ x dt ∂ y dt
= (2x cos y)(2t + 2) − (x2 sin y)(cost)
= 2(t 2 + 2t) cos(sint)(2t + 2) − (t 2 + 2t)2 sin(sint) cost

If z = f (x, y), where f is a differentiable function of x and y and where both x = x(s,t) and y = y(s,t)
have first order partial derivatives. Then we have the chain rules

dz ∂ f dx ∂ f dy
= . + . and
ds ∂ x ds ∂ y ds
dz ∂ f dx ∂ f dy
= . + .
dt ∂ x dt ∂ y dt


 Example 6.43 For f (x, y) = x2 ey , x(t) = t 2 −1 and y(t) = sint, find the derivative of g(t) = f (x(t), y(t)).
∂f ∂f dx dy
Solution: We first compute the derivatives ∂x = 2xey , ∂y = x2 ey , dt = 2t and dt = cost. The chain rule
then gives us

∂ f dx ∂ f dy
g0 (t) = +
∂ x dt ∂ y dt
= 2xey (2t) + x2 ey cost
= 2(t 2 − 1)esint (2t) + (t 2 − 1)2 esint cost

 Example 6.44 Suppose that f (x, y) = exy , x(u, v) = 3u sin v and y(u, v) = 4v2 u. For g(u, v) = f (x(u, v), y(u, v)),
∂g ∂g
find the partial derivatives and
∂u ∂v .
∂f ∂f ∂x ∂y
Solution: We first compute the derivatives ∂x = yexy , ∂y = xexy , ∂u = 3 sin v and ∂u = 4v2 . The chain

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.7 Relative Extrema of Function of Two Variables 21

rule then gives us

∂g ∂ f ∂x ∂ f ∂y
= +
∂u ∂x ∂u ∂y ∂u
= yexy (3 sin v) + xexy (4v2 )
2 v2 sin v 2 v2 sin v
= (4v2 u)e12u (3 sin v) + (3u sin v)e12u (4v2 )
∂x ∂y
For the partial derivative of g with respect to v, we compute ∂v = 3u cos v and ∂v = 8uv, Here the chain
rule gives us

∂g ∂ f ∂x ∂ f ∂y
= +
∂v ∂x ∂v ∂y ∂v
= yexy (3u cos v) + xexy (8uv)
2 v2 sin v 2 v2 sin v
= (4v2 u)e12u (3u cos v) + (3u sin v)e12u (8uv)

6.6.2 Implicitly Differentiation

Theorem 6.6.1 If the evaluation F(x, y, z) = 0 defines z implicitly as a differentiable function of x and
y; that is, z = f (x, y) and if Fz 6= 0, then

∂z Fx ∂z Fy
= − and =−
∂x Fz ∂y Fz

 Example 6.45 Find ∂∂ xz and ∂∂ yz , given that F(x, y, z) = xy2 + z3 + sin(xyz) = 0.

Solution: First, note that using the usual chain rule, we have

Fx = y2 + yz cos(xyz)
Fy = 2xy + xz cos(xyz) and
Fz = 3z2 + xy cos(xyz)

We now have
∂z Fx y2 + yz cos(xyz)
= − =− 2 and
∂x Fz 3z + xy cos(xyz)
∂z Fy 2xy + xz cos(xyz)
= − =− 2
∂y Fz 3z + xy cos(xyz)


Theorem 6.6.2 Implicit Function Theorem

If Fx ,Fy and Fz are continuous inside a sphere containing the point (a, b, c), where F(a, b, c) = 0 and
Fz (a, b, c) 6= 0, then the equation F(x, y, z) = 0 implicitly defines z as a function of x and y near by the
point (a, b, c).

6.7 Relative Extrema of Function of Two Variables

Definition 6.7.1 Let f (x, y) be a function of two variables. Then,
• f (x, y) is said to have relative maximum value at (x0 , y0 ) if and only if there exists a disk
centered at (x0 , y0 ) such that f (x, y) ≤ f (x0 , y0 ) for all (x, y) in the interior of the disk.
• f (x, y) is said to have relative minimum value at (x0 , y0 ) if and only if there exists a disk centered
at (x0 , y0 ) such that f (x, y) ≥ f (x0 , y0 ) for all (x, y) in the interior of the disk.

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
22 Differential Calculus of Functions of Several Variables

In either case f (x0 , y0 ) is called a local extreme of f (x, y).

Figure 6.11: Local and absolute extremum values

Definition 6.7.2 Suppose f is a function of two variables. Let (x0 , y0 ) be in the domain of f . Then,
(x0 , y0 ) is called the critical point of f if and only if fx (x0 , y0 ) = fy (x0 , y0 ) = 0 or either fx (x0 , y0 ) or
fy (x0 , y0 ) doesn’t exist.

Local extrema can occur at critical points, every critical points need not correspond to a local extreme.
For this reason, we refer to critical points as candidates for local extrema.

Theorem 6.7.1 If f (x, y) has a local extreme at (x0 , y0 ), then (x0 , y0 ) must be a critical point of f .

Definition 6.7.3 A critical point of a function f that is neither a point of relative maximum nor a point
of minimum is called a saddle point of f .

Theorem 6.7.2 If f has relative extreme at (x0 , y0 ) and fx (x0 , y0 ) and fy (x0 , y0 ) exist, then

fx (x0 , y0 ) = fy (x0 , y0 ) = 0

 Example 6.46 Consider f (x, y) = y2 − x2 . Then, we see that fx (x, y) = fy (x, y) = 0 implies that
x = 0 and y = 0. For any circular disk centered at (0, 0) in the xy plane, we can easily see that f is positive
on the y axis and f is negative on the x axis. This shows that the point (0, 0) is neither a point of relative
maximum nor a point of relative minimum of f .


Figure 6.12: f (x, y) = y2 − x2

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.8 Largest and Smallest Values of a Function on a Given Set 23

Theorem 6.7.3 Second Derivatives Test

Let f be a function of two variables with continuous second-order partial derivatives in some disk
centered at the critical point (x0 , y0 ) . Let

D(x0 , y0 ) = fxx (x0 , y0 ) · fyy (x0 , y0 ) − ( fxy (x0 , y0 ))2

1. If D > 0 and fxx (x0 , y0 ) > 0, then f has relative minimum at (x0 , y0 ).
2. If D > 0 and fxx (x0 , y0 ) < 0, then f has relative maximum at (x0 , y0 ).
3. If D < 0, then f has saddle point at (x0 , y0 ).
4. If D = 0, no conclusion can be drawn.

 Example 6.47 Consider f (x, y) = x2 + xy − 2y − 2x + 1. Find all the critical points and determine the
relative maximum and relative minimum values of f .
Solution: Let (x, y) be a relative maximum or relative minimum point of f . Then, we get fx (x, y) = 0
and fy (x, y) = 0; that is, we get the system of equations with solution (2, −2). Thus, (2, −2) is the only
critical point of f .
Now, it is easy to see that D < 0. Hence, by the second partial test, we conclude that (2, −2)is a saddle
point of f . 

Example 6.48 Locate all the critical points and find all the points of relative maximum and relative
minimum of f (x, y) = 2x2 − y3 − 2xy.
Solution: We first compute the first partial derivatives

fx = 4x − 2y and fy = −3y2 − 2x

Since both partial derivatives are defined for all (x, y), the critical points are solutions of the two equations.

fx = 4x − 2y = 0 and fy = −3y2 − 2x = 0

Solving the first equation for y, we get y = 2x. Substituting this in to the second equation, we have

0 = −3(4x2 ) − 2x
= −12x2 − 2x
= −2x(6x + 1)

So that x = 0 or x = −1 6 .
The corresponding y-values are y = 0 and y = − 13 . The only two critical points are then (0, 0) and
(− 16 , − 13 ). To classify these points, we first compute the second partial derivatives

fxx = 4, fyy = −6y and fxy = −2

then test the discriminant. We have

D(0, 0) = (4)(0) − (−2)2 = −4

−1 −1
D( , ) = (4)(2) − (−2)2 = 4
6 3
From the above theorem. We conclude that there is a saddle point of f at (0, 0), since D(0, 0) < 0. Further,
there is a local minimum at (− 16 , − 13 ). Since D(− 16 , − 13 ) > 0 and fxx (− 61 , − 13 ) > 0. 

6.8 Largest and Smallest Values of a Function on a Given Set

Definition 6.8.1 Let f (x, y) be a function of two variables. Then
• f (x, y) is said to have absolute maximum at (x0 , y0 ) if and only if f (x, y) ≤ f (x0 , y0 ) for all (x, y)
in the domain of f .
• f (x, y) is said to have absolute minimum at (x0 , y0 ) if and only if f (x, y) ≥ f (x0 , y0 ) for all (x, y)
in the domain of f .

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
24 Differential Calculus of Functions of Several Variables

Theorem 6.8.1 If f (x, y) is continuous in a closed and bounded subset S of the plane, then f has both
absolute maximum(minimum)value in S. Further, the absolute extrema must occur at either at a critical
point in R or on the boundary of R.

Remark 6.8.2 The following steps helps us to find absolute extreme values of f in a bounded region
S.
• Step 1: Find critical points that lie in the interior of S.
• Step 2: Obtain the maximum(minimum) value of f on the boundary of S.
• Step 3: Compare the values of f obtained in step 1 and step 2.

2
 Example 6.49 Let f (x, y) = xy − x and let R be the square region in the first quadrant having vertex

(0, 0), (1, 0), (1, 1) and (0, 1). Find the extreme values of f on R.
Solution: By the Maximum-minimum Theorem f has extreme values on R. Since

fx (x, y) = y − 2x and fy (x, y) = x

it follows that fx (x, y) = 0 if y = 2x and fy (x, y) = 0 if x = 0. Thus the only critical point of f is (0, 0),
which happens to be a boundary point of R. Therefore the extreme values of f on R must occure on
the boundary of R, which is composed of the four line segments. On the first line segment, x = 0 and
0 ≤ y ≤ 1, and since f (0, y) = 0, the maximum and minimum values of f are both 0. On the second line
segment, y = 0 and 0 ≤ x ≤ 1, and since

f (x, 0) = −x2

the maximum value of f on this line is 0 and the minimum value of f is −1. On the third line segment,
x = 1 and 0 ≤ y ≤ 1, and since
f (1, y) = y − 1

the maximum value of f on this line segment is 0 and the minium value is −1. Finally, on the fourth line
segment, y = 1 and 0 ≤ x ≤ 1, and since

f (x, 1) = x − x2

the maximum value of f is 14 and the minium value is 0. By comparing the extreme values, we conclude
that the maximum value of f on R is 14 and the minimum value is −1. 

 Example 6.50 Consider the function f (x, y) = x2 + y2 . It can be easily shown that f has an absolute
minimum at the origin.(see the graph below).


Figure 6.13: f (x, y) = x2 + y2

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.9 Extreme Value under Constraint Condition 25

6.9 Extreme Value under Constraint Condition

Lagrange Multiplier

6.9.1 Method of Lagrange Multipliers for One Constraint

Consider the problem

Maximum(minimum) of f (x, y, z)
subject to
g(x, y, z) = 0

(assuming that these extreme values exist and ∇g 6= 0 on the surface g(x, y, z) = 0)
• Find all values of x, y, z and λ such that ∇ f (x, y, z) = λ ∇g(x, y, z) and g(x, y, z) = 0.
• Evaluate f at all points (x, y, z) that result from (a).
The largest of these values is the maximum value of f ; the smallest is the minimum value of f .
 Example 6.51 Find the maximum and minimum of the function f (x, y) = x2 + y2 on the ellipse
x2 + 2y2 = 1.
Solution: Let g(x, y) = x2 + 2y2 − 1 and suppose (x, y) is a point of extreme value of f on the ellipse.
Then, there exists a real number λ such that

∇ f (x, y) = λ ∇g(x, y)

This is equivalent to solving 

2x = λ (2x)
2y = λ (4y)
for x, y and λ . Since (x, y) is on the ellipse, both x and y can not be zero at the same time. Hence, we get
either x = 0, λ = 12 or y = 0, λ = 1. Since we seek points on the ellipse, x = 0 implies y = ± √12 and y = 0
implies that x = ±1.
Moreover,we see that f (0, ± √12 ) = 12 , f (±1, 0) = 1. We conclude that if f has maximum, then it occurs
at (±1, 0) and if it has minimum, then it occurs at (0, ± √12 ). 

Example 6.52 Find the point on the line y = 3 − 2x that is closest to the origin.
Solution: the distance from the origin to the point (x, y) on the line is expressed as
p
h(x, y) = x2 + y2

but before we continue, we must point out that the distance is minimized at exactly the same point at
which the square of the distance is minimized.
Notice that if we instead minimize the square of the distance, given by

f (x, y) = x2 + y2

for f (x, y) = x2 + y2 , we have ∇ f (x, y) = (2x, 2y) and for g(x, y) = 2x + y − 3, we have ∇g(x, y) = (2, 1).
The vector equation ∇ f (x, y) = λ ∇g(x, y) becomes

(2x, 2y) = λ (2, 1)

from which it follows that

2x = 2λ and 2y = λ
Solving the second equation for λ , we have λ = 2y. The first equation then gives us x = λ = 2y.
Substituting x = 2y in to the constraint equation y = 3 − 2x, we have

y = 3 − 2(2y) or 5y = 3

The solution is y = 53 , giving us x = 2y = 65 . The closest point is then ( 65 , 53 ). 

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
26 Differential Calculus of Functions of Several Variables
ft
 Example 6.53 A rocket is launched with a constant thrust corresponding to an acceleration of u s2 .
1 2
Ignoring air resistance, the rocket’s height after t second is given by f (t, u) = 2 (u − 32)t feet. Fuel
2
usage for t seconds is proportional to u t, so the limited fuel capacity of the rocket can be expressed by an
equation of the form u2t = 10, 000. Find the value of u that maximized the height that the rocket reaches.
Solution: Now we have to find solutions of ∇ f (t, u) = λ ∇g(t, u) where g(t, u) = u2t − 10, 000 = 0 is the
constriant equation. We have

∇ f (t, u) = λ ∇g(t, u) and

∇g(t, u) = (u2 , 2ut)

Now we must have

1
((u − 32)t, t 2 ) = λ (u2 , 2ut)
2
For some constant λ . It follows that
1
(u − 32)t = λ u2 and t 2 = λ 2ut
2
solving both equations for λ , we have
1 2
(u − 32)t t
λ= 2
= 2
u 2ut
This gives us
1
2u(u − 32)t 2 = t 2 u2
2
Solution include t = 0 and u = 0 on physical ground, we can argue that these solutions represent minimum
heights, since zero thrust or zero time would correspond to zero height. Canceling the factors t 2 and
u, we have 4(u − 32) = u. The solution to this is u = 1283 , with this value of u, the engines can burn
10,000 10,000
t = 2u2 = 128 2 ≈ 5.5s with the rocket reaches height z = 21 ( 128 2
3 − 32)(5.5s) ≈ 161 feet. 
( 3 )
2
 Example 6.54 Suppose that the temperature of a metal plate is given by T (x, y) = x + 2x + y , for
2

points (x, y) on the elliptical plate defined by x2 + 4y2 ≤ 24. Find the maximum and minimum temperature
on the plate.
Solution: The plate corresponds to the region R on side x2 + 4y2 ≤ 24. We first look for critical points of
T (x, y) in side the region R, we have ∇T (x, y) = (2x + 2, 2y) = (0, 0) if x = −1 and y = 0. At this point,
T (−1, 0) = −1. We next look for the extrema of T (x, y) on the ellipse x2 + 4y2 = 24. We first rewrite the
constraints equation as g(x, y) = x2 + 4y2 − 24 = 0.
Now any extrema will satisfy the Lagrange multiplier equation

∇T (x, y) = λ ∇g(x, y) or
(2x, 2y) = λ (2x, 8y)
= (2λ x, 8λ y)

This occurs when 2x + 2 = 2λ x and 2y = 8λ y

Notice that the second equation hold when y = 0 or λ = 41 . If y = 0, the constraint x2 + 4y2 = 24 gives
√
x = ± 24. If λ = 14 , the first equation becomes 2x + 2 = 12 x so that x = − 34 . The constraint x2 + 4y2 = 24
√
now gives y = ± 350 . We now need only compare the function values at all these points (the one interior
critical point and the candidates for boundary extrema) as follows

T (−1, 0) = −1
√ √
T ( 24, 0) = 24 + 2 24 ≈ 33.8
√ √
T (− 24, 0) = 24 − 2 24 ≈ 14.2
√
−4 50 14
T( , ) = ≈ 4.7 and
3√ 3 3
−4 −50 14
T( , ) = ≈ 4.7
3 3 3

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.9 Extreme Value under Constraint Condition 27

From this√list, it is easy to identify

√ the minimum value of −1 at the point (−1, 0) and the maximum value
of 24 + 2 24 at the point ( 24, 0). 

 Example 6.55 For a business that produces three products, suppose that when producing x, y and z
thousand units of the products, the profit of the company (in thousand of dollars) can be modeled by
P(x, y, z) = 4x + 8y + 6z. Manufacturing constraints force x2 + 4y2 + 2z2 ≤ 800. Find the maximum profit
for the company.
Solution: We start with ∇P(x, y, z) = (4, 8, 6) and note that there are no critical points. This says that
the extrema must lie on the boundary of the constraint region. That is, they must satisfy the constraint
equation
g(x, , y, z) = x2 + 4y2 + 2z2 − 800 = 0
Then the Lagrange multiplier equation is
∇P(x, y, z) = λ ∇g(x, y, z) or
(4, 8, 6) = λ (2x, 8y, 4z)
= (2λ x, 8λ y, 4λ z)
This occurs when 4 = 2λ x, 8 = 8λ y and 6 = 4λ z
From the first equation, we get x = λ2 . The second equation gives us y = λ1 and finally the third equation
3
gives us z = 2λ . From the constraint equation x2 + 4y2 + 2z2 = 800.
We know have
2 1 3 25
800 = ( )2 + 4( )2 + 2( )2 =
λ λ 2λ 2λ 2
2 25 1
so that λ = 1600 and then λ = 8 .
The values of x, y and z are positive since they represent a number of quantity, so since x > 0, we must
have λ = 2x > 0.
The only candidate for an extrema is then
2 1 3
x= = 16, y = = 8 and z = = 12
λ λ 2λ
Where the corresponding profit is P(16, 8, 12) = 4(14) + 8(8) + 6(12) = 200
Observe that this is the maximum profit, while λ = − 18 gives the minimum value of the profit function. 

Consider the problem

Maximum(minimum) of f (x, y, z)
subject to
g(x, y, z) ≤ 0
we first find the critical points of f (x, y, z) that satisfy the constraint, then find the extrema of the function
on the boundary g(x, y, z) = 0 (the constraint curve or set ) and finally, compare the function values.

6.9.2 Method of Lagrange Multipliers for Two Constraints

Consider the problem
Maximum(minimum) of f (x, y, z)
subject to
g(x, y, z) = 0
h(x, y, z) = 0
(assuming that these extreme values exist, ∇g 6= 0 and ∇h 6= 0 on the surface g(x, y, z) = 0 and h(x, y, z) = 0,
respectively and ∇g is not parallel with ∇h)
• Step 1: Find all values of x, y, z, λ and µ such that
∇ f (x, y, z) = λ ∇g(x, y, z) + µ∇h(x, y, z)
g(x, y, z) = 0 and h(x, y, z) = 0

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
28 Differential Calculus of Functions of Several Variables

Or

fx (x, y, z) = λ gx (x, y, z) + µhx (x, y, z)

fy (x, y, z) = λ gy (x, y, z) + µhy (x, y, z)
fz (x, y, z) = λ gz (x, y, z) + µhz (x, y, z)
g(x, y, z) = 0 and h(x, y, z) = 0

• Step 2: Evaluate f at all points (x, y, z) that result from Step 1. The largest of these values is the
maximum value of f ; the smallest is the minimum value of f .
 Example 6.56 The plane x + y + z = 12 intersects the paraboloid z = x2 + y2 in an ellipse. Find point
on the ellipse that is closest to the origin.
Solution: We know that the minimizing the distance to the origin is equivalent to minimizing

f (x, y, z) = x2 + y2 + z2

We have constraints

g(x, y, z) = x + y + z − 12 = 0 and
h(x, y, z) = x2 + y2 − z = 0
So, f (x, y, z) = λ ∇g(x, y, z) + µ∇h(x, y, z) or
(2x, 2y, 2z) = λ (1, 1, 1) + µ(2x, 2y, −1)

Together with the constraint equations, we now have the system of equations.

2x = λ + 2µx
2y = λ + 2µy
2z = λ − µ
x + y + z − 12 = 0
and x2 + y2 − z = 0

From the above equations (2, 2, 8) and (−3, −3, 18) are the only candidates for extrema. Finally, since
f (2, 2, 8) = 72 and f (−3, −3, 18) = 342. Therefore, the closest point on the intersection of the two
surfaces to the origin is (2, 2, 8).
By the same reasoning, observe that the furthest point on the intersection of the two surfaces from the
origin is (−3, −3, 18). 

Consider the problem

Maximum(minimum) of f (x, y, z)
subject to
g(x, y, z) ≤ 0
h(x, y, z) ≤ 0

we first find the critical points of f (x, y, z) that satisfy the constraint, then find the extrema of the function
on the boundary g(x, y, z) = 0 and h(x, y, z) = 0 (the constraint curve or set) and finally, compare the
function values.

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
6.10 Exercise 29

6.10 Exercise
1. Let f (x, y) = x2 e3xy .
(a) Evaluate f (1, 1)
(b) Find the domain of f
(c) Find the range of f

Ans. 4, R2 , [0, ∞)
√
2 2
2. Let f (x, y, z) = e z−x −y .
(a) Evaluate f (2, −1, 6)
(b) Find the domain of f
(c) Find the range of f

Ans. e, {(x, y, z)|z ≥ x2 + y2 }, [1, ∞)

y4
3. Find the limit lim x 4 +3y4 if it exist.
(x,y)→(0,0)

Ans. Does not exist

4. Find the limit lim √ xy if it exist.

(x,y)→(0,0) x2 +y2

Ans. 0
xy+yz2 +xz2
5. Find the limit lim 2 2 2 if it exist.
(x,y,z)→(0,0,0) x +y +z

Ans. Does not exist

√
6. Find h(x, y) = g( f (x, y)) and the set on which h is continuous, where g(t) = t 2 + t, f (x, y) =
2x + 3y − 6.
√
Ans. h(x, y) = (2x + 3y − 6)2 + 2x + 3y − 6, {(x, y)|2x + 3y ≥ 6}

7. Determine the set of points at which the function is continuous, where

( 23
x y
2x2 +y2
if (x, y) 6= (0, 0)
f (x, y) =
1 if (x, y) = (0, 0)

Ans. {(x, y)|(x, y) 6= (0, 0)}

3 3
8. Use polar coordinate to find the limit lim x2 +y2 .
(x,y)→(0,0) x +y

Ans. 0
x−y
9. Find the first partial derivatives of the function f (x, y) = x+y .

2y −2x
Ans. fx = x+y , f y = x+y

10. Find the first partial derivatives of the function f (x, y, z) = ln(x + 2y + 3z).
1 2 3
Ans. fx = x+2y+3z , f y = x+2y+3z , f z = x+2y+3z

∂3 f ∂3 f x
11. Find ∂ z∂ y∂ x and ∂ x2 ∂ y
for a function f (x, y, z) = y+2z .

4
Ans. (y+2z)3
, 0

12. Find an equation of the tangent plane to the surface z = y cos(x − y) at (2, 2, 2).

Ans. z = y

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]
30 Differential Calculus of Functions of Several Variables
√
13. Find the linearization of the function f (x, y) = x y at (1, 4).

Ans. L(x, y) = 2x − 41 y − 1
y
dw
14. Use chain rule to find dt for the functions w = xe z , x = t 2 , y = 1 − t, z = 1 + 2t.
y
Ans. dw
dt = e z (2t − 2x − 2xy
z2
)
dz
√
15. Use chain rule to find dt for the functions z = er cos θ , r = st, θ = s2 + t 2 .
dz
Ans. dt = er (t cos θ − √ 2s sin θ )
s +t 2
√
16. Find the directional derivative of the function f (x, y) = 1 + 2x x at the point (3, 4) in the direction
of v = (3, 4).
23
Ans. 10
√
17. Find the directional derivative of the function f (x, y) = xy at the point (2, 8) in the direction of
v = (5, 4).
2
Ans. 5

18. Find the maximum rate of change of f (x, y) = sin(xy) at the point (1, 0) and the direction in which
it occurs.
Ans. 1, (0, 1)
19. Find the local maximum and minimum values and saddle point(s) of the function f (x, y) = x3 −
12xy + 8y3 .

Ans. Minimum f (2, 1) = −8, saddle point at (0, 0)

20. Use Lagrange multipliers to find

Maximum(Minimum) of x2 + y2 + z2
subject to
x4 + y4 + z4 − 1 = 0
√
Ans. Maximum 3, Minimum 1

Any one can get this soft-copy from Google site Exodus4Wisdom c Natnael Nigussie
[email protected]
[email protected]