14.

3: Partial Derivatives,
14.4: The Chain Rule, and
14.5: Directional Derivatives and Gradient Vectors
TA: Sam Fleischer
November 10

Section 14.3: Partial Derivatives

Definition: Partial Derivative with respect to x, y
The partial derivative of f (x, y) with respect to x at the point (x0 , y0 ) is
 
∂f  d  f (x0 + h, y0 ) − f (x0 , y0 )
fx = = f (x, y0 ) = lim
∂x (x0 ,y0 )
 dx 
(x0 ,y0 )
h→0 h

The partial derivative of f (x, y) with respect to y at the point (x0 , y0 ) is

 
∂f  d  f (x0 , y0 + h) − f (x0 , y0 )
fy = = f (x0 , x) = lim
∂y (x0 ,y0 )
 dy (x0 ,y0 )
h→0 h

Example: Taking Partial Derivatives

f (x, y) = y sin (xy)

To find fx (the partial derivative of f with respect to x), treat y as a constant.

fx = y cos (xy)y = y 2 cos (xy)

To find fy (the partial derivative of f with respect to y), treat x as a constant.

fy = (1) sin (xy) + y cos (xy)x = sin (xy) + xy cos (xy)

Example: Implicit Differentiation

∂z
Find if the equation
∂x
yz − ln z = x + y

defines z as a function of two independent variables x and y and the partial derivative exists.
d d
(yz − ln z) = (x + y)
dx dx
1
 ∂z 1 ∂z
y − =1+0
∂x  z ∂x
1 ∂z
y− =1
z ∂x
∂z x
=
∂x yz − 1

Definition: Second-Order Partial Derivatives

When we differentiate a function f (x, y) twice, we produce its second-order derivatives. Nota-
tion:

∂ 2f ∂ 2f
or f xx , or fyy
∂x2 ∂y 2
∂ 2f ∂ 2f
or fyx , and = fxy
∂x∂y ∂y∂x
The defining equations are

∂ 2f ∂ 2f
   
∂ ∂f ∂ ∂f
= , =
∂x2 ∂x ∂x ∂x∂y ∂x ∂y

∂ 2f
Note that fyx = means you first take the derivative with respect to y, and then take the
∂x∂y
derivate with respect to x.

In general, fxy 6= fyx .

Theorem 2: Mixed Derivative Theorem

If f (x, y) and its partial derivatives fx , fy , fxy , and fyx are defined throughout an open region
containing a point (a, b) and are all continuous at (a, b), then

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

Example: Partial Derivatives of Higher Order

Find fyxyz if f (x, y, z) = 1 − 2xy 2 z + x2 y

f (x, y, z) = 1 − 2xy 2 z + x2 y
fy = −4xyz + x2
fyx = −4yz + 2x
fyxy = −4z
fyxyz = −4

2
Definition: Differentiability
A function z = f (x, y) is differentiable at (x0 , y) if fx (x0 , y0 ) and fy (x0 , y0 ) exists and ∆z satisfies
an equation of the form

∆z = fx (x0 , y0 )∆x + fy (x0 , y0 )∆y + 1 ∆x + 2 ∆y

in which each of 1 , 2 → 0 as both ∆x, ∆y → 0. We call f differentiable if it is differentiable at

every point in its domain, and say that its graph is a smooth surface.

Theorem 3: The Increment Theorem for Functions of Two Variables

Suppose that the first partial derivatives of f (x, y) are defined throughout an open region R con-
taining the point (x0 , y0 ) and that fx and fy are continuous at (x0 , y0 ). Then the change

∆z = f (x0 + ∆x, y0 + ∆y) − f (x0 , y0 )

in the value of f that results from moving from (x0 , y0 ) to another point (x0 + ∆x, y0 + ∆y) in R
satisfies an equation of the form

∆z = fx (x0 , y0 )∆x + fy (x0 , y0 )∆y + 1 ∆x + 2 ∆y

in which each of 1 , 2 → 0 as both ∆x, ∆y → 0.

Corollary of Theorem 3
If the partial derivatives fx and fy of a function f (xy) are continuous throughout an open region
R, then f is differentiable at every point of R.

Theorem 4 - DIfferentiability Implies Continuity

If a function f (x, y) is differentiable at (x0 , y0 ), then f is continuous at (x0 , y0 ).

Section 14.4: The Chain Rule

Theorem 5: Chain Rule for Functions of One Independent Variable and Two Interme-
diate Variables
If w = f (x, y) is differentiable and if x = x(t), y = y(t) are differentiable functions of t, then the
composite w = f (x(t), y(t)) is a differentiable function of t and

dw
= fx (x(t), y(t)) · x0 (t) + fy (x(t), y(t)) · y 0 (t)
dt
or
dw ∂f dx ∂f dy
= +
dt ∂x dt ∂y dt

3
Theorem 6: Chain Rule for Functions of One Independent Variable and Three Inter-
mediate Variables
If w = f (x, y, z) is differentiable and x, y, and z are differentiable functions of t, then w is a
differentiable function of t and
dw ∂w dx ∂w dy ∂w dz
= + +
dt ∂x dt ∂y dt ∂z dt

Example
dw
Find if
dt
w = xy + z, x = cos t, y = sin t, z=t

Use Theorem 6:
dw ∂w dx ∂w dy ∂w dz
= + +
dt ∂x dt ∂y dt ∂z dt
= (y)(− sin t) + (x)(cos t) + (1)(1)
= (sin t)(− sin t) + (cos t)(cos t) + 1
= − sin2 t + cos2 t + 1
= cos 2t + 1

What is the derivative at t = 0?

 
dw 
= 1 + cos 0 = 2
dt t=0

Theorem 7: Chain Rule for Two Independent Variables and Three Intermediate Vari-
ables
Suppose that w = f (x, y, z), x = g(r, s), y = h(r, s), and z = k(r, s). If all four functions are
differentiable, then w has partial derivatives with respect to r and s given by the formulas
∂w ∂w ∂x ∂w ∂y ∂w ∂z
= + +
∂r ∂x ∂r ∂y ∂r ∂z ∂r

∂w ∂w ∂x ∂w ∂y ∂w ∂z
= + +
∂s ∂x ∂s ∂y ∂s ∂z ∂s

Theorem 8: A Formula for Implicit Differntiation

Suppose that F (x, y) is differentiable and that the equation F (x, y) = 0 defines y as a differentiable
function of x. Then at any point where Fy 6= 0,

dy Fx
=−
dx Fy

4
Example
dy
Find if y 2 − x2 − sin xy = 0.
dx
dy Fx
=−
dx Fy
−2x − y cos xy
=−
2y − x cos xy
2x + y cos xy
=
2y − x cos xy

Expansion of Theorem 8 to Three Variables

Suppose F (x, y, z) = 0 and z = f (x, y). Assuming F and f are differentiable functions, then

∂z Fx ∂z Fy
=− and =−
∂x Fz ∂y Fz

Expansion of Chain Rule to Functions of n variables

In general, suppose z = f (x1 , x2 , . . . , xn ) is a differential function of the intermediate variables
x1 , x2 , . . . , xn where n is a positive integer (n ∈ N). Also suppose each xi is a differentiable function
of the independent variables t1 , t2 , . . . , tm , with m ∈ N. In equation form,

x1 = g1 (t1 , t2 , . . . , tm )
x2 = g2 (t1 , t2 , . . . , tm )
..
.
xn = gn (t1 , t2 , . . . , tm )

Then w is a differential function of each of the independent variables t1 , t2 , . . . , tm , and the partial
derivatives of w with respect to each ti are
n
∂w ∂w ∂x1 ∂w ∂x2 ∂w ∂xn X ∂w ∂xi
= + + ··· + =
∂t1 ∂x1 ∂t1 ∂x2 ∂t1 ∂xn ∂t1 i=1
∂xi ∂t1

n
∂w ∂w ∂x1 ∂w ∂x2 ∂w ∂xn X ∂w ∂xi
= + + ··· + =
∂t2 ∂x1 ∂t2 ∂x2 ∂t2 ∂xn ∂t2 i=1
∂xi ∂t2

..
.
n
∂w ∂w ∂x1 ∂w ∂x2 ∂w ∂xn X ∂w ∂xi
= + + ··· + =
∂tm ∂x1 ∂tm ∂x2 ∂tm ∂xn ∂tm i=1
∂xi ∂tm

More compactly,
n
∂w X ∂w ∂xi
= for j = 1, 2, . . . , m
∂tj i=1
∂xi ∂tj

5
Directional Derivatives and Gradient Vectors
Definition: Directional Derivative
The derivative of f at P0 (x0 , y0 ) in the direction of the unit vector u = u1 i = u2 j is the
number
 
df f (x0 + su1 , y0 + su2 ) − f (x0 , y0 )
= lim
ds u,P0 s→0 s

provided the limit exists. The directional derivative is also denoted

 
df
= (Du f )P0
ds u,P0

and is read “The derivative of f at P0 in the direction of u”.

Definition: Gradient Vector

The gradient vector (gradient) of f (x, y) at a point P is the vector
∂f ∂f
∇f = i+ j
∂x ∂y

Theorem 9: The Directional Derivative is a Dot Product

If f (x, y) is differentiable in an open region containing P0 (x0 , y0 ), then

(Du f )P0 = (∇f )P0 · u

In words, the derivative of f at P0 in the direction of u is the dot product of the gradient ∇f at
P0 and u. In brief,

(Du )f = ∇f · u

Example
Find the derivative of f (x, y) = xey + cos xy at the point (2, 0) in the direction of v = 3i − 4j.
First, find the unit direction vector
v v 3 4
u= = = i− j
kvk 5 5 5

Then we need to find the partial derivatives of f at (2, 0) because together, they make up the
gradient, ∇f .

fx = ey − y sin xy
fx (2, 0) = e0 − 0 sin(2 · 0) = 1 − 0 = 1
fy = xey − x sin xy
fy (2, 0) = 2e0 − 2 sin(2 · 0) = 2 − 0 = 2

6
Plug these values into the definition of gradient.
∇f |(2,0) = fx (2, 0)i + fy (2, 0)j
= i + 2j
Then the directional derivative of f at (2, 0) in the direction of u is
(Du f )(2,0) = ∇f |(2,0) · u
 
3 4
= i − j · (i + 2j)
5 5
3 4
= −2·
5 5
= −1

Properties of the Directional Derivative

1. The function f increases most rapidly when cos θ = 1, i.e. when θ = 0, i.e. when u is the
direction of ∇f . That is, at each point P in its domain, f increases most rapidly in the
direction of the gradient vector ∇f at P . The derivative in this direction is
Du f = k∇f k cos 0 = k∇f k

2. Similarly, f decreases most rapidly in the direction of −∇f . The derivative in this direction
is
Du f = k∇f k cos π = −k∇f k

3. Any direction u orthogonal to a gradient ∇f =

6 0 is adirection of zero change in f because
π
θ = and
2
π
Du f = k∇f k cos = k∇f k · 0 = 0
2
Example
x2 y 2
Let f (x, y) = + , and consider the point (1, 1).
2 2
The function increases most rapidly in the direction of ∇f .
(∇f ) = xi + yj =⇒ (∇f )(1,1) = i + j
The unit vector of (∇f )(1,1) is
i+j 1 1
u= √ = √ i+ √ j
2 2 2
The function decreases most rapidly in the direction −(∇f )(1,1)
1 1
−√ i − √ j
2 2
The directions of zero change at (1, 1) are the directions ofthogonal to ∇f :
1 1 1 1
n = −√ i + √ j and −n= √ i− √ j
2 2 2 2

7
Important Concept
At every point (x0 , y0 ) in the domain of a differentiable function f (x, y), the gradient of
f is normal to the level curve through (x0 , y0 ).

Tangent Line to a Level Curve

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

Notice this is the same as point-slope form from elementary algebra.
y − y0 = m(x − x0 )
where
fx dy
m=− =
fy dx
by Theorem 8.

Algebra Rules for Gradients

1. Sum Rule: ∇(f + g) = ∇f + ∇g

2. Difference Rule: ∇(f − g) = ∇f − ∇g
3. Constant Multiple Rule: ∇(kf ) = k∇f
4. Product Rule: ∇(f g) = f ∇g + g∇f
 
f g∇f − f ∇g
5. Quotient Rule: ∇ =
g g2

Gradients of Functions of n variables

For a differential function f (x1 , x2 , . . . xn ) and a unit vector u = hu1 , u2 , . . . , un i in space, we have
 
∂f ∂f ∂f
∇f = , ,...,
∂x1 ∂x2 ∂xn
and
n
∂f ∂f ∂f X ∂f
Du f = ∇f · u = u1 + u2 + . . . un = ui
∂x1 ∂x2 ∂xn i=1
∂xi

The Derivative Along a Path

Let r(t) = x(t)i + y(t)j + z(t)k be a smooth path C and w = f (r(t)) a scalar function along C.
Then
dw ∂w dx ∂w dy ∂w dz
= + +
dt ∂x dt ∂y dt ∂z dt
or in vector notation,
d
f (u(t)) = ∇f (r(t)) · r0 (t)
dt