MH1100/MTH112: Calculus I.

Solutions for the Week 8 tutorial problems.

Solutions

Problem 1: (#2.3.21 from [Stewart].)

Dierentiate the function u(t) =

p
5

p
t + 4 t5 .

Solution
d
[u(t)] =
dt
=
=
=
=

p i
d hp
5
t + 4 t5
dt
1i
d h 1
t 5 + 4 t5 2
dt
i
5
d h 1
t 5 + 4t 2
dt
5
1 1 1
5
t5 + 4 t2
5
2
3
1 4
t 5 + 10t 2 .
5

Problem 2: (#2.3.22 from [Stewart].)

Dierentiate the function v(x) =
Solution

x+

1
p
3x

The quickest way to solve this is to use the chain rule. Strictly speaking, we
havent met this in our development of calculus yet, so instead Ill use the

product rule:
d
[v(x)]
dx

p
p
d
1
1
p
=
x+ p
x
+
3
3
dx
x
x

p
p
p
p
d
1
1
1
d
1
=
x+ p
x+ p
+
x+ p
x+ p
3
3
3
3
dx
x
x
x dx
x
1

h
i
1
1
1
d
= 2 x2 + x 3
x2 + x 3
dx

1
1 1 1 1
1
= 2 x2 + x 3
x 2
x 3 1
2
3
1
1 2 4
1
2
3
=
x +x
x 2
x 3 .
3

Problem 3: (#2.3.24 from [Stewart].)

4

Dierentiate the function F (x) = x 5xx2 + x in two dierent ways: by using

the quotient rule, and by simplifying the expression first. Check you get the
same answer.
Solution
Using the quotient rule, the computation is:

p
d
d x4 5x3 + x
[F (x)] =
dx
dx
x2
h
i

d
d
4
3 + x 12 x2
4
3 + x 12
2
x
5x
x
5x
dx
dx x
=
(x2 )2

1
1
4x3 15x2 + 12 x 2 x2
x4 5x3 + x 2 (2x)
=
x4
3
1 32
5
4
5
4x
15x + 2 x
2x + 10x4 2x 2
=
x4

3 3
4
5
4
= x 2x
5x
x2
2
3 5
= 2x 5
x 2.
2
2

Alternatively, if we simplify first, then dierentiate, then the computation looks like:

p
d
d x4 5x3 + x
[F (x)] =
dx
dx
x2
h
i
3
d
2
2
=
x
5x + x
dx
3 3 1
= 2x 5
x 2
2
3 5
= 2x 5
x 2.
2
We get the same thing, which means we didnt make any mistakes. The
lesson is that it always pays to stop for a moment and think before you
launch into a detailed calculation!

Problem 4: (#2.3.43 from [Stewart].)

Dierentiate the function
f (x) =

x
.
x + xc

Solution
There is a little trick to notice here, which is that beginning by dividing top
and bottom by x will shorten the calculation a little.

d
x
d
1
=
dx x + xc
dx 1 + xc2

d
c
d
1 dx
1 + cx 2
dx [1] 1 + x2
=
2
1 + xc2
=

( 2c)x

1 + xc2
x

= 2c

(c + x2 )2

Problem 5: (#2.3.58 from [Stewart].)

Find the equation of the tangent line and the normal line to the curve
p
x
y=
x+1
at the point (4, 0.4).
Solution
If we set f (x) =

p
x
x+1

then:

the tangent line at (4, 0.4) is the straight line going through this point
with gradient f 0 (4),
and the normal line at (4, 0.4) is the straight line going through this
point with gradient f 01(4) .
The derivative f 0 (x) is computed:
h 1i
1
d
d
2 (x + 1)
x 2 dx
[x + 1]
dx x
0
f (x) =
(x + 1)2
=
=

1
2

1
2x

(x + 1) x 2 1
(x + 1)2

1 12
2x

+ 12 x
(x + 1)2

1
2

The value at x = 4 is:

f 0 (4) =

1 + 14
=
(5)2

3
.
100

The tangent line at (4, 0.4) is given by

y = 0.4 + f 0 (4)(x

4) = 0.4

(0.03)(x

And the normal line at (4, 0.4) is given by:

1
4
y = 0.4 +
(x
4)
=
f 0 (4)
10

4) = 0.52

400
3

0.03x.

100
x.
3

Problem 6: (#2.3.74 from [Stewart].)

If f (x) is a dierentiable function, find an expression for the derivative of
each of the following functions:
(a) y = x2 f (x)
(c) y =

(b) y =

x2
f (x)

(d) y =

f (x)
x2
1+xf (x)
p
x

Solution
Part (a):

d 2
d 2
d
x f (x) =
x f (x) + x2
[f (x)]
dx
dx
dx
= 2xf (x) + x2 f 0 (x).
Part (b):

d f (x)
dx x2

=
=

d
dx

[f (x)] x2

2
d
f (x) dx
x

(x2 )2
2f (x)
.

f 0 (x)x
x3

Part (c):
2
d
x
dx f (x)

=
=

Part (d):

d 1 + xf (x)
p
dx
x

=
=
=

d
dx

d
x2 f (x) x2 dx
[f (x)]
2
(f (x))
2xf (x) x2 f 0 (x)
.
(f (x))2
d
dx

d
x (1 + xf (x)) dx
[x 2 ]
1 2
x2

1
1
(f (x) + xf 0 (x)) x 2 (1 + xf (x)) 12 x 2

[1 + xf (x)]

p
1
2 f (x) x

1
2x

1
2

x
3
+ f 0 (x)x 2

x
f (x)
p
2 x

1
x
2

3
2

p
+ f 0 (x) x.

Problem 7: (#2.3.77 from [Stewart].)

Show that the curve y = 6x3 + 5x

3 has no tangent line whose slope is 4.

Solution
The slope of a tangent line is given by the derivative of f (x) = 6x3 + 5x
at the point of tangency. That derivative is given by

f 0 (x) = 18x2 + 5.
This quantity is obviously always
5, so we conclude that there is no
tangent line of slope 4.
A plot of this curve confirms that it is indeed a very steep curve at every
point:

Problem 8: (#2.3.80 from [Stewart].)

Find equations of the tangent lines to the curve y =
to the line x 2y = 2.

x 1
x+1

that are parallel

Solution
Note that a straight line y = mx + c will be parallel to the line x 2y = 2
if and only if it has the same gradient, which is clearly (rearranging the
expression to put it in standard form) 12 . So we have to search for those x
solving the equation f 0 (x) = 12 , where we have set
f (x) =
6

x 1
.
x+1

The derivative we need is:

f 0 (x) =
=
=

d
dx

[x

1] (x + 1) (x
(x + 1)2
(x + 1) (x 1)
(x + 1)2
2
.
(x + 1)2

d
1) dx
[x + 1]

This equals 12 precisely when (x + 1)2 = 4, which has two solutions

x = 12. So this graph will have exactly two tangents parallel to the given
line, namely the tangents at (1, f (1)) = (1, 0) and at ( 3, f ( 3)) = ( 3, 2).
The tangent at (1, 0) is given by the rule:
1
y = 0 + (x
2

1).

And the tangent at ( 3, 2) is given by the rule

1
y = 2 + (x
2

( 3)) =

7 x
+ .
2 2

Here is a plot of the arrangement:

Problem 9: (#2.3.87 from [Stewart].)

Find a degree 2 polynomial P (x) such that P (2) = 5, P 0 (2) = 3, P 00 (2) = 2.
Solution
Every degree 2 polynomial can be written in the form
P (x) = a0 + a1 x + a2 x2 ,
where a0 , a1 and a2 are variables which we will use the three given conditions
to determine. The derivatives of this expression are:
P 0 (x) = a1 + 2a2 x
and
P 00 (x) = 2a2 .
The three conditions give us three equations:
P (2) = 5 ) a0 + 2a1 + 4a2 = 5.
P 0 (2) = 3 ) a1 + 4a2 = 3.
P 00 (2) = 2 ) 2a2 = 2.
We can solve these three equations in three unknowns by back-substitution.
We get:
a2 = 1 , a1 = 1 , a0 = 3.
So the desired polynomial is: P (x) = 3

x + x2 .

Problem 10? : (#2.3.98 from [Stewart].)

Consider a function F (x) = f (x)g(x), where f and g are assumed to have
derivatives of all orders f (n) (x) and g (n) (x).
(i) Show the following formula
00

F (x) = f 00 (x)g(x) + 2f 0 (x)g 0 (x) + f (x)g 00 (x).

(ii) Find analogous formulas for F 000 (x) and F (4) (x).
(iii) Guess a formula for the case of general n, F (n) (x), then use induction
to prove your guess is correct.
Solution
The first derivative of F (x) is exactly given by the product rule:
F 0 (x) = f 0 (x)g(x) + f (x)g 0 (x).
The second derivative is then calculated from this:
F 00 (x) =
=
=

d 0
F (x)
dx

d 0
f (x)g(x) + f (x)g 0 (x)
dx
f 00 (x)g(x) + f 0 (x)g 0 (x) + f 0 (x)g 0 (x) + f (x)g 00 (x)

= f 00 (x)g(x) + 2f 0 (x)g 0 (x) + f (x)g 00 (x).

We then calculate the third derivative in the same way:
F 000 (x) =
=
=

d 00
F (x)
dx

d 00
f (x)g(x) + 2f 0 (x)g 0 (x) + f (x)g 00 (x)
dx
f 000 (x)g(x) + f 00 (x)g 0 (x)
+2 f 00 (x)g 0 (x) + f 0 (x)g 00 (x)
+ f 0 (x)g 00 (x) + f (x)g 000 (x)

= f 000 (x)g(x) + 3f 00 (x)g 0 (x) + 3f 0 (x)g 00 (x) + f (x)g 000 (x).

And then we move onto the fourth derivative. (From here on well use
the notation f (n) (x) to refer to the n-th derivative of f (x). Note that f (0) (x)
refers to f (x) itself.)
d h (3) i
F (x)
dx
h
i
d
=
f (3) (x)g(x) + 3f (2) (x)g (1) (x) + 3f (1) (x)g (2) (x) + f (x)g (3) (x)
dx

F (4) (x) =

f (4) (x)g(x) + f (3) (x)g (1) (x)

+3 f (3) (x)g (1) (x) + f (2) (x)g (2) (x)

+3 f (2) (x)g (2) (x) + f (1) (x)g (3) (x)

+ f (1) (x)g (3) (x) + f (x)g (4) (x)

= f (4) (x)g (0) (x) + 4f (3) (x)g (1) (x)

+6f (2) (x)g (2) (x) + 4f (1) (x)g (3) (x) + f (0) (x)g (4) (x).
This should be enough data to guess a general formula for F (n) (x). The
thing to notice is that the coefficients appearing in this sequence of expressions are exactly the same coefficients that appear in the binomial theorem.
This leads us to make the guess:
F

(n)

n
X
n
(x) =
f (i) (x)g (n
i

i)

(x).

(?)

i=0

Proof of Equation (?) by induction.

Every induction proof has two steps: checking that the base (in other
words, the first case) of the induction is true, and then checking that if the
n-th case of the induction is true, then the n + 1-st case is true as well.
The case n = 1 of the statement (?) is exactly the product law, so the
base case is true.
Now well assume that the statement is true for some n, and show that
the n + 1 case follows from it.

10

The calculation is:

d h (n) i
F (n+1) (x) =
F (x)
dx
" n

d X n
=
f (i) (x)g (n
i
dx

i)

(x)

i=0

(Assuming the case n of (?) is true.)

n
i
X
d h (i)
n
=
f (x)g (n i) (x)
i
dx
i=0

n
X
n (i+1)
=
f
(x)g (n
i

i)

(x) + f (i) (x)g (n

i+1)

(x)

i=0

n
X
n
=
f (i+1) (x)g (n
i
i=0

n+1
X
j=1

n
j

(j)

(x)g

i)

n
X
n
(x) +
f (i) (x)g (n
i

i+1)

(x)

i=0

(n j+1)

n
X
n
(x) +
f (i) (x)g (n
i

i+1)

(x)

i=0

(Changing the summation variable to j = i + 1 in the first sum

to allow us to combine the two sums.)

n+1
X
i=0

n+1
X
i=0

n+1
X
i=0

n
1
n
i

n+1
i

(i)

(x)g

(n i+1)

(x) +

n+1
X
i=0

n
i

f (i) (x)g ((n+1)

f (i) (x)g ((n+1)

i)

n
i
i)

f (i) (x)g (n

i+1)

(x)

(x).

(Note that to get the final equation we have used a standard identity of the
binomial coefficients. This is an algebraic version of the familiar rule for
building Pascals triangle row-by-row. If you have never seen this before,
see the Wikipedia entry for Pascals triangle.)

11

(x)

Problem 11: (#2.3.102 from [Stewart].)

A tangent line is drawn to the hyperbola xy = c at a point P .
(a) Show that the midpoint of the line segment cut from this tangent line
by the co-ordinate axes is P .
(b) Show that the triangle formed by the tangent line and the co-ordinate
axes always has the same area, no matter what point P is chosen.
Solution
To begin, note that the hyperbola is the graph of the function f (x) = xc .
The first thing well do to understand the described arrangement is to
find an equation for the tangent line to the hyperbola at some point a, ac .
The derivative at that point is:
f 0 (a) =

c
.
a2

The tangent line through the point a, ac is the straight line through that
point with slope ac2 . That straight line is given by the equation:
y=

c c
+
(x
a
a2

a).

The next thing well do to understand this is find the points where this
straight line intercepts the x and y-axes. To find the x-intercept we set
y = 0. This gives us x = 2a. Then to find the y-intercept we set x = 0.
This gives us: y = 2 ac .

12

Here is a picture of what we have calculated:

To get the midpoint of the line segment that goes from the point 0, 2 ac
to the point
(2a, 0) you take the average of the co-ordinates. The point you

c
0+2a 2 a +0
get is:
= a, ac . This is always the point of tangency, which
2 , 2
is what we were asked to show in Part (a).
Part (b) of the problem concerned the area of the triangle appearing in
the diagram above. Its area is 12 (2a) 2 ac = 2c. This confirms the second
fact we were asked to show, that this triangle always has the same area, no
matter what the point of tangency a, a2 is chosen to be.

13

Problem 12? : (#2.3.103 from [Stewart].)

Using standard facts, find a simple evaluation of the limit limx!1

x1000 1
x 1 .

Solution
The trick here is to notice that this is exactly the limit you would write
down to calculate the derivative of f (x) = x1000 at the point x = 1. Then
you can just use the power rule to calculate it:
x1000 1
x!1 x
1
lim

=
=

d 1000
x
dx
1000x

x=1

999

x=1

= 1000.

14

Problem 13? : (#2.3.106 from [Stewart].)

Using your intuition, guess if there is a straight line that is tangent to both
of the parabolas y = x2 and y = x2 2x + 2. Then decide if this line exists,
and then find it, using equations.
Solution
To understand the geometry here, complete the square in y = x2 2x + 2
to get y = (x 1)2 + 1. So we have two curves: a standard parabola y = x2
(call this curve A) and a parabola obtained from it by shifting it one to the
right and one up (call this curve B). The picture is:

Here is an intuitive way of understanding that there is going to exist a

straight line which is tangent to both curves. Consider the tangent to curve
A at the point x = 0. It is a straight line lying on the x-axis. Notice that
it doesnt intersect curve B. But as we move the point of tangency to to
the right along curve A, the tangent line gradually gets steeper and steeper,
and at some point it will touch curve B for the first time. We expect that
straight line will be simultaneously a tangent line to both A and B.
Now lets check our intuition with equations. First well find an equation
for the tangent line to curve A at an arbitrary point on A: (a, a2 ). Note that
f 0 (a) = 2a, so the tangent line at that point is given by
y = a2 + (2a)(x

a) =
15

a2 + 2ax.

This line will intersect curve B at points where (x, a2 + 2ax) satisfies
the equation that determines curve B:
( a2 + 2ax) = x2

2x + 2.

Rearranging:
x2

2(1 + a)x + (2 + a2 ) = 0.

Completing the square:

(x
This has solutions

(1 + a))2 + (1
p
x = 2a

2a) = 0.

1 + (1 + a).

We guess that the line will be tangent to curve B when this equation has
exactly 1 solution, which is when a = 12 . The corresponding straight line is:
y = 14 + x. What we know so far is this line is tangent to curve A at 12 , 14
and intersects curve B exactly once at the point x = 32 .
The only thing we have to check is that this straight line is indeed a
tangent to curve B where it intersects curve B, at x = 32 . Well, the tangent
to curve B as x = 32 has slope
d 2
x
dx

2x + 2

= [2x
x= 32

2]|x= 3 = 1
2

so this straight line is the one we want.

Here is a MATLAB plot of the arrangement:

16

Problem 14: (#2.4.34 from [Stewart].)

Find the points on the curve y =
zontal.

cos x
2+sin x

at which the tangent line is hori-

Solution
cos x
This curve is the graph of the function f (x) = 2+sin
x . The points where the
tangent line is horizontal are precisely the solutions of the equation

f 0 (x) = 0.
The derivative is:
0

f (x) =
=
=
=
=

d
cos x
dx 2 + sin x
d
dx

d
[cos x] (2 + sin x) cos x dx
[2 + sin x]
2
(2 + sin x)
( sin x)(2 + sin x) cos x cos x
(2 + sin x)2
2 sin x sin2 x cos2 x
(2 + sin x)2
2 sin x + 1
(Because sin2 x + cos2 x = 1.)
(2 + sin x)2

This is zero if and only if sin x = 12 . Using elementary trigonometry

we deduce that the set of solutions is:

n
o 5
1
x 2 R sin x =
=
+ 2n n 2 N [
+ 2n n 2 N .
2
6
6
The corresponding points on the curve are:

1
5
+ 2n, p
n2N [
+ 2n,
6
6
3

1
p
3

n2N

17

Problem 15: (#2.4.41 from [Stewart].)

Determine the limit

tan 6t
.
t!0 sin 2t
lim

Solution
A straightforward application of the quotient rule wont work, because you
would get 00 . So the thing you need to do is rearrange the expression into a
form where you can apply the limit laws. One way of doing it is:
tan 6t
t!0 sin 2t
lim

= lim

sin 6t
cos 6t

sin 2t

1
sin 6t
lim

t!0 cos 6t
sin 2t

1
sin 6t
2t
6t
lim

t!0 cos 6t
6t
sin 2t 2t
1
sin 6t
1
3
lim

limt!0 cos 6t t!0 6t

limt!0
3111
t!0

=
=
=
=

sin 2t
2t

= 3.

Problem 16: (#2.4.45 from [Stewart].)

Determine the limit

sin
.
!0 + tan
lim

Solution
A calculation is:
sin
!0 + tan
lim

sin

!0 1 + 1 sin
cos

lim!0 sin
1
sin
1 + lim!0
cos lim!0

= lim
=
=
=

1+
1
.
2

1
1

18

Problem 17: (#2.4.48 from [Stewart].)

Determine the limit

sin (x 1)
.
x!1 x2 + x
2
lim

Solution
Well start by replacing the limit as x ! 0 with a limit as h ! 0, via the
standard identity
lim F (x) = lim F (a + h).
x!a

h!0

Then the computation of the limit becomes:

sin (x 1)
x!1 x2 + x
2
lim

=
=
=
=
=
=

sin (x 1)
x!1 (x + 2)(x
1)
sin (1 + h 1)
lim
h!0 (1 + h + 2)(1 + h
1)

sin h
1
lim

h!0
h
h+3
sin h
1
lim
lim
h!0 h
h!0 h + 3
1
1
3
1
.
3
lim

Problem 18? : (#2.4.56 from [Stewart].)

Consider the function

x
.
1 cos 2x
Is f continuous at x = 0? If not, what type of discontinuity does it have
there? Justify your answer using standard properties of limits. It may be
interesting to check your answer with a graph of this function.
f (x) = p

Solution
The first observation is that 0 isnt in the domain of f (x) (because both
top and bottom line go to zero at x = 0), so f (x) definitely wont be continuous there. The question is whether it is just a removable discontinuity
or a more substantial form of discontinuity.
19

The way we understand the behavior at this point is to consider the

1-sided limits. First lets use a double angle formula to make the expression
more transparent:
f (x) = p

x
=p
1 cos 2x
1

(1

So we need to study the 1-sided limits of:

=p

x
p

.
2 sin x)
2 sin2 x
p
The obvious mistake here is to replace: sin2 x by sin x, but that is only
true if sin x > 0. The correct replacement is:
p
sin2 x = |sin x| .
2

p x
2| sin x|

at x = 0. To determine

the limit from the right, note that when x > 0, and close to 0, then sin x > 0,
so:
1
1
lim f (x) = lim p sin x = p .
+
+
x!0
x!0
2 x
2
To determine the limit from the left, note that when x < 0, and close to 0,
then sin x < 0, so:
lim f (x) = lim p

x!0

x!0

1
2 sinx x

1
p .
2

The two 1-sided limits exist, but they disagree, so this function has a
jump discontinuity at x = 0. Here is a plot of the function:

20

Problem 19? :
Determine the limit
lim

1 + tan x

1 + sin x

x3

x!0

Solution
One approach is:
p
p
1 + tan x
1 + sin x
lim
x!0
x3
p
p
p
p

1 + tan x
1 + sin x
1 + tan x + 1 + sin x
p
= lim
p
x!0
x3
1 + tan x + 1 + sin x

tan x sin x
1
p
p
= lim

x!0
x3
1 + tan x + 1 + sin x

sin x 1 cos x
1
1
p
= lim

p
.
x!0
x
x2
cos x
1 + tan x + 1 + sin x
x
All of these factors have standard limits at 0, except possibly 1 xcos
. We
2
can understand that factor by using a double-angle formula to rewrite it

2 sin2
cos x
=
x2
x2

x
2

1 sin x2 sin x2

.
2 x/2
x/2

So this factor will have limit 12 at x = 0.

Putting it all together we obtain a limit:
1

1 1
1
1
p
p
= .
2 1
4
1+0+ 1+0

21

Problem 20? : (#10.27 from [Spivak].)

Let f be a function which is dierentiable at 0, and assume that f (0) = 0.
Prove that there exists a function g(x) which is continuous at 0 and such
that
f (x) = xg(x).
Solution
The obvious thing to do is just divide: g(x) = f (x)
x . But that expression
does not give you a value at x = 0, so it cant be the complete answer.
Anyway it has to be the formula at points away from x = 0, so well extend
it so as to make g(x) continuous at x = 0:
(
f (x)
if x 6= 0,
x
g(x) =
f (x)
limx!0 x
if x = 0.
Observe that:
The limit limx!0 f (x)
x does actually exist because f (x) is assumed to
be dierentiable at 0.
This function is continuous at zero because
f (x)
= g(0).
x!0 x

lim g(x) = lim

x!0

The equation f (x) = xg(x) is true at every point, including x = 0.

22

Problem 21: (#2.5.5 from [Stewart])

Identify the following function as a composition h(x) = f (g(x)) then use the
chain rule to dierentiate it:
p
h(x) = sin x.
Solution
To present h(x) as a composition set:
p
f (x) = x,
g(x) = sin x.
The derivative is computed using the chain rule via:
i
d hp
sin x
dx
d p
d
=
x
[sin x]
dx
sin x dx

1 1
=
x 2
cos x
2
sin x
1
1
=
p
cos x.
2
sin x

23

Problem 22: (#2.5.8 from [Stewart])

Identify the following function as a composition h(x) = f (g(x)) then use the
chain rule to dierentiate it:
h(x) = 4x

x2

100

Solution
To identify this as a composition h(x) = f (g(x)) we set:
f (x) = x100 ,
g(x) = 4x

x2 .

Using the chain rule, the derivative is:

i
d h
100
4x x2
dx
d 100
=
x
dx
4x
= 100(4x

x2
2 99

d
4x
dx

x ) (4

2x)

x2

Problem 23: (#2.5.23 from [Stewart])

Dierentiate the following function
f (x) = sin (x cos x) .
The chain rule computation is:
f 0 (x) =
=

d
d
[sin x]
[x cos x]
dx
x cos x dx
(cos x)|x cos x (cos x x sin x)

= cos (x cos x) (cos x

x sin x) .

24

Problem 24: (#2.5.25 from [Stewart])

Dierentiate the following function
F (z) =

z 1
.
z+1

Solution
The derivative is:
0

F (z) =
=

d h 1/2 i
d z 1
z
dz
z+1
z 1 dz

z+1
1 1/2
1 (z + 1) (z
z
2
(z + 1)2
z 1

1) 1

z+1

1
1
q
2
z

1
z+1

1
(z + 1)2

2
(z + 1)2

z+1
.
z 1

This could also be written: F 0 (z) =

(z+1) 2 (z 1) 2

Problem 25: (#2.5.25

29 from [Stewart])
Dierentiate the following function
y = sin
Solution

p
1 + x2 .

p
This is the composition of three functions: namely sin x, x, and 1 + x2 .
We dierentiate working from the outside in:
i
dy
d
d hp
=
[sin x] p

1 + x2
dx
dx
dx
1+x2

d h 1/2 i
d
2
2
= cos 1 + x
x
1+x
dx
1+x2 dx
p
1
= cos 1 + x2 p
2x
2 1 + x2
p
x cos 1 + x2
p
=
.
1 + x2
25

Problem 26: (#2.5.42 and #2.5.43 from [Stewart])

Dierentiate the functions:
p
p
(i) y = 1 + x.
q
p
p
(ii) y = 1 + 1 + x.
Solution to (i).
We calculate:
dy
dx

=
=
=

Solution to (ii).

p
d p
d
x

1
+
x
p
dx
dx
1+ x
1
1 1
p
p x 2
2 1+ x 2
1
p p
p .
4 x 1+ x

We can use our computation to part (i) to simplify this case a little bit:

q
p
dy
d p
d
=
x

1
+
1
+
x
p
p
dx
dx
dx
1+ 1+ x
=
=

1
1
q
p p
p
p
p
2 1+ 1+ x 4 x 1+ x

1
1
q

.
p
p
3
p
p
p
2
x 1+ x 1+ 1+ x

Question: What is the general rule for this pattern?

26

Problem 27: (#2.5.53 from [Stewart])

Find the tangent line to the curve
y = sin sin x
that goes through the point (, 0).
Solution
The derivative is:
dy
d
= (cos x)|sin x
[sin x] = cos x cos (sin x).
dx
dx
At the point this has the value 1.
The required tangent line is the straight line through the point (, 0)
with gradient 1. It is given by:
y = 0 + ( 1) (x

) =

x.

Problem 28: (#2.5.59 from [Stewart])

Find the points on the graph of the following function
f (x) = 2 sin x + sin2 x
at which the tangent line is horizontal.
Solution
We need the points x where the derivative is zero. In other words we
want the solutions to:
f 0 (x) = 2 cos x + 2 sin x cos x = 0.
This can be factorized:
2 cos x (1 + sin x) = 0.
The set of solutions is the set of the points where cos x = 0 union with the
set of points where sin x = 1. The required set is:
n
o
+ n n 2 Z .
2
27

It may help to write this set out explicitly:

...,

5
,
2

3
,
2

3 5
, , , ,...
2 2 2 2

The precise thing we were asked for is the set of points on the graph
where the tangent line is horizontal. There are two types of points in this
set. If x = 2 + 2n for some n 2 Z then sin x = 1 and f (x) = 3. The
corresponding points are
n

o
+ 2n, 3 n 2 Z .
2
On the other hand if x = 3
2 + 2n for some n 2 Z then sin x =
f (x) = 1. The corresponding set of points is:

3
+ 2n, 1 n 2 Z .
2

1, and

Problem 29: (#2.5.68 from [Stewart])

Suppose that f is dierentiable at every point of R and that 2 R. Let
F (x) = f (x ) and let G(x) = [f (x)] . Find expressions for:
(a) F 0 (x), and
(b) G0 (x).
Solution
The first derivative asked for is:
F 0 (x) = f 0 (x )

d
[x ] = x
dx

1 0

f (x ).

The other derivative asked for is:

G0 (x) = [f (x)]

f 0 (x).

28

Problem 30: (#2.5.70 from [Stewart])

If g is a twice-dierentiable function, and f (x) = xg(x2 ), then find an expression for f 00 (x) in terms of the derivatives of g(x).
Solution
An expression for the first derivative is:
f 0 (x) =

d
xg(x2 )
dx

d 2
g(x )
dx
= g(x2 ) + x g 0 (x2 ) (2x)
= 1 g(x2 ) + x

= g(x2 ) + 2x2 g 0 (x2 ).

And then the second derivative is:
f 00 (x) =
=

d 2
d 2 0 2
g(x ) +
2x g (x )
dx
dx
g 0 (x2 ) (2x) + 2 2x g 0 (x2 ) + 2x2 g 00 (x2 ) 2x

= 2xg 0 (x2 ) + 4xg 0 (x2 ) + 4x3 g 00 (x2 ).

Problem 31: (#2.5.71 from [Stewart])

If F (x) = f (3f (4f (x))), where f (0) = 0 and f 0 (0) = 2, determine F 0 (0).
Solution
A standard chain rule computation leads to:
d
[3f (4f (x))]
dx
d
= f 0 (3f (4f (x))) 3f 0 (4f (x))
[4f (x)]
dx
= 12f 0 (3f (4f (x))) f 0 (4f (x)) f 0 (x).

F 0 (x) = f 0 (3f (4f (x)))

Substituting the given values we get:

F 0 (0) = 12 f 0 (3 f (4 f (0))) f 0 (4 f (0)) f 0 (0) = 12 2 2 2 = 96.

29

Problem 32? : (#2.5.83 from [Stewart])

Use the chain rule to give proofs that:
(a) The derivative of an even function is an odd function.
(b) The derivative of an odd function is an even function.
Solution
Well make use of the function R(x) = x. Note that a function f (x) is
even if and only if:
(f R) (x) = f (x).
And note that a function f (x) is odd if and only if
(f

R) (x) =

f (x).

Now assume that some f (x) is dierentiable everywhere and even. This
means that (f R) (x) = f (x). If we dierentiate the LHS of this equation
using the chain rule we get:
d
[f (R(x))] = f 0 (R(x)) R0 (x) = f 0 (R(x)) ( 1).
dx
The derivative of the RHS is just f 0 (x). Equating these two derivatives we
deduce that the derivative of an even function satisfies:
f 0 (R(x)) =

f 0 (x).

This says that f 0 (x) is odd, which completes Part (a).

On the other hand, if we start with an odd function f (x), then we have
the equation f (R(x)) = f (x). If we dierentiate the two sides of this
equation we deduce
f 0 (R(x)) = f 0 (x).
This says that f 0 (x) is even, which completes Part (b).

30

Problem 33: (#2.5.86 from [Stewart])

Suppose that f (x) is a function which is dierentiable everywhere and
that y = f (x) is a curve that always lies above the x-axis, and never has
a horizontal tangent. For what value of y is the rate of change of y 5 with
respect to x eighty times the rate of change of y with respect to x?
Solution
The information we are given is that:
f (x) has a derivative at every x.
f (x) > 0 for all x.
f 0 (x) 6= 0 at any x.
We are asked to find a value f (x) such that

d
f (x)5 = 80 f 0 (x).
dx

If we do the dierentiation on the LHS this equation becomes

5f (x)4 f 0 (x) = 80f 0 (x).
This is true if and only if
f 4 (x) = 16,
which means f (x) = 2. (Note that f 0 (x) 6= 0 by assumption.)

31

Problem 34? : (#2.5.87 from [Stewart])

Use the chain rule to show that if is measured in degrees, then:
d

(sin ) =
cos .
d
180
Solution
Trigonometric functions are always written in radians. If the variable of
some trigonometric function is valued in degrees then we really should write
the function with a dierent name, such as:
sindeg. (x).
(For example: sindeg. (45) = p12 .)
Note that the formula that transforms from degrees to some quantity
2
R() of radians is R() = 360
. This means that the two versions of sin x
are related by the formula:

sindeg. () = sin (R()) = sin
.
180

Thus the requested derivative can be determined using the chain rule in
the following way:
d
[sindeg. (x)] =
dx
=
=

d h i
sin
x
dx
180

cos
x
180
180

cosdeg. (x).
180

This is the formula we were asked to obtain.

32

Problem 35:
If it is known that

d
[f (2x)] = x2
dx

then what is f 0 (x)?

Solution
Computing the LHS this equation becomes:
2f 0 (2x) = x2 .

(?)

We can use this to determine f 0 (x) by:

f 0 (x) =
=
=

x
1
2f 0 2
2
2
1 x 2

(Using (?).)
2
2
x2
.
8

33