2.

Limits and Continuity

2.1.1 Definition The limit of the function f (x) as x approaches a is the number L, written as
lim f (x) = L, (2.2)
x!a

provided that f (x) is arbitrarily close to L for all x sufficiently close to, but not equal to, a.

For a 2 — Limit of Functions

given function f (x), if there is a number L satisfying the above informal definition, we say that
1. the limit of f (x) exists and equals L; or
2. f (x) converges to L; or
3. f (x) has the value L as its limit,
when x is sufficiently close to, but not equal to, a. Otherwise, if there exists no such L, we say that
1. the limit of f (x) does not exist; or
2.1
2. Limit:
f (x) Intuitive
diverges; or Meaning and Computations
3. f (x) has no limit.
In this chapter, we introduce the concept of limit of functions, which is fundamental to the notion
differentiation that we will introduce later. We will first introduce limits using a more intuitive
Remark. We emphasize that, when finding a limit, we are concerned not with what happens to f (x) when
approach. After learning its geometric meaning and some examples of computing limits, we will
x equals a, but only with what happens to f (x) when x is close to a. Be very careful of this di↵erence.
then
Moreover, briefly
a limit introduce
must the rigorous
be independent of thedefinition of limit
way in which that will be
x approaches a. taught
In otherinwords,
MATHthe2033.
limit must
be the same whether x approaches a from the left or from the right (i.e., for x < a or for x > a, respectively).
2.1.1 Idea of limits
Let f : R ! R be a function whose graph can be found in Figure 2.1 (left).
⌅ Example 2.1.3 (Limits)
y y = f (x) y y = g(x) y y = h(x)

L L L
h(a)

x x x
0 a 0 a 0 a

The functions f , g and h all approach the limit L when x approaches a. Note that g(x) is not defined at
Figure 2.1: graphs of f , g and h
x = a and both g(x) and h(x) have a point of discontinuity at x = a (indicated by the hollow dot in the
figures). However, both functions g(x) and h(x) still get closer and closer to the value L as x gets closer and
According
closer to a from the lefttoand
thefrom
graph,
the when approaches a, the value f (x) approaches L. Using limit
right. xThus,
notations, we can write:
lim f (x) = L, lim g(x) = L, lim h(x) = L.
x!a x!a lim f (x) = L.x!a
x!a
The crucial point here is that each limit depends on the values of the function sufficiently close but not equal
A natural
to a. The function question
value at x =isathat
itselfhow
hasisno
it influence
different from
at all simply
on the writing
limit. f (a) = L? The key difference
2
is that lim f (x) concerns about the value of f when x is near a, whereas f (a) is the value of
x!a
f (x) when
In summary x isare
there exactly equal to a.points
three important The limit
aboutlim
limits of is
f (x) telling us what f (x) is close to, when x
functions.
x!a
is close to (but not yet equal) to a.
1. Saying that the limit of f (x) approaches L as x approaches a means that the value of f (x) may be
In Figure 2.1 (left), lim f (x) and f (a) are both the same. When move x toward a, the value
made arbitrary close to the
x!anumber L by choosing x closer and closer to a.
of f (x) moves toward L. Also when x is exactly at a, the value of f (x) is also exactly L.
2. For a limit to exist, you must allow x to approach a from either side of a. If f (x) approaches a
di↵erent number as x approaches a from the left than it does as x approaches a from the right,
then the limit does not exist.
3. The value of f (x) when x = a has no influence on the existence or non-existence of the limit of f (x)
as x approaches a. In fact, a may not even be in the domain of f . However, the function must be
defined on both sides of a.

32
40 Limit of Functions

However, it is different in the middle and right figures. Consider the function in the middle
graph: g(x) which is not defined at x = a. When x is approaching a but not yet equal to a, the
value of g(x) is still approaching L even though g(a) is not defined. Therefore, we can also write:

lim g(x) = L.
x!a

When we determine the value of a limit lim f (x), the function f (x) is not required to be
x!a
defined at x = a, and even if the function is defined at x = a, the limit and f (a) can be different
too! See the two examples below.
Even if the function is defined at x = a, its limit when x ! a can also be different from its
value at x = a. Consider this piecewise defined function (see Figure 2.1 (right))
(
f (x) if x 6= a
h(x) =
L 1 if x = a

Again, h and f coincide when x 6= a, but f (a) = L while h(a) = L 1. The value of lim h(x)
x!a
is determined by h(x) near x = a, but not at x = a. Therefore, we still have:

lim h(x) = L.
x!a

⌅ Example 2.1 Consider the function

x2 4
f (x) =
x 2
Is f (2) defined? How about lim f (x)?
x!2

⌅ Solution When x = 2, the denominator x 2 equals 0. Division by zero is NOT allowed, so

you cannot put x = 2 into the function f (x). In other words, f (2) is not defined.
However, lim f (x) is determined by the values of f (x) when x is close to, but not equal
x!2
to 2. When x 6= 2, we have

x2 4 (x 2)(x + 2)
f (x) = = = x + 2.
x 2 x 2
We can cancel out the x 2 factor as x 2 is close to, but not equal to, 0. Clearly, when x is
approaching to 2, we have x + 2 approaching to 2 + 2 =. This shows lim f (x) = 4.
x!2
We explain the reasons in the above paragraph just to clarify the concepts. In practice, we
may simply write down:

x2 4 (x 2)(x + 2)
lim f (x) = lim = lim
x!2 x!2 x 2 x!2 x 2
= lim (x + 2) (by cancelling x 2)
x!2
= 4.

In the last step we wrote lim (x + 2) = 4, but it is helpful to think of it as lim (x + 2) ! 4. It is

x!2 x!2
just a convention (a somewhat misleading one) to write “=”.
 2.1 Limit: Intuitive Meaning and Computations 41

⌅ Example 2.2 Consider the function

(
x+2 if x 6= 2
f (x) =
10 if x = 2

Is f (2) defined? How about lim f (x)?

x!2

⌅ Solution It is clear from the definition of f that f (2) = 10. For lim f (x), we only need to
x!2
look into the values of f (x) when x is close to, but not equal to 2. Therefore, the value of f (2)
is irrelevant.
When x 6= 2, we have f (x) = x + 2, so

lim f (x) = lim (x + 2) = 4.

x!2 x!2

⌅ Exercise 2.1 Consider the function

x3 8
f (x) =
x 2
Is f (2) defined? How about lim f (x)?
x!2

⌅ Exercise 2.2 Consider the function

(
x3 8
x 2 if x 6= 2
f (x) = .
a if x = 2

What is lim f (x) when a = 3? How about when a = 4?

x!2

2.1.2 Limit rules

We can make use of the following rules to compute its limit. We will omit the proofs here,
as it involve the rigorous ", -definition of limits, which will be covered formally in MATH
2033.
Proposition 2.1 If the limits lim f (x) and lim g(x) both exist, then we have:
x!a x!a
1. lim f (x) ± g(x) = lim f (x) ± lim g(x)
x!a x!a x!a
2. lim f (x)g(x) = lim f (x) · lim g(x)
x!a x!a x!a
3. lim cf (x) = c lim f (x) where c is any real constant.
x!a x!a
f (x) limx!a f (x)
4. lim = if limx!a g(x) 6= 0
x!a g(x) limx!a g(x)
5. lim f (x) = lim f (x) for any positive integer n
n n
x!a x!a
q
p
6. lim n f (x) = n lim f (x) for any odd positive integer n, and for any even positive
x!a x!a
integer n if limx!a f (x) 0.

In MATH 2033, we will define rigorous what is meant by “limit exists”. Meanwhile, you may
simply make sense of it through examples and by taking some given facts for granted. Here are
some examples that the limit does not exist:
42 Limit of Functions

1 1
lim does not exist (since when x is approaching to 0, approaches to infinity)
x
x!0 x
lim sin x does not exist (since sin x is oscillating between 1 and 1 as x increases)
x!+1

However, the limit of many elementary functions (such as polynomials, square roots) can be
shown (in MATH 2033) to exist, and so we can simply compute its limit by directly applying the
above rules. For instance, consider the limit:

⌅ Example 2.3 Compute the following limit, and explain what rules have been used in Proposi-
tion 2.1. p
lim 3x2 + x
x!9

⌅ Solution As x ! 9, we have lim x2 = 92 by Rule #5, and so lim 3x2 = 3 lim x2 = 3 · 92 = 243
x!9 px!9 x!9
p
by Rule #3. Furthermore, by Rule #6, we have lim x= 9 = 3. Now that both limits
x!9
p
lim 3x2 and lim x
x!9 x!9

exist, so by Rule #1 we conclude that

p p
lim 3x2 + x = lim 3x2 + lim x = 243 + 3 = 246.
x!9 x!9 x!9

i After having enough practice of using these limit rules, it is not necessary to always mention
what limit rules have been used. You could simply check by yourself that limit exists before
you apply the above rules, and simply write:
p p p
lim 3x2 + x = lim 3x2 + lim x = 3(9)2 + 9 = 246.
x!9 x!9 x!9

However, all these limit rules do not apply when the required condition is not met. For
example, it is NOT correct to apply Rule #4 like the following:

(3 + x)2 9 lim (3 + x)2 9

x!0
lim = WRONG!
x!0 x lim x
x!0

It is because the condition lim g(x) 6= 0 is needed when using the rule
x!a

f (x) limx!a f (x)

lim = .
x!a g(x) limx!a g(x)

In such case, some algebraic manipulations in order to find out the limit. See the examples
below:
⌅ Example 2.4 Compute the following limits:
(3 + x)2 9
1. lim
x!0 p x
x2 + 9 3
2. lim
x!0 x2
4x2 + 2x
3. lim1 2
x! 2 4x 8x + 3
2.1 Limit: Intuitive Meaning and Computations 43

⌅ Solution The denominators of all three functions approach 0, so we cannot simply apply the
limit rules #4 stated in Proposition 2.1. They can be computed as follows:
1.
(3 + x)2 9 9 + 6x + x2 9
lim = lim (expansion)
x!0 x x!0 x
6x + x2
= lim
x!0 x
x(6 + x)
= lim (factorization)
x!0 x
= lim 6 + x
x!0
=6

2.
p
x2 + 9 3
lim
x!0 x2
p p
x2 + 9 3 x2 + 9 + 3
= lim · p
x!0 x2 x2 + 9 + 3
p 2
x2 + 9 32
= lim p using (a b)(a + b) = a2 b2
x!0 x2 x2 + 9 + 3
x2 + 9 9
= lim p
x!0 x2 x2 + 9 + 3
x2
= lim p
x!0 x2 x2 + 9 + 3
1
= lim p
x!0 2
x +9+3
1
=p
02 + 9 + 3
1
= .
6
3.
4x2 + 2x 2x( 2x + 1)
lim1 2
= lim1 (factorization)
x! 2 4x 8x + 3 x! 2 (2x 1)(2x 3)
2x(2x 1)
= lim1
x! 2 (2x 1)(2x 3)
2x
= lim1
x! 2 2x 3
2( 12 )
=
2( 12 ) 3
1 1
= = .
2 2

i The identity (a b)(a + b) = a2 b2 is often helpful for computing limits which involve
square roots.
 44 Limit of Functions

⌅ Exercise 2.3 Evaluate the following limits:

x2 16
(a) lim
x!4 x 4
x2 + 5x 6
(b) lim
x!1 x2 1
x+3
(c) lim 3
x! 3 x + 27
p
1+x 1
(d) lim
x!0 x

2.1.3 One-sided limits

As x ! a, some functions (especially those which are piecewise defined) may approach to
different values depending on whether x approaches a from the left or from the right. In this
regard, we introduce the notion of one-sided limits:
Definition 2.1 — Left- and Right-Limits.

lim f (x) = L means: f (x) approaches L as x approaches a from the left

x!a
lim f (x) = M means: f (x) approaches M as x approaches a from the right
x!a+

i L and M may be different, and can be different from f (a).

i The notations a and a+ simply refer to left or right toward a, and do not mean whether a
is positive or negative.

One-sided limits appear often when the function is piecewise defined. See the example below:
Let: 8
<3x + 1 if x < 1
>
f (x) = 3 if x = 1 (2.1)
>
: 2
2x 1 if x > 1

4
3

1
x
1

Figure 2.2: left- and right-hand limits are different

The graph of f can be found in Figure 2.2. As you can see from the graph, the function f (x)
approaches to 4 if x comes close to 1 from the left, but it approaches to 1 if x tends to 1 from the
left. We denote:

lim f (x) = 4
x!1
lim f (x) = 1.
x!1+
Mark your answers on the Multiple Choice Item Answer Sheet provided.

Do not forget to put your name, student ID number, and the color
version of your exam paper on the Multiple Choice Item Answer Sheet.

2.1 Limit: Intuitive Meaning and Computations 45

Each of the following MC questions is worth 2 points. No partial credit.
If the left- and right-hand limits are the 8
same as x ! a, we say:
< 4ex + x k if x  0
Definition 2.2 The limit lim f (x) exists and equals to a finite number L if and only if
x!a
1. For which constant k can the function f (x) = sinnumber.
kx be continuous everywhere?
lim f (x) = lim f (x) = L where L is a :
finite if x > 0
x!a x!a+
x
(a) 1 i (b)discuss
We will 2 what happen
(c) 3if L is infinite(d) 4 next subsection.
in the (e) None of the previous

For the function f (x) defined in (2.1), since

lim f (x) 6= lim+ f (x),

x!1 x!1

we say that lim f (x) does not exist.

x!1
|f (|x|) 2|
2. Find lim according to the given graph of f below.
⌅ Example 2.5 Consider the graph of a function f below:
x! 2 f (x) + 1
y
y = f (x) 3

x
-3 3

(a) What
0 are f (2), lim
(b)x!23 x!2(c)
+
1 and lim f (x)? How
1 f (x), lim f (x)
2 x!2 (d) 23 about f(e)
(0), Does
lim fnot
x!0
(x), exist
lim f (x)
x!0+
and lim f (x)?
x!0

⌅ Solution The solid dot at the point (2, 0) indicates that f (2) = 0. When x ! 2 from the left,
f (x) is approaching 2, so
lim f (x) = 2.
x!2
⇣ 1 3 ⌘⇣ 1⌘
When
3. Find the x ! 2 from
horizontal the right,
asymptote (x) is
offthe approaching
function so
y = 3,cos 1 + x sin .
2x x2 x
lim f (x) = 3.
(a) 0 (b) 1 (c) 2 x!2+ (d) 3 (e) Does not exist
Since lim f (x) 6= lim+ f (x), the limit lim f (x) does not exist.
x!2 x!2 x!2
However, when x = 0, the value of f (x) is exactly 3, so f (0) = 3. Also, when x ! 0 either
from the left or from the right, we see that f (x) approaches 3, so we have

lim f (x) = 3 = lim f (x).

x!0 x!0+

Since the left and right limits are equal and finite, we can say lim f (x) exists, and lim f (x) = 3.
x!0 x!0

⌅ Exercise 2.4 Consider the graph y = f (x) is the above example. Discuss what are f ( 2),
lim f (x), lim f (x) and lim f (x)?
x! 2 x! 2+ x! 2
46 Limit of Functions

⌅ Example 2.6 Consider the graph y = f (x) in Example 2.5, determine whether

lim f (f (x))
x!2

exists or not.

⌅ Solution We need to determine separately lim f (f (x)) and lim f (f (x)).

x!2 x!2+
When x ! 2 (from the left), f (x) ! 2 from below. In other words, f (x) is slightly less
than 2 when x ! 2 . Now that f (x) ! 2 , we have f (f (x)) ! 2 as well. Therefore,

lim f (f (x)) = 2.
x!2

On the other hand, when x ! 2+ (from the right), f (x) ! 3 from below. We can see that
when z ! 3 , we have f (z) ! 1.2 (approximately). So we conclude that

lim f (f (x)) = 1.2

x!2+

Since lim+ f (f (x)) 6= lim f (f (x)), the limit

x!2 x!2

lim f (f (x))
x!2

does not exist.

⌅ Exercise 2.5 Consider the graph y = f (x) in Example 2.5, fill in the following table:

value value value

lim f (x) lim f (x) f (2)
x!2 x!2+
lim f (x) lim f (x) f ( 2)
x! 2 x! 2+
lim f (f (x)) lim f (f (x)) f (f (2))
x!2 x!2+
lim f (f (x)) lim f (f (x)) f (f ( 2))
x! 2 x! 2+
lim f ( f (x)) lim f ( f (x)) f ( f (2))
x!2 x!2+
lim f ( f (x)) lim f ( f (x)) f ( f ( 2))
x! 2 x! 2+

⌅ Example 2.7 Consider the function f : R ! R given by:

(
2x + 5 if x < 2
f (x) =
(x a)2 if x 2

where a is an unknown real number. Find the value(s) of a such that lim f (x) exists.
x!2

⌅ Solution This limit exists if and only if the left- and right-hand limits are equal. Let’s first
compute them out: As x ! 2 (from the left), x is regarded as less than 2. Therefore,

lim f (x) = lim 2x + 5

x!2 x!2
= 2(2) + 5 = 9
 2.1 Limit: Intuitive Meaning and Computations 47

On the other hand, as x ! 2+ (from the right), x is regarded as greater than 2, and so:

lim f (x) = lim+ (x a)2

x!2+ x!2
= (2 a)2

In order for lim f (x) to exist, we need:

x!2

9 = (2 a)2 =) 2 a = 3 or 3 =) a = 1 or 5.

⌅ Exercise 2.6 (a) Draw the graph of y = f (x) = |x 2| |x + 4|.

(b) Determine whether the limits lim f (x) and lim f (x) exist.
x!2 x! 4

(
x2 x + 2 if x 1
⌅ Exercise 2.7 Define f (x) = .
x
2 + 3
2 if x < 1
(a) Draw the graph of y = f (x) for 3  x  1.
(b) Determine whether the limit of lim f (x) exists.
x! 1

⌅ Exercise 2.8 Suppose f and g are functions defined by

( (
x2 4 if x < 0 5 if x < 0
f (x) = and g(x) = .
x+5 if x 0 x2 + a if x 0

f (x)
Find the value(s) of a such that lim exists.
x!0 g(x)

⌅ Exercise 2.9 Determine whether

|x|
lim
x!0 x
exists or not.

2.1.4 Infinite limits

1
Consider the function f : ( 1, 0) [ (0, 1) ! R given by: f (x) =
x

Figure 2.3: graph of f (x) = 1

x

The graph of f is shown in Figure 2.3. As you can see, as x is getting closer to 0 from the
48 Limit of Functions

right, the value of 1

x is getting larger and larger indefinitely. We denote:

1
lim = +1.
x!0+ x
Here, +1 should be regarded as a symbol rather than an ordinary number. Some arithmetic
rules of real numbers may not apply to the infinity.
Likewise, as x ! 0 , the value of x1 becomes more and more negative indefinitely, and so we
say:
1
lim = 1.
x!0 x

Our convention would be 1 (without the sign) means +1, but please note that other authors
may have different conventions. When either one of the left- or right-hand limits is 1 or 1,
we say that the limit does not exist. Graphically, having an infinite limit when x ! a means the
graph of the function is almost “touching” the vertical line x = a. Formally, we say:
Definition 2.3 — Vertical Asymptotes. The vertical line x = a is called a vertical asymptote of
the function f (x) if either one of lim f (x) or lim f (x) is +1 or 1.
x!a x!a

A function f has infinite limit as x ! a when it has the form of non-zero

0 when x approaches to
a, i.e.
g(x)
f (x) = where lim g(x) 6= 0 while lim h(x) = 0.
h(x) x!a x!a

Also, if the limit is in the form of finite

1
, including 1
0 , the resulting limit is also ±1 – whether it is
+1 or 1 will be determined by the sign of the function. For example:
1
lim = +1
x!0+ x
1
lim = 1
x!0 x
1
lim = ⌥1
x!0± x
1
lim 2 = +1
x!0 ± x

On the other hand, if the limit is of the form finite

1 , the resulting limit would be 0 since when
the denominator gets very large, the whole fraction gets smaller. For example:
1 1
lim 1 = = 0.
x!0 x+ x
0±1

However, if the limit is of the form 00 or 1

1 , then the limit may be finite or infinite, as you can
see in the examples below: all of the following functions are of the form 00 when x ! 0, but the
first two have finite limits whereas the third one tends to infinity.

x2
lim = lim x = 0
x!0 x x!0
2
2x
lim = lim 2 = 2
x!0 x2 x!0
x 1
lim = lim 2 = +1.
x!0 x3 x!0 x

Recall that when lim f (x) = ±1, we say the limit does not exist. Therefore, all limit rules
x!a
listed in Proposition 2.1 cannot be used. If lim f (x) = +1 and lim g(x) = +1, it is also
x!a x!a
incorrect to write:

lim f (x) g(x) = lim f (x) lim g(x) WRONG!

x!a x!a x!a
2.1 Limit: Intuitive Meaning and Computations 49

The limits below are all of the form 1 1, but the outcomes are different:
✓ ◆
1 1 x4 1
lim 2 2 4
= lim 2 2 = lim =1
x!0 x x +x x!0 x (x + x4 ) x!0 1 + x2
✓ ◆
1 2 1 1
lim = lim 2 = + = 1
x!0 x2 x2 x!0 x 0
✓ ◆
2 1 1 1
lim = lim 2 = + = +1
x!0 x2 x2 x!0 x 0
You can see that all three possibilities: finite, +1, and 1 could happen in case the limit is of
the form 1 1.
To summarize, if the limit is of the form 00 , 1 1 , or 1 1, we cannot determine its limit
directly and some algebraic manipulation is needed to find its value. We call any of these limits
to be of indeterminate form.
It is incorrect to write, for instance, that
r ! r
1 1 1 1
lim 1+ 2 = lim 1 + 2 lim WRONG!
x!0 x |x| x!0 x x!0 |x|

r
1 1
because the limits lim and lim
1+ are infinity, so they do not exist.
x!0 x2 x!0 |x|
Instead, one should perform some “rationalization” trick in order to compute this limit:
r ! r ! q
1 1 1 1 1 + x12 + |x|
1
lim 1+ 2 = lim 1+ 2 ·q
x!0 x |x| x!0 x |x| 1+ 1 + 1 x2 |x|
1 1
1+ x2 x2
= lim q
x!0 1 1
1+ x2 + |x|
1
= lim q .
x!0 1 1
1+ x2 + |x|
r
1 1
When x ! 0, we have 1
|x| ! +1, so the denominator, 2
+1+ , which is even larger than
x |x|
finite
|x| , would tend to +1 too. Therefore, the limit is of the form +1 , so we conclude that
1

r !
1 1 1
lim 1+ 2 = lim q = 0.
x!0 x |x| x!0
1+ 1
+ 1
x2 |x|

However, one should be more careful with the limit

r !
1 1
lim 1+ 2
x!0 x x

even thought it looks similar to the previous example. By the similar rationalization trick, we get
r !
1 1 1
lim 1+ 2 = lim q .
x!0 x x x!0
1+ 1 + 1 x2 x

Let’s show that the limit does not exist by looking into its left and right limits.
p
As x ! 0 , we have x12 ! +1, so 1 + x12 ! +1 as well. Since the square root sign always
refers to the positive square root, we conclude that
r
1
lim 1 + 2 = +1.
x!0 x
50 Limit of Functions

1
However, we have lim = 1 since 1
x < 0 when x approaches 0 from the left. Therefore, the
x!0 x
denominator r
1 1
1+ 2
+
x x
is of indeterminate form 1 1 as x ! 0 . We cannot conclude what is the limit
1
lim q .
x!0 1 1
1+ x2 + x

However, if we look back at r !

1 1
lim 1+ 2 ,
x!0 x x
p
it is not of indeterminate form, since x1 ! +1 as x ! 0 . Recall that must be positive, so
q
1 + x12 x is of the form 1 + 1 as x ! 0 . We can conclude that
1

r !
1 1
lim 1+ 2 = +1.
x!0 x x

Hence, even the left-limit does not exist. It is already even though to argue that
r !
1 1
lim 1+ 2
x!0 x x

does not exist.

⌅ Exercise 2.10 Determine whether the following limit is finite:
r !
1 1
lim 1+ 2 .
x!0+ x x

⌅ Exercise 2.11 Determine!whether or the following limits exist:

r
1 1
(a) lim 1
x!0 x2 x
r !
1 1
(b) lim 1
x!0 x2 |x|

⌅ Exercise 2.12 Find the following limits, or show that they do not exist:
x 1
(a) lim p
x!1
✓x 1 ◆
1 2
(b) lim +
x!0 x x(x 2) ◆
✓
1 2
(c) lim +
x!0 x
✓ x |x 2| ◆
1 2
(d) lim
x!0 x x |x 2|
 2.1 Limit: Intuitive Meaning and Computations 51

⌅ Exercise 2.13 Evaluate the following limit. If it is infinite, indicate whether it is +1 and 1.
1
(a) lim =
x!1 x 1
1
(b) lim =
x!1+ x 1
2
(c) lim =
x!1 x2 1
2
(d) lim =
x!1+ x2
✓ 1 ◆
2 1
(e) lim 2
+ =
x!1
✓x 1 x 1◆
2 1
(f) lim+ 2
+ =
x!1
✓x 1 x 1◆
2 1
(g) lim 2
=
x!1
✓x 1 x 1◆
2 1
(h) lim =
x!1+ x2 1 x 1

2.1.5 Limit at infinities

In many practical applications, we need to talk about the trend of a function f (x) when x is very
large or very negative. These applications include the analysis of computational complexity (in
CS) and the study of equilibrium states. In this regards, we have:
Definition 2.4 — Limit at Infinity. Let f : R ! R be a function, we say:
• lim f (x) = L if f (x) tends to L when x increases indefinitely
x!+1
• lim f (x) = M if f (x) tends to M when x decreases indefinitely
x! 1

Geometrically, lim f (x) = L means that the graph y = f (x) is getting closer and closer to
x!+1
the horizontal line y = L as x increases indefinitely; and lim f (x) = M means that y = f (x)
x! 1
gets closer and closer to the line y = M as x decreases indefinitely. In either case, we call that
horizontal line to be a horizontal asymptote.

Figure 2.4: the two dotted lines are horizontal asymptotes of the function

Easy examples of limit at infinity include the following (recall that 1 means +1):
1
• lim = 0 where p is any positive number.
xp
x!1 p
• lim x=1
x!1
• lim a = 1 (where a > 1)
x
x!1
• lim ax = 0 (where 0 < a < 1)
x!1
52 Limit of Functions

Some typical techniques of finding limits at infinity of rational functions (i.e. polynomial
divided by polynomial) are illustrated in the example below:

⌅ Example 2.8 Evaluate the following limits:

x2 1
1. lim
x!1 x2 + 2
x
2. lim 2
x!1 x + 1
x3 1
3. lim 2
x!1 x + 1

⌅Solution A standard technique for evaluating these limits is so-called division by highest
power of x:
1.
x2 1
x2 1 x2
lim 2 = lim x2 +2
(division by x2 )
x!1 x + 2 x!1
x2
1
1 x2
= lim 2
x!1 1 +
x2
1 0 1
= (since ! 0 as x ! 1)
1+0 x2
=1

2.
x 1
x x2 x
lim = lim x2 +1
= lim 1
x!1 x2 + 1 x!1 x!1 1+
x2 x2
0
= =0
1+0
3.
x3 1
x3 1 x3
lim = lim x2 +1
x!1 x2 + 1 x!1
x3
1
1 x3
= lim
x!1 1 + 1
x x3

Since the numerator 1 1

x3 ! 1 while the denominator 1
x + 1
x3 ! 0, the quotient tends to 1,
and so:
x3 1
lim = 1.
x!1 x2 + 1
2.1 Limit: Intuitive Meaning and Computations 53

⌅ Example 2.9 Evaluate the limit:

2x + 3 x
lim .
x!1 1 5x

⌅Solution We employ similar technique as in the previous example, by dividing both numerator
and denominator by the largest term, which is 5x in this example:

2x + 3x
x x
2 +3 5x
lim = lim
x!1 1 5x x!1 1 5x
5x
( ) + ( 3 )x
2 x
= lim 5 1 x 5
x!1 ( ) 1
5

Since 15 , 25 , 3
5 are all in (0, 1), we have
✓ ◆x ✓ ◆x ✓ ◆x
1 2 3
lim = 0, lim = 0, and lim = 0.
x!1 5 x!1 5 x!1 5

This concludes that

2x + 3x ( 52 )x + ( 35 )x 0+0
lim x
= lim 1 x
= = 0.
x!1 1 5 x!1 (5) 1 0 1

When handling the limit when x ! 1, we need to bare in mind that we would have instead
(
x 0 if a > 1
lim a = .
x! 1 +1 if 0 < a < 1

This result follows from substituting y = x, then as x ! 1, we have y ! +1, so

1
lim ax = lim a y
= lim .
x! 1 y!+1 y!+1 ay
When a > 1, we will have lim ay = +1, and so
y!+1

1 1
lim ax = lim = = 0.
x! 1 y!+1 ay +1
When 0 < a < 1, we then have ay ! 0 as y ! +1. As ay is positive, we can conclude that
1 1
lim ax = lim = + = +1.
x! 1 y!+1 ay 0

⌅ Exercise 2.14 Suppose p(x) = am xm +am +· · ·+a1 x+a0 and q(x) = bn xn +bn 1 xn 1 +
1x
m 1

· · · + b1 x + b0 are two polynomials of degrees m and n respectively, where am , bn 6= 0. Show

that 8
>0 if m < n
p(x) < am
lim = bn if m = n .
x!1 q(x) >
:
±1 if m > n
54 Limit of Functions

⌅ Exercise 2.15 Find the limit:

2x + 3 x
lim .
x! 1 1 5x

⌅ Exercise 2.16 Find the limits

3x3 4x2 + 1 3x3 4x2 + 1

lim p and lim p .
x!1 25x6 4x2 + 8 x! 1 25x6 4x2 + 8

Here is an important fact concerning limit at infinity and the Euler’s number e:
Definition 2.5 — Euler’s Number. ✓ ◆x
1
e = lim 1+
x!1 x

In MATH 1014 (L1) or MATH 1023, we will explain why this limit exists, but let’s assume so
for this course. By putting a very large number into x, we can find that

e ⇠ 2.71828183 . . .

⌅ 2.10 Evaluate
Example ✓ ◆ the limits:
x
2
1. lim 1+
x!1
✓ x◆
x
1
2. lim 1
x!1 x

⌅ Solution This example shows one can evaluate limits by substitution.

1.
✓ ◆x ✓ ◆x
2 1
lim 1+ = lim 1+
x!1 x x!1 x/2
✓ ◆2y
1
= lim 1 + (Let y = 2x)
y!1 y
✓ ◆y 2
1
= lim 1+ (recall that (ab )c = abc )
y!1 y
= e2

2. The second limit is actually trickier than it seems. The key trick is to do a substitution
y = x 1. As x ! 1, we also have y ! 1.
✓ ◆x ✓ ◆x ✓ ◆y+1
1 x 1 y
lim 1 = lim = lim (Let y = x 1)
x!1 x x!1 x y!1 y+1
1 1
= lim ⇣ ⌘y+1 = y!1
lim ⇣ ⌘y+1
y!1 y+1 1
y 1+ y
1
= lim ⇣ ⌘y ⇣ ⌘
y!1 1
1+ · 1 + y1
y
1 1
= = .
e·1 e
 2.1 Limit: Intuitive Meaning and Computations 55

⌅
✓ Find
Exercise 2.17 ◆ the limits below
x+1
x+1
(a) lim
x!1 x
✓ ◆x 2
1
(b) lim 1+
x!1
✓ x ◆
x
1
(c) lim 1 + 2
x!1
✓ x◆
x
x+3
(d) lim
x!1 x + 4

2.1.6 Limits of logarithms and trigonometric functions at infinity

ln x and ex are inverse functions of each other in a sense that

y = ln x () x = ey .

Recall that lim ex = 0. Think of it informally as e 1

= 0 (which is not conceptually correct),
x! 1
then we expect that ln(0) = 1, or formally, we have

lim ln x = 1.
x!0+

It can be rigorously proven in MATH 1023/2033, yet we will omit the formal proof here.
Likewise, by the fact that lim ex = +1, we also expect that
x!+1

lim ln x = +1,
x!+1

which can also be rigorously proven in MATH 1023/2033.

Similar argument can be used to conclude the limit of lim tan 1

x. Before we get into that,
x!±1
let’s explain more about these inverse trigonometric functions. One common confusion about
these inverse sine, cosine, tangent functions are that sin x, cos x, tan x are not injective functions,
but why we can make sense of sin 1 x, cos 1 x and tan 1 x?
Take sin x as an example. It is not injective because sin(0) = sin(⇡) = 0. However, if one
restricts the domain of sin x to [ ⇡/2, ⇡/2], then it is strictly increasing on [ ⇡/2, ⇡/2] making it
injective. Denote
SIN : [ ⇡/2, ⇡/2] ! [ 1, 1]
56 Limit of Functions

to be the restricted sine function, i.e. SIN(x) = sin x for any x 2 [ ⇡/2, ⇡/2]. Then, SIN is a
bijective function. The symbol sin 1 is actually introduced with abuse of notations. It is NOT the
inverse function of sin x, but is the inverse function of SIN(x).
Since sin 1 is in fact SIN 1 in disguise, the domain and range of sin 1 is:
1
sin : [ 1, 1] ! [ ⇡/2, ⇡/2].

Therefore, the value of sin 1 x is always in the range [ ⇡/2, ⇡/2]. For instance, sin 1 (1/2) =
⇡/6 because sin(⇡/6) = 1/2. However, even if we have sin(3⇡/2) = 1, it does NOT imply
sin 1 ( 1) = 3⇡/2. Instead, sin 1 ( 1) is the value c 2 [ ⇡/2, ⇡/2] such that sin(c) = 1. This
value c can be easily seen to be ⇡/2. Therefore, sin 1 ( 1) = ⇡/2. Similarly, cos x is not
injective when its domain is R, but when we restrict the domain to be [0, ⇡], then it becomes
injective. Define
COS : [0, ⇡] ! [ 1, 1]
with COS(x) = cos x for any x 2 [0, ⇡], then COS is a bijective function. By what we meant cos 1

is the inverse function of COS. Therefore, the domain and range of cos 1 is:
1
cos : [ 1, 1] ! [0, ⇡].

Figure 2.5: although sin x and cos x are not injective on the whole real line, they would become
injective if one restricts their domains on the yellow region in the above diagram.

For tan x, it becomes injective if we restrict its domain on ( ⇡/2, ⇡/2). The inverse tangent
tan 1
is in fact the inverse function of the restricted tangent:

TAN : ( ⇡/2, ⇡/2) ! R

where TAN(x) = tan(x) for any x 2 ( ⇡/2, ⇡/2). The domain and range of tan 1
is
1
tan : R ! ( ⇡/2, ⇡/2).
2.1 Limit: Intuitive Meaning and Computations 57

Recall that lim

⇡
tan x = +1, or informally tan( ⇡2 ) = +1, we expect that informally we
x! 2
have tan 1
(+1) = 2,
⇡
or more formally:

1 ⇡
lim tan x= .
x!+1 2

Likewise, we have (again informally) tan( ⇡2 + ) = 1 and so tan 1

( 1) = 2,
⇡
so we should
have
⇡
lim tan 1 x = .
x! 1 2
On the other hand, all of the following limits do not exist:

lim sin x, lim cos x, lim tan x, and their x ! 1 counterparts.

x!1 x!1 x!1

An informally explanation for sin x and cos x are that they are oscillating between 1 and 1 so
they do not approach to a specific number when x ! 1. It can be proven more rigorously in
MATH 1023/2033.
 58 Limit of Functions

2.2 Continuity
2.2.1 What is a continuous function?
Loosely speaking, continuity of a function f means that the graph y = f (x) is not broken.
Formally, it is defined using limit:
Definition 2.6 — Continuous Functions. Given a function f :
• f is said to be continuous at x = a if
1. lim f (x) exists; and
x!a
2. f (x) is defined at x = a (in particular, f (a) needs to be finite); and
3. lim f (x) = f (a).
x!a
• f is said to be continuous on an interval I if it is continuous at every point in I
• f is said to be continuous (everywhere) if it is continuous at every point in R.
• If f is not continuous at x = a, we say f is discontinuous at x = a.

Figure 2.6: illustration on concept of continuity

Figure 2.7: this function is discontinuous at x = 1, x = 3 and x = 5.

2.2 Continuity 59

⌅ Example 2.11 For each function below, find all value(s) of a (if any) such that the function is
continuous at x = 0:
1. (
1
if x 6= 0
f (x) = x
a if x = 0
2. (
3x2 + x + a if x 6= 0
g(x) =
a2 if x = 0
3. 8
> 2
<x + 1 if x < 1
h(x) = a if x = 1
>
:
2x if x > 1

⌅ Solution 1. Since
1
lim f (x) = lim = +1.
x!0+ x!0+ x
The limit lim f (x) does not exist. The function f cannot be continuous at x = 0 no matter
x!0
what value a is.
2.
lim g(x) = lim 3x2 + x + a = 3(0)2 + 0 + a = a
x!0 x!0

Hence, g is continuous at x = 0 if and only if:

a2 =
|{z} a
|{z}
g(0) limx!0 g(x)

Solving for a, we get a = 0 or 1. Hence g is continuous at x = 0 if a = 0 or 1.

3.
lim h(x) = lim x2 + 1 = 1
x!0 x!0

lim h(x) = lim 2x = 0

x!0+ x!0+

Therefore, the left- and right-hand limits do not match and the limit lim h(x) does not exist.
x!0
There is no a such that h(x) is continuous at x = 0.

Many elementary functions are continuous on their domain. Examples include:

• polynomials are continuous on R
• sin x and cos x are continuous on R
• sin 1 x, cos 1 x are continuous on their domains
• tan
p
1
x is continuous on R
• x is continuous on [0, 1)
• |x| is continuous on R
• ex is continuous on R
• loga x (where a > 1) is continuous on (0, 1)
By the following proposition, we can see that the above functions can generate a lot of other
continuous functions on their domain:
Proposition 2.2 If f and g are both continuous at x = a, then all of the following are continuous
at x = a:
f + g, f cf (where c is a constant), f g.
g,
p
Furthermore, if f (x) 0 when x is close to a, then f is also continuous at x = a. If g(a) 6= 0
(and f (a) can be any real number), then fg is also continuous at x = a.
 60 Limit of Functions

Using this proposition, one can see quickly that a complicated function like:
1
ln x + tan x
f (x) =
x2 1
is continuous on (1, 1) for instance – since at every a in (1, 1), the functions ln x, tan 1 x and
x2 1 are all continuous, and the denominator x2 1 is non-zero. However, as x ! 1, we have
1 1 1 ⇡
lim (ln x + tan x) = lim ln x + lim tan x = 0 + tan (1) = ,
x!1 x!1 x!1 4
finite
but lim (x2 1) = 12 1 = 0. Therefore, f is of the form 0 as x ! 1, hence lim f (x) = ±1.
x!1 x!1
The function f is not continuous at x = 1.
Yet, f is continuous on (0, 1), since ln x and tan 1
x are so, and lim (x2 1) = a2 1 6= 0 if
x!a
a 2 (0, 1). Also, the function ln x is not defined at any point in ( 1, 0], so f is not continuous at
any point in ( 1, 0].
To summarize, f is continuous on (0, 1) [ (1, 1)

⌅ Exercise 2.18 Determine the maximum possible set on which the following functions is
continuous.
x2 + 2
(a) f (x) =
x2 4
|x|
(b) f (x) =
x
(c) f (x) = ln
p |x|
4 + x2 2
(d) f (x) =
x

⌅ Exercise 2.19 Consider the function f whose graph is given as in Example 2.5. Which of the
following functions is/are continuous at x = 2? How about continuous at x = 2?

f (x), f (f (x)), f ( f (x)).

2.2.2 Can we put the limit sign into a function?

When we evaluate a limit such as ✓ 2 ◆
x +1
lim sin ,
x!0 x 1
we may ask whether it is legitimate to put the limit sign lim into the sin function, so that
x!0
✓ ◆ ✓ ◆
x2 + 1 x2 + 1
lim sin = sin lim ?
x!0 x 1 x!0 x 1

The following theorem tells us that we can do that if the functions is continuous.
Theorem 2.3 If lim g(x) = b and f is continuous at b, then
x!a
⇣ ⌘
lim f (g(x)) = f lim g(x) = f (b).
x!a x!a

The result also holds for left- and right-limits lim , and for limits at infinity lim .
x!a± x!±1

⌅ Example 2.12 Evaluate the limit:

lim tan (sin x) .

x!⇡/2
2.2 Continuity 61

⌅ Solution Since lim sin x = 1 and |{z}

tan is continuous at 1, we have:
x!⇡/2 |{z}
g(x) f

✓ ◆
lim tan (sin x) = tan lim sin x = tan 1.
x!⇡/2 x!⇡/2

⌅ Example 2.13 Evaluate the limit:

✓ ◆
1
lim x ln 1 + .
x!1 x

Hence, derive that:

ah 1
lim = ln a
h!0 h
where a > 1 is a real constant.

⌅ Solution First observe that:

✓ ◆ ✓ ◆x
1 1
x ln 1 + = ln 1 + .
x x

We know that ✓ ◆x
1
lim 1+ =e
x!1 x
and ln is continuous at e, so:
✓ ◆ ✓ ◆x  ✓ ◆x
1 1 1
lim x ln 1 + = lim ln 1 + = ln lim 1 + = ln e = 1.
x!1 x x!1 x x!1 x

The second part is quite tricky. The key trick is to let y = ah 1, then

ln(y + 1)
h = loga (y + 1) = .
ln a
As h ! 0, y ! 0 too. Hence

ah 1 y
lim = lim
h!0 h y!0 ln(y+1)
ln a
y
= lim · ln a.
y!0 ln(1 + y)

Now it looks closer to the previous limit. By letting x = 1

y (which tends to 1 as y ! 0), we
get:
1
y x
lim · ln a = lim · ln a
y!0 ln(1 + y) x!1 ln 1 + 1
x
1
= lim · ln a
x!1 x ln 1 + 1
x
ln a
= (from previous results)
1
= ln a.
62 Limit of Functions

In particular, we have
eh 1
lim = ln e = 1. (2.2)
h!0 h
You are probably being “told” about this result in DSE M1/M2. Now you know the proof of it.
Using the fact that ln x is continuous on (0, 1), one can find the limit of functions of the form
f (x)g(x) :

⌅ Example 2.14 Find the limit

2
lim (1 + x3 )1/x
x!0+

2
⌅ Solution Let y = (1 + x3 )1/x , then ln y = 1
x2 ln(1 + x3 ). As x ! 0+ , we have

1
lim ln y = lim+ ln(1 + x3 ).
x!0+ x!0 x2

By letting z = x3 ,
1
then z ! 1, and so
✓ ◆ ✓ ◆
1 3 1 1 2/3 1
lim+ ln(1 + x ) = lim q ln 1 + = lim z ln 1 +
x!0 x2 z!1 3 1 2
( ) z z!1 z
z

which looks quite✓similar◆(but not exactly) to the previous example, except that the previous
1
result is lim z ln 1 + . To tackle this, we relate z 2/3 with z:
z!1 z
✓ ◆ ✓ ◆
1 1 1
lim z 2/3 ln 1 + = lim 1/3 · z ln 1 + = 0 · 1 = 0.
z!1 z z!1 z
| {z } | z
{z }
!0 !1

This conclude that

✓ ◆
1 3 2/3 1
lim+ ln y = lim+ ln(1 + x ) = lim z ln 1 + = 0.
x!0 x!0 x2 z!1 z

Remember that our goal is to find lim+ y. Using the fact that ln is continuous on (0, 1), we
x!0
have ✓ ◆
ln lim y = lim ln y = 0.
x!0+ x!0+

This shows lim y = 1.

x!0+

⌅ Exercise 2.20 Find the following limits:

3x + 7
(a) lim x log
x!1 3x + 5
3x + 5
(b) lim x log
x!1 3x + 7
1 3x + 7
(c) lim log
x!0 x 3x + 5
1 3x + 5
(d) lim log
x!0 x 3x + 7
 2.2 Continuity 63

⌅ Exercise 2.21 Find the limit:

ex ea
lim
x!a x a
where a is any real number.

⌅ Exercise 2.22 Show that

8
>
<0 if p > q > 0
p 1/xq
lim (1 + x ) = e if p = q > 0 .
x!0+ >
:
+1 if q > p > 0

⌅ Exercise 2.23 Consider the function

(
|x|
x if x 6= 0
f (x) = .
2 if x = 0

Show that ⇣ ⌘
lim f (sin2 x) 6= f lim sin2 x .
x!0 x!0

Does the result contradict Theorem 2.3? Why or why not?

2.2.3 Squeeze theorem

When a the graph of a function f is being “trapped” by two other functions g and h, and that g
and h have the same limit as x ! a, one should naturally expect that f should have the same
limit as x ! a. It is in fact true, and the result is called:
Theorem 2.4 — Squeeze Theorem, or Sandwich Theorem. Let f, g, h : R ! R be three functions
which satisfy
g(x)  f (x)  h(x) for any x in ( 1, a) [ (a, 1).
If we have
lim g(x) = L = lim h(x),
x!a x!a

then lim f (x) also exists and

x!a
lim f (x) = L.
x!a

Here a can be +1 or 1. For finite a, all the above limits can be one-sided.

⌅ Example 2.15 Let f : R ! R be defined by:

(
x sin x1 if x 6= 0
f (x) =
0 if x = 0

Evaluate
lim f (x).
x!0

⌅ Solution For any x 6= 0, we have:

1
1  sin 1
x
64 Limit of Functions

and so for any x > 0, we have:

1
x x sin x
| {z }x
=f (x) when x 6= 0

Since both x and x approaches to 0 as x ! 0+ , by squeeze theorem, we have:

lim f (x) = 0.
x!0+

A similar argument can be applied to show lim f (x) = 0. However, one needs to be
x!0
careful that when x < 0, we have instead:
1
x  x sin  x.
x
A useful fact about limit of trigonometric functions stated below can be proved using squeeze
theorem:
Proposition 2.5
sin x
lim =1
x!0 x

Proof. Assume x > 0, consider the diagram below:

Clearly, we have [4OBC] < [sector OBC] < [4OBD]. Here [ ] denotes the area. By elementary
trigonometry, we get
1 1 1
sin x < x < tan x.
2 2 2
sin x
The left inequality implies < 1. The right inequality can be written as
x
sin x sin x
x< =) > cos x.
cos x x
To summarize, we have
sin x
cos x < < 1.
x
Since lim+ cos x = 1 = lim+ 1, by squeeze theorem, we can conclude that
x!0 x!0

sin x
lim = 1.
x!0+ x
sin x
Since is an even function, we also have
x
sin x
lim = 1,
x!0 x
completing the proof of the proposition. ⌅
2.2 Continuity 65

⌅ Example 2.16 Evaluate the following limit using Proposition 2.5:

sin 2x
1. lim
x!0x
1 cos x
2. lim
x!0 x2
tan x
3. lim
x!0 x

⌅ Solution 1.
sin 2x sin y
lim = lim y (Let y = 2x)
x!0 x y!0
2
sin y
= 2 lim
y!0 y

=2·1=2

2.

1 cos x sin2 x2
lim = lim (Half-angle formula)
x!0 x2 x!0 x2
sin2 y x
= lim (Let y = )
y!0 (2y)2 2
sin2 y
= lim
y!0 4y 2
✓ ◆2
1 sin y
= lim
y!0 4 y
1 2 1
= ·1 =
4 4
3.
tan x sin x
lim = lim
x!0 x x!0 x cos x
sin x 1
= lim ·
x!0 x cos x
1
=1·
cos 0
1
=1· =1
1

⌅ Example 2.17 Show that

(
n2 m 2
cos mx cos nx 2 if m2 =
6 n2
lim = .
x!0 x2 0 if m = n2
2

⌅ Solution If m2 = n2 , then m = n or m =
n. When m = n, clearly we have cos mx = cos nx.
When m = n, we recall that cos is an even function, so we also have cos mx = cos( nx) =
cos nx. Therefore, when m2 = n2 , we have
cos mx cos nx 0
lim = lim 2 = 0.
x!0 x2 x!0 x

Note that as x ! 0, we shall assume x is just very close to, but not equal to 0. Therefore 0
x2 is
66 Limit of Functions

exactly equal to 0 when x ! 0.

Now when m2 6= n2 , we use the sum-to-product formula for cos:
cos mx cos nx
lim
x!0 x2
2 sin (m+n)x
2 sin (m 2n)x
= lim
x!0 x2
(m+n)x
sin 2 sin (m 2n)x
= lim 2 · · .
x!0 x x

(m+n)x sin (m+n)x

Note that although 2 goes to 0 as x ! 0, we cannot claim that lim 2
=1
x!0 x
(m+n)x
because 2 and x are not the same. However, we can rewrite the limit as:

sin (m+n)x
2 sin (m+n)x
2 m+n
lim = lim (m+n)x · .
x!0 x x!0 2
2

(m+n)x sin (m+n)x

Since 2 ! 0 as x ! 0, so lim 2
(m+n)x
= 1 too, and this shows
x!0
2

sin (m+n)x
2 m+n
lim = .
x!0 x 2
Likewise, we can also show
sin (m 2n)x m n
lim = .
x!0 x 2
Combining all the above results, we finally can conclude that when m2 6= n2 , we have

cos mx cos nx sin (m+n)x

2 sin (m 2n)x m+n m n n2 m2
lim = lim 2 · · = 2· · = .
x!0 x2 x!0 x x 2 2 2

⌅ Exercise 2.24 Find the limits below:

sin x
(a) lim
x!1 x
1
(b) lim x sin
x!1 x
1
(c) lim x sin
x!0 ✓ x ◆
1
(d) lim x 1 cos
x!1
✓ x◆
1
(e) lim x 1 cos
x!0 x
cos x cos a
(f) lim where a 2 R.
x!a x2 a2
sin x sin a
(g) lim where a 2 R.
x!a x a

⌅ Exercise 2.25 — HKAL Pure Math 2008, Paper II Q1. Answer the following questions:
1
(a) Prove that lim x cos = 0.
x!0 x
(b) Find the limit
x2 cos x1
lim .
x!0 tan x
Exercise 2 .
24

( a) Fiud l( e
lixt s ∞
刘
) ox ( tos
similarly 箱 像
xsin
弛)
golution : ≤ sinx ≤ 1 -

1 sinx 1
≤≤

O ≤ sin ( x) ≤ 1
) ≤ s坊 |
-

1 x 1 ≤ x sin
^

( x) ≤ x1

limxt ∞ ⼀⽚ ≤∞≤+xl | limxxo - x |☆ 洲i

{
☆
0λ
l lx 1
sin [ x ] ≤ ☆ 0

: ∞⼀⽚ |= x |= 它
o ∵ 1 。
-

1x 1
=01 xl = 0

∴∞= On . ∴ 。 x ( 1- os = )

0
λ sin
=

(x) =
06

( bxxsin 1 1
H
怎⼼
(⽚ )

s timly
Put =
⽅
y ,

λ→∞ → 0
,
y

“ “
品 x m 忙
吃

2 xlx 立 xlx 当⼆⼀当

1
⼆ -

= 1
,

( gasiua 2 c
“
)s
0 s

《
Find ) xsin (

(
Solontion sin ≤ 1
⼆
2 ws ()
xsm 吃

| x1 ≤λ sin ( x) ≤ ( x1 = 2 cash ] xlx 点

。 。xsin ( ) ≤ 。lx 1 Cosacs

=
⼀

x≤

∵ 1 。
-

1 x |= 。 1
x1 = 0

∴ 。xsin () =
oa

@
xx 1 - ( 以⽂ )

我 sinosimo氢
xx ano cs)
=
oso
2

os
(
_

= -

⼆⾄ gimx

xxsin
Patg 就
⼆蕊 smx 范 sx 立 xsinyx
=

立 xsin 。

, 0
→∞ f
=
-

-
1 × 当 x 0

= O
4
 2.2 Continuity 67

⌅ Exercise 2.26 Let n be a positive integer.

(a) Prove by induction that:
n
X cos nx cos(n + 1)x
1+2 cos kx =
1 cos x
k=1

for any n 2 N and x 2 R such that cos x 6= 1.

(b) Prove by induction that:

n
!2 n
!2
X X 1 cos nx
sin kx + cos kx =
1 cos x
k=1 k=1

for any n 2 N and x 2 R such that cos x 6= 1.

(c) Hence, or otherwise, find the limit:
1 cos nx
lim
x!0 1 cos x

2.2.4 Intermediate Value Theorem

A continuous function has an “connected” graph. Therefore, it is quite intuitive that if a function
f is continuous on [a, b] and satisfies f (a) > 0 while f (b) < 0, the function should have a zero,
i.e. f (c) = 0, for some c in (a, b). This is commonly called:
Theorem 2.6 — Intermediate Value Theorem (IVT). Let f : [a, b] ! R be a continuous function
such that f (a) and f (b) have different signs, i.e. f (a) < 0 < f (b) or f (b) < 0 < f (a). Then,
there exists at least one c in the interval (a, b) such that f (c) = 0.

Figure 2.8: Intermediate Value Theorem

68 Limit of Functions

A quick example of using the IVT is to show that the polynomial f (x) = x3 + 2x 1 has at least
one real root. That is because f (0) = 03 + 2(0) 1 = 1 < 0 and f (1) = 13 + 2(1) 1 = 2 > 0.
Therefore, the value 0 is between f (0) and f (1). Any polynomial is continuous, so by the IVP
there exists at least one c in (0, 1) such that f (c) = 0.

⌅ Exercise 2.27 Show that the equation x3 + 2 = sin x has at least one solution.

⌅ Exercise 2.28 Let f : [0, 1] ! [0, 1] be a continuous function. Show that there exists c 2 [0, 1]
such that f (c) = c.

The proof of IVT is not easy. It will be covered in MATH 2033.

Exercie 2 27
.

Shon has least

'
that the sinx at solutin
equation X 4 2 =
one .

Proof ,Leeftx ) xez - sinx

=
hich is ontinnous :
f 1 .

f 10) = 0 +2 -
0 = 2 > 0

f( ) -

2 = -

8 + 2 -

siaE -
2 ]

=
-

6+ sin 2 < 0

∴ 雪 c ε ( -

2
,
0
) that f[ c
) = 0

has
'
The x siux at least solutin
eqhation
∴ 4 2 =

one .
 2.3 Toward Understanding Rigorous Definition of Limits 69

2.3 Toward Understanding Rigorous Definition of Limits

In this section, we will briefly introduce the true definition of limits. We have been interpreting
lim f (x) = L as “when x is very close to a, then f (x) is very close to L”. However, such an
x!a
interpretation is not mathematically rigorous. How do you define what it means by two numbers
being “close”?
The main theme of MATH 2033 is to rigorously introduce the definition of limit. Many things
that are not proved in this course, such the limit rules in Proposition 2.1, will be proven in MATH
2033. In this section, we first explain the graphical meaning of the rigorous definition of limits,
and will give a couple of simple examples.

i Will the content in this section be in the test of L1? Yes, but not too much. Math majors will
need to understand this section well, sooner or later.

The limit of a function is defined as the following complicated sentence:

Definition 2.7 — Rigorous definition of limits. We define lim f (x) = L, where a, L 2 R, to be:
x!a

“8" > 0, 9 > 0 such that if 0 < |x a| < , then |f (x) L| < ".”

2.3.1 Graphical meaning

If it does not make any sense for now, that’s normal. Even the best mathematicians in the world
found it hard to understand when they were students learning it the first time. The goal is of this
section is to help you understand it. It would be easier if you think of the above definition as:
“f (x) can be as close to L as you wish, provided that x is sufficiently close to a.”
We will later explain that this verbal description is in fact the same as Definition 2.7. Meanwhile,
let’s make sense of it using graphs:

It is very helpful to think of “(x, f (x)) in the blue strip” means “x is close to a”, and “(x, f (x))
in the red strip” means “f (x) is close to L”. In the above diagrams, we see that:
“no matter how narrow the red strip we draw around L, we can always draw a
small blue strip around a, such that if (x, f (x)) is in the blue strip, then (x, f (x)) is in
the red strip.”
 70 Limit of Functions

Let’s try to translate the above verbal sentence into mathematical language:
verbal mathematical
no matter how narrow the red strip 8" > 0 (which is the width of the red strip)
we can draw a small blue strip 9 > 0 (which is the width of the blue strip)
if (x, f (x)) is in the blue strip if 0 < |x a| <
then (x, f (x)) is in the red strip then |f (x) L| < "

tE > 0 雪 S 70 ,
suchthd 0 < 1x a
-
|<δ→| f (x) -

L| Kε
,

i 0 < |x a| < is just equivalent to |x a| < and x 6= a. Note that the limit of a function
is irrelevant to the value f (a), so we exclude the consideration of x = a.

2.3.2 Examples with an explicitly given "

Now let’s try to get a sense of the rigorous definition of limits through some concrete examples.
Consider the function:
f (x) = x2 1.
We all know that lim f (x) = 3, and now let’s see what it exactly means. Given that the red strip
x!2
has width 0.1, or in other words, take⽽ " = 0.1, we try to find a narrow blue strip with a certain
width , such that
δ
if 0 < |x 2| <Δ, then |f (x) 3| < 0.1 L
1
→
3 =

a 2 =

To find such a , we usually first leave it blank, and adjust its value later.
If 0 < |x 2| < ??, then

|f (x) 3| = x2 1 3
2
= x 4 = |x 2| |x + 2|
< ?? |x + 2| .
2.3 Toward Understanding Rigorous Definition of Limits 71

Note that 0 < |x 2| < ?? implies ?? < x 2 < ??, and so ?? + 4 < x + 2 < ?? + 4. If we
choose ?? so that ?? < 4 , then 0 < x + 2 < 8 and so |x + 2| < 8. This will give

|f (x) 3| < ?? ⇥ 8.

Hence, if we choose ?? < 1

80 as well, we then have

1
|f (x) 3| < ⇥ 8 = 0.1
80

To conclude, if we can make sure both ?? < 4 and ?? < 80 ,

1
or simply just ?? < 1
80 , then this
number can be chosen to be our . For instance, we choose = 1
100 . Let’s tidy up the above
“rough work”:
Given " = 0.1, we let = 1
100 such that if 0 < |x 2| < = 100 ,
1
then

|f (x) 3| = x2 1 3
2
= x 4 = |x 2| |x + 2|
1
< |x + 2| .
100

Note that 0 < |x 2| < 1

100 < 4 implies 4<x 2 < 4, and so 0 < 4 < x + 2 < 8. This shows
|x + 2| < 8, and it gives

1 1
|f (x) 3| < |x + 2| < ⇥ 8 = 0.08 < 0.1
100 100

i Attention: In order to find such a , do NOT start by solving the inequality |f (x) 3| < 0.1
to show that 0 < |x 2| < certain number, and then taking that number to be . In so
doing, you are attempting to prove the converse “if |f (x) 3| < 0.1, then 0 < |x 2| < ”,
which is not logically the same.

Now if we take an even smaller ", say " = 0.01, then how shall we choose a suitable ?
As before, if 0 < |x 2| < 4, then |x + 2| < 8 and so

|f (x) 3| = |x 2| |x + 2| < 8 |x 2| .

If we further have 8 |x 2| < 0.01, then it would imply |f (x) 3| < 0.01 as desired.
Now we are clear about the choice of a suitable , we need both of these hold:
0.01
0 < |x 2| < 4 and |x 2| < .
8

We pick = 0.01
10 = 0.001, which is both less than 4 and 8 . Now if 0 < |x
0.01
2| < = 0.001,
then 0 < |x 2| < 4 and |x 2| < 8 , and it implies |x + 2| < 8, and so
0.01

0.01
|f (x) 3| = |x 2| |x + 2| < 8 |x 2| < 8 ⇥ = 0.01
8
To summarize, we have proved that given " = 0.01, we can let = 0.001 so that

0 < |x 2| < = 0.001 =) |f (x) 3| < 0.01 = ".

⌅ Exercise 2.29 Let f (x) = x2 1 again. Now given " = 0.005, find a suitable > 0 such that

0 < |x 2| < =) |f (x) 3| < 0.005 = ".

flx)
≥
=
X -

悠 flx)
,
= 0

E
⼀
=
0 01
.
8 i ks
suchthart
号
if oit -
Gx
<δ→
<
-

1 -

→ 2 -

8 < x 11 <δ+ <

…-
2
4

∴
fx)
|= -
1
x ≡1 |=| x +1 ① lfl 1
<δ 1 x +1 |=δ [ x+1 ) < 48 =
Q 04
-

5 < 0 01
- .

1 fx )
-
o< Qo 1

P )
(
- 1 3
嘴 x2 =

Pe
θ E> 0 .
⽇
8 = min ( 4 信) ,

suchthol O < 1
x

- 21 <
-

δ< x- 22

→ 4 -

8 < xt 2 < 8t 4288 ]

( x≥ - 1

- 3 |=|xt 21 |λ - 2 |<δ 1 xt 2 | <88

≤ 8 。
哥 =ε
if 0 < lx 21
-
<δ= min (4 , λ ) then,
|(-
x
-

31 <ε

∴ 1洲
x2
[x ≡ 1 ) =
3 .
 72 Limit of Functions

2.3.3 Examples of rigorous proofs of limit

The above examples and exercise take " to be an explicit positive number, and we demonstrate
how to find a suitable given that choice of ". Now let’s regard " to be an arbitrary positive
number, which is can be as small as we wish, and we will demonstrate how do we find a suitable
.

⌅ Example 2.18 Let f (x) = x2 1. Prove, using the rigorous definition of limit, that

lim f (x) = 3.
x!2

⌅ Solution Rough work: Let’s do some rough work first. Given any " > 0, we need to find a
number ?? such that if 0 < |x 2| < ??, then |f (x) 3| < ".
Recall that |f (x) 3| = |x 2| |x + 2| < ?? · |x + 2|. If we can bound |x + 2| by a constant
C, then |f (x) 3| < ?? · C. In such case, in order for ?? · C < ", we just need ?? < C" . Then
how to bound |x + 2| by a constant? We have discussed that before, namely by choosing ?? < 4,
then |x + 2| < 8. Overall, we need
"
?? < 4 and ?? < .
8
One such as choice is ?? = min{3, 10
"
}, then we have both

?? = min{3, "/10}  3 < 4, and

" "
?? = min{3, "/10}  < .
10 8
Formal proof : Now let’s tidy up the above rough work and turn it into a formal proof.
8" > 0, we let = min{3, 10 "
} such that if 0 < |x 2| < , we then have |x 2| < 3 < 4,
and so
4 < x 2 < 4 =) 0 < x + 2 < 8 =) |x + 2| < 8.
Therefore, it implies

|f (x) 3| = x2 1 3 = |x 2| |x + 2|
"
< 8 |x 2| < 8  8 ·
10
< ".

To conclude, we have proved that lim f (x) = 3.

x!2

⌅ Example 2.19 Let f (x) = x3 1. Prove, using the rigorous definition of limits, that
lim f (x) = 0.
x!1

⌅ Solution Now we skip the “rough work” and present the formal proof. In practice, you
should first do the “rough work” to get a hint of how to choose a suitable .
8" > 0, we let = min{1, 8" } such that if 0 < |x 1| < , then we first have

0 < |x 1| <  1 =) 1<x 1 < 1 =) 0 < x < 2.

2.3 Toward Understanding Rigorous Definition of Limits 73

Then, we have

|f (x) 0| = x3 1 = |x 1| x2 + x + 1
< x2 + x + 1
= ( x2 + |x| + 1) (triangle inequality)
2
< (2 + 2 + 1) (since 0 < x < 2)
"
= 7  7 · < ".
8
To conclude, we have lim f (x) = 0.
x!1

⌅ Exercise 2.30 Let f (x) = x2 4.

(a) Find a suitable > 0 such that if 0 < |x 1| < , then |f (x) + 3| < 0.1
(b) Find a suitable > 0 such that if 0 < |x 1| < , then |f (x) + 3| < 0.05
(c) Prove, using the rigorous definition of limits, that lim f (x) = 3.
x!1

⌅ Exercise 2.31 Let f (x) = x3 1.

(a) Find a suitable > 0 such that if 0 < |x 2| < , then |f (x) 7| < 0.1
(b) Find a suitable > 0 such that if 0 < |x 2| < , then |f (x) 7| < 0.05
(c) Prove, using the rigorous definition of limits, that lim f (x) = 7.
x!2

Let’s see one more trickier example:

⌅ Example 2.20 Prove, using the rigorous definition of limit, that

✓ ◆
1
lim x+ = 2.
x!1 x

⌅ Solution Let’s do some rough work first:

✓ ◆ 2
1 x2 2x + 1 |x 1|
x+ 2 = = .
x x |x|

As x ! 1, we are considering 0 < |x 1| < , so the numerator is a good term because it can
2
be as small as we wish by adjusting the . Now that we want to bound |x|x|1| from above, we
should bound the denominator |x| from below (Recall that the larger the denominator, the
smaller the fraction). If 0 < |x 1| < 12 , then 12 < x 1 < 12 , and so 12 < x < 32 . This give
|x| = x > 12 , and so |x|
1 1
< 1/2 = 2. That would make
✓ ◆ 2
1 |x 1|
x+ 2 = < 2 2.
x |x|
p
To make 2 2
< ", we simply require < "/2. To summarize, we need both

1 p
< and < "/2.
2
p"
Now we get to the formal proof: 8" > 0, we take = 1
2 min{ 12 , 2} such that 0 < |x 1| <
 74 Limit of Functions

, then we first have

⇢ r
1 " 1 1 1 1 3
0 < |x 1| < < min ,  =) <x 1< =) <x< .
2 2 2 2 2 2 2

With this, it implies that

✓ ◆ 2
1 x2 2x + 1 |x 1|
x+ 2 = =
x x |x|
2
< 1 (since |x 1| < and |x| = x > 1/2)
2
2
p 2
=2 <2· "/2 = ".
✓ ◆
1
To conclude, we have proved that lim x+ = 2.
x!1 x

r
1
⌅ Exercise 2.32 Let f (x) = . x+
x p
(a) Find a suitable > 0 such that if 0 < |x 1| < , then f (x) 2 < 0.1
p
(b) Find a suitable > 0 such that if 0 < |x 1| < , then f (x) 2 < 0.05
p
(c) Prove, using the rigorous definition of limits, that lim f (x) = 2.
x!1

It is helpful to think about the phrase “9 > 0 such that if 0 < |x a| < , then something
holds” to be:

“something holds when x is sufficiently close to a and x 6= a.”

Then, we can rephrase the rigorous definition of limits as follows, by lim f (x) = L it means
x!a

“8" > 0, we have |f (x) L| < " when x is sufficiently close to a and x 6= a.”

In other words, it means f can be as close to L as we wish when x is sufficiently close to a and
x 6= a.

2.3.4 Limits and inequalities

Given two functions f and g, we hope to discuss the relation between the following
(A) f (x) > g(x) for all x 2 R
(B) f (x) g(x) for all x 2 R
(C) f (x) > g(x) when x is sufficiently close to a and x 6= a
(D) f (x) g(x) when x is sufficiently close to a and x 6= a
(E) lim f (x) > lim g(x)
x!a x!a
(F) lim f (x) lim g(x)
x!a x!a
We would like to ask, for example, whether (A) =) (E), or (A) =) (F), etc., and whether the
converse (E) =) (A), or (F) =) (A), etc. holds. Some of them are common misconceptions
even among math majors. However, if one has a good conceptual understanding of the rigorous
definition of limit, it will help you to recall which implications are correct, and which are wrong.
Let’s first consider two correct results.
Proposition 2.7 Suppose f and g are two functions such that both lim f (x) and lim g(x) exist.
x!a x!a
Then, the following holds:
(a) If lim f (x) > lim g(x), then f (x) > g(x) when x is sufficiently close to a and x 6= a.
x!a x!a
(b) If g(x) f (x) when x is sufficiently close to a and x 6= a, then lim g(x) lim f (x)
x!a x!a
Exercie 2 32.

(a 3 θε= 0 .
1 雪 δ < 0 .
1
,

such that Oala -

( la 多
-

与 < x -

1 <δ

. { 10
< x <δ

fx 分部了⼀下不
11 = x

州
“

*
(坏 )》
<01
:
州
“
∴
Ifo <1 x- 1 |<δ< 0 . 1 ,
then | f [ x ) ☆|<
-

0 .
1

( Similor
3 froom (a ,

H FE >
0 ,雪 位
min δ

such that O < 1 λ - 11 <

⼀系是 λ -

1 <δ

ax
当 1 δ -
-

. 8 + -

Fomld , |f )

-p 1

xllx
=|

π← 可 ∵ 安可 x

π<ε
+
6

If 0 < 1 x -

1 |<δ= min ( , )位

,
then 1 f[ x ) -
☆ I <ε

∴ 在 fla ) = 可
Example :Prove that悠比 +÷) = 2

θε> 0

,
雪 δ= mim 位 ] ,

if o < lx -

1 lag

=☆| x

- 19 a δ
主

⼀定 ≤λ - 1 往

⼆当会 λ≤ 童

到 =| ^|
x
→| ( x +÷)
-

=警
= 28 < E
2.3 Toward Understanding Rigorous Definition of Limits 75

To explain part (a), it’s best to look at some diagrams. For simplicity we denote

lim f (x) = L, and lim g(x) = l.

x!a x!a

Suppose L > l, then we have the following diagrams:

From the last diagram, we see that if x is sufficiently close to a so that x is still within the
dark blue strip, then f (x) > g(x). This proves (a).
To prove (b), we are given that g(x) f (x) when x is sufficiently close to a and x 6= a, we
need to show lim g(x) lim f (x). We prove by contradiction. Assume on the contrary that
x!a x!a

lim g(x) < lim f (x).

x!a x!a

Then from (a), we have f (x) > g(x) when x is sufficiently close to a and x 6= a. However, it clearly
contradicts to the given condition of part (b). Therefore, our assumption lim g(x) < lim f (x) is
x!a x!a
false, and it shows we must have

lim g(x) lim f (x).

x!a x!a

This proves (b).

Let’s discuss some uses of the above proposition. Recall that
sin x
lim = 1.
x!0 x
In particular, it implies
sin x
0.99 < lim < 1.01.
x!0x
By (a) of Proposition 2.7 (just taking one of the functions to be constant), we deduce that
sin x
0.99 < < 1.01 when x is sufficiently close to 0, and x 6= 0.
x
 76 Limit of Functions

In particular,

0.99x < sin x < 1.01x when x is sufficiently close to 0, and x > 0,

and the reverse inequalities hold when x is sufficiently close to 0 and x < 0. No wonder some
non-math majors (especially engineering majors) could treat “sin x = x” when x is very small.
Proposition 2.7 can also be applied when x ! 1, but then “x is sufficiently close to a” will be
replaced by “x is sufficiently large”. Consider an odd degree polynomial say p(x) = x11 5x9 100.
With the negative terms 5x9 and 100, it is not true that p(x) > 0 for all x > 0. However, one
can show p(x) > 0 when x is sufficiently large using Proposition 2.7.
Consider the limit:
✓ ◆
x11 5x9 100 5 100
lim = lim 1 = 1.
x!1 x11 x!1 x2 x11

Therefore, we have
1 x11 5x9 100
< lim < 2.
2 x!1 x11
Then, part (a) of Proposition 2.7 shows when x is sufficiently large (say when x C when C > 0
is a very big number), we have

1 x11 5x9 100 1 11

< < 2 =) x < x11 5x9 100 < 2x11 .
2 x11 2
Since x11 > 0 when x C > 0, we conclude that x11 5x9 100 > 12 C 11 > 0 when x C.
A similar argument could be used to show x11 5x9 100 < 0 when x is sufficiently negative.
Note that by the same calculation, we still have

x11 5x9 100 1 x11 5x9 100

lim = 1 =) < lim < 3.
x! 1 x11 3 x! 1 x11
By part (a) of Proposition 2.7, we deduce that there exists B < 0 such that if x < B, then

1 x11 5x9 100

< < 3.
3 x11
However, recall that when x < B, the denominator x11 is also negative. When multiplying it
over, one should reverse the inequality sign:
1 11
x > x11 5x9 100 > 3x11 .
3
Since 13 x11 < 0 when x < B < 0, we conclude that x11 5x9 100 < 0 when x < B.

⌅ Exercise 2.33 Show that there exists a constant B < 0 such that x1014 + 1013x1013
1003x 1012
> 0 when x < B.

⌅ Exercise 2.34 Show that for any odd degree polynomial p(x), there exist constants B < 0 <
C such that p(x) has different signs on the intervals ( 1, B) and (C, 1). Hence, deduce
that any odd degree polynomial must have at least one real root.

2.3.5 Some misconceptions about limits

Part (b) of Proposition 2.7 shows that assuming both lim f (x) and lim g(x) exist, then if
x!a x!a
f (x) g(x) when x is sufficiently close to a and x 6= a, one has lim f (x) lim g(x). However,
x!a x!a
if one of the limits does not exist, the proposition may not hold. For instance,

sin x  1 8x 2 R.
2.3 Toward Understanding Rigorous Definition of Limits 77

However, it is not correct to say that

lim sin x  lim 1 = 1 (WRONG!)

x!1 x!1

In fact, the above statement is meaningless since lim sin x does not exist.
x!1
Moreover, even if we have f (x) > g(x) when x is sufficiently close to a and x 6= a, and that
both lim f (x) and lim g(x) exist, the best one can argue is that
x!a x!a

lim f (x) lim g(x).

x!a x!a

It can proved by part (b) of Proposition 2.7:

f (x) > g(x) when x is sufficiently close to a and x 6= a

=) f (x) g(x) when x is sufficiently close to a and x 6= a
=) lim f (x) lim g(x).
x!a x!a

The counterexample below shows that even if f (x) > g(x) when x is sufficiently close to a
and x 6= a, it could still happen that

lim f (x) = lim g(x).

x!a x!a

Consider ( (
x2 if x 6= 0 x2 if x 6= 0
f (x) = g(x) = .
1 if x = 0 1 if x = 0
Clearly, no matter whether x 6= 0 or x = 0, we have f (x) > g(x). However, we have both

lim f (x) = lim x2 = 0

x!0 x!0
lim g(x) = lim ( x2 ) = 0
x!0 x!0

⌅ Exercise 2.35 Find a counterexample to the following statement:

“If both lim f (x) and lim g(x) exist, and
x!a x!a

lim f (x) lim g(x),

x!a x!a

then f (x) g(x) when x is sufficiently close to a and x 6= a.”