Chapter 2

Limits and continuity of functions

of one variable

2.1 Limits
To determine the behavior of a function f as x approaches a finite value c, we use the
concept of limit. We say that the limit of f is L, and write limx→c f (x) = L, if the values
of f approaches L when x gets closer to c.
Definition 2.1.1. (Limit when x approach a finite value c). We say that limx→c f (x) = L
if for any small positive ϵ, there is a positive δ such that

|f (x) − L| < ϵ

whenever 0 < |x − c| < δ.

We can split the above definition in two parts, using one–sided limits.
Definition 2.1.2.

1. We say that L is the limit of f as x approaches c from the right, limx→c+ f (x) = L,
if for any small positive ϵ, there is a positive δ such that

|f (x) − L| < ϵ

whenever 0 < x − c < δ.

2. We say that L is the limit of f as x approaches c from the left, limx→c− f (x) = L, if
for any small positive ϵ, there is a positive δ such that

|f (x) − L| < ϵ

whenever 0 < c − x < δ.

Teorema 2.1.3. limx→c f (x) = L if and only if

lim f (x) = L and lim f (x) = L.

x→c+ x→c−

1
 We can also wonder about the behavior of the function f when x approaches +∞ or
−∞.
Definition 2.1.4. (Limits when x approaches ±∞)

1. limx→+∞ f (x) = L if for any small positive ϵ, there is a positive value of x, call it x1 ,
such that
|f (x) − L| < ϵ
whenever x > x1 .

2. limx→−∞ f (x) = L if for any small positive ϵ, there is a negative value of x, call it x1 ,
such that
|f (x) − L| < ϵ
whenever x < x1 .

If the absolute values of a function become arbitrarily large as x approaches either a

finite value c or ±∞, then the function has no finite limit L but will approach −∞ or +∞.
It is possible to give the formal definitions. For example, we will say that limx→c f (x) = +∞
if for any large positive number M , there is a positive δ such that

f (x) > M

whenever 0 < |x − c| < δ. Please, complete the remaining cases.

Note 2.1.5. Note that it could be c ∈ D(f ), so f (c) is well defined, but limx→c f (x) does
not exits or limx→c f (x) ̸= f (c). Consider for instance the function f that is equal to 1 for
x ̸= 0, but f (0) = 0. Then clearly the limit of f at 0 is 1 ̸= f (0).
Example 2.1.6. Consider the following limits.
1. lim x2 − 2x + 7 = 31.
x→6

2. lim x2 − 2x + 7 = ∞, because the leading term in the polynomial gets arbitrarily

x→±∞
large.

3. lim x3 − x2 = ∞, because the leading term in the polynomial gets arbitrarily large
x→+∞
for large values of x, but lim x3 − x2 = −∞ because the leading term in the poly-
x→−∞
nomial gets arbitrarily large in absolute value, and negative.
1
4. lim = 0, since for x arbitrarily large in absolute value, 1/x is arbitrarily small.
x→±∞ x

1
5. lim does not exists. Actually, the one–sided limits are:
x→0 x

1
lim = +∞.
x→0+ x
1
lim = −∞.
x→0− x

2
 The right limit is infinity because 1/x becomes arbitrarily large when x is small and
positive. The left limit is minus infinity because 1/x becomes arbitrarily large in
absolute value and negative, when x is small and negative.

6. lim x sen x does not exist. As x approaches infinity, sen x oscillates between 1 and
x→+∞
−1. This means that x sen x changes sign infinitely often when x approaches infinity,
whilst taking arbitrarily large absolute values. The graph is shown below.

100

50

−50

−100
0 20 40 60 80 100

 2
 x , if x ≤ 0;
7. Consider the function f (x) = −x2 , if 0 < x ≤ 1; limx→0 f (x) = f (0) = 0, but
x, if x > 1.

limx→1 f (x) does not exist since the one–sided limits are different.

lim f (x) = lim x = 1,

x→1+ x→1+
lim f (x) = lim −x2 = −1.
x→1− x→1−

|x|
8. lim does not exist, because the one–sided limits are different.
x→0 x
|x| x
lim = lim = 1,
x→0+ x x→0+ x
|x| −x
lim = lim = −1 (when x is negative, |x| = −x).
x→0− x x→0− x

In the following, lim f (x) refer to the limit as x approaches +∞, −∞ or a real number
c, but we never mix different type of limits.

2.1.1 Properties of limits

f and g are given functions and we suppose that all the limits below exist; λ ∈ R denotes
an arbitrary scalar.
1. Product by a scalar: lim λf (x) = λ lim f (x).

2. Sum: lim(f (x) + g(x)) = lim f (x) + lim g(x).

3
 3. Product: lim f (x)g(x) = (lim f (x))(lim g(x)).
f (x) lim f (x)
4. Quotient: If lim g(x) ̸= 0, then lim = .
g(x) lim g(x)
Teorema 2.1.7 (Squeeze Theorem). Assume that the functions f , g and h are defined
around the point c, except, maybe, for the point c itself, and satisfy the inequalities

g(x) ≤ f (x) ≤ h(x).

Let limx→c g(x) = limx→c h(x) = L. Then

lim f (x) = L.
x→c
 
1
Example 2.1.8. Show that lim x sen = 0.
x→0 x

Solution: We use the theorem above with g(x) = −|x| and h(x) = |x|. Notice that for
every x ̸= 0, −1 ≤ sen (1/x) ≤ 1 thus, when x > 0

−x ≤ x sen (1/x) ≤ x,

and when x < 0

x ≤ x sen (1/x) ≤ −x.
These inequalities mean that −|x| ≤ x sen (1/x) ≤ |x|. Since

lim −|x| = lim |x| = 0,

x→0 x→0

we can use the theorem above to conclude that limx→0 x sen x1 = 0.

f (x)
2.1.2 Techniques for evaluating lim
g(x)
1. Use the property of the quotient of limits, if possible.

2. If lim f (x) = 0 and lim g(x) = 0, try the following:

f (x)
(a) Factor f (x) and g(x) and reduce g(x) to lowest terms.
(b) If f (x) or g(x) involves a square root, then multiply both f (x) and g(x) by the
conjugate of the square root.
Example 2.1.9.

x2 − 9 (x − 3)(x + 3)
lim = lim = lim (x − 3) = 0.
x→3 x + 3 x→3 x+3 x→3

√ √  √ 
1− 1+x 1− 1+x 1+ 1+x −x −1 1
lim = lim √ = lim √ = lim √ =− .
x→0 x x→0 x 1+ 1+x x→0 x(1 + 1 + x) x→0 1 + 1 + x 2

4
 3. If f (x) ̸= 0 and lim g(x) = 0, then either lim fg(x)
(x)
does not exist or lim fg(x)
(x)
= +∞ or
−∞.
4. If x approaches +∞ or −∞, divide the numerator and denominator by the highest
power of x in any term of the denominator.
Example 2.1.10.
1 2
x3 − 2x x − x3 0−0
lim 4
= lim 2 = = 0.
x→∞ −x + 2 x→∞ −1 + 4 −1 + 0
x

2.1.3 Exponential limits

Let the limit
lim [f (x)]g(x)
x→c
be an indetermination. This happens if
ˆ limx→c f (x) = 1 and limx→c g(x) = ∞ (1∞ ).
ˆ limx→c f (x) = 0 and limx→c g(x) = 0 (00 ).
ˆ limx→c f (x) = ∞ and limx→c g(x) = 0 (∞0 ).
Noting that
lim [f (x)]g(x) = lim eg(x) ln f (x) = elimx→c g(x) ln f (x) ,
x→c x→c
all cases are reduced to the indetermination 0 · ∞, since we have to compute the limit
lim g(x) ln f (x).
x→c

In the first indetermination, 1∞ , it often helps to use the identity

lim g(x) ln f (x) = lim g(x)(f (x) − 1).
x→c x→c

since when x is close to 0, ln(1 + x) ≈ x, or, ln x ≈ x − 1

x 1 1
Example 2.1.11. limx→∞ 1 + x1 = limx→∞ ex ln (1+ x ) = ex x = e.
 x
Example 2.1.12. Let a, b > 0. Calculate limx→∞ 1+ax 2+bx .
If a > b, then the basis function tends to a/b > 1, thus the limit is ∞. If a < b, then
the basis function tends to a/b < 1, thus the limit is 0. When a = b
1 + ax x
 
1+ax −x
lim = elimx→∞ x( 2+ax −1) = elimx→∞ 2+ax = e−1/a .
x→∞ 2 + ax

2.1.4 Remarkable limit

Recall that
sen x
lim
= 1.
x→0 x
Example 2.1.13. Evaluate the following limits:
tan x sen x 1 sen x 1
1. lim = lim = lim lim = 1 · 1 = 1.
x→0 x x→0 x cos x x→0 x x→0 cos x

5
 sen 3x {z=3x} sen z sen z
2. lim = lim z = 3 lim = 3.
x→0 x z→0
3
z→0 z

2.2 Asymptotes
An asymptote is a line that the graph of a function approaches more and more closely until
the distance between the curve and the line almost vanishes.
Definition 2.2.1. Let f be a function

1. The line x = c is a vertical asymptote of f if limx→c+ |f (x)| = ∞ or limx→c− |f (x)| =

∞.

2. The line y = b is a horizontal asymptote of f if limx→+∞ f (x) = b or limx→−∞ f (x) =

b.

3. The line y = ax + b is an oblique asymptote of f if

f (x)
(a) lim = a and lim (f (x) − ax) = b, or
x→+∞x x→+∞
f (x)
(b) lim = a and lim (f (x) − ax) = b.
x→−∞ x x→−∞

Notice that a horizontal asymptote is a particular case of oblique asymptote with a = 0.

(1 + x)4
Example 2.2.2. Determine the asymptotes of f (x) = .
(1 − x)4

Solution: Since the denominator vanishes at x = 1, the domain of f is R − {1}. Let

us check that x = 1 is a vertical asymptote of f :

(1 + x)4
lim = +∞
x→1± (1 − x)4
On the other hand
(1 + x)4 (1/x + 1)4
lim = lim =1
x→+∞ (1 − x)4 x→+∞ (1/x − 1)4

hence y = 1 is a horizontal asymptote at +∞. In the same way, y = 1 is a horizontal

asymptote at −∞. There is no other oblique asymptotes.
3x3 − 2
Example 2.2.3. Determine the asymptotes of f (x) = .
x2
Solution: The domain of f is R − {0}. Let us check that x = 0 is a vertical asymptote
of f .
3x3 − 2 2 2
lim 2
= lim (3x − 2 ) = lim 3x − lim 2 = −∞.
x→0 ± x x→0 ± x x→0 ± x→0 x
±

Thus, x = 0 is a vertical asymptote of f . On the other hand

3x3 − 2 2
lim 2
= lim (3x − 2 ) = ±∞
x→±∞ x x→±∞ x

6
thus, there is no horizontal asymptote. Let us study now oblique asymptotes:

3x3 − 2
 
f (x) 2
a = lim = lim = lim 3 − 3 = 3,
x→±∞ x x→±∞ x3 x→±∞ x
 3   
3x − 2 2
b = lim (f (x) − 3x) = lim − 3x = lim − 2 = 0.
x→±∞ x→±∞ x2 x→±∞ x

We conclude that y = 3x is an oblique asymptote both at +∞ and −∞.

2.3 Continuity
The easiest limits to evaluate are those involving continuous functions. Intuitively, a func-
tion is continuous if one can draw its graph without lifting the pencil from the paper.

Definition 2.3.1. A function f : R −→ R is continuous at c if c ∈ D(f ) and

lim f (x) = f (c).

x→c

Hence, f is discontinuous at c if either f (c) is undefined or limx→c f (x) does not exist
or limx→c f (x) ̸= f (c). Moreover, we can define one-sided continuity of f at c,

Definition 2.3.2. A function f : R −→ R is right continuous at c, if c ∈ D(f ) and

lim f (x) = f (c).

x→c+

f is left continuous at c, if c ∈ D(f ) and

lim f (x) = f (c).

x→c−

Obviously, a function f is continuous at c when is both, right and left continuous at c.

2.3.1 Properties of continuous functions

Suppose that the functions f and g are both continuous at c. Then the following functions
are also continuous at c.

1. Sum. f + g.

2. Product by a scalar. λf , λ ∈ R.

3. Product. f g.

4. Quotient. f /g, whenever g(c) ̸= 0.

7
2.3.2 Limit y continuity of the composite function
Teorema 2.3.3. Let f, g be functions from R to R and let c ∈ R. If g is continuous at L
and limx→c f (x) = L, then
lim g(f (x)) = g(lim f (x)) = g(L).
x→c x→c

If the function f is continuous at c, then, calling L = f (c) the result above becomes:
Corollary 2.3.4. Let f be a continuous function at c and g continuous on f (c). Then ,
the composite function g ◦ f is also continuous at c.
Example 2.3.5. Compute the following limits:
ln (1 + x)  
ˆ lim = lim ln (1 + x)1/x = ln lim (1 + x)1/x = ln e = 1.
x→0 x x→0 x→0

Note that the function ln (·) is continuous at e, then we can apply 3.3.4.
ax − 1
 
z z
ˆ lim = lim ln (1+z) = ln a lim = ln a.
x→0 x z→0 z→0 ln (1 + z)
ln a
We have used the substitution z = ax − 1, so that x = ln (1 + z)/ ln a, and we have
used the value of the limit computed before.

2.3.3 Continuity of elementary functions

A function is called elementary if it can be obtained by means of a finite number of arith-
metic operations and superpositions involving basic elementary functions. The functions
y = C = constant, y = xa , y = ax , y = ln x, y = ex , y = sen x, y = cos x, y = tan x,
y = arctan x are examples of elementary functions. Elementary functions are continuous in
their domain.
Example 2.3.6.
√
1. The function f (x) = 4 − x2 is the composition of the functions y = 4 − x2 and
f (y) = y 1/2 , which are elementary, thus f is continuous in its domain, that is, in
D = [−2, +2].
1
2. The function g(x) = √4−x 2
is the composition of function f above and function
g(y) = 1/y, thus it is elementary and continuous in its domain, D(g) = (−2, +2).

2.3.4 Continuity of the inverse function

A one-to-one function (also named bijectve) does not have to be continuous. For example,
the following function
 1, si x = 0;
f (x) = x, si 0 < x < 1; is bijective considering that its domain and image are the
0, si x = 1.

interval [0, 1].

8
 It can be shown that neither f (x) is continuous nor f −1 (x), which coincidentally happens
to be the same f .
This will not be the case should the function f (x) be continuous, as the following theorem
proves:

Teorema 2.3.7. Let f : I −→ J be continuous and bijective Then:

a) f is strictly increasing (or decreasing), and

b) The inverse f −1 is a continuous function as well.

Observation: Obviously, f −1 is also strictly increasing (o decreasing), depending on

f having the same nature as well.
 2 
x +x−2 π
Example 2.3.8. Prove that lim arctan = .
x→1 3x2 − 3x 4

Solution: The function arctan = tan−1 is continuous from what we just have seen
above. Then applying theorem 3.3.3:
 2
x2 + x − 2
  
x +x−2
lim arctan = arctan lim
x→1 3x2 − 3x x→1 3x2 − 3x
 
(x − 1)(x + 2)
= arctan lim
x→1 3x(x − 1)
 
x+2
= arctan lim
x→1 3x

= arctan 1
π
= .
4

2.3.5 Continuity theorems

Continuous functions have interesting properties. We shall say that a function is continuous
on the closed interval [a, b] if it is continuous at every point x ∈ (a, b), is right-continuous
at a and left-continuous at b.

Teorema 2.3.9 (Bolzano’s Theorem). If f is continuous in [a, b] and f (a) · f (b) < 0, then
there exists some c ∈ (a, b) such that f (c) = 0.

Example 2.3.10. Show that the equation x3 + x − 1 = 0 admits a solution, and find it
with an error less than 0.1.
Solution: With f (x) = x3 + x − 1 the problem is to show that there exists c such that
f (c) = 0. We want to apply Bolzano’s Theorem. First, f is continuous in R. Second, we
identify a suitable interval I = [a, b]. Notice that f (0) = −1 < 0 and f (1) = 1 > 0 thus,
there is a solution c ∈ (0, 1).
Now, to find an approximate value for c, we use a method of interval–halving as follows:
consider the interval [0.5, 1]; f (0.5) = 1/8 + 1/2 − 1 < 0 and f (1) > 0, thus c ∈ (0.5, 1).
Choose now the interval [0.5, 0.75]; f (0.5) < 0 and f (0.75) = 27/64 + 3/4 − 1 > 0 thus,

9
c ∈ (0.5, 0.75). Let now the interval [0.625, 0.75]; f (0.625) ≈ −0.13 and f (0.74) > 0 thus,
c ∈ (0.625, 0.75). The solution is approximately c = 0.6875 with a maximum error of 0.0625.
The previous theorem, known as Bolzano’s Theorem can be generalized for every inter-
mediate value between f (a) and f (b), since it is proved in the following theorem.

Teorema 2.3.11 (Intermediate Value Theorem). Let f be a continuous function on the

closed interval [a, b]. Then, for any intermediate real number k between f (a) and f (b),
there is at least a number xk ∈ [a, b] satisfying f (xk ) = k.

Notice: An intermediate value means any real number k with f (a) < k < f (b) or
f (b) < k < f (a).

Proof. Consider the function g(x) = f (x) − k. Then, g(a) < 0 < g(b) or g(b) < 0 < g(a).
Applying Bolzano’s Theorem to the function g, there is xk ∈ [a, b] such that g(xk ) = 0.
Similarly, there exists a xk ∈ [a, b] such that f (xk ) = k.

The following result is very useful when you are trying to find the image of a continuous
function.

Corollary 2.3.12. Let f be a continuous, non-constant function defined on any interval I

(not necessarily closed or bounded). Then, J = Im(f ) is also an interval.

Notice: J does not always satisfy the same properties of the interval I.

Example 1: f (x) = 1/x is continuous on the bounded interval I = (0, 1], but J =
Im(f ) = [1, ∞) is not bounded.

Example 2: f (x) = 1/x is continuous on the closed interval I = [1, ∞), but J = Im(f ) =
(0, 1] is not closed.

Nevertheless, if the interval I is compact, ie: it is closed and bounded, then J is also
compact.
This last result is called Weierstrass’ Theorem, and it is the most important of chapter
2.

Teorema 2.3.13 (Weierstrass’ Theorem). If f is continuous in [a, b] , then there exist

points c, d ∈ [a, b] such that
f (c) ≤ f (x) ≤ f (d)
for every x ∈ [a, b].

The theorem asserts that a continuous function attains over a closed interval a minimum
(m = f (c)) and a maximum value (M = f (d)). The point c is called a global minimum of
f on [a, b] and d is called a global maximum of f on [a, b].
Example 2.3.14. Show that the function f (x) = x2 + 1 attains over the closed interval
[−1, 2] a minimum and a maximum value.
Solution: The graph of f is shown below.

10
 10

0
−2 −1 0 1 2 3

We can see that f is continuous in [−1, 2], actually f is continuous in R, and f attains
the minimum value at x = 0, f (0) = 1, and the maximum value at x = 2, f (2) = 5.

Example 2.3.15. The assumptions in the Theorem of Weierstrass are essential.

ˆ The interval is not closed, or not bounded.

– Take I = (0, 1] and f (x) = 1/x; f is continuous in I, but it does not have global
maximum.
– Take I = [0, ∞) and f (x) = 1/(1 + x); f is continuous in I, but it does not have
global minimum, since limx→∞ f (x) = 0, but f (x) > 0 is strictly positive for
every x ∈ I.

x, if 0 ≤ x < 1;
ˆ The function is not continuous. Take I = [0, 1] and f (x) = ; f
0, if x = 1.
has a global minimum at x = 0, but there is no global maximum since limx→1 f (x) = 1
but f (x) < 1 for every x ∈ I.

11