Chapter 3: Limits and Continuity: Calculus
Chapter 3: Limits and Continuity: Calculus
Chapter 3: Limits and Continuity: Calculus
Calculus is the study of functions. Specially, it studies two important operations on functions,
called differentiation and integration. A key concept that underlies both operations is the concept of limit. In
Chapters 3, 4 and 5 we deal with calculus of functions of the form 𝑓: 𝐷 ⊆ ℝ → ℝ.. Such functions are
called real valued functions of real variables and part of calculus that deals with them is called single-variable
calculus.
Let a and L be real numbers and 𝑓 be a real valued function defined at each point of an open interval
containing a except possibly at a itself. We say that the limit of 𝑓(𝑥) as 𝑥 approaches 𝑎 is L, written
𝑙𝑖𝑚
𝑥 →𝑎 𝑓(𝑥) = L, if f(x) can be made arbitrarily close to L by taking all x sufficiently close to a but without letting
x = a.
The given formula defines 𝑓 for all real numbers x except x = 1. For 𝑥 ≠ 1, 𝑓(𝑥) = 𝑥 + 1. The
graph of 𝑓 is thus the line 𝑦 = 𝑥 + 1 with the point (1, 2) removed. It may be intuitively clear that we
can make the value of f as close to 2 as we want by choosing x close enough to 1 (but we cannot make
𝑥 = 1). Hence
𝑥2 − 1
lim =2
𝑥→1 𝑥 − 1
From the following tables, you can see the trend of the values of f as two sequences of values of x
approach 1.
2 x 1, if x 3
Example 2: Let f ( x)
6, if x 3
1
Find lim f ( x) , by intuition (by evaluating values of 𝑓 at several numbers near 3, but not at 3).
x3
Solution: Let’s investigate the behavior of f(x) near 3(but not at 3), by tabulating values of 𝑓 at
several numbers near 3 (some less than 3 and some greater than 3).
x f(x) x f(x)
⋮ ⋮ . .
. .
x approaches f(x)
. .
3 and x<3 approaches 5
x approaches 3 and x>3 f(x) approaches 5
Intuitively, it is clear that when x is close to 3 but x 3, then f(x) is close to 5. Moreover, it seems f(x) can
be made as close to 5 as we wish by taking all x’s sufficiently close to 3.
lim f ( x) = 5.
x3
The intuitive definition of limit is ambiguous because the word “close” in the definition is vague. The
statement “𝑓(𝑥) can be made arbitrarily close to 𝐿” means if any positive number 𝜀 is given (however
small) we can make |f(x) - L| <𝜀. The statement “f(x) can be made arbitrarily close to L by taking all x
sufficiently close to a but without letting x = a” means if any positive number 𝜀 is given, there exists some
positive number 𝛿 such that for all x satisfying |𝑥 − 𝑎| < 𝛿 and 𝑥 ≠ 𝑎, we have |𝑓(𝑥) − 𝐿| < 𝜀. This
paves way to the following formal definition of limit.
2
Definition 1 [The Formal Definition of Limit]:
We say
𝑙𝑖𝑚
𝑥 →𝑎 𝑓(𝑥) = L,
if for any positive number 𝜀 (however small), there is a positive number 𝛿 (that depends on 𝜀 in
general)such that for all 𝑥 in the domain of 𝑓 the following implication is true:
if 0 < |𝑥 − 𝑎| < 𝛿 , then | 𝑓(𝑥) − 𝐿| < 𝜀
Note (Intuitive vs Formal definition): An intuitive definition (informal definition) of a term is the literal meaning of the term. It is
ambiguous; different readers can interpret it differently. But the formal definition of a term is more precise. For instance,
congruence of triangles is defined informally as triangles having the same shape and same size and formally as triangles whose
corresponding sides and angles have the same measures.
𝑙𝑖𝑚
Remark 1: If 𝑥 →𝑎 𝑓(𝑥) = L, then we briefly say L is the limit of f at a. If there is no real number L
𝑙𝑖𝑚
satisfying 𝑥 →𝑎 𝑓(𝑥) = L, then we say f has no limit at a.
Note: Recall from logic that in order to prove a quantified proposition of the form
(∀𝜀) (∃𝛿) P (𝜀, 𝛿),
we must assume that 𝜀 is given and we must show that there exists 𝛿 for which P (𝜀, 𝛿) is true. For example, (∀𝜀) (∃𝛿) (𝜀 + 𝛿 =
0) is true but (∀𝜀) (∃𝛿) (𝜀𝛿 = 1) is false.
𝑙𝑖𝑚
Remark 3: To Show that 𝑥→𝑎 𝑓(𝑥) = L, you can proceed as follows.
Assuming that any positive number 𝜀 is given find a positive number 𝛿 (as a function of 𝜀 in general)
such that the implication
0 < |𝑥 − 𝑎| < 𝛿 ⇒ | f(x) - L| < ∈
3
is true.
This can be done in two steps:
Step 1(Scratch work/preliminary Analysis): Guess 𝛿 by some means.
Step 2(Formal Work): Verify that the guessed 𝛿 works.
Remark 4: There is no hard and fast rule to find 𝛿. The following observations may be helpful..
a) The implication “if 0 < 𝑧 < 𝛿, then 𝑧 < 𝛿” is true, but its converse is false. Thus if
|𝑓(𝑥) − 𝐿| = |𝑥 − 𝑎| we can choose 𝛿 to be 𝜀 itself.
b) If there exists a constant 𝑘 such that |𝑓(𝑥) − 𝐿| ≤ 𝑘 |𝑥 − 𝑎| ∀𝑥 ∈ 𝐷𝑜𝑚𝑓\{𝑎} , then any
𝜀
positive number less than or equal to 𝑘 can be chosen to be a value of δ (Why?). See Example 4.
c) If there exists a constant 𝑘 and a positive number 𝛿1 such that |𝑓(𝑥) − 𝐿| ≤ 𝑘 |𝑥 − 𝑎|∀𝑥 ∈
𝜀
𝐷𝑜𝑚𝑓\{𝑎} and ∀𝑥 ∈ (𝑎 − 𝛿1 , 𝑎 + 𝛿1 ), then the minimum of and 𝛿1 can be chosen to be a
𝑘
Remark 5: Let 𝛿, 𝜀, 𝛿1 and 𝜀1 be positive real numbers such that 𝛿 > 𝛿1 , 𝜀 < 𝜀1 and 0 < |𝑥 − 𝑎| <
𝛿 ⇒ |f(x) - L| < ∈. Then
a) 0 < |𝑥 − 𝑎| < 𝛿1 ⇒ | 𝑓(𝑥) − 𝐿| < 𝜀 and
b) 0 < |𝑥 − 𝑎| < 𝛿 ⇒ | 𝑓(𝑥) − 𝐿| < 𝜀1 .
𝑙𝑖𝑚
Example 3: Show that 𝑥→2 7𝑥 − 4 = 10.
Solution: We are given 𝑎 = 2, 𝑓(𝑥) = 7𝑥 − 4 and 𝐿 = 10.
Let 𝜀 > 0 be given. We must find a 𝛿 > 0 such that
0 <|𝑥 − 2| < 𝛿 ⇒ |(7𝑥 − 4) − 10| < 𝜀
Step 1 (Scratch Work): Guess𝛿.
First, we need to find a way of relating 0 <|𝑥 − 2| < 𝛿 and |(7𝑥 − 4) − 10| < 𝜀
We will use algebraic manipulation to get this relationship. Remember that the whole point of this
manipulation is to find a δ in terms of 𝜀 so that if |x − 2| < δ is true for some x then it forces |(7x − 4)
− 10| < 𝜀 to be true for that x. So we will start with the 𝜀 − term and manipulate it until we find the
delta term in there. So |(7𝑥 − 4) − 10| < 𝜀 ⇔ |7𝑥 − 4 − 10| < 𝜀 ⇔ |7𝑥 − 14| < 𝜀 ⇔
𝜀
|7(𝑥 − 2)| < 𝜀 ⇔ |7| · |𝑥 − 2| < 𝜀 ⇔ 7|𝑥 − 2| < 𝜀. Now 7|x − 2| < 𝜀 ⇔ |x − 2| < 7 .
𝜀
Now with the above string of equivalences, we can see that if we let δ = 7 , then we have found a δ that
𝜀
will fulfill the definition because if |x−2| < δ then |x−2| < 7 ⇔ 7|x−2| < 𝜀 ⇔ |7x−14| < 𝜀 ⇔
𝜀
|7𝑥 − 4 − 10| < 𝜀 ⇔ |(7𝑥 − 4) − 10| < 𝜀. So we have that for any 𝜀 given to us if we let δ =
7
and |x − 2| < δ then this will force |(7x − 4) − 10| < 𝜀 thus showing that the definition is fulfilled.
Now let’s write the formal proof.
4
𝜀 𝜀
Let 𝜀 ² > 0 be given. Let δ = . So if 0 < |x − 2| < δ then |x − 2| < ⇒ 7|x − 2| < 𝜀 ⇒ |7x − 14| <
7 7
𝑙𝑖𝑚
𝜀 ⇒ |7x − 4 − 10| < 𝜀 ⇒ |(7x − 4) − 10| < 𝜀. Therefore we have shown that 𝑥→2 7𝑥 − 4 = 10.
2 x 1, if x 3
𝑙𝑖𝑚
Example 4: Show that 𝑥→3 𝑓(𝑥) = 5, where f ( x)
6, if x 3
Solution: Let 𝜀 > 0 be given. We must find a 𝛿 > 0 such that
0 <|𝑥 − 3| < 𝛿 ⇒ |(2𝑥 − 1) − 5| < 𝜀
(Even if 𝑓 is defined by two formulae, we used only the first formula for 𝑓 because the inequality
0 <|𝑥 − 3| implies that 𝑥 ≠ 3.)
Step 1 (Scratch Work): Guess𝛿.
Note that |(2𝑥 − 1) − 4| = |(2𝑥 − 6| = 2|𝑥 − 6|
Thus, we must find a 𝛿 > 0 such that
0 <|𝑥 − 3| < 𝛿 ⇒ 2|𝑥 − 6| < 𝜀
𝜀
Remark (4b) suggests that we can choose 𝛿 = 2.
Step 2: Formal Proof
0 <|𝑥 − 3| < 𝛿 ⇒ |𝑥 − 3| < 𝛿 (Remark 4a)
𝜀
⇒ |𝑥 − 3| < 2 (Transitivity)
⇒ 2|𝑥 − 2| < 𝜀
⇒ ||(2𝑥 − 1) − 5|| < 𝜀 (Transitivity and *)
𝑙𝑖𝑚 2
Example 5: Show that 𝑥→2 𝑥 = 4.
Solution: We are given 𝑎 = 2, 𝑓(𝑥) = 𝑥 2 and 𝐿 = 4.
Let 𝜀 > 0 be given. We must find a 𝛿 > 0 such that
0 <|𝑥 − 2| < 𝛿 ⇒ |𝑥 2 − 4| < 𝜀
Unlike the previous examples 𝛿 cannot be immediately chosen. We do in two steps.
Step 1 (Scratch Work): Guess𝛿.
We start with the left inequality and do a little simplification and see if that suggests a choice for 𝛿.
In view of Remark 4c), we proceed as follows. We find 𝑘 and 𝛿1 such that |x2 - 4| ≤ k |x - 2| ∀𝑥 ∈
𝜀
(2 − 𝛿1 , 2 + 𝛿1 ) . Then the minimum of 𝛿1 and can be chosen as a value of 𝛿.
𝑘
5
0 <|𝑥 − 2| < 𝛿 ⇒ |𝑥 − 2| < 𝛿 (Remark 3a)
𝜀
⇒ |𝑥 − 2| < 5 (Transitivity)
⇒ 5|𝑥 − 2| < 𝜀
⇒ |𝑥 2 − 4| < 𝜀 (Transitivity and *)
Step-1: Guess .
x 2 x2
Note that | x - 2 | = ( x 2 ) = x2
x 2 x 2
1 1 1
( because x 0 x 2 2 1 1 )
x 2 2 x 2
This suggests we can choose to be any positive number . Question: If = 10, what can be
chosen for ? Any positive number ≤ 10? For example, if we choose =8, the interval (a- , a +
)\{2} becomes (-6, 10)\ {2}. Now if x=-5, then the inequality 0 < |x-2| < 8 is fulfilled but the
inequality| x - 2 | <10 is not fulfilled. So, the selected doesn’t work. The reason is that the interval
(-6, 10)\{2} is not with in domf. So we must select so that 2.
6
𝒍𝒊𝒎
Remark 7(Negation of the statement 𝒙 →𝒂 𝒇(𝒙) = L):
lim
x →a f(x) ≠ L ⇔ ( ∃𝜀 > 0) ∋(∀𝛿 > 0) (∃𝑥 > 0)∋[0 < |𝑥 − 𝑎| < 𝛿 & |𝑓(𝑥) − 𝐿| ≥ 𝜀.
This says whenever 𝑥 ∈ (2 − 𝛿, 2 + 𝛿), we have |(2𝑥 + 5) − 7| < 1. But this is not true for 2 < 𝑥 <
2 + 𝛿.
Exercises 1.1
1. Determine how small a positive number 𝛿 must be in order that
a) 0 < |𝑥 − 2| < 𝛿 ⇒ |𝑥 2 − 4| < 0.05
b) 0 < |𝑥 − 2| < 𝛿 ⇒ |𝑥 2 − 4| < 6
c) 0 < |𝑥 − 2| < 𝛿 ⇒ |𝑥 2 − 4| < 𝜀
𝑙𝑖𝑚
𝑥 2 +1
a) =−1
𝑥→0 2𝑥−1
𝑥 2 −1
b) lim𝑥→1 𝑥−1 = 2
1
c) lim𝑥→1 = 2.
2 𝑥
𝑙𝑖𝑚 3
d) 𝑥→23𝑥 =24
𝑥−1
e) lim𝑥→1 =2
√𝑥 −1
7
𝑙𝑖𝑚 1 𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
b) 𝑥 →0 f(𝑥) where 𝑓(𝑥) = {
−1 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
𝑙𝑖𝑚 1
c) 𝑥 →0 𝑥2
𝑙𝑖𝑚 1 𝑖𝑓 𝑥 ≥ 0
d) 𝑥 →0 𝑓(𝑥) where 𝑓(𝑥) = {
−1 𝑖𝑓 𝑥 < 0
𝑙𝑖𝑚
e) 𝑥 →0 sin(1/𝑥)
4. Prove that 𝑥 𝑙𝑖𝑚
→2 𝑥 2 ≠ 3.99
Proof:
8
The rest are left as exercises.
Theorem 3: If lim𝑥→𝑎 𝑓(𝑥) and lim𝑥→𝑎 𝑔(𝑥) exist and k is a constant, then
Proof:
Remark (Some Consequences of the Rules): Theorem 3 (a) and (b) can be extended to any finite
number of functions.
Example 1: Suppose lim𝑥→2 𝑓(𝑥) =4, lim𝑥→2 𝑔(𝑥) =5 and lim𝑥→2 ℎ(𝑥) = 8. Then
𝑥𝑓(𝑥)−𝑔(𝑥) 3
a) lim𝑥→2 =8
ℎ(𝑥)
b) lim𝑥→2 (6𝑓(𝑥)𝑔(𝑥) +
= 128
Limit theorems can also be used in showing that a limit does not exist.
1
Example 2: Show that lim𝑥→0 does not exist.
𝑥
1
Solution: Suppose it exist and let lim𝑥→0 𝑥 = 𝐿. Then by the product rule of limits we have
𝑥
1 =lim𝑥→0 = 𝐿 × 0 = 0.
𝑥
Note that we cannot use the quotient rule of limits, if lim𝑥→𝑎 𝑔(𝑥) = 0.
The following law is often used along with simplifications and rationalizations in evaluating certain limits
where the quotient rule cannot be directly used. It follows trivially from the formal definition of limits.
Remark 8 (Cancellation Law): If f(x) = g(x) for every x in some neighborhood of a real number a
except possibly at a itself, then
lim 𝑓(𝑥) = lim 𝑔(𝑥)
𝑥→𝑎 𝑥→𝑎
9
Example 3: Evaluate
𝑥2 − 1
lim
𝑥→−1 𝑥 + 1
Solution: Since the limit of the denominator is 0 at -1, so we cannot apply Quotient Rule. Using
simplification and the above remark we obtain
𝑥 2 −1 (𝑥−1)(𝑥+1)
lim𝑥→−1 𝑥+1
= lim𝑥→−1 𝑥+1
= lim𝑥→−1 𝑥 − 1 = -2.
Example 4: Evaluate
√𝑥 − 1
lim
𝑥→1 𝑥 − 1
Solution: We cannot apply Quotient Rule to evaluate this limit. Using rationalization and the above
remark we obtain
√𝑥−1 1
lim𝑥→1 = lim𝑥→−1 = ½.
𝑥−1 √𝑥+1
Solution:
Example 6: Show that
1 1
lim𝑥→𝑎 𝑥 𝑛 = 𝑎𝑛 (𝑎 > 0)
Example 8: Let R be a rational function and a be a number in the domain of R. Show that
𝑙𝑖𝑚𝑥→𝑎 𝑅(𝑥) = 𝑅(𝑎)
10
Limits of Composite Function:
If 𝑓(𝑎) = 𝑐 and 𝑔(𝑐) = 𝐿, then (𝑔𝑜𝑓)(𝑎) = 𝐿. But if lim𝑥→𝑎 𝑓(𝑥) = 𝑐 and lim𝑦→𝑐 𝑔(𝑦) =𝐿, it is
not always true that lim𝑥→𝑎 (𝑔𝑜𝑓)(𝑥) =𝐿. The difference lies in the fact that (𝑔𝑜𝑓)(𝑎) is defined only
if 𝑓(𝑎) is in the domain of 𝑔 whereas lim𝑥→𝑎 (𝑔𝑜𝑓)(𝑎) can exist even if 𝑓(𝑎) is not in the domain of
𝑔. A counter example can be easily constructed as follows. Assume 𝑓 to be the constant function
𝑓(𝑥) = 𝑐 and let 𝑔 be any function such that lim𝑦→𝑐 𝑔(𝑦) ≠ 𝑔(𝑐). Then 𝑔(𝑓(𝑥)) = 𝑔(𝑐) for every x
and hence lim𝑥→𝑎 𝑔(𝑓(𝑥)) = 𝑔(𝑐). Thus, lim𝑦→𝑐 𝑔(𝑦) ≠ lim𝑥→𝑎 𝑔(𝑓(𝑥)).
then
lim𝑥→𝑎 𝑔(𝑓(𝑥)) = 𝐿.
Remark:
a) For computational convenience substitute 𝑦 for the inner function, 𝑓(𝑥). Then 𝑦 → 𝑐 as 𝑥 → 𝑎
and lim𝑥→𝑎 𝑔(𝑓(𝑥)) = lim𝑦→𝑐 𝑔(𝑦).
b) This theorem is sometimes called the substitution rule for limits.
c) We may write the conclusion of the theorem briefly as:
0 if x = −1 .
Example 9 (condition (iii) of Theorem 4 is essential): Let f(x) = −1 & g(x) = { 1
if x ≠ −1 . Show
x+1
by direct computation that lim𝑥→0 𝑔(𝑓(𝑥)) ≠ lim𝑦→𝑐 𝑔(𝑦) where c =lim𝑥→0 𝑓(𝑥).
11
1
Solution: Clearly, lim𝑥→0 𝑓(𝑥) = −1 ; hence 𝑐 = −1. Since is negative for numbers
x+1
smaller than -1 and it is positive for numbers greater than -1, so lim𝑦→−1 𝑔(𝑦) does not exist. On the
other hand, 𝑔(𝑓(𝑥)) = 0 ∀𝑥. Hence, lim𝑥→0 𝑔(𝑓(𝑥)) = 0 ≠ lim𝑦→−1 𝑔(𝑦).
Example 10: Use the Substitution Rule of limits to show that limℎ→0 𝑓(𝑎 + ℎ) = lim𝑥→𝑎 𝑓(𝑥)and use
it to show that
Solution (a): lim𝑥→𝑎 sin 𝑥 = limℎ→0 sin (𝑎 + ℎ) = limℎ→0 [sin a cos ℎ + cos 𝑎 sin ℎ] = sin 𝑎.
12
1 1
Solution: Note that −1 ≤ 𝑠𝑖𝑛 ≤ 1 for every x ≠ 0. Thus −𝑥 ≤ 𝑠𝑖𝑛 ≤ 𝑥 𝑓𝑜𝑟 𝑥 > 0 and
𝑥 𝑥
1 1
−𝑥 ≤ 𝑠𝑖𝑛 𝑥 ≤ 𝑥 𝑓𝑜𝑟 𝑥 < 0. Therefore, −|𝑥| ≤ 𝑥𝑠𝑖𝑛 ≤ |𝑥| for x ≠ 0. Moreover, lim𝑥→0 −|𝑥| = 0 =
𝑥
1
lim𝑥→0 |𝑥|. By the Squeezing law, lim𝑥→0 𝑥𝑠𝑖𝑛 𝑥 = 0.
The following limit is useful in evaluating many other limits involving trig functions.
Exercises 2.2
𝑙𝑖𝑚 𝑙𝑖𝑚 2𝑓(𝑥)−𝑔(𝑥)
1. Suppose 𝑥→0 𝑓(𝑥) = 1 and 𝑥→0 𝑔(𝑥) = −5. Find lim𝑥→0 2 .
[𝑓(𝑥)+7]3
Symbolically,
𝑙𝑖𝑚𝑥→𝑎+ 𝑓(𝑥) = 𝐿 ⇔ (∀𝜀 > 0) (∃𝛿 > 0) ∋ x ∈ (a, a + 𝛿 ) ⇒ f(x) ∈ (L - 𝜀, L + 𝜀)
If 𝑙𝑖𝑚𝑥→𝑎+ 𝑓(𝑥) = 𝐿 , we also say L is the right hand limit of f at a.
13
Similarly we can define the left hand limit of f at a ,denoted by 𝑙𝑖𝑚𝑥→𝑎− 𝑓(𝑥), symbolically as follows.
𝑙𝑖𝑚𝑥→𝑎− 𝑓(𝑥) = 𝐿 ⇔ (∀𝜀 > 0) (∃𝛿 > 0) ∋ x ∈ (a - 𝛿, a ) ⇒ f(x) ∈ (L - 𝜀, L + 𝜀)
Right hand limits and left hand limits are called one sided limits and limits of the form - lim𝑥→𝑎 𝑓(𝑥) - are
called two sided limit.
Notation: ⟦𝑥⟧ ≔ 𝑡ℎ𝑒 𝑔𝑟𝑒𝑎𝑡𝑒𝑠𝑡 𝑖𝑛𝑡𝑒𝑔𝑒𝑟 𝑙𝑒𝑠𝑠 𝑡ℎ𝑎𝑛 𝑜𝑟 𝑒𝑞𝑢𝑎𝑙 𝑡𝑜 𝑥. For instance, ⟦𝜋⟧ = 3,
⟦3⟧ = 3 , etc.
Remark [Limit Theorems for One Sided Limits]: The basic limit theorems (sum, difference, product,
quotient, substitution and squeezing rules) are valid for one sided limits, too.
Remark: This theorem is useful in evaluating limits of functions defined by two or more formulas.
Functions involving absolute values are functions of such sort.
|𝑥| 𝑙𝑖𝑚 𝑙𝑖𝑚 𝑙𝑖𝑚
Example 3: Let (𝑥) = 𝑥
. Find 𝑥 →0+ 𝑓(𝑥), 𝑥 →0 − 𝑓(𝑥) and 𝑥 →0 𝑓(𝑥).
Example 4: Given
4𝑥 − 3 𝑖𝑓 𝑥 ≤ −1
𝑓(𝑥) = { 2
𝑥 + 𝑘 𝑖𝑓 − 1 < 𝑥
𝑙𝑖𝑚
Find 𝑘 such that 𝑥 →−1 𝑓(𝑥) exists.
14
Exercise 3.3
1. Prove the following limits by the precise definition of one sided limits.
𝑙𝑖𝑚 1 𝑙𝑖𝑚 1
a) 𝑥 →0+ 1 =1 b) 𝑥 →0− 1 =0
− −
1+2 𝑥 1+2 𝑥
Definition 1: Let 𝑎 be a real number and 𝑓 be a function defined on an open interval containing a
except possibly at 𝑎 itself. We say the limit of f as x approaches 𝑎 is infinity, written lim f ( x) , if
x a
for every real number M there is a positive real number such that
15
Definition 2: Let 𝑎 be a real number and f be a function defined on an open interval containing a except
possibly at a itself. We say the limit of f as x approaches a is negative infinity, written lim f ( x) , if
xa
for every real number M there is a positive real number such that
The two limits defined above are called infinite limits. Similarly we can define the following one sided
infinite limits:
𝑙𝑖𝑚𝑥→𝑎+ 𝑓(𝑥) = ∞, 𝑙𝑖𝑚𝑥→𝑎+ 𝑓(𝑥) = −∞, 𝑙𝑖𝑚𝑥→𝑎− 𝑓(𝑥) = ∞ and 𝑙𝑖𝑚𝑥→𝑎− 𝑓(𝑥) = −∞.
1
Example 1: Show that lim
x 0 x2
3x
Example 2: Show that lim
x 2 2 x
Remark [Limit Theorems for Infinite Limit]: The basic limit theorems (the sum rule, difference rule,
product rule, quotient rule, power rule, substitution rule and squeezing law) are not valid for infinite
limits. In mathematics, the symbols ∞ and - ∞ are not numbers; they do not obey the usual rules of
arithmetic. Expressions involving these symbols may or may not have meanings. For instance, we can
define ∞ + 1 = ∞ ,∞ − 1 = ∞, 2. ∞ = ∞, etc. But we cannot properly define ∞. 0, ∞ − ∞, etc.
Ambiguous expressions involving infinity are called indeterminate forms.
Let f be a function defined on some interval (𝑎, ∞). Then, lim f ( x) L means that the values 𝑓(𝑥) can
x
be made arbitrarily close to 𝐿 by taking 𝑥 sufficiently large. Similarly, lim f ( x) L means that the
x
values of 𝑓(𝑥) can be made arbitrarily close to 𝐿 by taking 𝑥 sufficiently large negative.
Definition 3: Let 𝑎 be a real number and 𝑓 be a function defined on (𝑎, ). We say the limit of 𝑓 as 𝑥
approaches infinity is 𝐿, written lim f ( x) L , if for every positive real number 𝜀 there is a real number
x
16
The definition can be written compactly as follows.
Definition 4: Let 𝑎 be a real number and f be a function defined on (- , a). We say the limit of f as x
approaches negative infinity is L, written lim f ( x) L , if for every positive real number 𝜀 there is a
x
The two limits defined above are called limits at infinity. Similarly we can define the following infinite limits
at infinity:
𝑙𝑖𝑚𝑥→∞ 𝑓(𝑥) = ∞, 𝑙𝑖𝑚𝑥→∞ 𝑓(𝑥) = −∞, 𝑙𝑖𝑚𝑥→−∞ 𝑓(𝑥) = ∞ and 𝑙𝑖𝑚𝑥→−∞ 𝑓(𝑥) = −∞.
1
a) lim 0
x x
1
b) lim 0
x x
Remark 14(Limit Theorems for Limits at Infinity): The basic limit theorems (sum, difference,
product, quotient, substitution and squeezing rules) are valid for finite limits at infinity.
3𝑥 2 −𝑥−2
Example 4: Evaluate lim𝑥→∞ 2
5𝑥 +4𝑥+1
Solution: To evaluate the limit we divide both the numerator and the denominator by the highest power
of x that occurs. So we have,
3𝑥 2 −𝑥−2
lim𝑥→∞ 5𝑥 2 +4𝑥+1
17
Example 5: Evaluate
18
1 0 ∞
, ,0
∞ ∞
b) A limit in any of the following forms must be ∞:
1 ∞ ∞
, , ,∞ + ∞, ∞. ∞, ∞∞ and ∞1 .
0+ 0+ 1
Example 6:
1 𝑦 4
(a) lim𝑦→∞ (𝑦) = 0. (b) lim𝑥→∞ 𝑙𝑛𝑥 = 0 (c) lim𝑥→∞ (3𝑥 + 1)𝑥 = ∞.
Theorem 2:
∞
1 ∞
lim (1 + 𝑥) = 𝑒 = lim (1 + ) .
𝑥→0 𝑦→∞ 𝑦
2𝑥 3𝑥 2𝑥 3𝑥 1
Example 7: Evaluate: a) lim𝑥→∞ (2𝑥+1) b) lim𝑥→−∞ (2𝑥+1) c) lim𝑥→0 (1 + 2𝑥)2𝑥 .
3.4.4 Asymptotes
Infinite limits and limits at infinity are useful in sketching the graphs of functions.
Definition 7: The line 𝑥 = 𝑎 is called a vertical asymptote of the curve 𝑦 = 𝑓(𝑥) if either
For a given function f(x), there are four cases, in which vertical asymptotes can present themselves:
𝑠𝑖𝑛𝑥
𝑓(𝑥) = 𝑥(𝑥 2 −1).
Definition 8: The line 𝑦 = 𝐿 is called a horizontal asymptote of the curve 𝑦 = 𝑓(𝑥) if either
lim𝑥→∞ 𝑓(𝑥) = 𝐿 or lim𝑥→−∞ 𝑓(𝑥) = 𝐿.
19
A function can have at most two horizontal asymptotes, why? But it can have even infinitely many
vertical asymptotes.
Example 9:
Definition 9 (oblique asymptote): The line 𝑦 = 𝑎𝑥 + 𝑏 is called an oblique asymptote of the curve
𝑦 = 𝑓(𝑥) if either lim𝑥→∞ [𝑓(𝑥) − (𝑎𝑥 + 𝑏)] = 0 or lim𝑥→−∞[𝑓(𝑥) − (𝑎𝑥 + 𝑏)] = 0.
Definition 10 (general asymptote): The curve 𝑦 = 𝑔(𝑥) is called an asymptote of the curve 𝑦 = 𝑓(𝑥) if
either lim𝑥→∞ [𝑓(𝑥) − 𝑔(𝑥)] = 0 or lim𝑥→−∞ [𝑓(𝑥) − 𝑔(𝑥)] = 0.
𝑃
Let 𝑅 = be a rational function such that the degree of P and Q are respectively m and n and the
𝑄
𝑎
(i) If n=m , then 𝑦 = 𝑏 𝑛 is the horizontal asymptote.
𝑚
(iii) If n > m, then there is no horizontal asymptote, but rather a slant (oblique) asymptote and can be
found be using long division.
Exercise 3.4:
20
𝑥 2 +1
i) lim𝑥→∞
2𝑥−3
3 𝑥 2 +3
j) lim𝑥→∞ √27𝑥2 −1
k) lim𝑥→∞ 2𝑥 + 1 − √4𝑥 2 + 5
𝑥
5. Find the vertical and horizontal asymptotes for the graph of 𝑓(𝑥) = .
√𝑥 2 +1
5𝑥 2
6. Find the vertical and horizontal asymptotes for the graph of 𝑓(𝑥) = .
√𝑥 4 −16
7. Find the vertical and horizontal asymptotes for the graph of 𝑓(𝑥) = √𝑥 2 + 𝑥 + 1 − √𝑥 2 − 𝑥.
3.5.1 Continuity
There are two commonly used intuitive definitions of continuous functions:
(1)Intuitively, a function is continuous if when two points of the domain are close to each other, then their
images should also be close to each other.
(2)Intuitively, we can say that a function is continuous when you can draw its graph without lifting
your pencil. The second intuitive definition is more misleading, because there exist continuous
functions with strange graphs.
Remarks:
21
𝑓 is continuous at 𝑥0 ⇔ ∀𝜀 > 0, ∃𝛿 > 0 ∋ all 𝑥 in the domain of 𝑓 the following
0 < |𝑥 − 𝑎| < 𝛿 , ⇒ | 𝑓(𝑥) − 𝐿| < 𝜀
1
Example 1: Let 𝑓(𝑥) = 𝑥−2. Then 𝑓 is discontinuous at 𝑥0 = 2 , because condition (i) fails.
3 + 𝑥 𝑖𝑓 𝑥 ≤ 1
Example 2: Let 𝑓(𝑥) = { . Then 𝑓 is discontinuous at 𝑥0 = 1, because condition (ii)
3 − 𝑥 𝑖𝑓 𝑥 > 1
fails.
3 + 2𝑥 𝑖𝑓 𝑥 ≠ 1
Example 3: Let 𝑓(𝑥) = { . Then 𝑓 is discontinuous at 𝑥0 = 1, because condition
2 𝑖𝑓 𝑥 = 1
(iii) fails.
3 + 2𝑥 𝑖𝑓 𝑥 ≠ 1
Example 4: Let 𝑓(𝑥) = { . Then 𝑓 is continuous at 𝑥0 = 1.
5 𝑖𝑓 𝑥 = 1
1 𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
Example 5[Dirichlet Function]: Let 𝑓(𝑥) = { .
0 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
Just as we had rules for combining limits, we can also combine continuous functions to create continuous
functions.
a) 𝑓 + 𝑔 is continuous at 𝑥0 ,
b) 𝑓 − 𝑔 is continuous at 𝑥0 ,
22
c) 𝑓 𝑔 is continuous at 𝑥0 and
𝑓 𝑓
d) 𝑔
is continuous at 𝑥0 (provided that 𝑔 (𝑥0 ) ≠ 0).
We extend the local property of continuity at a point into a global property with the next definitions.
a) A function f is said to be continuous on an open interval (a, b) if it is continuous at 𝑐 for every 𝑐 ∈ (𝑎, 𝑏).
b) A function f is said to be continuous on a closed interval [a, b] if f is continuous at every c ∈ (𝑎, 𝑏),
lim𝑥→𝑎+ 𝑓(𝑥) = f(a) and lim𝑥→𝑏− 𝑓(𝑥) = f(b).
c) In general, we say that a function is continuous on an interval I if it is defined on that interval and
continuous at every point of that interval; where at the endpoints, we only use the appropriate one-sided
limit.
Classification of discontinuities:
b) a jump discontinuity at a if both one sided limits exist and are finite at a but they are unequal.
c) an essential discontinuity at a if only one of the two one-sided limits needs not exist or be infinite
23
3.5.2 The Intermediate Value Theorem
Functions that are continuous on closed intervals possess a number of remarkable properties. The so
called Intermediate Value Theorem (I.V.T) is one among them.
ii) p is a number between f(a) and f(b) (that is, f(a) ≤ 𝑝 ≤ f(b) or f(b) ≤ 𝑝 ≤ f(a) ) ,
a c
b
Remark: Not all functions satisfying the intermediate value property are continuous.
Corollary: If 𝑓 is continuous on [𝑎, 𝑏] and if 𝑓(𝑎) and 𝑓(𝑏) have different signs (that is, 𝑓(𝑎)𝑓(𝑏) < 0
), then f has a root (a zero) in (a, b).
Proof: Let p =0. Since f(a) and f(b) have different signs and p is zero, so p is between f(a) and f(b). By the
IVT there is a number r in [a, b] such that f(r) =p=0. Since f(a)and f(b) cannot be zero, r ∈ (𝑎, 𝑏).
∎
Note: This corollary is useful in locating the zeroes (roots) of functions and in showing existence of
solutions (roots) to certain equations. Recall that a zero (root) of a function f is a number r for which f(r)
=0. Recall also that a number r is called a solution (root) of the equation f(x) =0 if f(r) =0. Observe the
following chain of equivalencies.
Example 7: Given f(x) = 𝑥 5 − 2𝑥 3 + 𝑥 2 + 2 , show that there exist c such that f(c)= -1.
Solution: The function is continuous on any closed interval. By trial and error we can easily find numbers
a and b so that -1 is between f(a) and f(b). For instance, f(-2) < -1 < f(-1). Therefore, it follows from IVT
that there exists 𝑐 𝜖[−2, −1] such that f(c) = -1. //
24
Solution: The function 𝑓(𝑥) ≔ 4𝑥 3 − 6𝑥 2 + 3𝑥 − 2 is continuous on any closed interval. Moreover,
f(2) f(1) < 0. Therefore, it follows from the corollary of the IVT that there exists 𝑟 𝜖(1, 2) such that f(r)
= 0. But this implies 4𝑟 3 − 6𝑟 2 + 3𝑟 − 2=0 which in turn implies 4𝑟 3 − 6𝑟 2 + 3𝑟 = 2.
Exercise 3.5:
1. Let 𝑎 and 𝑏 be real numbers and 𝑓 be a function. Define what is meant by 𝑓 is continuous on
a) [𝑎, ∞) b) [𝑎, 𝑏)
2. Determine whether or not the specified functions are continuous at the specified numbers.
a)
3. Determine whether or not the specified functions are continuous. If they are, prove it. If they are
not, identify the subset X in the domain of the function such that for every x ∈ X, f is continuous
at x.
𝑥 𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
(a) 𝑓(𝑥) = {
0 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
1
sin(𝑥) 𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
(b) 𝑓(𝑥) = {
0 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
4. Let
5.
Show that f is continuous for all values of x. Show that f is differentiable for all values of x, but
that the derivative, f’ is not continuous at x=0
𝑥 𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
6. Let f : R → R be defined by 𝑓(𝑥): = { 2
𝑥 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
Use the 𝜀 − 𝛿 definition of continuity directly to show that f is continuous at 1 and
discontinuous at 2.
7. Give an example of functions f : R → R and g : R → R such that the function defined by h(x) :=
f(x) + g(x) is continuous but f and g are not continuous.
𝑝 𝑝
𝑖𝑓 𝑥 𝑖𝑠 𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑎𝑛𝑑 𝑞 𝑖𝑠 𝑖𝑡𝑠 𝑠𝑖𝑚𝑝𝑙𝑖𝑓𝑖𝑒𝑑 𝑓𝑜𝑟𝑚
8. (Tomae Function): Let 𝑓(𝑥) = {𝑞
0 𝑖𝑓 𝑥 𝑖𝑠 𝑖𝑟𝑟𝑎𝑡𝑖𝑜𝑛𝑎𝑙
Show that f is continuous at every irrational and discontinuous at every rational number.
9. Show that 𝑝(𝑥) = 2𝑥 3 − 5𝑥 2 − 10𝑥 + 5 has a root somewhere in the interval [−1, 2].
10. Determine all values of the constants A and B so that the following function is continuous for all
values of x.
25
11. Use the IVT to show that the curves 𝑦 = 𝑒 𝑥 and 𝑦 = 𝑒 −𝑥 intersect each other in (−2, 2).
12. Use the IVT to show that there exists a real number that is exactly one less than its cube.
13. Prove: If 𝑓: [𝑎, 𝑏] → [𝑎, 𝑏] is continuous, then there exists some number p in [a, b] such that f(p)
= 0.
14. Prove: If 𝑓, 𝑔: [𝑎, 𝑏] → ℝ is continuous, f(a) < g(a) and g(b)< g(b), then there exists a number c in
(𝑎, 𝑏) such that 𝑓(𝑐) = 𝑔(𝑐).
26