The Denite Integral

Complete Key-Concepts Reference for Exercises

1 · Subdivisions of an Interval

Denition
A subdivision (partition) of [a, b] is a nite sequence
a = x0 < x1 < · · · < xn = b.

The set of all subdivisions of [a, b] is denoted S(a, b).

The norm (pas) of s = (x ) is i

|s| = max x1 − x0 , x2 − x1 , . . . , xn − xn−1 .

A subdivision is regular if all sub-intervals have equal length: x − x i i−1 =

b−a
n
for all i.
Proposition
, such that |s| < ε. (Take n large enough so (b − a)/n < ε.)
∀ ε > 0 ∃ s ∈ S(a, b)
s is ner than s (write s ⊂ s ) if s contains all points of s.
′ ′ ′

Always: s ⊂ s ∪ s and s ⊂ s ∪ s .
′ ′ ′

2 · Darboux Sums

Let f : [a, b] → R be bounded (m ≤ f ≤ M ) and s ∈ S(a, b). Set

Mi = sup f (x), mi = inf f (x).
x∈[xi−1 ,xi ] x∈[xi−1 ,xi ]

Denition
(upper Darboux sum)
n
X
D(f, s) = Mi (xi − xi−1 )
i=1

(lower Darboux sum)

n
X
D(f, s) = mi (xi − xi−1 )
i=1

Key Inequalities  memorise these

(a) Bound: For any subdivision s,
m(b − a) ≤ D(f, s) ≤ D(f, s) ≤ M (b − a).
(b) Renement: If s ⊂ s then
′

D(f, s) ≤ D(f, s′ ) ≤ D(f, s′ ) ≤ D(f, s).

Meaning: adding points increases lower sums and decreases upper sums.
(c) Any two subdivisions: ∀ s , s ∈ S(a, b),
1 2

D(f, s1 ) ≤ D(f, s2 ).

1
 Proof trick: form s = s1 ∪ s2 and apply (b) twice.

3 · Riemann Integrability

Since {D(f, s)} is bounded above and {D(f, s)} is bounded below:
I∗ = sup D(f, s), I ∗ = inf D(f, s), I∗ ≤ I ∗ .
s s

Denition
fis Riemann-integrable onZ[a, b] if I = I . ∗
∗

The common value is written f (x) dx.

b

Characterisation (the go-to criterion)

f integrable on [a, b] ⇐⇒ ∀ε > 0, ∃ s ∈ S(a, b) : D(f, s) − D(f, s) < ε.

Key Tip
Strategy for proofs: take the regular subdivision of step (b − a)/n. Compute D − D, simplify
(often a telescoping sum), and show it → 0.
For a monotone increasing f :
n
b − a X  b − a  n→∞
D(f, s) − D(f, s) = f (xi ) − f (xi−1 ) = f (b) − f (a) −−−→ 0.
n n
i=1

Classes of integrable functions

Proposition
The following functions are always Riemann-integrable on [a, b]:
ˆ Every monotone function.
ˆ Every continuous function.
ˆ Every bounded function with at most nitely many discontinuities.
ˆ (More generally) every bounded function whose discontinuity set is countable.

Classic non-integrable example

on [0, 1].
(
1 x∈Q
f (x) =
0 x∈
/Q

Every interval contains both rationals and irrationals, so ∀s: D(f, s) = 0 and D(f, s) = 1, hence
I = 0 ̸= 1 = I .
∗
∗

2
4 · Riemann Sums

Denition
Choose any ξ ∈ [x
i i−1 , xi ] . The Riemann sum is
n
X
R(f, s, ξi ) = f (ξi ) (xi − xi−1 ).
i=1

It always satises D(f, s) ≤ R(f, s, ξi ) ≤ D(f, s) .

Limit theorem (most used in exercises)
f integrable on [a, b] if and only if ∃ I ∈ R such that
∀ε > 0, ∃ δ > 0 : |s| < δ ⇒ R(f, s, ξi ) − I < ε for all choices of ξ .
i

Consequence  regular-subdivision formula:

b n  
b−aX b−a
Z
f (x) dx = lim f a+i· .
a n→∞ n n
i=1

Recognising a limit as an integral

If you see , this equals f (x) dx.
n   Z 1
1X i
lim f
n→∞ n n 0
i=1

More generally, factor out and identify f nk .

 
1
n
n n Z 1
1 1X 1 dx π
Example:
X
lim 2 2
= lim = = .
n→∞ n +k n→∞ n 1 + (k/n)2 0 1+x 2 4
k=1 k=1

Worked example: R 2
0
x2 dx

Use the regular subdivision x = , step , with ξ = x :

i
2i
n
2
n i i

Z 2 n  2 n
2X 2i 8 X 2
x2 dx = lim = lim i
0 n→∞ n n n→∞ n3
i=1 i=1
8 n(n + 1)(2n + 1) 8
= lim 3 · = .
n→∞ n 6 3

Useful summation formulas: , ,

n n n
X n(n + 1) X n(n + 1)(2n + 1) X
i = i2 = i3 =
2 6
i=1 i=1 i=1

.
 2
n(n + 1)
2

5 · Properties of the Riemann Integral

Throughout: f, g integrable on [a, b], λ, β ∈ R.

3
 Property Statement Notes
Linearity Rb
a (λf + βg) = λ
Rb
a f + β
Rb
a g λf + βg is integrable
Positivity Rb
f ≥0⇒ af ≥0 follows from D ≥ 0
Monotonicity Rb
f ≤g⇒ af ≤ ag
Rb
from linearity + positivity
Triangle ineq. Rb Rb
a f ≤ a |f | |f | is also integrable
Chasles' relation Rb Rc Rb
a f = a f + c f ∀ c ∈ [a, b]
Max/Min ,
max(f, g) min(f, g) integrable use max = f +g+|f −g|
2

Key Tip
Warning: |f | integrable does not imply f integrable. Counter-example: f = 1 on Q, f = −1
on R \ Q; then |f | = 1 (integrable) but f is not.
6 · Mean Value Theorem for Integrals

Proposition
Let f be integrable on [a, b], m = inf f , M = sup f .
f (x) dx ≤ M .
Z b
1
(1) m ≤
b−a a
(2) If f is continuous on [a, b], then ∃ c ∈ [a, b] such that
Z b
f (x) dx = f (c) · (b − a).
a

7 · Exam Checklist

1. Prove integrability: use the ε-criterion  exhibit s with D(f, s) − D(f, s) < ε.
2. Compute R f from scratch: regular subdivision, evaluate f (a + i(b − a)/n), sum, take
b

n → ∞.
a

3. Evaluate a limit of a sum: factor 1/n, identify f (k/n), write R f (x) dx. 0
1

4. Show non-integrability: nd a xed gap D(f, s) − D(f, s) ≥ c > 0 for every s.
5. Renement inequalities: know them cold  they underpin every other proof.
6. Chasles relation proof: add point c to any subdivision s; use renement.
7. Linearity proof: write R(λf + βg, s, ξ ) = λR(f, . . .) + βR(g, . . .), bound each term sepa-
rately with ε/2.
i