Pure Maths Guide From Love of Maths

This book is available (in whole or by chapter) in pdf format via digital
download from our website loveofmaths.co.uk.

Feedback of any kind – via the website – would be most welcome, including
any errors you find (there will be some I’m afraid) and suggestions for
improvement. Please comment too on anything you found hard to
understand: you should regard this as a failing on my part, not yours.
"The true spirit of delight, the exaltation, the sense of being more
than man, which is the touchstone of the highest excellence, is to
be found in Mathematics as surely as in poetry."
Bertrand Russell
"Sometimes a theorem can just be so elegant. It can draw together.

It can take a problem which seems to be intractable, almost
impossible to get your head around. And yet, within a few lines,
you can discover that you can demonstrate that this is absolutely
true and beyond dispute, and always will be. A sense of power and
beauty in that which can’t be replicated anywhere else."
John Waterman
Acknowledgements
I am hugely grateful for numerous helpful comments and suggestions from
Ian Christie and Mike Fairbank, Abbey College London, Alan Gregory MBE,
Prof Alastair McDonald, Barry Ainge, Sam Waterman, Steve Gooch, Ilona
Hurrell, Joe Waterman and Mike McGuire.
I particularly want to thank Matt Joy who, as a seasoned journalist, patiently
pointed the need for simplicity and succinct brevity. And Victor
Lunn-Rockliffe, with whom I spent many hours exploring how best to make
the layout and presentation as inviting as possible.
I want to thank Dan Balis and two of my grandchildren, Ethan and Sekai,
who road-tested some of it in the latter stages of development.
Most importantly, I need to acknowledge the hundreds of other students
whom I have had the privilege to teach, and who have contributed
immeasurably, albeit unwittingly, to this book.
Finally to my wife Anne who, despite having not the slightest interest in
Mathematics, patiently encouraged me over the many times it looked like it
would never get finished.
John Waterman, May 2019
Love of Maths Pure Maths Guide First edition published by Love of Maths Ltd
June 2021.
Contents
0 WELCOME 0-1
0.1 Is this book for you? . . . . . . . . . . . . . . . . . . . . . . . . . . . 0-1
0.2 Navigating your way around . . . . . . . . . . . . . . . . . . . . . 0-2
0.3 Solving problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0-6
1 FUNDAMENTAL CONCEPTS 1-1

1.1 Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.1.1 Natural Numbers . . . . . . . . . . . . . . . . . . . . . . . . 1-2
1.1.2 Prime numbers . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1.1.3 Integers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
1.1.4 Rational Numbers . . . . . . . . . . . . . . . . . . . . . . . 1-5
1.1.5 Irrational Numbers . . . . . . . . . . . . . . . . . . . . . . . 1-7
1.1.6 Real Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
1.1.7 Normal numbers . . . . . . . . . . . . . . . . . . . . . . . . 1-8
1.2 The ∞ symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9
1.3 Commutativity and Associativity . . . . . . . . . . . . . . . . . . . 1-12
1.4 Set language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-14
1.4.1 Elements or members of a set . . . . . . . . . . . . . . . . . 1-14
1.4.2 Subsets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-15
1.4.3 Intersection and Union of sets . . . . . . . . . . . . . . . . 1-17
1.5 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-21
1.5.1 Solving equations . . . . . . . . . . . . . . . . . . . . . . . . 1-21
1.5.2 The zero product principle . . . . . . . . . . . . . . . . . . 1-22
1.5.3 Polynomial equations . . . . . . . . . . . . . . . . . . . . . 1-24
1.5.4 Quadratic formula . . . . . . . . . . . . . . . . . . . . . . . 1-25
1.6 Inequalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-26
1.7 Identities and Equivalence . . . . . . . . . . . . . . . . . . . . . . 1-28
1.8 Function notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-31
1.9 Reasoning and proof . . . . . . . . . . . . . . . . . . . . . . . . . . 1-36
1.10 Directed Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-43
1.10.1 Adding and subtracting directed numbers . . . . . . . . . 1-43
1.10.2 The Box Principle . . . . . . . . . . . . . . . . . . . . . . . . 1-45
1.10.3 Combining signs . . . . . . . . . . . . . . . . . . . . . . . . 1-47
1.10.4 Dots and brackets . . . . . . . . . . . . . . . . . . . . . . . . 1-52
2 INDICES AND SURDS 2-1

2.1 Multiplying and dividing powers of the same base . . . . . . . . . 2-3
2.2 The product and quotient rules for indices . . . . . . . . . . . . . 2-8
2.3 Raising one index to the power of another . . . . . . . . . . . . . 2-10
2.4 Negative indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18
2.5 Rational (or fractional) indices . . . . . . . . . . . . . . . . . . . . 2-26
2.6 Negative fractional indices . . . . . . . . . . . . . . . . . . . . . . 2-29
2.7 Factorising the base . . . . . . . . . . . . . . . . . . . . . . . . . . 2-37
1
CONTENTS
2.8 Surds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51

2.8.1 What is a surd? . . . . . . . . . . . . . . . . . . . . . . . . . 2-51
2.8.2 Multiplying surds . . . . . . . . . . . . . . . . . . . . . . . . 2-56
2.8.3 Decomposing surds . . . . . . . . . . . . . . . . . . . . . . 2-59
2.8.4 Rationalising surds . . . . . . . . . . . . . . . . . . . . . . . 2-65
3 BRACKETS AND FACTORISING 3-1

3.1 Expressions, Terms and Coefficients . . . . . . . . . . . . . . . . . 3-2
3.2 Multiplying out brackets . . . . . . . . . . . . . . . . . . . . . . . . 3-4
3.3 Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
3.3.1 Adding and subtracting polynomials . . . . . . . . . . . . 3-24
3.3.2 Multiplying polynomials . . . . . . . . . . . . . . . . . . . . 3-27
3.4 Factorising & Highest Common Factor . . . . . . . . . . . . . . . 3-33
3.5 One-bracket factorising . . . . . . . . . . . . . . . . . . . . . . . . 3-35
3.6 Two-bracket factorising: ax 2 + bx + c . . . . . . . . . . . . . . . . 3-43
3.6.1 When c = 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-43
3.6.2 When a = ±1 . . . . . . . . . . . . . . . . . . . . . . . . . . 3-44
3.6.3 When a 6= 1 or −1 . . . . . . . . . . . . . . . . . . . . . . . . 3-49
3.6.4 The difference of two squares. . . . . . . . . . . . . . . . . 3-53
3.6.5 When b = 0 and c > 0. The sum of two squares? . . . . . . 3-57
3.6.6 Perfect squares . . . . . . . . . . . . . . . . . . . . . . . . . 3-58
3.6.7 Disguised quadratics . . . . . . . . . . . . . . . . . . . . . . 3-61
4 ALGEBRAIC DIVISION AND FACTOR THEOREM 4-1

4.1 Introduction to algebraic fractions . . . . . . . . . . . . . . . . . . 4-2
4.2 An important notation issue . . . . . . . . . . . . . . . . . . . . . 4-3
4.3 Cancelling algebraic fractions . . . . . . . . . . . . . . . . . . . . . 4-5
4.4 Multiplying algebraic fractions . . . . . . . . . . . . . . . . . . . . 4-11
4.5 Dividing algebraic fractions: hammer and boot rule . . . . . . . 4-20
4.6 Adding and subtracting algebraic fractions . . . . . . . . . . . . . 4-26
4.7 More algebraic fractions . . . . . . . . . . . . . . . . . . . . . . . . 4-35
4.8 Dividing polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . 4-41
4.9 The Factor Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 4-55
5 COORDINATE GEOMETRY 5-1

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
5.2 Linear equations and straight lines . . . . . . . . . . . . . . . . . . 5-6
5.2.1 Gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-6
5.2.2 The equation of a straight line: y = mx + c . . . . . . . . . 5-13
5.2.3 Methods for finding the equation of a line . . . . . . . . . 5-22
5.2.4 The equation of a straight line . . . . . . . . . . . . . . . . 5-31
5.2.5 Perpendicular lines . . . . . . . . . . . . . . . . . . . . . . . 5-34
5.2.6 Intersecting lines . . . . . . . . . . . . . . . . . . . . . . . . 5-40
5.3 An extraordinary puzzle . . . . . . . . . . . . . . . . . . . . . . . . 5-43
5.4 Distances between points and lines . . . . . . . . . . . . . . . . . 5-44
5.4.1 The distance between two points . . . . . . . . . . . . . . 5-44
5.4.2 The distance between a point and a line . . . . . . . . . . 5-47
5.5 The mid-point of a line . . . . . . . . . . . . . . . . . . . . . . . . . 5-49
5.6 Circle Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-54
PURE MATHS: AS LEVEL GUIDE 2

CONTENTS
6 LINEAR RELATIONSHIPS 6-1

6.1 Linear Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
6.2 Formulating linear equations . . . . . . . . . . . . . . . . . . . . . 6-14
6.3 Changing the subject of a formula . . . . . . . . . . . . . . . . . . 6-22
6.4 Linear Inequalities . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-32
6.5 Linear Simultaneous Equations . . . . . . . . . . . . . . . . . . . . 6-39
6.6 Combining inequalities . . . . . . . . . . . . . . . . . . . . . . . . 6-56
6.6.1 Examples involving one variable . . . . . . . . . . . . . . . 6-56
6.6.2 Examples involving two variables . . . . . . . . . . . . . . 6-59
6.7 Linear modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-68
7 QUADRATIC RELATIONSHIPS 7-1

7.1 Completing the square. . . . . . . . . . . . . . . . . . . . . . . . . 7-2
7.1.1 x 2 coefficient = 1 . . . . . . . . . . . . . . . . . . . . . . . . 7-3
7.1.2 x 2 coefficient 6= 1 . . . . . . . . . . . . . . . . . . . . . . . . 7-6
7.2 Maximum and minimum values . . . . . . . . . . . . . . . . . . . 7-10
7.3 Solving quadratic equations . . . . . . . . . . . . . . . . . . . . . . 7-20
7.3.1 The zero product principle . . . . . . . . . . . . . . . . . . 7-21
7.3.2 Writing quadratic equations in the form f (x) = 0 . . . . . 7-23
7.3.3 Quadratic equations. Method 1: factorising . . . . . . . . 7-25
7.3.4 Quadratic equations. Method 2: completing the square . 7-30
7.3.5 Quadratic equations. Method 3: the Formula. . . . . . . . 7-36
7.4 The Discriminant . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-40
7.5 Disguised quadratic equations . . . . . . . . . . . . . . . . . . . . 7-48
7.6 Sketching quadratic functions . . . . . . . . . . . . . . . . . . . . 7-53
7.6.1 Repeated roots . . . . . . . . . . . . . . . . . . . . . . . . . 7-58
7.7 Solving quadratic inequalities . . . . . . . . . . . . . . . . . . . . . 7-72
7.7.1 Linear inequalities . . . . . . . . . . . . . . . . . . . . . . . 7-72
7.7.2 Quadratic inequalities . . . . . . . . . . . . . . . . . . . . . 7-74
7.7.3 Harder quadratic inequalities . . . . . . . . . . . . . . . . . 7-86
7.8 Inequalities with an algebraic denominator . . . . . . . . . . . . 7-96
7.9 Non-linear simultaneous equations . . . . . . . . . . . . . . . . . 7-100
7.10 Non-linear simultaneous inequalities in two variables . . . . . . 7-108
7.11 Circle Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-117
7.12 Equation of a circle . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-119
7.12.1 From geometry to equation . . . . . . . . . . . . . . . . . . 7-119
7.12.2 The equation of a circle with any centre and radius . . . . 7-121
8 SKETCHING AND TRANSFORMING CURVES 8-1

8.1 Sketching Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
8.1.1 Cubic functions and beyond . . . . . . . . . . . . . . . . . 8-5
8.1.2 Multiple roots . . . . . . . . . . . . . . . . . . . . . . . . . . 8-11
1 1
8.2 The functions y = and y = 2 . . . . . . . . . . . . . . . . . . . 8-24
x x
8.3 Transforming functions and curves . . . . . . . . . . . . . . . . . 8-34
8.3.1 Translation y = f (x) + a . . . . . . . . . . . . . . . . . . . . 8-35
8.3.2 Translation y = f (x + a) . . . . . . . . . . . . . . . . . . . . 8-37
8.3.3 Stretch y = a f (x) . . . . . . . . . . . . . . . . . . . . . . . . 8-39
8.3.4 Stretch y = f (ax) . . . . . . . . . . . . . . . . . . . . . . . . 8-41
8.3.5 Reflection in the axes: y = − f (x) and y = f (−x) . . . . . . 8-43
8.4 Periodic functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-54
8.5 Transforming functions: an elegant approach . . . . . . . . . . . 8-62

CONTENTS
9 BINOMIAL EXPANSION 9-1

9.1 Binomial coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . 9-2
9.2 Expansion of (1 + x)n . . . . . . . . . . . . . . . . . . . . . . . . . . 9-3
9.3 Binomial notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-7
9.4 Expansion of (y + x)n . . . . . . . . . . . . . . . . . . . . . . . . . . 9-10
9.5 Approximating with the Binomial expansion . . . . . . . . . . . . 9-18
10 CALCULUS: DIFFERENTIATION AND INTEGRATION 10-1

10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-2
10.2 Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-5
10.2.1 Differentiation from first principles . . . . . . . . . . . . . 10-9
10.2.2 Differentiating a polynomial function . . . . . . . . . . . . 10-14
dy
10.2.3 dx
notation . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-17
10.2.4 Differentiating harder polynomial functions . . . . . . . . 10-20
10.2.5 Differentiating non-polynomial functions . . . . . . . . . 10-23
10.2.6 Increasing and decreasing functions . . . . . . . . . . . . . 10-30
10.2.7 Using differentiation . . . . . . . . . . . . . . . . . . . . . . 10-36
10.2.8 Tangents, Normals and Turning Points . . . . . . . . . . . 10-37
10.2.9 Stationary values and turning points . . . . . . . . . . . . 10-43
d2y
10.2.10 Second derivative f 00 (x) or d x 2 . . . . . . . . . . . . . . . 10-47
10.2.11 Practical examples . . . . . . . . . . . . . . . . . . . . . . 10-58
10.3 Indefinite Integration . . . . . . . . . . . . . . . . . . . . . . . . . . 10-75
10.3.1 Converting to and from index form . . . . . . . . . . . . . 10-83
10.4 Definite Integration and Areas . . . . . . . . . . . . . . . . . . . . 10-93
10.4.1 Definite integration and area: the theory . . . . . . . . . . 10-98
10.4.2 Finding areas: harder examples . . . . . . . . . . . . . . . 10-99
11 EXPONENTIAL EQUATIONS AND LOGARITHMS 11-1

11.1 Exponential equations . . . . . . . . . . . . . . . . . . . . . . . . . 11-3
11.1.1 Quadratic Exponential Equations . . . . . . . . . . . . . . 11-6
11.2 Exponential graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-13
11.3 The function y = e x . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-17
11.4 Transforming exponential functions . . . . . . . . . . . . . . . . . 11-22
11.5 Logarithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-28
11.5.1 Logarithms and exponential equations . . . . . . . . . . . 11-31
11.5.2 The fundamental laws of logarithms . . . . . . . . . . . . . 11-32
11.5.3 Solving exponential equations using logarithms . . . . . . 11-36
11.5.4 Simplifying expressions involving logs . . . . . . . . . . . 11-39
11.5.5 More problems involving logs . . . . . . . . . . . . . . . . . 11-44
11.6 Exponential modelling . . . . . . . . . . . . . . . . . . . . . . . . . 11-52
11.7 Logs and non-linear data . . . . . . . . . . . . . . . . . . . . . . . 11-56
12 TRIGONOMETRY 12-1
12.1 SOHCAHTOA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-3
12.2 Two important trig relationships . . . . . . . . . . . . . . . . . . . 12-6
12.3 Re-defining the trig functions . . . . . . . . . . . . . . . . . . . . . 12-8
12.4 Angles outside the range 0◦ to 360◦ . . . . . . . . . . . . . . . . . . 12-13
12.5 Trig values for 0◦ and 90◦ . . . . . . . . . . . . . . . . . . . . . . . . 12-19
12.6 The sine, cosine and tan curves . . . . . . . . . . . . . . . . . . . . 12-22
12.7 The Sine Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-26
12.8 The Cosine Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12-34
12.9 More trigonometrical equations . . . . . . . . . . . . . . . . . . . 12-49
12.9.1 The sine, cosine and tangent curves . . . . . . . . . . . . 12-49

12.9.2 Identities and the notion of equivalence . . . . . . . . . . 12-51
12.9.3 The identity sin2 x + cos2 x ≡ 1 . . . . . . . . . . . . . . . 12-52
12.10 A formula for the area of a triangle . . . . . . . . . . . . . . . . . 12-69
12.11 More on trigonometrical identities . . . . . . . . . . . . . . . . . 12-70
12.12 Other types of trig problem . . . . . . . . . . . . . . . . . . . . . 12-74
13 VECTORS 13-1
13.1 What are vectors? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-2
13.2 Column vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-7
13.3 i, j notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-9
13.4 Magnitude of a vector . . . . . . . . . . . . . . . . . . . . . . . . . 13-11
13.5 Direction of a vector . . . . . . . . . . . . . . . . . . . . . . . . . . 13-12
13.6 Position Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13-15
13.7 Solving geometry problems using vectors . . . . . . . . . . . . . . 13-18
13.8 Modelling with vectors . . . . . . . . . . . . . . . . . . . . . . . . . 13-20
14 GRANDE FINALE 14-0

14.1 Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-0
14.2 Binomial expansion: turbo-charged version . . . . . . . . . . . . 14-3
14.3 Power series of e x . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-5
14.4 Power series and logs . . . . . . . . . . . . . . . . . . . . . . . . . . 14-6
14.5 Power series and the trig functions . . . . . . . . . . . . . . . . . . 14-7
14.6 p
Maclaurin’s incredible theorem . . . . . . . . . . . . . . . . . . . . 14-9
14.7 −1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-11
14.8 Euler’s equation: e i π + 1 = 0 . . . . . . . . . . . . . . . . . . . . . . 14-15
A Notes and points of interest A-1

A.1 An infinity of primes . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
A.2 The largest known prime number . . . . . . . . . . . . . . . . . . A-2
A.3 A formula for the primes? . . . . . . . . . . . . . . . . . . . . . . . A-3
A.4 More on recurring decimals . . . . . . . . . . . . . . . . . . . . . . A-4
A.5 Is 0.9̇ = 1? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-5
A.6 Complex Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . A-6
A.7 Normal numbers
p . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-6
A.8 Proof that 2 is irrational . . . . . . . . . . . . . . . . . . . . . . . A-7
A.9 Solution to the chess board puzzle . . . . . . . . . . . . . . . . . . A-8
A.10 A cautionary tale . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-9
A.11 Irrational indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-10
A.12 Combining terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-11
A.13 Factorising quadratics: a shortcut . . . . . . . . . . . . . . . . . . A-12
A.14 Difference of two squares: harder examples . . . . . . . . . . . . A-13
A.15 “Proof" that 2 = 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-14
A.16 Harder algebraic fractions . . . . . . . . . . . . . . . . . . . . . . . A-15
A.17 Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-17
A.18 Formulating linear equations: a harder example . . . . . . . . . . A-18
A.19 Simultaneous equations in more than two variables . . . . . . . A-19
A.20 Derivation of the quadratic formula . . . . . . . . . . . . . . . . . A-21
A.21 Quadratic formula with algebraic coefficients . . . . . . . . . . . A-22
A.22 Inverse functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-23
A.23 Hard problems involving surds . . . . . . . . . . . . . . . . . . . . A-25
A.24 Sets: HCF and LCM . . . . . . . . . . . . . . . . . . . . . . . . . . . A-26
Chapter 0
WELCOME
0.1 Is this book for you?

Welcome to the utterly beautiful world of advanced Maths. It will not be like
anything you have met before. That may come a surprise or as a shock. Either
way, if you stick with it it will reward you immeasurably.
This ebook is intended to cover all the Pure Maths required for current (post
2017) AS Level and Year 1 A Level syllabuses. It takes as its starting point a
good (but by no means perfect) grounding in any GCSE Maths syllabus. More
specifically, you’ll need to be comfortable with GCSE level exam questions
involving:
• Directed numbers.
• Fractions, decimals, percentages; ratios and proportions.
• Indices (e.g. x 5 × x 2 = x 7 ).
• The geometry of straight lines, triangles and circles.
• Linear equations. Rearranging simple formulae. Simple quadratic

factorising. (Simultaneous equations not required.)
• Pythagoras’ Theorem. Basic trigonometry and SOHCAHTOA.
• Coordinate axes, but not the equation of a straight line.
If this looks like quite a low baseline it is because most students have some
blind-spots, so it does no harm to recap important ideas — not least because
an understanding which may be perfectly adequate for GCSE is frequently
too flimsy for AS level. A good example of this would be fractions: doing
them in arithmetic, and understanding them sufficiently to apply the
principles algebraically, are two different things!
Importantly the book assumes that the reader is a mere mortal and would
like to have things explained in some detail. Here we part company with
many textbooks.
Chapter 1 is ideal painless post-exam reading before starting the course. It
covers the basic concepts you will need. The first few chapter after that are
very much about getting your algebra up to speed. They are particularly
aimed at UK student—the GCSE Maths syllabuses do not prepare students
sufficiently for A Level, a fact that comes as a great shock to many.
0-1
CHAPTER 0. WELCOME
0.2 Navigating your way around

This ebook is intended as a manual—for looking something up when you
want to learn about it. If you are using it in conjunction with another
textbook, the chapters won’t match. However you should be able to find your
way around without too much trouble.
If you are viewing it electronically the hyperlinks in the table of contents on
page 2 take you to the relevant page. The links in the table below take you to
the start of each chapter, where you will find a detailed summary of what
each chapter covers.
Ch Title Topics covered

1 Fundamental Concepts Numbers
Commutativity & Associativity
Set language
The language of Equations and Inequalities
Identities and the notion of Equivalence
Function notation
Reasoning and proof
Directed Numbers
2 Indices & Surds The Basic Rules
Negative indices
Rational (fractional) indices
Surds
3 Brackets & Factorising Expressions, Terms, Coefficients
Multiplying out brackets
Polynomials
Factorising and the Highest Common Factor
One-bracket factorising
Two-bracket factorising
Difference of two squares
Perfect squares
Disguised quadratics
4 Algebraic Fractions Cancelling algebraic fractions
Polynomial division Multiplying & dividing algebraic fractions
Factor Theorem Adding / subtracting algebraic fractions
Dividing polynomials (linear divisor)
Factor Theorem
5 Coordinate Geometry Linear equations and straight lines
y = mx + c
Perpendicular lines
Distance between two points
The mid-point of a line segment
The centres of a triangle
Circle properties and theorems
6 Linear Relationships Linear equations
Formulating linear equations
Changing the subject of a formula
Linear inequalities
Simultaneous linear equations
Combining Linear inequalities
Linear modelling
PURE MATHS: AS LEVEL GUIDE 0-2

CHAPTER 0. WELCOME
Ch Title Topics covered

7 Quadratic relationships Completing the square
Maximum and minimum values
Quadratic equations
The discriminant
Disguised quadratic equations
Sketching quadratics. Repeated roots
Quadratic inequalities
Inequalities with algebraic denominators
Non-linear simultaneous equations
The equation of a circle
8 Sketching and transforming Sketching polynomials
1 1
y = , y = 2 . Asymptotes.
x x
Translating, stretching, reflecting curves
Periodic functions
9 Binomial expansion Binomial coefficients and notation
Expansion of (1 + x)n and (y + x)n
Approximation using binomial expansion
10 Calculus: differentiation Rate of change and gradient
& integration Differentiation (including from 1st principles)
Increasing & decreasing functions
Tangents & Normals
Stationary points
Second derivative
Indefinite integration
Definite integration and areas
11 Exponential Equations Index Rules
and Logs Exponential equations
Quadratic exponential equations
Exponential graphs & transformations
y = e x properties & derivatives
Logarithm Laws
Solving equations using logs
Exponential modelling
Non-linear to linear transformation
12 Trigonometry SOHCAHTOA
Trig functions: angles outside 0◦ − 90◦
The sin, cos and tan of 0◦ and 90◦
Sine Rule & Cosine Rule
Area of triangle formula 12 ab sinC
Trig equations using identities
13 Vectors Column vectors & i - j notation
Magnitude & direction
Position vectors
Solving geometry problems using vectors
Modelling with vectors
14 Grand Finale Power series
Advanced Binomial Expansion
Powers series: e x ,ln and trig functions
p
−1, Complex Numbers

CHAPTER 0. WELCOME
To make it easy to pick out crucial material
• rules, formulae and key ideas are emphasised using a bold typeface
and sometimes using a box A
!
A
• particularly important stuff has this motif alongside it. A
A
The body of text covers the essentials. For those aiming for a top grade, or if
you want a bit more detail, there is a lot of optional extra material:
• Footnotes, where we think a short comment could be useful or

interesting, but we don’t want to disturb the flow of the text;
• References, usually by page number, reminding you about something

you want to be clear about. If you have downloaded the entire book you
may be able to click on the link to get to it1 .
• Harder worked examples as part of the explanation of the concepts;
• Tacked onto the exercises, problems designed to test the depth of your
understanding. You will find them harder, but more stimulating and
satisfying.
• Proofs that won’t be examined on, but you may want to see;
• Intriguing (and sometimes amazing!) diversions that relate in some

way to the topic the chapter is about.
Where this optional material is extensive it is shown in blue, so you can skip
it if you wish, or come back to it later. Exercises
1
The link will show as a question mark if you have opted to download a single chapter
and the reference is to another chapter. The ability to jump to another part of the book via a
hyperlink will depend on the platform you are using.

CHAPTER 0. WELCOME
Quick Quizes
From time to time you’ll come across a few questions, topped and tailed...
QQ
...between double lines like this
If you’ve understood what went before they’ll mostly be quick to do. None of
them are designed to trick you. They are there to help you check whether
you have understood what you have just read.
It is impossible to overstate the importance of Exercises. Whether the ones

you find here or ones your teacher sets. It is remarkably easy to read through
explanations of the principles and say “yes, that makes sense, I get it".
It is only when you tackles questions for yourself that you find out how
thoroughly you understand. The acid test will be the harder question. Once
you are comfortable with the more straightforward type, stretch yourself
with more difficult ones.
When you get stuck on one, which you inevitably will, go back over the
previous pages to see what you missed. If necessary refer back to earlier
parts of the chapter or to other chapters. There will be references to help
you.
T RY T H E QU E ST I ON
GE T ST UC K
?
@
ST U DY P REV I OU S P AGE S
?
?
RE F E R B AC K I F N EC E SS ARY
Sometimes you will find yourself going round this loop a number of times!
This is good because it means you are the boundary of your understanding;
you are breaking new ground. If you find it too easy you are probably not
learning anything very new.
The process work like this because you are training your brain to think in
ways that don’t come naturally—to any of us! Stick with it and be patient
with yourself.
We complete this introduction with some practical tips on solving Maths

problems in general, in other words some things to try when you are stuck.

CHAPTER 0. WELCOME
0.3 How your teacher solves Maths problems

If you are feeling confident and excited about studying Advanced Level
Maths, feel free to go straight on to the first chapter. But remember this is
here, and if you ever get bogged down or frustrated with the subject, come
back and take a browse. Although it’s the obvious place to put this section, it
may mean more to you if you have already encountered some of the
difficulties it describes.
For most of the Maths you have met so far the journey from problem to
solution has been a short one, and generally speaking either you know how
to find the answer or you don’t. At A Level the journey will almost always be a
longer one: you will have to think more, there will be more steps in the
reasoning and you may need to try some alternative approaches, or string
together several techniques. The upshot is you will often get stuck. In this
section we will be giving you some pointers to help you understand what is
going on.
We start with a familiar scenario. You have been attempting some

homework. Much of it is fine but there some of it you got stuck on, and you
are back in class looking to your teacher to enlighten you. She is happy to do
so. Unfortunately, when she’s finished you’re left with the thought: that’s all
very well, I followed what she did, but how was I supposed to think of doing
that? Somehow the teacher just seemed to know what to do. What’s her
secret?
Let’s “rewind the tape" and see if we can peer into her brain. To her the
process of solving the problem might have looked something like this:
“What does this problem remind me of? Have I seen one like it before?"
If it fits a familiar pattern (and usually it will), the solution follows easily. If it
doesn’t, she might study the question more carefully: "have I missed
something?" Or she may decide to draw a diagram to help her see what’s
going on. If it’s an algebraic expression she’s stuck on: “how could I write it in
a different way? Hmm? Now does that remind me of anything?"
Sometimes she will need a piece of paper; often it will just be in her head.
The important thing to recognise is that she is not just scratching her head
and hoping for inspiration, it’s a practical thing. She is doing stuff to explore
what she has in front of her. She doesn’t know for sure if any of it’s going to
help—and sometimes it doesn’t, although sooner or later if it’s a topic she’s
very familiar with she will almost certainly hit on a way through. It’s
impressive watching her performance in real time because the
thinking-about-it phase seems to be pretty much instantaneous. It’s all the
more impressive because she rounds it off by writing out the solution in a
beautifully logical and organised way. What a mind!

CHAPTER 0. WELCOME
What can we learn from this? First of all it tells us that the process of solving
a mathematical problem is likely to involve one or more of these:
• drawing a diagram ;
• looking for a familiar pattern or relationship ;
• experimenting with writing things in different ways.
Sometimes it’s a case of suck-it-and-see, of trial and error, and the concise
way the solution is written out often belies the element of randomness in the
process of getting there. Don’t be fooled into thinking otherwise!
But there is more to it than that. Maybe you did try different things when you
were attempting the problem, and you are wondering: how come what she
tried worked and what I tried didn’t? Well experience has taught her that
some alternatives are more likely to be fruitful that others; it’s much less of a
random process for her than it is for a newcomer like you. But hey, how did
your teacher get that experience? By trying things that didn’t work. A major
part of the learning process for us humans is getting our fingers burnt—in
the case of Maths that means going up time consuming blind alleys that we
then learn to avoid.
To summarise, success with Maths is not a matter of serendipitous

brain-wiring or spooky intuition, but repeated mulling over the theory and
copious amounts of practice. As in everything there is an element of raw
talent, but like being a writer or an artist, it’s a matter of 1% inspiration and
99% perspiration.
One final very practical point: use a pencil2 —one with a rubber on the end.
This is because you will make many mistakes, and crossings out distract the
eye from spotting patterns. If there’s a lot to correct, it may be better to put a
line through it and write it out again neatly. This is not for the benefit of your
teacher, (although she won’t complain); it’s for your benefit. Try it, you’ll be
surprised how helpful it is.
2
Up until the exams, when for obvious reason you are required to write in ink.

Chapter 1
FUNDAMENTAL CONCEPTS
Overview
Each chapter begins with a detailed description of its contents like the one
below, so you can skip any part you don’t need. If you are not sure what you
need the exercises will tell you.
1.1 Some of the terminology and notation you’ll need around number
systems: integers, prime numbers, rational & irrational numbers, Real
numbers. Converting a recurring decimal into a fraction.
Also a fascinating glimpse into a special type of number with mind-blowing
properties.
1.2 ∞ (infinity).
1.3 Commutativity and Associativity, two very important principles.
1.4 The language and notation of sets. The symbols ∈, ⊆, ∩ and ∪.
Subsets, union and intersection of sets
1.5 Equations, some language and an important principle.
Polynomial equations. The quadratic formula.
1.6 Inequalities. Basic principles.
1.7 Identities and the notion of equivalence (≡).
Difference between an equation and an identity.
1.8 A reminder about function notation. For example:
If f (x) = 1 + x2 , find f (−1) and solve the equation f (x) = x.
1.9 Introduction to mathematical reasoning and proof.
The symbols ⇐, ⇒, ⇐⇒ and ∴ An example of using elegant reasoning to
effortlessly solve a very difficult problem.
1.10 A review of directed numbers. For example:
Why is −4 − 6 = −10 when a rule says “two minuses make a plus"?
p
Simplifying an expression like b 2 − 4ac, if a = k, b = −k and c = −2k.
1-1
CHAPTER 1. FUNDAMENTAL CONCEPTS
1.1 Numbers
First some definitions, and a bit of background on number systems. Most of
this will be very straight forward, but behind the obvious all is not as simple
as it first appears, and you will get a glimpse of this here.
1.1.1 Natural Numbers

These are the whole numbers, or the counting numbers, and we use the
symbol N to represent them:
N = 1, 2, 3, 4, 5, . . .
The row of dots at the end is our way of saying that the list goes on for ever.
For many thousand of years these were the only numbers humans used, so
there could be no notion of subtracting a bigger number from a smaller one,
and doing calculations with fractions would be out of the question. It would
only be in relatively recent times that a symbol for nothing (i.e. zero) was
introduced, without which even simple arithmetic is extremely difficult. This
was a crucial breakthrough1 .
1
The first recorded zero is attributed to the Babylonians in the 3rd century BC. A long
period followed when no one else used a zero place holder, until the Mayans independently
invented zero in the fourth century AD. The final independent invention of zero in India was
long debated by scholars, but seems to be set around the middle of the fifth century. It spread
to Cambodia around the end of the 7th century. From India it moved into China and then to
the Islamic countries. Zero finally reached western Europe in the 12th century (see Kristen
McQuillin, January 2004). The West got there in the end.

1.1.2 Prime numbers

Those of you familiar with Prof. Marcus du Sautoy’s wonderful book “The
Music of the Primes" will need no persuading that prime numbers are utterly
fascinating. We usually think of a prime number as a natural number that
only divides by itself and one (without a remainder).
So for example 7 is a prime number and so is 23.
But 50 is not a prime because it can also be divided by 2, 5, 10 and 25. We say
it is has altogether 6 factors2 .
In fact mathematicians define a prime number in a subtly different way: it is
a number that has precisely two factors. The point in defining it in this way
is to exclude the number 1. We won’t go into the reason this for this here; all
you need to remember is that
the number 1 is not a prime number.
You are most likely to meet prime numbers in the context of finding the
prime factorisation of a number. For example 2400 = 3 × 25 × 52 .
If this is a new idea to you, find the prime factorisation of 504 (the answer is
in the footnote3 .)
How many prime numbers are there? In fact there is an infinite

number. The Greek mathematician Euclid proved this very
elegantly. To see his proof turn to Appendix A (page A-1), or
click on the page number.
On page A-2 you will find the biggest prime number so far
discovered. You will also learn about the $150,000 reward being
offered to find an even larger one.
2
The four factors we just listed plus 1 and 50
3
504 = 2 × 252 = 22 × 126 = 23 × 63 = 23 × 7 × 9 = 23 × 32 × 7
We can break the number down in any order, but it’s obviously easiest to start with
low numbers like 2, 3 etc.

1.1.3 Integers
The Integers are all the Natural Numbers, together with their negative
counterparts and 0. We use the symbol Z for these:
Z = · · · − 4, −3, −2, −1, 0, 1, 2, 3, 4, . . .
The dots at each end mean that the list extends indefinitely in both directions.
Without the negative numbers and zero, all but the most basic arithmetic is
very difficult (try multiplying Roman numerals). Once they were included,
whole new branches of mathematics started to blossom (excuse the pun).
Sometimes you will see the symbols Z+ and Z− . These stand respectively for
the positive integers and the negative integers, excluding zero:
Z+ = 1, 2, 3, 4, . . . and Z− = −1, −2, −3, −4, . . .
Notice that Z+ and Z− combined do not make up Z, because the number 0

belongs to neither. You will have realised that Z+ = N.
The integers from 0 to 9 are referred to as digits. So for example there are 4
digits in the number 3.916, and 5 digits in the number 72001.

1.1.4 Rational Numbers

As soon as we start dividing numbers we need fractions. That leads us to the
next category of numbers, the rational numbers. A rational number is any
number that can be formed by dividing two integers. We use the symbol Q
for these. It is possible to list4 all the rational numbers, although not as
simply as listing N or Z. Happily we don’t need to.
The integers involved could be positive or negative, and the top number
could be zero (obviously the bottom one can’t be zero, because it’s
impossible to divide by zero).
Some examples of rational numbers are:
2 376 1 1
, , − , 4 and 13
5 2117 3047 2
1 −1 4 1 27
because − 3047 = 3047 , 4 = 1 and 13 2 = 2 .
Most decimals that come to mind are rational numbers.

4
For example 0.4 = 10 = 25 and 3.6901 = 36901
10000 .
4 1 1 2 1 2 3 1 2 3 4
This gives you a clue as to how it can be done: 2 , 3 , 3 , 4 , 4 , 4 , 5 , 5 , 5 , 5 ...

Recurring5 decimals are also rational, a simple example being

1
0.333 . . . =
3
We can show that any recurring number is rational by demonstrating how to
convert a recurring decimal into a fraction. Take the number 0.10351035 . . .,
normally written 0.1̇035̇, the dots indicating the start and finish of the
repeating sequence. Let’s call this number x.
x = 0.1̇035̇ = 0.10351035 . . .
10, 000x = 1035.1̇035̇ = 1035.10351035 . . .
We have chosen to multiply x by 10, 000 because that gives us a new number
with exactly the same digits after the decimal point, and which therefore
differs from x by a whole number.
9, 999x = 1035
hence 1035
x= .
9999
Cancelling6 this gives:
115
0.1̇035̇ = .
1111
If this technique is new to you, try it out for yourself using the number
0.297297 . . .. The answer is in the footnote7 .
If you want to have a go at a slightly trickier one, try 0.094̇28571̇.
For the answer turn to page A-4 of Appendix A, or click on the page number.
While on the subject of recurring decimals, here is a question:

is 0.9̇ = 1?
The answer is obviously no, isn’t it. Or is it? Find out by going to page
A-5, or clicking on the page number.
5
A recurring decimal has the property that after a certain point the same digit, or
sequence of digits, is constantly repeated, e.g. 0.7777 . . . or 0.061137419419419 . . ..
6
You may recall that a number is divisible by 3 if the sum of its digits is divisible
by 3. Similarly, a number is divisible by 9 if the sum of its digits is divisible by 9.
These are the only two numbers for which this simple “sum of digits” test works.
7 11
0.2̇97̇ =
37

1.1.5 Irrational Numbers

So now we know that every fraction, every decimal with a finite number of
digits and every recurring decimal is a rational numbers. Is that it? Does that
leave any numbers out in the cold? Are there any decimals with an infinite
number of digits that never recur? And if there are, how could we possible
know thy never recurred!? The question was debated by mathematicians for
thousands of years.
It was the Greeks who showed there are such numbers, by proving that the
square roots of certain whole numbers cannot be written as a fraction, i.e.
cannot be rational. Such numbers are (not surprisingly) called irrational
numbers.
p
For a proof that 2 is irrational turn to Annex A page A-7, or click on the
page number.
We now know that if the square roots p of an integer is not itself an integer,
then it is an irrational number, e.g. 17.
Some Greeks suspected π was also an irrational number but they couldn’t
prove it. In fact it would be well into the 19th century before π was proved to
be irrational. Establishing whether or not a number is irrational, can be
fiendishly difficult (well beyond degree level Maths).
Now prepare for a surprise.
You might expect that the irrational numbers are very much the
exception, i.e. most numbers are rational. In fact you couldn’t
be more wrong. There is an infinite number of rationals and
an infinite number of irrationals. But it doesn’t end there: it
turns out that there are so many more irrationals, the only way
we can describe the difference is to say that the kind of infinity
involved is of a higher order. So there we have it: not only does
the answer turn our intuition on its head, we discover that
some infinities are bigger than others!

1.1.6 Real Numbers

So every number is either rational or irrational. Put another way, every point
on the number line will correspond to either a rational number or an
irrational number, and we refer to the combination of rational and irrational
numbers as the Real numbers. The symbols for the Real numbers is R.
Real numbers = Rational numbers + Irrational numbers
When you see the term Real number, you can (for now) just think of it as any
number.
So why add the word “Real" at all? It’s because there are numbers that are not
real. The story of how they came to be discovered (or you might say invented)
is a fascinating one. Such numbers are called Complex numbers and to find
out what these peculiar numbers are you’ll find a brief description if you turn
to Appendix A page A-6, or click on the page number.
1.1.7 Normal numbers

You don’t need to know about these, but if you want to have your mind
blown away on the subject of numbers there is a special kind of irrational
number called a Normal number. These are described on the same page in
Appendix A.
You are probably beginning to realise that there is a lot more to the idea of
numbers than meets the eye. There are still many unanswered questions
and many mathematicians devote their lives to delving ever deeper into it
Number Theory to try and find the answers.

1.2 The ∞ symbol

This is the symbol we use to stand for infinity. Infinity refers to a quantity that
is boundless or limitless. It isn’t really a number. This becomes obvious if we
think about adding another number to it: because of what is means we would
have to say: ∞+1 = ∞. Indeed ∞+∞ = ∞ or even ∞×∞ = ∞. We can go on:
for example ∞∞ = ∞.
Having a symbol for infinity is useful. For example, we can write x → ∞ (we
say “x tends to infinity") as shorthand for “the value of x becomes bigger and
bigger without limit".
We sometimes even write infinity as if it were a number. Look at the fraction
1
.
0
What’s the answer? Does it have an answer? Strictly speaking no—because
it is the answer to the question: “how many 0’s do we need to add together
to get 1?"! However many 0’s we add together we would never get there, so
we could say the answer is an infinite number. If this isn’t clear, look at these
divisions:
1 1 1 1
= 10, = 100, = 1, 000, = 10, 000 and so on.
0.1 0.01 0.001 0.0001
As the denominator gets smaller, the answer gets bigger.
More succinctly we say that as the denominator → 0 the fraction → ∞. Or,
1
in shorthand = ∞.
0
Treating ∞ as a number, what about this fraction?
1
∞
It’s asking the question “how many infinities are there in 1?" We can make
sense of his question if we use the same approach as before:
1 1 1 1
= 0.1, = 0.01, = 0.001, = 0.0001 and so on.
10 100 1000 100000
This time as the denominator → ∞ the fraction → 0. Summarising (and
using a shorthand that is strictly speaking illegal)
1 1
A
=∞ and = 0.
!
A
0 ∞ A
A
The statements are strictly speaking meaningless (because ∞ isn’t a number),

but you will find them useful memory aids.

Exercises 1
1. In each case, determine whether the number is irrational. If it is

rational write it in fraction form (i.e one integer divided by another).
(a) −11 (b) 1.5̇ (c) 0.03̇

q q
π
(d) 2
(e) 11 7
(f) 25 4
q
112
p p p p p
(g) 7 (h) 2+ 3 (i) 2( 2 + 1) − 2
1
p
(j) 0.6̇ (k) 0.1̇
2. (You may need to refer to Appendix A, page A-4.)
(a) Express the number 32.70924924924 . . . as a fraction.

(b) Cancel this fraction to its simplest form.
3. ∗ There is well-known puzzle that goes like this:
• Pick any three digit number.

• Form a 6-digit number by repeating the same three digits.
(e.g. 291 becomes 291, 291.)
• Divide the 6-digit number by 7.
(Assure the participant that there won’t be a remainder.)
• Divide the answer by 11. (Again there won’t be a remainder.)
• Divide the answer by 13. (Again there won’t be a remainder.)
• The result will be the 3-digit number the participant started with.
Show how this “trick" works.

Solutions 1
−
11
1. (a) 1 (b) 1 59 = 14
9 (c) 31 ÷ 10 = 30
1
(d) Irrational (e) Irrational (f) 52 , since 25 × 25 = 25

4
p p p
(g) 16 = 4 (h) Irrational (i) 2 + 2 − 2 = 2
q
(j) 1 ÷ 23 = 32 (k) 19 = 13
3,267,654 544,609
2. (a) 99,900 (b) 166,500
It easy to see that 2 and 3 divide into numerator and denominator. To look
for other possible factors find the prime factors of the denominator. We
could use the numerator but factorising the denominator is much easier.
None of its factors divide into the numerator. (Notice we don’t need the
complete prime factorisation of the denominator, just which primes divide
into it.)
3. The process of repeating the original 3 digits to form a 6-digit number

amounts to multiplying it by 1001. For example 291 × 1001 = 291000 + 291.
Dividing by 7, then 11 and then 13 is equivalent to dividing by
7 × 11 × 13 = 1001.

1.3 Commutativity and Associativity

Take the calculation 3 × 5 × 2. We know that the answer can be worked out in
a number of ways. For example:
3 × 5 × 2 = (3 × 5) × 2 = 15 × 2 = 30
= 3 × (5 × 2) = 3 × 10 = 30
= 5 × (2 × 3) = 5 × 6 = 30 etc...
The order in which we do the calculations makes no difference to the answer.
Being able to group the multiplications differently (as indicated by the
brackets) is called Associativity; being able to swap the order, e.g. 5 × 2 or
2 × 5, is called Commutativity.The upshot is this:
when lots of numbers are multiplied together, we can

jumble them up and multiply them in any order.
We are using these principles8 when we simplify an expression like 7y × 2,

7y × 2 = 7 × y × 2 = 7 × 2 × y = 14y
By the way, it wouldn’t be wrong to write the answer y14, but it’s conventional
to put any numbers first. Here are some more examples:
4a × 3 × 2a = 4 × 3 × 2 × a × a = 24a 2
8 × 13y × 0.125 = 13y [because 8 × 0.125 = 1]
1 1
10x × x 2 = 10 × × x × x 2 = 5x 3
2 2
1 2 2
× y × 42 = 6y
7
8
It’s not difficult to see that, like multiplication, addition is also associative and
commutative. For example
(2 + 3) + 4 = 2 + (3 + 4) (associativity); 1 + 5 = 5 + 1 (commutativity).
Can you see that the same is not true of subtraction or division? For example
(7 − 1) − 3 6= 7 − (1 − 3); 6 − 2 6= 2 − 6.
Check for yourself that the same applies to division.

QQ
Simplify these for yourself (the answers are in the footnote9 )

(a) 5 × 798 × 2 (b) 13 × 11 × 27 (c) 0.7 × 0.03 × 2000
(d ) 1.25 × 7 × 16 (e) 3a × 5 × 2a 2 ( f ) 12 y × 8
(g ) 6x(x + 1) × 13 (h) 8 × 43p × 0.25
9
(a) 7980 (b) 3 × 11 = 99 (c) 42 (d) 1.25 × 8 × 2 × 7 = 140
3
(e) 30a (f) 4y (g) 2x(x + 1) (h) 2 × 43p = 86p

1.4 Set language

The study of sets is a field in its own right and has been pivotal in some of
the most profound developments in the philosophical foundations of
Mathematics10 . For AS Level all we need are some basic concepts and the
terminology that goes with them.
There are something like eight distinct meanings of the word “set" in the
English language.
In Mathematics we use the term set simply to describe a

collection of things. The “things" might be numbers, triangles,
functions,. . ., anything we like.
We might, for example, have a set called K that consists of the numbers −7.5,
0, 4 and 99, or it might be a collection of numbers we already have a name for,
like N. We use curly brackets when listing what’s in the set.
K = {−7.5, 0, 4, 99}
N = {1, 2, 3, 4, 5, . . .}
In the case of N, we can’t list them all but we can see how they are generated
and the dots tell us the sequence goes on indefinitely. Sometimes it’s
impossible to write what’s in the set in the form of a list—even an infinite
one, for example
R = {Real numbers}
1.4.1 Elements or members of a set

If something belongs to a set we call it an element of the set or
a member of the set,
and we have a shorthand way of writing this using the symbol ∈. For example,
using the sets we talked about in the previous section
99 ∈ K , 99 ∈ N and 0.146 ∈ R.
We would say “99 belongs to K " or “99 is a member of K ", and ‘5 belongs to
the set of Natural numbers", or better “5 is a Natural number" and “0.146 is a
Real number". We occasionally come across the symbol 6∈, meaning “is not a
member of", for example −1 6∈ N and π 6∈ Q.
Set language provides a powerful tool for precise thinking.
10
If you want to know more about this, and enjoy a bit of mental gymnastics,
Google “Russell’s paradox".

1.4.2 Subsets
A subset, as the name suggests, is a set within a set. Consider the set of
integers A = {4, 5, 6}, for example. One subset of these might be the set {4, 6};
let’s label this B . Another would be the set consisting of the single number
C = {4}. The symbol for “is a subset of" is ⊆, and the statement
B ⊆ A means that every element of B is also an element of A
represented in a Venn diagram by showing B entirely inside A.
A

4

6
5 HH
Y
H
B
Can you see that we could also write C ⊆ B , and C ⊆ A?

What’s less obvious is that B ⊆ B , because every element of B is (of course)
an element of B . This is an example of sticking slavishly to the precise
wording of a definition. In this way mathematicians are rather like lawyers.
How many subsets does A have? Common sense would tell us 6: the subsets
{4, 5}, {5, 6}, {6, 4}, {4}, {5} and {6}. In fact, given what we said above,
mathematicians would also include the whole of A. Perhaps more
surprisingly, they would also include the set with nothing in it {}. This is
called the empty set and there is a special symbol for it: φ. So according to
this definition of a subset, A has 8 subsets.
How does this subset notion apply to the number sets we have defined? Well
N⊆Z⊆Q⊆R⊆C
In Venn diagram form
C
R
Q
Z
N
This hasn’t told us anything new, it just adds a bit of clarity. For example
where do the irrationals fit into this picture? They are the in the region inside
R, but outside Q.
A simple non-numerical example of a subset might be P ⊆ T , where
T = {Triangles}
P = {Polygons}
You can no doubt think of many other examples, mathematical and
non-mathematical.

Why do we use set language?
Why are we going to all this trouble to use fancy language and invent new
symbols to convey dead simple ideas? The answer is to avoid ambiguity. If
someone asked you to “pick a number between 1 and 10", they would
probably be surprised if you said “6 14 ", or “π". This is because all spoken
languages rely heavily on context to convey additional information (in this
case that the number should be a whole number). Even if the question says
“pick a whole number. . . ", it’s still not entirely clear whether the numbers 1
and 10 are allowed—the word “between" is used in both ways. If we rely on
spoken language we can’t avoid running into this kind of problem: that’s why
in mathematics we have to invent a certain amount of new language, and
define mathematical terms precisely.
If we were asking a mathematician to select the number we could ask her to
select an element of the set
{x ∈ Z : 1 ≤ x ≤ 10}.
We include a variable simply to avoid having to say “a number" (we’ve
chosen x here but we could have chosen any variable). The colon sign : is
read “given that" or “such that". The whole thing we therefore read as “the
set of integers greater than or equal zero and less than or equal to 10".
Naturally we alter the detail to convey exactly what we mean. Suppose the
solution to an inequality problem is “any number greater than 7.5" (i.e. not
including 7.5 itself), we would describe the solution as
{x ∈ R : x > 7.5}.
In this case we specify R rather than Z because there’s nothing to say we are
restricted to whole numbers. It’s not that we must always use set language,
but it is beautifully concise and precise.
For a mathematician, there’s an altogether more compelling reason: it is a
powerful tool for thinking deeply. Set language plays an absolutely central
role in Bertrand Russell’s famous book Principia Mathematica, which
established for the first time a rigorous foundation for the whole of
Mathematics.
The following section introduces two of the key concepts upon which set
theory is built.

1.4.3 Intersection and Union of sets

Intersection
The intersection of two sets A and B consists of the elements they have in
common. Those elements form a new set, which we call A ∩ B . A simple
example using the set K and N that we saw on page 1-14, i.e.
K = {−7.5, 0, 4, 99}
N = {1, 2, 3, 4, 5, . . .}
would be K ∩ N = {4, 99}.
But the answer may not be quite so obvious. The way two sets overlap will
always happen in one of three ways. We can show the possible
configurations using Venn diagrams:
A B (a) The sets have no elements in common

—there is no overlap.
We say A and B are mutually exclusive, or
A ∩ B = φ (where φ is the empty set).
A
B (b) The sets overlap.

The set of common elements (the shaded
area) we call the intersection, written A ∩ B .
A (c) One set is a subset of the other.

B The intersection is simply the smaller set:

A ∩B = B

Let’s take a simple example. Suppose the sets D, E and F are defined by
D = {0, 2, 4, 6, 8, 10}, E = {1, 2, 3, 4} and F = {6, 8}.
Starting with D ∩ E = {2, 4}, the Venn would look like this
0 E
D 6 2 1
8 4 3
10
E and F are mutually exclusive, so E ∩ F = φ and the Venn diagram would be
E 1 2 F
3 6
8
4
F is a subset of D, so F ∩ D = {6, 8} = F giving the Venn diagram
0
F
D 2 6
4 8
10

We won’t always we be able to mark the elements on the Venn diagram,

because there may be too many of them. Take these two sets
X = {x ∈ R : x > 2} and Y = {x ∈ R : x ≤ 5}
X ∩ Y = {x ∈ R : 2 < x ≤ 5}
Let’s see what this looks like on a number line diagram. X and Y are shown
by the lines with blobs and X ∩ Y is where they both cover the number line.
X ◦ H
H

Y

H
H •
0 1 2 3 4 5 6
The blobs on the ends of the lines are a neat device for indicating whether
the end point is included or not. For example the top line has an empty blob
at the end because is represents the interval x > 2, so 2 is not included; the
line below it, representing x ≤ 5, has a solid blob at the end because 5 is
included. The absence of a blob at the end of a line signifies that it extends
indefinitely at that end (i.e. to +∞ or −∞).
Using this convention we would show the set X ∩ Y like this
X ∩Y ◦ •
0 1 2 3 4 5 6

Union
The union of two sets A and B is the aggregation of all the elements in both
sets. To be more precise: the elements that belong either to A, or to B or to
both A and B .
It is written A ∪ B and once again there are three ways the union of two sets
can be formed.
A B

(a) The sets have no elements in common.

A ∪ B is the whole of A plus the whole of B .

(b) The sets overlap.

A

B A ∪ B is everything lying in A or B or both.

If we were listing the elements of A ∪ B ,

the elements of A ∩ B (the intersection)

would only appear once.
A

(c) One set is a subset of the other.

B

The union is the whole of the larger set.

In set language A ∪ B = A.

before let’s see what this looks like for the sets D, E and F , defined by
D = {0, 2, 4, 6, 8, 10}, E = {1, 2, 3, 4} and F = {6, 8}
We start with D and E . D ∪ E = {0, 1, 2, 3, 4, 6, 8, 10}.

The Venn diagram showing D and E is
0 E
D 6 2 1
8 4 3
10
The union we can think of as “aggregating". To describe it as “adding" could

be misleading: D has 6 elements and E has 4, but the set D ∪ E contains 8
element, not 10—we don’t double-count the two elements that are in both.
E and F are mutually exclusive. E ∪ F = {1, 2, 3, 4, 6, 8}.
The Venn diagram looks like
E 1 2 F
3 6
8
4
In the case of F and D, F is a subset of D, D ∪ F = {0, 2, 4, 6, 8, 10} = D

The Venn diagram is
0
F
D 2 6
4 8
10

Can you see that for the sets X and Y defined on the previous page,
X ∪ Y = R?
Does the notion of sets and set language stills seems a bit pointless to you?
Well, you’ll need to have a working knowledge of it because it will be used in
exam questions. As for using it yourself, the area where you’ll definitely need
it is in probability theory.
We’ll come across some other applications of the union and intersection of
sets later in this book. We end with a puzzle that illustrates in a light-hearted
way the issue of muddled thinking around sets.
A boy taken to see a doctor turns out to be the doctor’s son. But the
doctor is not the boy’s father. How is this possible?
It’s not a trick question in the sense that there are no step-relationships,
adoptions or IVF treatments involved, and there is only one doctor in the
consulting room. If you can’t work it out, the explanation is in the footnote11 .
QQ
Say whether each of these statements is true or false. (The answers are in the
footnote12 .)
π
(a) N ⊆ Z+ (b) 3
∈Q (c) Q ⊆ R
(d ) {2, 4, 6, 8} ⊆ {4, 6, 8, 10} (e) φ ∈ {1, 2, 3}
( f ) {all real numbers less than 1} ⊆ {all real numbers less than 2}
(g ) If x ∈ R, then x 2 ∈ Q
If you want to explore sets a little further or try some questions turn to
Appendix B page A-26, or click on the page number.
11 The doctor is the boy’s mother. The set of doctors is not a subset of the set of men.
(a) T (The two sets are identical.) (b) F (π is not a rational number.) (c) T. (d ) F.
12 (e) T (φ is the empty set and an element of every set.) ( f ) T. (g ) T (Z− ⊆ Z ⊆ Q)
p p
(h) F (Sometimes true e.g. x = 51 , x = 2, but not always e.g. x = 1 + 2.)

1.5 Equations
It’s extraordinary to think that the = sign was not invented until the 16th
Century, when today equations lie at the heart of Mathematics. Designing an
aircraft wing or forecasting the weather both involve solving equations.
Sophisticated equations underpin the so-called “financial engineering" that
underpins some aspects of the modern global economy. Sometimes the
equation is so complicated that we can only solve it by trial and error
(generally using a computer), but here we’ll be studying equations that we
can solve directly using some kind of algorithm13 .
In this section we are just reminding you of the basics. You’ll find more
through explanations later on: Chapter 5 for linear equations, Chapter 7 for
quadratic equations, you’ll meet polynomial equations again in Chapters 4
and 8, and Chapter 12 for trigonometrical equations.
1.5.1 Solving equations

Sometimes we can simply spot the solution. An example would be the
equation
2x + 1 = 13.
We are looking for a number which, when we double it and add one, gives 13.
The value x = 6 “fits” the equation. We say the solution is x = 6 .
It is much harder to spot the solution of an equation like
x3
3(1 − 5x) = x + 7 or = 12 − x 2 .
x −1
That’s when we need to learn some techniques.
The solutions of an equation are often called its roots;
the terms solution and root can be used interchangeably.
Sometimers an equation will only have one root, sometimes it will have
more than one. Occasionally it will have an infinite number.
13
A series of steps that lead to the answer.

1.5.2 The zero product principle

When an equation has more than one root, there is a very important
principle that may allow us to find them all. We’ll call it the zero product
principle.
Imagine there are two boxes. In each one there is a card with a number
written on it. We don’t know either number, but we are told their product is
zero.
? × ? =0
What could the numbers be? Think about it.

It can’t work unless one of the numbers is zero. Or both could be zero.
A very simple application is solving an equation like
5x = 0.
One of the two numbers must be zero. The left hand one clearly isn’t, so
x = 0.
A common application of the zero product principle is with quadratic
equations. For example if
x(x − 7) = 0
we can think of his as the two numbers x and x − 7 multiplied together14 . So
either
x =0 or x − 7 = 0
i.e. x =0 or x = 7.
Or we may have two brackets multiplied together, like this15
(x − 10)(x + 2) = 0 ⇒ x − 10 = 0 or x + 2 = 0
⇒ x = 10 or x = −2.
14
Being adjacent to each other means they are multiplied. For example 5a means 5 × a.
15
⇒ means “implies"

It is not difficult to see that the zero product principle applies however many
numbers are multiplied together: the product of the numbers can’t be zero
unless at least one of the numbers is zero.
For example
x(x − 2)(x − 4) = 0.
This is a cubic equation: when we multiply it out we get
x 3 − 6x 2 + 8x = 0
but we can easily solve it. The zero product principle tells us that
x = 0, x − 2 = 0, or x − 4 = 0
so it has three solutions: x = 0, x = 2 and x = 4.
This leads us onto other types of equation.

1.5.3 Polynomial equations

A type you will meet a lot is the polynomial equation, which is a general
name which some we have already met.
A linear equation is one in which the highest power of x is 1,

for example. 3(2 − x) = 5x.
In a quadratic16 equation the highest power of x is 2,

for example 3x 2 = x + 4.
In a cubic equation the highest power of x is 3,

for example x 3 + 2x 2 − 7x + 1 = 0.
In a quartic equation the highest power of x is 4,

for example 7x 4 − x 2 + 1 = 0.
and so on. Notice that some of the powers of x may be missing.
A polynomial equations can only involve terms like
1 5
5x 3 , −3x, x , 17, and so on
3
To be precise each term must be of the form ax n , where a is a fixed number
(called a constant) and n is an integer greater than or equal to zero.
Equations like
1p
2x = 32, sin x = 0.5, x+
+1 = 0
x3
are therefore not polynomial equations.
The order of a polynomial
The highest power of x we call the order of the polynomial.

So the order of a quadratic is 2, the order of a cubic is 3, and so on.
Can you see what the order of this polynomial is?
x 2 (1 − x 2 ) = 7
The answer17 is in the footnote.
17
Multiplying the bracket out gives x 2 − x 4 so the order is 4.

1.5.4 Quadratic formula

Not all quadratics can be factorised. In Chapter 7 you will learn two methods
for solving them if they don’t. One which you may have met is the quadratic
formula.
The quadratic formula is:

p A
−b ± b 2 − 4ac
!
A
x= A
2a A
where the a, b and c in the formula refer to the coefficients of the quadratic
equation when it is written in the form
ax 2 + bx + c = 0
i.e. with zero on one side of the equation.
For example
2x 2 − 4x − 3 = 0.
a = 2, b = −4, c = −3, so
p
−b ± b 2 − 4ac
x =
2a
p
4 ± (−4)2 − 4(2)(−3)
=
4
p
4 ± 16 + 24
=
4
p p
4 + 40 4 − 40
= or
4 4
We could reach for our calculator at this point, and get the answers to as
many decimal points as we wish.
Alternatively we could keep the answer exact but simplify it:
p p p p p
40 = 4 × 10 = 4 10 = 2 10
so p
4 ± 2 10
x =
4
p p
2 + 10 2 − 10
= or
2 2
p
If you are unsure about how we rewrote 40, this is covered in Chapter 2.

1.6 Inequalities
An inequality18 is the same as an equation except the equals sign is replaced
by the sign
<, ≤, >, or ≥ .
Typical wording used to describe these relationships are:
< “less than"

≤ “less than or equal to"; “no greater than"; “at most"
> “greater than"
≥ “greater than or equal to"; “no less than"; “at least".
Notice that the phrase “at least" includes the value itself. So for example a
score of “at least 10" means a score of 10 or above. Similarly a distance of at
most 50 miles means 50 miles or less.
Almost always inequalities have more solutions than equations do— usually
an infinite number. For example the solution of the inequality
2x + 1 < 7
i.e. x <3
Or in set notation {x : x < 3}.
Every single value of x smaller than 3 satisfies this inequality. And of course
there is an infinite number of such values.
We could represent the solution using a number line, where the empty circle
indicates that the value x = 3 is not included.

H
H ◦
−2 −1 0 1 2 3 4
The set {x : x ≤ 3} would be represent using a filled circle like this:

H
H ◦
−2 −1 0 1 2 3 4
18
You may occasionally see it referred to as an inequation.

Analytical vs Numerical solutions

When an equation is solved by following a series of logical steps, this is
referred to as an analytical solution.
In real life situations the equations can be so complex that often no such
method exists. In this case we have to use a so-called numerical method, the
simplest example of which is trial and error, i.e. trying out lots of values until
we hit upon a value that fits the equation or equations close enough.
Happily we can invariable do better than just trial and error, because
mathematicians have developed a battery of techniques, which, though still
basically number-crunching, home in the solution much more efficiently.
Obviously computers are used for this; you may learn about some of these
techniques in later modules.
A mathematician will always prefer an analytical solution—a general
formula that works with different numbers. For that she needs to know
what’s going on.
But before you dismiss numerical methods as “inferior", spare a thought for
the
p engineer whose clever formula tells him the dimension he is after is, say,
3(1 + sin 5π
17
) mm. If his laptop can find the answer numerically in
milliseconds, to as many decimal places as he needs, a formula giving him
the exact answer is of little value.

1.7 Identities and Equivalence

Look at these two statements:
x +1 = 7 and 2(x + 1) = 2x + 3.
They both express equality but there is a key difference.
x +1 = 7 is an equation: it only applies to one value of x, namely x = 6
2(x + 1) = 2x + 3 is an identity: it applies to all values of x
We say that the two sides of an identity are equivalent, and when we want to
make this explicit we use the sign ≡, rather than the = sign:
2(x + 3) ≡ 2x + 6
You have already met this concept in arithmetic: an equivalent fraction is the
same fraction just written a different way, so strictly speaking we should use
the ≡ sign when we write:
2 4
≡ .
3 6
The term identity is related to the word “identical"— a stronger term than
“equal". It is not essential to use the ≡ notation every time, in fact we usually
don’t. The key thing is to recognise the difference: with an equation we are
looking for the value (or values) of the variable that satisfies it. An identity
gives us an alternative way of writing something. Therein lies it’s power: it
provides a different way of looking at something.
QQ
Say which of these is an equation and which is an identity:
10 − 5x = x
3y − 9 = 3(y − 3)
x2 = 9
x(x + 1) − (x − 1) = x 2 + 1
The answers19 are in the footnote.
There is one important identity that crops up again and again in different
contexts. We will look at it now and remind you about it from time to time.
19
First and third are equations, the other two are identities.
x(x + 1) − (x − 1) = x 2 + x − x + 1 = x 2 + 1. If this is rusty don’t worry for now.

This identity is known as the difference of two squares. It can be expressed

as:
x 2 − y 2 ≡ (x + y)(x − y)
It’s greatest value is in algebraic manipulation: you’ll be expected to spot

when it would be useful to replace the left hand side with the right hand side,
or vice versa.
For example
x 2 − 9 = (x + 3)(x − 3)
We will see many more examples in the coming chapters.
The difference of two squares identity is a simple but very valuable algebtraic
tool. It can also be used in some arithmetic situations. Of course we don’t
need to use it for arithmetic calculations since we can use a calculator. Our
main purpose here is to help you embed the relationship in your mind.
Suppose we want to find the value of 2000.22 − 1999.82 .

The identity tells us that
2000.22 − 1999.82 = (2000.2 + 1999.8)(2000.2 − 1999.8) = 4000 × 0.4 = 1600.
It turns the original calculation into one we can do in our head. The identity
can also be used in reverse.
Suppose we want to evaluate 1996 × 2004.

We can rewrite the calculation as
1996 × 2004 = (2000 − 4)(2000 + 4) = 20002 − 42 = 4, 000, 000 − 16 = 3, 999, 984

Another very important identity that crops us a lot in trigonometry is this:
sin2 x + cos2 x ≡ 1
It says that if we take the sine of an angle and the cosine of the same angle and
square each of them, the two numbers add up to 1—regardless of the angle
we choose.
For example if x = 30◦ , sin x = 0.5 and cos x = 0.866025 (to 6 DP), so we have
0.52 + 0.8660252 = 0.25 + 0.75 = 1
Try an example for yourself. It’s powerful because it’s far from obvious,
although as it happens it is easy to prove.
Notice that
when we write sin2 x and cos2 x, we mean (sin x)2 and (cos x)2 .
This is the convention that is adopted. It is certainly better than sin x 2 and
cos x 2 , which would mean that it is the quantity x that is being squared.

1.8 Function notation

The more you study Maths the more you will come to value the concept of a
function. Simple though it is, few ideas are more central to mathematical
thinking.
If you are very familiar with the function notation you may be able to skip this
section. The questions at the end of the chapter will tell you if you can safely
do this.
The concept of a function is very important in Maths. A simple example20
would be:
f (x) = x + 5.
We use a formula to describe the function. For example, if we start with the
number 2, f turns it into 7.
We would write f (2) = 7, f (−4) = 1 or f (−5) = 0.
For a reason that is about to become clear, it is actually better to think of a
function as a process rather than a formula: the process in this case is that of
“adding 5" to a number.
Imagine a magic box: a number is fed in at one end (the input) and a new
number (the output) comes out at the other end.
2 - f - 7
We could make the process notion even clearer in this case by labelling the
box like this:
−4 - +5 - 1
Of course most functions will be more interesting than this one!
20
Because brackets can be used to indicate multiplication, f (x) could mean f × x.
The context will always tell you which it is.

The reason it is best to think of a function as a process is because the number

we feed in could be algebraic. (It’s still a number of course—one whose value
we don’t yet know.) The process is the same: whatever goes in we add 5 to it.
For example f (a 2 ) = a 2 + 5.
a2 - f - a2 + 5
Or f (y − 3) = (y − 3) + 5 = y + 2
y −3 - f - y +2
The important thing to notice here is that the x in f (x) simply stands for
whatever number (arithmetic or algebraic) the function is going to operate
on. It’s like a place keeper; we input whatever value we want and f operates
on it to give an output.
Bearing this in mind an expression like f ( f (2)) makes sense. The function f
is going to operate on the number f (2). We know f (2) = 7, so
f ( f (2)) = f (7) = 12.
Another way you will see the same function described is like this:
f : x → x + 5.
We shall use both notations.
By the way since the x in the formula is just a “place keeper"; we could use
any algebraic symbol we like. We could equally well describe f by saying, for
example
f (t ) = t + 5 or f : t → t + 5.
Or
f (z) = z + 5 or f : z → z + 5.
It is this that separates functions from formulae. A formula describes the
relationship between certain variables, whereas
a function decribes a process
that can be applied to any number or any variable.

Sometimes we want to think of the process in reverse. What number would

we have to input to get a certain output? For example to find x, given that
f (x) = 12.
This defines an equation: in this case the equation x + 5 = 12. The answer of
course is x = 7. Viewed as a process, we are asking what the input to f would
have to be if the output is 12.
? - f - 12
We can extend this idea to a less obvious question, for example to solve the
equation
f (x) = 3x.
In plain English what number would we have to start with if the effect of
adding 5 to it is the same as trebling it? A little bit of algebra will tell us.
Since f (x) = x + 5, the statement f (x) = 3x is equivalent to saying
x + 5 = 3x
Subtracting x from both sides of the equation, we have 2x = 5, so x = 2.5.
We can verify this is correct: f (2.5) = 2.5 + 5 = 7.5 = 3 × 2.5.

What if the function is not quite so simple?
Example
Solve the equations g (x) = 1, where
g (x) = 2x − 5
This is the same as saying 2x − 5 = 1 , in other words 2x = 6

the solution of which is obviously x = 3.
Solving an equation defined using a function (like we just did) involves

reversing the process. “What would we have to start with to get this answer?"
That reverse process we call the inverse function.
You won’t need to know any more about inverse functions at this stage, but if
want to delve a bit deeper go to page A-23 in Appendix A (or click on the page
number).

The unique nature of Mathematics

This section is recommended reading but missing it out won’t prevent you
from understanding what follows.
Who could fail to be impressed by, and hugely grateful for, the monumental
advances brought to us by scientists. From microelectronics to DNA testing
to Black Holes the history of science reflects man’s ingenuity and dogged
dedication. But none of it is fact. For thousands of years some of the truly
great minds thought the Sun went around the Earth. Then Newton came
along and gave us his beautiful Laws of Motions. Along with numerous other
developments, they enabled humans to travel to the Moon.
Then Einstein came along and told us his Laws are not quite correct—and in
some cases far from correct. We find the same story in all branches of
science. What we understand today to be “scientific truth" is simply the
latest view. We see this very clearly in the field of cosmology where new
observations are routinely causing scientists to re-think their theories.
Mathematics is different. Once a mathematical
p fact has been established, it
is immutable. The Greeks established that 2 cannot be written as a fraction
using whole numbers. If you could find a way of doing it you’d become
famous, but it isn’t going to happen, because there is a mathematical proof
that it can never be done. It’s the absolute, timeless nature of Maths that
makes it unique,and what makes the notion of “proof" so important. A
proven mathematical statement we call a Theorem, which in turn can be
used as a building block for a further Theorem, and so on. The development
of Mathematics has been largely about doing this again and again, building
ever more sophisticated bodies of mathematical knowledge that now stretch
into almost every area of our lives. The whole edifice would be vulnerable to
collapse if we had a single doubt about any one of the building blocks.
Hence the need for rigour21 .
What makes mathematicians highly sought after is their ability to bring
rigorous, clear thinking to very complex problems. Calculations may be
needed too, but in real life situations that is often the easy bit. So an essential
part of mathematical training is developing an ability to think
rigorously—that means not jumping to invalid conclusions.
For a fascinating example of how something can looks true but turns
out to be wrong, see Appendix A page A-3 (or click on the p page
number). You’ll also find a beautiful proof, on page A-7, that 2 is an
irrational number.
21
Mathematicians haven’t always regarded it necessary; it would be well into the 19th
Century, after some serious blunders, before they woke up to the fact.

1.9 Reasoning and proof

In 1637 the Frenchman Pierre to Fermat claimed to have had proved an
interesting theorem, that
an + bn = c n has no integer solutions if n > 2.
We know it works for n = 2 e.g. 3, 4, 5 or 7, 24, 25 because when n = 2 this is

Pythagoras’ Theorem and these are example of so-called Pythagorean
triples. No proof was ever found however so many of the greatest
mathematicians tried to construct one, over a period spanning several
hundred years. Some even tried to show it was not true by looking for
solution using a computer. None were found but that didn’t mean there were
no solutions. It seemed we may never know for sure one way or the other.
It’s important to understand that Mathematics is unique among all
disciplines in that it only deals with irrefutable fact. Science is the closest to
it but even there what we often regard as “scientific facts" are only
hypotheses, theories that stand until they are of disproved or replaced by
something better.
Then in 1995 the Cambridge mathematician Andrew Wiles, after 7 years
working almost entirely on his own, came up with a 129 page proof of the
theorem.
The proof uses a number off different approaches and we are going to
explore some of them here.
Direct proof
This is the commonest type: using a series of logical steps, building on
previously proved results. For example the proof that the angles of a triangle
add up to 180◦ . Or indeed Pythagoras’ Theorem.
Proof by exhaustion
We can sometimes show that somethings is true for all possible cases. An
example would be that if n is an integer either n 4 or n 4 − 1 is divisible by 56.
This seemingly impossible task is easily cracked when we see that the last
digit of the fourth power comes from raising the final digit of the original
number to the power 4. So for example if the number n ends in 3, then the
last digit of n 4 is 34 = 81, i.e. it ends in 1. All we have to do is check this for all
the integers 0, 1, 2, 3, . . . 9. When we do this we find n 4 always ends in 0, 1, 5 or
6, and the result follows.
Proof by contradiction
This can sometimes be used if we can show that the opposite is not true. If
you ask a young child what the largest number is they will often say a billion,
or a trillion, or a trillion trillion. We know of course that there is no largest
number and don’t need to prove it. But it is easy to prove, by assuming that
there is a largest number, call it N . But then N + 1 is a larger one, which
contradicts our assumption that N is the largest number. Our assumption
must have been wrong: there is no largest number.

Disproof by counterexample
Sometimes we want prove that a statement is not true. This usually when it
seems to be true, or is evidently true in most cases. We want to know if it is
always true. For example the statement “3x = kx implies that k = 3". The
counterexample would be x = 0.
Proof by induction
This is another method, which you are likely to meet if you continue to study
maths to a higher level. Conceptually it is more difficult to understand than
other methods so we won’t attempt to illustrate it here.

The symbols ⇒, ⇐ and ⇔

You may or may not have seen this notation.
The symbol ⇒ means “implies".
So we could say
“This month is June" ⇒ “This month has 30 days"
It is sometimes useful, for the purposes of discussion, to label each statement,

like this:
A ⇒ B
where A and B are statements defined by
A: “This month is June" B: “This month has 30 days"
A simple algebraic example:
x =3 ⇒ x2 = 9
Implicit in the use of the symbols is the word “definitely" or “always". The
statement on the right is definitely true if the statement on the left is true.
Clearly
The symbol ⇐ means “is implied by",
with the same definiteness implied. So we could equally well say
“This month has 30 days" ⇐ “This month is June"
or
x2 = 9 ⇐ x =3
However it clearly wouldn’t be correct to say
“This month has 30 days" ⇒ “This month is June"
or
x2 = 9 ⇒ x =3
because neither follows for certain. “This month" could equally well be April,
September or November, and x could be −3.

The ⇐⇒ symbol is for when the implication does apply both ways, e.g.
x +1 = 5 ⇒ x =4
x +1 = 5 ⇐ x =4
in which case we can replace the two statements by the single statement
x +1 = 5 ⇐⇒ x =4
and we say that x + 1 = 5 is a necessary and sufficient condition for x to be

equal to 4. The necessary part is the ⇐ and the sufficient part is the ⇒.
Another way to express this is to say “x + 1 = 5 if and only if x = 4".
An if and only if example using the months of the year could be:
“I was born in June" ⇐⇒ “my birthday is in June"
QQ
For each of these statements fill in the appropriate sign: ⇒,⇐, ⇐⇒ or none
of them.
(a)“I have no background in Physics" `‘I believe the earth is flat"

(b) “I was born on 29th of February" “I don’t have a birthday every year"
(c) “The Prime Minister is a woman" “Men and women are paid equally
for the same job"
The answers are in the footnote22 .
The symbols ∴ and ∵

In practice, when one statement follows from another, we more use the
symbol ∴ meaning “therefore", rather than ⇒. For example
“I was born in the month of June" ∴ “My birthday is in June"
Similarly, when the implication goes the other way, we can use the the
symbol ∵ meaning “because", rather than ⇐, and write:
“My birthday is in June" ∵ “I was born in the month of June"

This is no less rigorous than using the symbols ⇒ and ⇐. There is no
equivalent symbol using little dots for ⇐⇒ .
22
(a) ⇐ (b) ⇐⇒ (c) none apply.

Exercises 2
In each case see if you think the statement is true. If so, try to prove it; if not
try to disprove it.
1. N is an integer. N is odd ⇒ N 2 is odd.
2. N is an integer. N 2 is odd ⇒ N is odd.
3. a 2 = ax ⇒ x = a
4. b x = 1 ⇒ x = 0
5. The sum of three consecutive integers is dividable by 3.
6. The sum of 4 consecutive odd numbers is divisible by 8.
7. Today did Monday. My birthday is in 100 days. Show that it will fall on
a Wednesday this year.
8. The sum of two consecutive odd numbers is always a multiple of 4.
9. If two integer differ by k, their squares differ by a multiple of k.
10. Their exit a pair of prime numbers other than 2 and 3 that differ by 1.
11. The difference between any two prime numbers greater than 2 is an
even number.
1 1
12. x > y ⇒ x < y .
1
13. If n is an integer 2 n(n + 1) is also an integer.
14. Every number of the form n 2 (n + 17) is even.
15. The 100th term of the Fibonacci sequence is an odd number. [This is
the sequence formed by string with two 1’s and adding the previous two
terms to get the next. So it starts 1, 1, 2, 3, 5, 8, 13, 21, . . . .]
m2
16. If m and n are even numbers, then n is even.
17. If a and b are rational, then ab is rational.
18. If a 2 is rational, a is rational.
19. No square number ends in an 8.
20. In a class of 28 students at least two have a surname starting with the
same letter.
21. If n is an integer, n(n 2 + 2) is a multiple of 3. [Clue: every integer n can

be written in the form 3m, 3m + 1 or 3m + 2 for some other integer m]
22. If n is an integer 9n 2 + 6n is a multiple of 12.
23. If p 1 and p 2 are prime numbers greater than 2, p 1 p 2 + 1 is not a prime

number.

Solutions 2
1. True. We can write N = 2n + 1 for some other integer n.

So N 2 = (2n + 1)2 = 4n 2 + 4n + 1 = 2(2n 2 + 2n) + 1 which is odd.
2. True. Proof by contradiction. Suppose N is not odd. Then it can be

written N = 2m for some other integer m. So N 2 = 4m 2 which is even,
contradicting the statement that N 2 is odd.
3. Not true. Counterexample a = 0.
4. Not true. Counterexample b = 1.
5. True. Let the number be n, n +1 and n +2. Their sum is 3n +3 = 3(n +1).
6. True. Let the number be 2n + 1, 2n + 3, 2n + 5 and 2n + 7. Their sum is

8n + 16 = 8(n + 2).
7. True. Every multiple of 7 brings us to another Monday. 100 = 14 × 7 + 2.
8. True. An odd number can be written 2n +1 for some integer n. The next
odd number is 2n + 3. Their sum is therefore 4n + 4 = 4(n + 1).
9. True. (n + k)2 − n 2 = 2nk + k 2 = k(2n + k)
10. Not true. For any two consecutive numbers one his odd, the other even.
But all prime numbers greater than 2 are odd numbers.
11. True. All primes > 2 are odd numbers.
12. Not true. Counterexample: any values for which x > 0 and y < 0.
13. True. Either n or n + 1 will be an even number.
14. True. If n is even so is n 2 (n + 17). If n is odd n + 17 is even.
15. True. Look at the parity of each term (whether it is odd or even). It goes
O, O, E, O, O, E, O, O, E etc. Every third term is even, so the 99th term is
even and the 100th is odd.
16. Not true. Counterexample m = 2, n = 4.
17. True. Since a and b are rational there will be integers p, q, r, and s such
p r pr
that a = q and b = s . Therefore ab = q s so ab is rational.
p
18. Not true. Counterexample a = 2.
19. True. By exhaustion. Last digit of any number is 0, 1, 2, 3 . . . 9, so any

square number ends is 0, 1, 4, 9, 6, 5, 6, 9, 4, 1.
20. True. Proof by contradiction. Suppose all the surnames start with a
different letter. We would need 28 letters in the alphabet.
21. True. Proof by exhaustion. If n = 3m then the result is proved.

If n = 3m+1, n(n 2 +2) = (3m+1)(9m 2 +6m+3) = 3(3m+1)(3m 2 +2m+1).
We leave the reader to check that it also works for n = 3m + 2.
22. Not true. Counterexample n = 1.
23. True. p 1 and p 2 are odd, so their product will be odd and p 1 p 2 + 1 is
even, so is not a prime.

An elegant proof
The following problem illustrates the power of simple reasoning.
At first sight it seems impossibly hard. Take a chessboard, which has
8 × 8 = 64 squares. Now suppose we have 32 dominoes, each of which covers
two squares of the chessboard exactly. So all 32 dominoes could cover the
entire chessboard. There are, of course, a great many ways of doing this.
What if we now cut away a square from two opposite corners, and discard
one of the dominoes. The question is can we cover the (mutilated)
chessboard with the 31 dominoes? Spend a few minutes thinking about the
problem.
I can tell you that any attempt is bound to fail. For a beautifully simple
reason why, turn to Appendix A (page A-8), or click on the page number.
Apart from the sheer joy of a powerful proof, there is another point to
describing it. It’s about elegance and creativity. It has been said that a good
mathematician is a lazy one—because she will always look for an easy
method. You can start straight away to look for quick methods for yourself.

1.10 Directed Numbers

Is a section on directed numbers really necessary? Well it depends. If you
find yourself using a calculator to help you with negative numbers—yes,
definitely. Or you may be very comfortable when the numbers are arithmetic
but less so when they are algebraic. Remember there is a big difference
between knowing the rules and having real understanding.
If you believe you are are already quite at home with them, by all means skip
this section. But first do the exercises at the end, without using a calculator,
and check your answers. If you are already an expert it won’t take you long; if
not it will tell you which aspects you need to brush up on.
1.10.1 Adding and subtracting directed numbers

You will have met the number line and be familiar with using it in the
context of directed numbers.
−5 −4 −3 −2 −1 0 1 2 3 4 5
Having this structure clearly one’s mind is fundamental to our understanding

of directed numbers. Without it, statements like
−
3−7 = 4
−
and 5 < 1
−
or 3 < −1
don’t immediately make sense23 .
23
The number − 5 is less than the number 1 because it is to the left of 1 on the number
line.
If you don’t readily remember that < means "less than" and > means "greater than”,
just remember that the thin end of the sign is always next to the smaller number, and the
fat end is always next to the larger number.

We use the notation

−
5 to mean the point on the number line 5 units to the left of zero;
−5 refers to the operation of subtracting 5.
We shall see that in practice the two notations are interchangeable.
A picture of a number line is always useful, and—until it is firmly embedded

in your mind’s eye—you will often find it helpful to draw it. Sometimes you
won’t need to—the idea it embodies will be enough.
Take the calculation
−71 + 50.
We can visualise the number line and see that the answer is going to be a
negative number, and its size is the difference between 71 and 50: the answer
is −21.
The same applies when the calculation is algebraic; the number line principle
tells us what the relationship is between the two quantities even though we
don’t mark them on an actual line, e.g.
3a − 7a = −4a
−71x + 50x = −21x
QQ
Make sure you are comfortable simplifying these in your head:
−10 + 700
29x − 31x
−100y − 90y
We are now in a position to home in on some of the common errors that arise
with directed numbers. We can avoid them all if we use what I call the box
principle.
24
690, −2x, −190y

1.10.2 The Box Principle

This neat idea is really useful when manipulating expressions involving
directed numbers, especially ones with algebra terms.
We shall illustrate it using a simple arithmetic expression:

−
2 − 3 + 9.
To someone new to directed numbers this might look like one subtraction
and one addition. Or is it two subtractions?
It’s best to think of it as a series of steps, moving up and down the number
line:
−5 −4 −3 −2 −1 0 1 2 3 4 5
−2: we start at − 2;
−3: we move 3 to the left, bringing us to − 5;
+9: we move 9 to the right, which brings us to 4.
−
2 − 3 + 9 = 4.
It would be quite nice if we could rearrange the order of the numbers if we
wanted to, but it’s not obvious where the minus signs would go.
The trick is to think of each number as being

inside a box that includes the sign in front of it.
Then we are always adding numbers—some of which may be positive and

some negative.
−2 −3 +9
In this form we can rearrange the boxes in any way we like. e.g.
−2 +9 −3 or +9 −2 −3
Notice that it doesn’t matter which sort of minus sign it is. This is why we say
they are interchangeable.
If there is no sign, insert a +. e.g.

17 − 25 + 29 = +17 −25 +29 = +29 −25 +17 = 21
We don’t need to physically put the boxes in. The important thing is to “see"
the numbers in this way—with the signs in front of them super-glued to them.
After a while it becomes so instinctive it’s impossible to think in any other
way. The benefit of being able to rearrange the numbers is very apparent in
the following example, where we can combine the numbers in convenient
pairs:
−17 + 15 + 19 − 13 − 9 + 11 = +19 −17 +15 −13 +11 −9

= +2 +2 +2
= 6

Turning to an algebraic example, we use the ability to rearrange in order to

group together all the a terms and all the b terms:
−2a + 6b − a − 7b + 4a = −2a +6b −a −7b +4a

= +4a −2a −a +6b −7b
= a −b
To summarise, in any expression involving minus signs:
• See the sign in front of each number as being stuck to it. A
!
A
A
• Think: number line. A
QQ
Try this out for yourself: simplify the expression:
2y − x − 5y + 8x
The answer25 is at the bottom of the page.
25
−3y + 7x or 7x − 3y

1.10.3 Combining signs

We start by looking at a very common error. What is
−5 − 2 ?
The answer is −7. But what about the rule that says “two minuses make a
plus"? That suggests the answer is +7, or is it +10? In the following sections
we’ll see exactly when that “rule" applies.
Combining adjacent signs

Two adjacent plus signs are equivalent to a single plus sign.
++ 5 = +5.
If the signs are different, it’s equivalent to having a single minus sign.
+− 5 = -5 −+ 5 = -5 .
We can think of −+ 5 as giving away $5 and +− 5 as receiving a bill for $5.

They amount to the same thing.
That leaves the case of two adjacent minus signs.

For example suppose we are substituting a = 12 and b =− 5 into the
expression a − b.
It gives us 12 −− 5. You probably know that the answer is 12 + 5 = +17, not 7,
but why do the two minuses combine to make a plus? In other words
how come −− 5 = +5 ?

Here’s a practical example. Imagine a department store,

6
whose floors go from floor − 2 (below ground) to floor 6.
5
To get from floor 1 to floor 5, we go up 5 − 1 = +4 floors. 4
3
To get from floor −2 to floor 3, we go up 3−− 2 = +5 floors.
2
In other words 1
3 −− 2 = 3 + 2. 0
−1
−− 2 has become +2 −2
Similarly −− 3 = +3
−− 7 = +7
−− 3x 2 = +3x 2 etc
Another example would be temperature change.

Warming up from 2o C to 9o C represents an increase of 9 − 2 = 7o . Warming
up from −4o C to 6o C represents an increase of 10o
6 −− 4 = 6 + 4 = 10.
Let’s put all this together: the box principle plus combining adjacent signs.

We are now in a position to confidently simplify an expression like

this—where minus signs abound
−
7 − 5 + 17 −− 11 = −7 −5 +17 +11
= +17 −7 +11 −5 = +16.
We have used the box principle, thinking of the calculation as four numbers
added together, two of which happen to be negative. And we’ve rearranged
them to make the mental arithmetic a bit easier.
QQ
Try these examples for yourself:

−
(i) 5 + 2 −− 4 − 8
(ii) 2a +− 8a −− 10a − 3a
26
(i) −7 (ii) a

Combining signs when multiplying or dividing

Suppose the a in 3a has a negative value e.g. − 5, we would have
3a = 3 ×− 5
You probably know the answer to this is − 15, which we can easily verify:
3 ×− 5 =− 5 +− 5 +− 5 = -5 -5 -5 =− 15
Similarly − 3 × 5 = 5 ×− 3 =− 15.
It remains to deal with − 3 ×− 5. The answer can’t also be − 15, since changing
the sign of one of the numbers must make a difference to the answer. So the
answer has27 to be
−
3 ×− 5 =+ 15.
The rules for dividing directed numbers come from the same rules. For
example, from relationships like 3 ×− 5 =− 15, we can see that
− −
15 − 15 − 15
= 5, −3
= 5 and −3
=+ 5.
3
27
Or if you are not persuaded by this:
−
3 ×− 5 = −(− 5) − (− 5) − (− 5) = +5 + 5 + 5 = +15.

Summarising, the sign rules for multiplying and dividing directed numbers
are:
+
3 × 5 = 15
− −
3 × 5 = 15
− −
3 × 5 = 15
− − +
3 × 5 = 15
15 +
= 5
3
−
15 −
= 5
3
15 −
−3
= 5
−
15 +
−3
= 5
So the rule: “two minuses make a plus" only applies when
• adjacent signs are combined into one , or

A
!
A
• two directed numbers are multiplied or divided. A
A
Don’t be tempted to apply it when adding directed numbers. For example

−4 − 6 = −10

1.10.4 Dots and brackets

When multiplying integers we sometimes use
a dot to stand for a multiplication sign.
For example
2.3.5 means 2 × 3 × 5 = 30.
Depending on how you write your x’s the notation avoids possible confusion
between x and ×.
But it may also strike you as odd: how do we know that the multiplication 5.2
doesn’t mean the decimal 5.2? In practice this is never an issue: we only adopt
the notation where the meaning is unambiguous.
In the expression
1 + 1.2 + 1.2.3 + 1.2.3.4
the terms follow a pattern which clearly relates to integers. Anyway we can
see from the later terms that the dots can’t be decimal points. The dots in
this context signify multiplication.
With negative integers, we can still use a dot, for example 7.− 2, meaning 7 ×−
2, but the dot could be overlooked—and the expression misread as 7 − 2. It
would be better to use a bracket28 —and write
7(− 2).
Similarly we would use bracket writing things like this:
¡1¢¡
− 12 − 32 .
¢¡ ¢
(−6)(−8) and 2
28
The meaning of y(−2) is clear because in algebra being adjacent to means multiply.

In algebra we rarely need to use the × (times) symbol—we know that

something like 2ab means 2 × a × b.
But when there are minus signs around we have to be a bit careful.
With this expression
y × −2
we can’t just miss out the × sign—it would look like y − 2. The neatest way to
write it is using a bracket:
y(−2)
or better still write is as −2y .
Here is another example. It is better to write
(3x 2 − 1) × −6x as (3x 2 − 1)(−6x).

Better still, since it is always better to have the minus sign at the front, we
could write this:
(−6x)(3x 2 − 1) or − 6x(3x 2 − 1).
This may seem like labouring a trivial point. It isn’t: tidying up isn’t just
about housekeeping it’s about maximizing our chance of seeing
connections. So much of maths is about spotting patterns and recognising
relationships, so
visual clarity about what’s on the page is very important.
But most important of all, whatever notation you use make sure it’s
unambiguous. Not just for another reader (like an examiner), but for
yourself.

Using bracket with powers.

If we write −24 , it may not be entirely clear whether we mean −(24 ) or (− 2)4 .
The brackets dispose of any ambiguity:
−(24 ) = −16 (− 2)4 = +16.
4
Here is another example. What is the value of 23 ? Does it mean
4
(23 )4 or 2(3 ) ?
There’s a big difference29
4
(23 )4 = 212 = 4, 096 ≈ 4 × 103 2(3 ) = 281 ≈ 2 × 1024
The more algebraic our work, the more helpful all of this is, as we shall see next.
29
Measured in kilograms, the first number is about half the the mass of a bus;
the second is about half the mass of the earth.

Using brackets when substituting.

Using brackets is especially useful when substituting negative numbers.
Example 1
Substitute a = 4 and b = −2 into the expression a 3 − 3b 3 .
a 3 − 3b 3 = 43 − 3(−2)3 = 64 − 3(−8) = 64 + 24 = 84.
However sure you are of the directed number rules—and brilliant at mental
arithmetic—doing all this entirely in your head would be risky.
Example 2
2a 2 −5b−c
Substitute a = −3, b = 5 and c = −1 into the expression .
1−ac
2a 2 − 5b − c 2(−3)2 − 5(5) − (−1)

=
1 − ac 1 − (−3)(−1)
18 − 25 + 1 −6
= =
1−3 −2
= 4
Using a bracket can also be helpful when substituting in algebraic values.
Example 3
Substitute x = 3a and y = −a into x 2 + 3x y − y 2 .
(3a)2 + 3(3a)(−a) − (−a)2 = 9a 2 − 9a 2 − a 2 = −a 2 .
Example 4
p
Simplify b 2 − 4ac, given that a = k, b = −k and c = −2k.
p p
b 2 − 4ac = (−k)2 − 4k(−2k)
p
= k 2 + 8k 2
p
= 9k 2 = 3k
You will find some examples to try for yourself in the Exercises that follow.

A cautionary tale
While on the subject of combining directed numbers we take some time out
for a diversion that we think you will find intriguing.
Look at this sum:
S = 1−2+3−4+5−6+7−8+9...
Notice that it goes on for ever.

What is the value of S?
If you think you have the answer, turn to Appendix A page A-9 (or
click on the page number). I promise you a surprise!

Exercises 3
Don’t use a calculator in questions 1 and 2.
1. Evaluate the following entirely in your head (use the box principle)
(a) − 4 + 7 (b) 2 − 6 + 11 (c) −1 − 2 − 3 − 4 + 10
(d) −1 − 5 (e) −12 − 27 + 50 (f) −9 + 11 − 1
(g) 26 − 73 + 44 − 1 (h) −150 − 90 + 310 − 20
(i) 2.3 − 18.9 + 11.6 − 10 (j) 1.4 − 1.5 + 1.6 − 1.7 + 1.8 − 1.9
2. Evaluate these. Not all of these are designed to be done completely in

your head.
(a) 3 −− 2 (b) 11 +− 5 (c) −7 +− 10
(d) 4 +− 3 −− 6 (e) −7 −− 5 (f ) 1 −− 3 +− 9
(g) −240 −− 350 (h) (−3)2 (i) −32
(j) 7 − (−2)2 (k) 2(−5) − 1 (l) −(−8)(−3)

−20 −30
(m) −12 ÷ 6 (n) −4
(o) 7.5
−0.8
(p) −0.04 (q) (−1)2 − (−4)2 (r) −(−2)2 −− 5
(s) 2(−10)2 (t) (−2)5 (u) 2(−3)2 − (−4)2
(v) (−1)100 (w) (−1)101
3. Use your calculator (and the BIDMAS rule) to evaluate these.

8+0.4×10 1.63×78 1
(a) (b) (c) 2 1
11−3×3 4−1.082 +
7 5
4. Simplify these expressions (remember the box principle).

(a) 6a + 8a − a (b) 4y − 9y
(c) 3b − 7b + 20b − 2b (d) −15x 2 + 33x 2 − 5x 2
(e) −8x y + 4y x + 4x y (f ) 6a − 8b − a + 2b
(g) −3p − 10q + p − 5q + 6p (h) x 2 y + 6x y 2 − 7x 2 y
(i) −4p 3 − 3p 2 − 2p
5. In each case substitute in the values x = 2, y = 4. They start fairly gently

but don’t be fooled! You will be doing very well if you can get them all
correct first time.
(a) 3x − 5y (b) x − 10y (c) y − x 2 (d) −7x − 2y
(e) y − (3x − 1) (f ) 1 + (x − y)2 (g) 2 y − 2x (h) 2 y+x

³ ´2
(i) x1 − 1y (j) 1
x2
− y12

6. We suggest you check your answers to the previous question before

going on to this last one—where pitfalls abound. Substitute in the
values a = 3, b = −5 and c = −2.
(a) 2a + b (b) a + 2b + c (c) 1 − 3c (d) 4a − b
(e) b + c (f) ac + b 2 (g) 5a − b 2 (h) c + (b − c)2

a b−1 a−b c
(i) b+2 (j) a (k) a+b (l) b−a
1
(m) b 3 (n) 1 − c 3 (o) ab 2 c (p) 2a 2
1
(q) (−c)4 (r) 12 (a + 7) − b (s) 5c − (1 − b 2 ) (t) p 2
b −a 2
−c a a+b−c
(u) 3 (v) c (w) 2
(x) [(2a + b)7 − 3(b − 4c − a)5 ]3 (A juicy one to finish with.)
7. In each case substitute as indicated and simplify the resulting

expression as far as you can.
(a) (3x + y)2 , y = −6x
(b) p 2 + 3pq − 2q 2 , p = 2t and q = −t
(c) x 2 + y 2 , y = −x
(d) x 3 + y 3 , y = −x
(e) b(b 2 − 2a 2 ) , b = −3a

p
(f) 4x y + 2y 2 , y = 2x
x−y
(g) , x = −y
5x+4y
(h) (2a − b)(a + b)2 , a = 2t and b = −t

Solutions 3
1. (a) 3 (b) 7 (c) 0 (d) −6 (e) 11 (f) 1 (g) −4

(h) 50 (i) −15 (j) −0.3 [Combine them in pairs.]
2. (a) 5 (b) 6 (c) −17 (d) 7 (e) −2
(f) −5 (g) 110 (h) 9 (i) −9 (j) 3
(k) −11 (l) −24 (m) −2 (n) 5 (o) −4
(p) + 80
4 = 20 (q)1 − 16 = −15 (r) −4 + 5 = 1 (s) 200
(t) (−2)2 (−2)2 (−2) = (+4)(+4)(−2) = −32
(u) 2(9) − (16) = 2 (v) 1 [50 pairs of (−1) × (−1)]
(w) −1 [(−1)100 × (−1)]
3. (a) 6. You can check this by hand.
(b) 44.89872 = 44.9 (3 SF) (c) 0.7714285 = 0.771 (3 SF)
4. (a) 13a (b) −5y (c) 14b (d) 13x 2 [Box principle.]
(e) 0 (x y = y x) (f) 5a − 6b (g) 4p − 15q (h) 6x y 2 − 6x 2 y

(i) −4p 3 − 3p 2 − 2p [The powers of p all differ.]
5. (a) 2 (b) −38 (c) 0 (d) −22 (e) −1

1 3
(f) 5 (g) 12 (h) 64 (i) 16 (j) 16
6. (a) 1 (b) −9 (c) 7 (d) 17 (e) −7
(f) 19 (g) −10 (h) 7 (i) −1 (j) −2

1
(k) −4 (l) 4 (m) −125 (n) 9 (o) −150
1 1
(p) 18
(q) 16 (r) 10 (s) 14 (t) 2
(u) 9 (v) −8 (w) 1 [20 = 1]
(x) 1 [(2a + b)7 = 17 = 1 and (b − 4c − a) = 0]

[In part (p) only the a is squared]
7. (a) (−3x)2 = 9x 2
(b) (2t )2 + 3(2t )(−t ) − 2(−t )2 = 4t 2 − 6t 2 − 2t 2 = −4t 2
(c) x 2 + (−x)2 = 2x 2 (d) x 3 + (−x)3 = 0
(e) (−3a)(9a 2 − 2a 2 ) = (−3a)(7a 2 ) = −21a 3

p p
(f) 8x 2 + 2(4x 2 ) = 16x 2 = 4x
−y−y −2y
(g) = = 2
−5y+4y −y
(h) (4t − (−t ))(t − 2t )2 = (5t )t 2 = 5t 3

Summary of key ideas

• Integers are whole numbers. Prime numbers are integers having
exactly two factors. So 1 is not prime numbers.
a
• A rational number has the form b , where a and b are integers.
1 1
• Infinity is an undefined number but we write 0 = ∞ and ∞ = 0
• Associativity and commutativity: the order can be changed when

multiplying algebraic numbers.
• The language of sets and the meaning of intersection and union.
• Solving equations. The zero product principle. Polynomial equations.

The quadratic formula. “Root" means the same as solution. The
difference between an equation and an identity.
• The notion of an inequality.
• Definition of a function. A function as a process not a formula.
• Methods of proof: direct, by exhaustion, by contradiction. Disproof

by counterexample.
• Directed numbers. The importance of the number line. The box

principle. “Combining signs" in different contexts e.g. −− 3 = +3, and
−2 − 5 = −7 but −2 × −5 = +10. Using brackets and dots when
multiplying.

Chapter 2
INDICES AND SURDS
Overview
2.1 The basic rules of indices. Brackets used with indices. Simplifying
expressions like
¡ x ¢4
10a × 12 a 2 , 6 3
2.2 Negative indices. The meaning of 2−3 , 5x −2 , 4−5y , a −x

2−3 12 ¡ 3x ¢−2
and simplifying expressions like , , 18 .
5 4−2 5
2.3 Rational or fractional indices. Evaluating expressions such as
4 ¡ 9 ¢1 p 10 p
3 3 1
83 , 25
2
, ( 2) , 8a 9 , 32− 5 , −1
.
9 2
Writing expressions involving roots in index form. For example
q
p
p1 = 3− 2 , 3 1 −2
5 1
( 2)5 = 2 2 , 2= x 3
3 x p
What would it mean to have an irrational index such as 7 3 or 10−π ?
2.5 Surds. What is the point of them? Combining, decomposing and

rationalising surds to simplify expressions like these:
p p p 2 p p p
p18 , 4 7 + 2(1 − 7), (1 + 3) , 6 + p12
, 75 × 48
( 3)4 6
p p
p p p
p p 24
, 2 27 + 12 − 75 p5−1 , 1p
+ 1p
3× 128 5+1 2− 3 2+ 3
2-1
CHAPTER 2. INDICES AND SURDS
The Basic Rules of Indices

Introduction
Indices provide a shorthand notation for whenever a number is multiplied
by itself:
25 = 2 × 2 × 2 × 2 × 2
a3 = a × a × a
In the first example 2 is the base and 5 is the index (plural indices). The
index is sometimes called the exponent and it tells us how many times the
base must be multiplied by itself. (We’ll need a different way of looking at it
later but for now this works.)
Expressions like 72 and 73 we read as “7 squared" and “7 cubed". There are
no corresponding words for 74 , 75 , etc.; we simply say “7 to the power 4", and
“7 to the power 5" (or just “7 to the 5").
Students occasionally confuse 3a and a 3 . To be absolutely clear
3a = 3 × a = a + a + a whereas a3 = a × a × a
(3 lots of a)
We have also met the idea of a root, for example this square root
p
16 = 4.
The connection with finding a power is clear: 16 = 42 .
In this chapter we will see that roots can be expressed as indices in the same
way that powers can.

2.1 Multiplying and dividing powers of the same

base
Take something like
24 × 23
where we are multiplying together two powers of the same number (i.e. the
same base).
You may know a rule for writing the product as a single power of 2. We can
see what the rule must be if we write out what we mean by 24 and 23 .
24 × 23 = (2 × 2 × 2 × 2) × (2 × 2 × 2) = 27
When the bases are the same we simply add the powers.
24 × 23 = 27 .
Similarly
25 2 × 2 × 2 × 2 × 2
= = 22
23 2 × 2 × 2
When dividing we subtract the powers (the top power minus the bottom
power)
25
= 22 .
23
In general
a m × a n = a m+n
am
= a m−n
an
If there are more than two terms multiplied, or we have a combination of

multiplication and division, we apply the rules several times.
k4 × k2 × k = k7 (because k = k 1 )
37 × 32
= 35
34
2x × 24x = 25x
107 × 102
= 104
10 × 104

a 0 = 1.
This follows directly from the rule for division. For example1
25 25
= 25−5 = 20 but =1
25 25
so 20 = 1.
We can see that a 0 = 1 for any a 6= 02 .
In words:
A
any number to the power of 0 is 1.
!
A
A
A
1
More generally
an an
= a n−n = a 0 but =1
an an
so a 0 = 1.
2
The “number" 00 is meaningless; mathematicians say it is “undefined".

Brackets, Commutativity and Associativity

We make no apology for reminding you of the principles of commutativity
and associativity that we met in Chapter 1. (We’ll be seeing how they apply
specifically to indices.)
Take the calculation 3 × 5 × 2. We know that the answer can be worked out in
a number of ways. For example:
3 × 5 × 2 = (3 × 5) × 2 = 15 × 2 = 30
= 3 × (5 × 2) = 3 × 10 = 30
= 5 × (2 × 3) = 5 × 6 = 30 etc...
The order in which we do the calculations makes no difference to the

answer.
Being able to group the multiplications differently (as indicated by the

brackets) is called Associativity; being able to swap the order, e.g. 5 × 2 or
2 × 5, is called Commutativity.
When lots of numbers are multiplied together,

we can jumble them up and combine them in any order.
The same principle applies to addition, of course, which is also associative

and commutative. Obvious though this is, we set it out below in the same
way as with multiplication so we can see the parallel:
3 + 5 + 2 = (3 + 5) + 2 = 8 + 2 = 10
= 3 + (5 + 2) = 3 + 7 = 10
= 5 + (2 + 3) = 5 + 5 = 10 etc...

The rule applies equally to algebraic numbers, of course.

We are using these principles3 when we simplify an expression like 7y × 2,
7y × 2 = 7 × y × 2 = y × 7 × 2 = 14y
Some more examples:
4a × 3 × 2a = 4 × 3 × 2 × a × a = 24a 2
8 × 13y × 0.125 = 13y
1 1
10x × x 2 = 10 × × x × x 2 = 5x 3
2 2
1 4
ab × 24ba = 3a 2 b 5
8
You will have met this idea when multiplying out brackets. For example
3a(4 + 5a) = 3×a ×4 + 3×a ×5×a

= 12a + 15a 2
and
2x y(y 2 − 3x) = 2x y 3 − 6x 2 y
By all means write the × sign if it helps, but try to get used to omitting it; it
will make life easier in the long run. Not least because the variable will often
by x, and ×x can look a bit confusing.
3
It’s not difficult to see that, like multiplication, addition is also associative and
commutative. For example:
(2 + 3) + 4 = 2 + (3 + 4) (associativity); 1 + 5 = 5 + 1 (commutativity).
Can you see that the same is not true of subtraction or division? For example:
(7 − 1) − 3 6= 7 − (1 − 3); 6 − 2 6= 2 − 6.

QQ
(a) 3a × 5 × 2a 2 (b) 12 y × 8 (c) 6x(x + 1) × 13 (d) 8 × 43p × 0.25
Make sure you are thoroughly familiar with these rules before moving on.
4 (a) 30a 3 (b) 4y (c) 2x(x + 1) (d) 2 × 43p = 86p

2.2 The product and quotient rules for indices

There are two very useful rules that apply when a product or a quotient is
raised to a power. The rules are summarised below but first we’ll see why they
work. Look at these two examples.
(ab)4 = ab × ab × ab × ab = (a × a × a × a) × (b × b × b × b) = a 4 b 4
³ a ´3 a a a a3
= × × = 3
b b b b b
You’ll see we are using the ideas of commutativity and associativity, in other
words when we have lots of numbers multiplied together we can jumble
them up in any order to do the calculation.
In general
(ab)n = anbn A
¡ a ¢n n
a
!
A
= A
b bn A
Some more examples:
(5y)2 = 52 y 2 = 25y 2
¡ 1 ¢3 13 1
= =
2 23 8
¡ x ¢4 x4 x4
= =
3 34 81

Pitfall: don’t be tempted to apply a rule like this if the bases are added or
subtracted:
(y − 1)2 6= y 2 − 1 (x + 3)2 6= x 2 + 9 A
!
A
(a + b)n 6= a n + b n (a − b)n 6= a n − b n .
A
A
We can easily see that they are not equivalent.

For example if y = 2, (y − 1)2 = 1 but y 2 − 1 = 3.
This is not to say they won’t sometimes be equal. For example if a and b
happen to be equal, (a − b)2 and a 2 − b 2 will also be equal.
Spend a few minutes mulling over these, substituting in your own numbers,
maybe trying to find values for which they are the same.

2.3 Raising one index to the power of another

From our understanding of what a bracket means:
(24 )3 = 24 × 24 × 24 = 212
In general
(a m )n = a mn
A
!
A
A
A
For example:
(x 2 )5 = x 10 and (3a )2 = 32a
Here are some more examples. Make sure they make sense to you. You may
want to write down examples of your own, until they start to “look" right.
(35 )2 = 310 (b 2 )4 = b 8
(410 )10 = 4100 (7a )2 = 72a

2
(x 5 ) y = x 5y (2x )x = 2x
The key thing is not to confuse this with multiplying powers of the same base:
b 2 × b 5 = (b × b) × (b × b × b × b × b) = b 7
whereas
(b 2 )5 = b 2 × b 2 × b 2 × b 2 × b 2 = b 10 .
QQ
Try simplifying these (leave your answers in index form):
(i) (32 )4 (ii) (x 2 )7 (iii) x 2 × x 7 (iv) (a b )3
5
(i) 38 (ii) x 14 (iii) x 9 (iv) a 3b .

A common pitfall
The mistake hinges on the principle that
A
!
A
an index applies to what it’s attached to. A
A
In the following example the index is attached to a bracket, so the index

applies to the whole bracket:
(2x 2 y)3 = (2x 2 y) × (2x 2 y) × (2x 2 y)

= 2 × 2 × 2 × x2 × x2 × x2 × y × y × y
= 8x 6 y 3
Or looked at another way, the index applies to everything inside the bracket:
(2x 2 y)3 = 23 × (x 2 )3 × y 3
= 8 × x2 × x2 × x2 × y 3
= 8x 6 y 3
We have cubed the 2, cubed the x 2 and cubed the y.
It’s important to be very clear about this.

Using a combination of the index rules

Often we need to use a combination of the rules. Some examples:
4a 3 × (3a)2 = 4a 3 × 9a 2 = 36a 5
(x y)4 × x 3 y = x4 y 4 × x3 y = x7 y 5
(pq 2 )3 = p 3 (q 2 )3 = p3q6
¡ t ¢3 t3 t3
= =
2s 23 s 3 8s 3
52 x 2 25x 2
³ ´2
5x
= =
y4 y8 y8
(x 3 )3 x9
= = x5
x4 x4
24(n 3 )2 24n 6
= = 3n 3
(2n)3 8n 3
¡ a ¢2 a 2 3a 4 a6
× 3a 4 = × =
3 9 1 3
You might want to test yourself by covering up all but the left hand column.

QQ
(a) 32 × 35 (b) 25 × 24 (c) x 4 × x a (d ) 4x × 4x
58 x9 3x a 3 ×a 5
(e) 52
(f ) x4
(g ) 32
(h) a4
(i ) (24 )3 ( j ) (a 3 )2 (k) (43 )x (l ) (b y )2
12x 2 (y 2 )6
(m) y 3 (3y)2 (n) (4a)2 2a (o) (2x)2
(p) y4
If you want a juicy example to try, have a go at simplifying this:
(4y)4 × x y 3
.
(2x y)6
The answer is in the footnote7 .
6
(a) 37 (b) 29 (c) x 4+a (d) 42x (e) 56 (f ) x 5 (g) 3x−2 (h) a 4
(i) 212 (j) a 6 (k) 43x (l) b 2y (m) 9y 5 (n) 32a 3 (o) 3 (p) y 8
7
Seeing that 4 = 22 helps with the cancelling: 44 = (22 )4 = 28 ,
(4y)4 ×x y 3 44 ×y 4 ×x y 3 28 x y 7 2y
so 6 = 6 6 6 = 6 6 6 = 5
(2x y) 2 ×x y 2 x y x

Some powers worth memorising

Apart from 10, the bases you will meet most often are 2, 3, 4 and 5, so it will be
extremely useful if you could have at your fingertips the powers in the table
below (powers of 4 can be derived from the powers of 2):
n 0 1 2 3 4 5 6 7 8 9 10
2n 1 2 4 8 16 32 64 128 256 512 1024
3n 1 3 9 27 81 243 729
5n 1 5 25 125 625
Take for example the equation
2x = 1024.
Being familiar with the common powers of 2 tells us immediately that x must
be 10.
And it is extremely useful to be able to spot immediately the solution to an
equation like
x 5 = 243.

Exercises 4
1. Simplify these.
(a) b 2 × b 5 (b) y 4 × y 2 × y (c) 3a × 34 (d ) 2x × 23x
(e) 10b × 10c ( f ) ab × ac (g ) 50 × 51 × 52 (h) t x × t x+1
a7 53 ×56 2a ×24 36x

(i ) (j) (k) (l )
a3 57 23 32x
x a 6y 10a+4 5x+2
(m) (n) (o) (p)
xa a 3y ×a 2y 10a+1 5x−1
2. Simplify these.
(a) (43 )6 (b) (102 ) y (c) (3x )4 (d ) (a 2 )x
(e) (y a )b ( f ) (5 y ) y (g ) (a 3 )2 × a 5 (h) (2b)3 × b 2

¡ y ¢3
(i ) (x 2 y)4 ( j ) (2p 3 )2 × p (k) y 2 × (x 2 y)3 (l ) 2 × 24
3. Show which of these expressions could be another way of writing

(pq 2 )x .
(a) p x (q 2 )x (b) p x q 2x (c) p x (q x )2 (d ) (pq x )2
(pq)2x
(e) (pq)2x ( f ) (pq)x q 2 (g ) (pq)x q x (h) px
4. In each case calculate the value in your head. (The purpose is not just
to find the answer but to practice exploiting the commutativity and
associativity properties of multiplication in the process.)
(a) 0.25 × 17 × 4 (b) 8 × 32 × 12.5 × 60
(c) (0.02)3 × 0.897 × (500)3 (d ) 160 × 49 × 0.625.
Hints. (i) To multiply something by 4, double it twice. To multiply

it by 8, double it 3 times; to multiply by 16, double it 4 times,
and so on.
(ii) In part (c) multiply the outer two numbers together first.
5. Write each of these expressions in their simplest form:
(a) 2y × 8y × 0.125 (b) 4a 3 × 12 a (c) 3x y × y 2
(d ) 31 p 3 × 4p × 6 (e) (7a)2 ( f ) (2x y)3
(g ) (5p 2 q)2 (h) 2(bc 4 )5 (i ) (x y 2 ) × (3x y)2

6. (a) Find the area of this triangle, expressing your answer in its simplest
form.
hhhh
1
x
h hhhh
2 hhhh
h hhh
4x
(b) Find the volume of this cuboid, whose sides are of length 4y, 12 y
and 32 y. Express each answer in its simplest form. (The diagram is not
to scale.)
HH HH
1
3
y H
HH
H
HH
H H
HH
5H
2
y HH
H
12y
7. If y = 3x , identify which of these expressions is equal to y 2 . Explain why

in each case.
2
(a) 6x (b) 32x (c) 9x (d ) 92x (e) 3x
8. Solve these equations.
(a) 2x = 32 (b) 3x = 243 (c) 2 y−1 = 1024 (d ) 2x + 5x = 133
9. Say which of these statements is incorrect, and explain why.
(a) (x + 2)2 = x 2 + 4 (b) (2a)2 + a 2 = 5a 2
(c) (3p)2 × 6p = 18p 3 (d ) (y + 1)2 − 1 = y 2
(e) (5x y 2 )2 + 10x 2 y 4 = 35x 2 y 4 ( f ) (4y 2 )3 = 4y 6
(g ) 2x × 3 y = 6x+y
a
10. If 2a+b = 1, and b 6= 0, find the value of b .

Solutions 4
1. (a) b 7 (b) y 7 (c) 3a+4
(d) 24x (e) 10b+c (f ) a b+c
(g) 53 (h) t 2x+1 (i) a 4
(j) 52 (k) 2a+1 (l) 34x
(m) x 1−a (n) a 6y−5y = a y (o) 10(a+4)−(a+1) = 103
(p) 5(x+2)−(x−1) = 53
2. (a) 418 (b) 102y (c) 34x

2
(d) a 2x (e) y ab (f ) 5 y
(g) a 11 (h) 8b 3 × b 2 = 8b 5 (i) x 8 y 4

y3
(j) 4p 6 × p 1 (k) y 2 × x 6 y 3 = x 6 y 5 (l) 8 ×24 = 3y 3
3. (a) and (b) from the rules on page 2-10. (c) Because q 2x = (q x )2
(f) Because (pq)x q x = p x q x q x = p x q 2x .

p 2x q 2x
(h) Since (pq)2x = p 2x q 2x , we have px
= p x q 2x
4. (a) 17 (b) 4000 [e.g. 8 × 12.5 × 32 × 60]
(c) 897 [e.g. 0.897 × (10)3 ] (d) 490 [e.g. (0.625 × 4) × 40]
5. (a) 2y 2 (b) 2a 4 (c) 3x y 3 or 34 x y 3 (d) (a 2 )x
(e) 49a 2 (f) 8x 3 y 3 (g) 25p 4 q 2 (h) 2b 5 c 20
(i) x y 2 × 9x 2 y 2 4 = 9x 3 y 4
1 1 1 5
6. (a) Area = 2 × 2 ×4x = x 2 . (b) Volume = 3 × 2 ×12 × y 3 = 10y 3 .
7. (b) From the rule on page 2-10. (c) Because 32x = (32 )x = 9x .
8. (a) x = 5 (b) x = 5 (c) y = 11 (d) x = 3
9. (a) Incorrect. (x + 2)2 means (x + 2) × (x + 2), not x 2 + 22 .
(b) Correct. (2a)2 = (2a) × (2a) = 22 × a 2 = 4a 2 .
(c) Incorrect. (3p)2 = 9p 2 . Correct answer is 54p 3 .
(d) Incorrect. Same reason as (a). (e) Correct.
(f) Incorrect. (4y 2 )3 = 43 (y 2 )3 = 64y 6
(g) Incorrect. The bases are not the same.

a
12. b
= −1.
20 = 1, so a + b = 0. Now divide every term by b (legal because

0 a
b 6= 0). Since b = 0 we have b + 1 = 0.

2.4 Negative indices

[In the AQA syllabus the rest of this chapter falls within Core 2.]
In describing the rules in the first section, we glossed over something rather
important. Suppose we want to simplify
24 am
using the rule = a m−n
27 an
The answer would be 2−3 . Is this complete nonsense? How can we multiply 2
by itself − 3 times? We are going to explore this conundrum in two ways. First
by drawing up a table of powers of 2.
n -5 -4 -3 -2 -1 0 1 2 3 4 5
2n 1 2 4 8 16 32
Notice that we have filled in the value for
n=0: 20 = 1
because saw on page 2-4 that any 8 number to the power 0 is equal to 1. In
fact this gives us a clue about completing the rest of the table. Going from
left to right the next number is obtained by multiplying the previous number
by 2. Going from right to left, therefore, we need to divide by 2 to get the next
number. That means we can fill in the blanks:
n -5 -4 -3 -2 -1 0 1 2 3 4 5
1 1 1 1 1
2n 32 16 8 4 2 1 2 4 8 16 32
So we see that negative indices give us fractions, for example:

1 1
2−3 = =
23 8
It means we can always get rid of a negative index by writing it underneath 1
and dropping the negative sign. This isn’t intuitive—it looks as if it ought to
be a negative number.
For example 1 1
5−2 = 2
=
5 25
In general (for any base and any index):

A
−n 1
a = n !
A
a
A
A
Of all the index rules this is probably the least intuitive, so there is another
way we can look at it.
8
The only exception to this rule being when the base is 0.

We know that, for any m and n
a m × a n = a m+n
Putting m = 3 and n = −3, for example
23 × 2−3 = 20 = 1
It follows that
1 1
2−3 = 3
=
2 8
just as before.
Some more examples:
1 1
7−2 = 49 , 10−2 = 0.01, a −3 =
a3
1 1 1
x −1 =
x
, 3−2y = 32y , (b + 3)−2 = (b+3) 2
QQ
Now do a few for yourself. Write each of these in a form that doesn’t use a
negative index:
(i) 4−2 (ii) 2−5 (iii) y −2 (iv) 3−x

Allowing the index to be negative opens up some new possibilities.
9 1 1 1 1
(i) 16 (ii) 32 (iii) y2
(iv) 3x .

Going back to the rule

am
= a m−n
an
the power on the bottom can be bigger than the power on the top. For
example
a3 1
5
= a 3−5 = a −2 = 2 .
a a
This makes perfect sense because
a3 a ×
a ×a 1
= =
a5 a × a × a × a a2
a ×
And taking it a step further, all the index rules we learned apply equally to
negative indices. For example
1
x 2 × x −7 = x −5 =
x5
1
a −1 × a −3 = a −4 =
a4
x −2
= x −2−6 = x −8
x6
x −2
= x −2+6 = x4
x −6
1
(y −2 )4 = y −8 =
y8
1
(2x )−3 = 2−3x =
23x
QQ
Do a few for yourself. Simplify these:
x −1
(i) y −2 × y −7 (ii) 2−5 × 26 (iii) (iv) (y 4 )−2 (v) (3−a )−2
x −7
We are now going to learn a very useful trick. This will be particularly useful
when dealing with rather more complicated expressions involving negative
indices.
10 1
(i) y −9 (ii) 2 (iii) x 6 (iv) (v) 32a .
y8

More on negative indices within fractions

When we come across a negative index, it’s usually a good idea to to change it
into a positive one. Doing so will introduce a fraction, as we have seen. If the
negative index already appears within fraction, however, things can get a bit
messy—like in this example
8−2 × 23
2−2
There is a very easy way of dealing with this situation, provided we think of
it in a slightly different way. We’ll illustrate how using 5−2 , by writing it as a
fraction (with denominator 1):
5−2 1
= 2.
1 5
We see that
turning a fraction upside down changes the sign of the index.
The beauty of this is simple rule is that it doesn’t matter if the negative index
is initially on the top or the bottom 11 :
1 52
= .
5−2 1
We can use this principle to effortlessly simplify the example we started with:
8−2 × 23 22 × 23 32 1
= = = .
2−2 82 64 2
This process doesn’t have a name. I will call it the switch rule.
QQ
Try simplifying these:
3 2−3 x −4 ×y 3
(i) (ii) (iii) .
4−2 −2
5 ×7 y −2 ×x
The answers 12 are in the footnote.
11 5−2 1
Turn both fractions upside down in = 2.
1 5
1 1 52
Or −2 = 1 ÷ 5−2 = 1÷ 2 = 1× .
5 5 1
12 25 y5
(i) 48 (ii) (iii) 5
56 x

Still more on negative indices within fractions

Now we take the idea a step further. In the following examples we have
fractions raised to negative indices.
¡ a ¢n a n
Combining the switch rule with the quotient rule13 i.e.
b
= bn
µ ¶−5
1 1−5 25
= −5 = 5 = 32
2 2 1
µ ¶−2
5 5−2 32 9
= −2 = 2 = =
3 3 5 25
This boils down to another simple rule. Look at the second example
µ ¶−2 µ ¶2
5 3
=
3 5
If we have a fraction raised to a negative power, all we do is drop the minus

sign from the index and turn the whole fraction upside down. We can think of
this as the switch rule applied to fractions.
Summarising:
5−2 1 1 23 3 3×7 3 3 × 42
= , = , = , =
2 2 × 52 2−3 1 7 −1 1 5 × 4−2 5
µ ¶−5 µ ¶5 µ ¶−2 µ ¶2
1 2 5 3
= , =
2 1 3 5
13
The quotient rule works all indices, whether positive or negative:
³ a ´−m 1 ³ a ´m am b m b m a −m
= ¡ a ¢m = 1 ÷ = 1 ÷ m = 1 × m = m = −m .
b b
b b a a b

Exercises 5
1. Write each of these as a single number, or if it algebraic, in a form that

doesn’t use a negative index.
(a) 3−2 (b) 5−1 (c) 10−4 (d ) a −5
(e) y −2 ( f ) x −1 (g ) 4−n (h) 7−2n
(i ) 3−a ( j ) 10−(n−1) (k) 10−n−1 (l ) 2−x+3
(m) t −n (n) p −a (o) q −(y+3) (p) q −y+3
2. Using negative indices, write each of these in a way that doesn’t require
the use of a fraction. Simplify the answer where possible.
1 1 1 1 1
(a) 34 (b) 210 (c) x 5 (d ) y 2 (e) 10x
1 1 1 43 2×27
( f ) 5y (g ) 62a (h) 7 (i ) 49 ( j ) 211
3x ay 1 16
(k) 35x (l ) a 2y (m) (2 y )3 (n) (2x)4
3. Solve these equations to find the value of x.

1 1
(a) 9 = 3x (b) 1000 = 10x (c) 0.01 = 10x
1 1
(d ) 16 = 2x (e) 0.5 = 2x+1 ( f ) 9− 2 = 3 x
1 1
(g ) 85 = 23x (h) 256 = 5x (i ) 0.125 = 2x
( j ) 0.04 = 5x (k) 0.3̇ = 3x
4. Write each of these in a form that doesn’t use any negative indices.
Simplify your answer where possible.
2−3 3−1 4×5−2

(a) 5 (b) 2 (c) 3
(d ) 2−1 × 3−2
1 1 6
(e) 10−3 × 7 ( f ) 5−2 (g ) 4−3 (h) 3−1
10 2 5 4
(i ) 2−3 ( j ) 3×4−2 (k) 7−1 ×2 (l ) 3−2 ×2−1
5−2 7−1 3−2 ¡ 3 ¢−2

(m) 2×3−1 (n) 2×4−2 (o) 5−2 (p) 5
¡ ¢−4
(q) 23
5. Write each of these expressions in the form ax n where a is a constant.

¡ x ¢3 (3x)3
(a) (2x)4 (b) 16 2
(c) 6
¡ 1 ¢2 7 ¡ 4 ¢3
(d ) 4x (e) x 2 (f ) x
¡ 5x ¢2 ³ ´−2
1 −1 6
¡ ¢
(g ) 3
(h) 2x
(i ) x2
¡1 ¢−1 ¡ x ¢−2 ³ ´2
1
(j) 3
x −1 (k) 7
(l ) x −2
¡ 1 ¢−1 (6x)−1 ¡ 3 ¢−4 ¡ 3 ¢4

(m) 2x (n) x −2 (o) x × x

6. Simplify each of these, leaving your answer in index form.

(There should be no fractions in the answer.)
(a) 3−2 × 3−3 (b) (4−2 )3 × 45 (c) (5−1 )−3
(d ) (x −4 )5 (e) (y −2 )−5 ( f ) (2−x )−3
32 6−1 6−1
(g ) 3−4 (h) 64 (i ) 6−4
x −3 1 2−1 ×2−5
( j ) x −1 (k) a 3 ×a −5 (l ) 23
(p −2 )2 ×p −1
(m) p
7. Say which of these expressions is equal to x 6 , giving your reasoning.
1 x4
(a) (x 3 )3 (b) x −6 (c) x −2
1 (2x)6
(d ) (x −2 )3 (e) 2
8. This question is about rigorous reasoning. (See Section 1.4.)

x is any real number.
(i) Is this statement true: “x < 0 ⇒ x n < 0 for any value of n"?
If you think it’s false, find a value for n that demonstrates it.
(ii) What about the statement: “x > 1 ⇒ x n > 1 for any value of n".
Is this true or false?

Solutions 5
1 1 1 1 1 1
1. (a) 9 (b) 5 (c) 10000 (d) a 5 (e) y 2 (f) x
1 1 1 1 1
(g) 4n (h) 72n (i) 3a (j) 10n−1 (k) 10−(n+1) = 10n+1
1 1 1 1 1
(l) 2x−3 (m) t n (n) p a (o) q y+3 (p) q y−3
2. (a) 3−4 (b) 2−10 (c) x −5 (d) y −2 (e) 10−x
(f) 5−y (g) 6−2a (h) 7−1 (i) 4−6 (j) 2−3
1 24
(k) 3−4x (l) a −y (m) 23y = 2−3y (n) 24 x 4 = x −4
3. (a) x = −2 (b) x = −3 (c) x = −2 (d) x = −4

1 1 1
(e) 0.5 = 2−1 , so x + 1 = −1, x = −2 (f ) 9− 2 = 1 = 3 = 3−1 , so x = −1
92
1 1 1 1
(g) 85 = (23 )5 = 2−15 , so x = −5 (h) 256 = (52 )6 = 5−12 , so x = −12
1 1
(i) 0.125 = 8 = 2−3 , so x = −3 (j) 0.04 = 25 = 5−2 , so x = −2
1
(k) 0.3̇ = 3 = 3−1 , so x = −1
1 1 1 1 4 1 1 1
4. (a) 23 ×5 = 40 (b) 3×2 = 6 (c) 75 (d) 2 × 9 = 18
7
(e) 1000 (f) 25 (g) 64 (h) 18
32 35
(i) 10 × 8 = 80 (j) 3 (k) 2 (l) 72
3 16 8 25 ¡ 5 ¢2
(m) 50 (n) 14 = 7 (o) 9 (p) 3
= 25
9
¡ ¢4
(q) 23 = 81
16
9 1
5. (a) 16x 4 (b) 2x 3 (c) 2 x 3 (d) 16 x 2 (e) 7x −2
25 1
(f) 64x −3 (g) 9 x 2 (h) 2x (i) 36 x −4 (j) 3x
1
(k) 49x −2 (l) x 4 (m) 2x −1 (n) 6 x (o) 1
6. (a) 3−5 (b) 4−1 (c) 53 (d) x −20 (e) y 10

(f) 23x (g) 36 (h) 6−5 (i) 63 (j) x −2
(k) a 2 (l) 2−9 (m) p −6
7. (b) Because we can invert the fraction and change the sign.
(c) Because x 4−(−2) = x 6
(d) Because (x −2 )3 = x −6 . Not (e) because (2x)6 = 26 x 6 .
8. (i) False. It is not true if n is any even number.

(ii) False. It is not true if n is a negative integer. e.g. x = 3, n = −2.

2.5 Rational (or fractional) indices

We have seen that a negative index implies the presence of a fraction. What if
the index itself is a fraction, such as
1 1
92 or 83 ?
It’s not too difficult to work what it must mean. By the multiplication rule
1 1
9 2 × 9 2 = 91 = 9.
Since 3 × 3 = 9, it follows that

1 p
92 = 9=3
Similarly,
1 1 1
8 3 × 8 3 × 8 3 = 81 = 8.
so since 2 × 2 × 2 = 8, 1 p
3
83 = 8=2
In general
1 p
n
an = a
which we describe as the n t h root of a. When n = 2 we call it the “square

root", of course, not the “2t h root" (which has dental overtones, and when
n = 3 it’s the “cube root". After that it’s “fourth root", “fifth root", etc.
Notice that fractional indices don’t give rise to fractions; it’s negative indices
that do this.
QQ
Write down the values of these:

1 1 1
(i) 25 2 (ii) 16 4 (iii) 1000 3
What if the index is a fraction with a numerator other than 1? That’s what
we’ll be looking at next.
14
(i) 5 (ii) 2 (iii) 10

We are now in a position to work out what the meaning must be with any
fraction as the index. Using the rule (a m )n = a mn
4 1 p
8 3 = (8 3 )4 = ( 8)4 = (2)4 = 16
3
m p m
p
n
A
In general a = ( a)
n
n
or am
!
A
A
A
Another example would be

3 1 p
25 2 = (25 2 )3 = ( 25)3 = 53 = 125
You might like to remember it like this:
m power
a n root
where the root is on the bottom (like in a tree).

When raising to a fractional index we can do the root first, and then the power,
or the other way round, e.g.
2 p
5 2
p
5 p
5
32 5 = ( 32)2 = 22 = 4 or 32 5 = 322 = 1024 = 4
It is generally easier to take the root first. Here’s another example

4 p
3 4
p
3 p
3
27 3 = ( 27)4 = 34 = 81 or 27 3 = 274 = 531441 = 81.
The first option

p (taking the root first) wins hand down, as most of us wouldn’t
3
know what 531441 is.
Here’s an example of a fraction raised to a fractional power.
¶3 p
( 16)3 43
µ
16 2 64
= p = 3=
25 ( 25)3 5 125
¡ a ¢n an
using the rule = .
b bn
QQ
Find the values of these:

3 3 5
¡ 27 ¢ 2
3
(i) 9 2 (ii) 16 4 (iii) 8 3 (iv)
8
It remains to combine what we learned about negative indices with what we

now know about fractional indices.
15 9
(i) 27 (ii) 8 (iii) 32 (iv) 2

2.6 Negative fractional indices

3
What if we have a fractional index that is also negative, such as 9− 2 ?
When simplifying it’s easiest to get rid of the negative index first, using the
switch rule (see page 2-22). Some examples
1
1 1
16− 2 = 1 = 4
16 2
3
1 1 1 1
9− 2 = 3 = 1 = 33
= 27
92 (9 2 )3
¡ a ¢n n
In this next example we use the bracket rule
b
= ba n
µ ¶− 3 3 3
9 2 9− 2 4 2 8
= 3 = 3 =
4 4− 2 9 2 27
or we could use the switch rule applied to fractions (see page 2-22):
µ ¶− 3 µ ¶ 3 3
9 2 4 2 42 8
= = 3 =
4 9 9 2 27
And two algebraic examples:
¢− 12 1 1
x6
¡
= =
(x 6 )
1
2 x3
N 1 1 1
16− 2 = = 1
=
16
N
2 (16 )N 2 4N
All these examples are merely applications of the basics index rules.
QQ
Find the values of these:

1 5 1 ¡ 25 ¢− 1
(i) 100− 2 (ii) 8− 3 (iii) (a 6 )− 3 (iv) 4
2
Sometimes we will want to go the other way: an expression that contains roots
needs to be expressed in fractional index form.
16 1 1 1 2
(i) 10 (ii) 32 (iii) a2
(iv) 5

Converting from root form to index form

Converting into index form sometimes requires a little thought. For example
p 1
28 = (28 ) 2 = 24 = 16
p 1 3
3 3 = 31 × 3 2 = 32
1
1
p
3 = 11 = 5− 3
5 53
p n 1 n 1
( y) = (y 2 )n = y2 or y 2 n
QQ
Write each of these in index form:

p p 1 p p
(i) 38 (ii) 2 2 (iii) p (iv) ( 3 y)6 (v) 8
5
3 1 1 3
17
(i) 81 (ii) 2 2 (iii) 5− 2 (iv) y 2 (v) (23 ) 2 = 2 2

Manipulating fractions: a quick reminder

Before we tie up the remaining loose ends, it’s very important to make sure
that the business of working with fractions doesn’t present you with any
problems. So a quick reminder about how fractions work, and then some
examples applied to indices.
To multiply two fractions, multiply numerators and denominators:

2
3
× 56 = 10
18
=
5
9
2 6
6× 3 = 1
× 23 = 12
3
= 4
One of the commonest mistakes students make (even very able ones) is to to
confuse multiplication with equivalent fractions:
2 12
6× 3 6= 18
We are after a fraction that is 6 times bigger, not one the same size written a
different way.
To divide fractions, turn the right hand one upside down and multiply.
Add or subtract fractions by making the denominators the same—using
equivalent fractions:
2
3
+ 56 = 4
6
+ 56 = 9
6
=
3
2
5 6
3− 2 = 2
− 52 = 1
2

Applying these to some common types of index calculation:

2 5 4 5 3
93 × 96 = 96+6 = 92 = 27
2 2
(2 3 )6 = 2 3 ×6 = 24 = 16
125 53 6 5 1 p
5 = 5 = 52−2 = 52 = 5
52 52
p 1
26 = (26 ) 2 = 23 = 8
p 3 1 4
3× 3 = 31 × 3 3 = 33
1 1 1 1 1
(7 2 × 4 4 )2 = (7 2 )2 × (4 4 )2 = 7 × 4 2 = 14
QQ
Try some for yourself. Simplify these:

1 1 p 1 p
2)6
3
(i) 9 2 × 9 6 (ii) 5 × 25 (iii) (3 3 ×
2 7
(iv) (2− 5 )10 (v) p
4
7
There is one more level of difficulty to go. But first some practice, bringing
together what we have covered so far.
1 5 2 2
18 1
(i) 9 2 = 3 (ii) 5 2 (iii) 32 × 22 = 36 (iv) (2− 5 )10 = 2−4 = 16 (v) 7 3

Exercises 6 (Questions 9 to 14 are optional)
1. Put these numbers in ascending order.

q
1 1 ¡ 2 ¢2 8
¢1
(a) 10−1 (b) (c) 9− 2 (e) 2−3 ( f )
¡
3
4
(d ) 3 27
2. Write each number in its simplest form.

1 1 2 3
(a) 27 3 (b) 10000 4 (c) 8 3 (d ) 25 2
5 3 1 3
(e) 4 2 ( f ) 16 4 (g ) 400 2 (h) 32 5
¡4¢1 ¡ 27 ¢ 1 ¡ 16 ¢ 3 1
(i ) 9
2
(j) 8
3
(k) 81
4
(l ) 16− 4
1 7 3 1
(m) 9− 2 (n) 4− 2 (o) 100− 2 (p) 900− 2
3 ¡ 25 ¢− 1 ¢ 2
1 −3
(q) 81− 4
¡
2
(r ) 36
(s) 27
3. Write these in fractional index form (i.e. without using any roots).
p p
q
3 p 1
(a) 7 (b) 112 (c) ( x)3 (d ) 53
q q
4
(e) a5 (f ) p1 (g ) 5
(h) p1
3 y
3 7
p p p
5
p
n
(i ) 3 a (j) 3a (k) 2y 2 (l ) k
p
3 1
q
3 1
p
n
(m) t2 (n) p
3 2 (o) t (p) ax 2
t
4. Simplify each expression as far as you can.

5 3 p
(a) (a 4 ) 2 (b) (4y 2 ) 2 (d ) ( 3 x)6
p
(c) 9y 8
x 1 y
2y)4 (g ) (25y 4 )− 2 (h) 1000− 3
p
(e) (3 ( f ) 64 2
5. Rewrite each of these expressions in the form ax n . The constant

coefficient a should be written as simply as possible.
2 2 2 4
For example (8x 2 ) 3 = (8) 3 (x 2 ) 3 = 4x 3 .
p 1 1 2
(a) 9x (b) (16x 6 ) 4 (c) (27x) 3 (d ) (8x 2 )− 3
q
15 2 1 x4
(e) x 3 (f ) x (g ) 2x (h) 25
³ ´3
7 1 4 p2
(i ) px (j) p (k) p
5 (l ) x
3x x
p
6. Write each of these in the form k 3, where k in an integer.
p p 1 3
(a) 12 (b) 75 (d ) 27 2 (c) 27 2
3 p p
7. Write the number 8 2 + 18 in the form k 2, where k ∈ Z.

8. Simplify each of these expressions and leave your answer in index form.
1 p 1 1 p
(a) 23 × 2 2 (b) 3 × 3 (c) x 3 × x 6 (d ) a × 3a
p p 5 1 1 1 1
(e) ( y)5 × y (f ) 52 ×52 (g ) 11 × 11 2 × 11 2 (h) x 2 × x − 2
3 1 p 1 1 3
(i ) (7 2 )4 × 72 ( j ) t 6 × ( 3 t )2 (k) b 4 × b − 2 (l ) 3− 2 × 3− 2
− − 1 2
1 1 1 1 4 3 ×4 3
(m) (x 3 )2 × (x 2 )3 (n) (y −2 )− 2 (o) (a − 3 )−6 (p) 42
1 1 µ p ¶1
26 x 2 ×x 2 1 a 2
(q) 1 5 (r ) 3 (s) 1 5 3 (t ) 3
(2 3 )−2 ×2 6 x− 2 x − 4 ×x − 4 ×x − 2 a− 2
9. Write these in index form and simplify where possible.

pp p p p
q
1
3 2 2 2 2−1
(a) 2 (b) (2 ) (c) (10 ) (d ) 16 (e) 11 n
p p
q
p p
qp
3 1 p k
y 2 a a
(f ) 6 a k (g ) (a ) (h) (x ) (i ) t (j) k
10. Determine which of these statements is true. Correct those that are
false by rewriting the left hand side in a simpler form.
1 p 1 1 1 1
(a) a 2 × a = a (b) x 2 × x 2 × x 2 = x 8
1 1
(c) 8 (2y)3 = y 3 (d ) 1 = x −2
x2
p
(e) ((y a )a )a = y 3a (f ) a9 = a3
q ³ ´−1 1
1 p1
(g ) 3
t3
= t −1 (h) x
= x− 2
a2b p 3
(i ) (ab)−1 = a 3 b 2 (j) x + 1(x + 1)3 = (x + 1) 2
¡ a ¢− 1 2
(k) 4
2 = pa
11. Simplify these.

r 48
¶1
p
q µ
−2 −1
4 3 2 − 12 16 4
(a) ((−3) ) (b)  3 (c) 3 +
9
1 1
12. (a) Without using a calculator determine which is larger: 2 2 or 3 3 .
¡ ¢1 ¡ ¢1
(b) Deduce which is the larger of 12 2 and 31 3 .
5
13. The volume of a cube is 6 cm3 . Show that its total surface area is 6 3 cm2 .
14. (a) The volume of a cube is V. Find an expression for its total surface
area, giving your answer in index form.
(b) The area of a circle is A. Find, in index form, expressions for (i ) its
radius, and (i i ) its circumference.
(c) A square has half the area of the circle in part (b). How long is each
side? Give your answer in the form k A n , where k and n are constants.

Solutions 6
1 1 1 4 1 2
1. In order: (a), (e), (c), (d ), (b), ( f ). i.e. 10 < 8 < 3 < 9 < 2 < 3 .
2. (a) 3 (b) 10 (c) 4 (d) 125 (e) 32

2 3
(f) 8 (g) 20 (h) 8 (i) 3 (j) 2
8 1 1 1 1
(k) 27 (l) 2 (m) 3 (n) 128 (o) 1000
1 1 6
(p) 30 (q) 27 (r) 5 (s) 9
1 2 2 1
3. (a) 7 2 (b) 11 3 (c) 11 3 (d) 5 6
2 1 1 1 1 1
(e) a 5 (f) 3− 2 (g) ( 57 ) 2 or 5 2 7− 2 (h) y − 3
1 1 1 1 1 2 1
(i) 3a 2 (j) (3a) 2 or 3 2 a 2 (k) 2 5 y 5 (l) k n
2
2 2 1 2 13 2 1 2
(m) t 3 (n) t − 3 (o) ( t ) 3 = 2 = t−3 (p) a n x n
t3
3 1 1
4. (a) a 10 (b) 4 2 × y 3 = 8y 3 (c) (9y 8 ) 2 = 3y 4 (d) (x 3 )6 = x 2
1 1
(e) (3 × 2 2 × y 2 )4 = 81 × 22 × y 2 = 324y 2
x 1 1 1 1
(f) (26 ) 2 = 23x (g) = (h) y =
1
(25y 4 ) 2 5y 2 1000 3 10 y
1 3 1 1 4
5. (a) 3x 2 (b) 2x 2 (c) 3x 3 (d) 4 x − 3 (e) 15x −3
1 1 1 1 1
(f) 2x −1 (g) 2 x −1 (h) 5 x 2 (i) 7x − 2 (j) 3− 2 x − 2
1 3
(k) 4x − 5 (l) 8x − 2
p p p p 3 1
6. (a) 2 3 (b) 5 3 (c) 3 3 (c) 81 3 [27 2 = 27 × 27 2 ]
p 3 9 1
p p p p p
2
7. 19 2. [(23 ) 2 = 2 2 = 24 × 2 2 =16 and 18 = 9 2 = 3 2]
7 3 1 3
8. (a) 2 2 (b) 3 2 (c) x 2 (d) 3a 2 (e) y 3
3 5
(f) 53 (g) 112 (h) x 2 (i) 78 (j) t 6
7 13
(k) b 2 (l) 3−2 (m) x 6 (n) y (o) a 2
(p) 4−3 (q) 1 (r) x 3 (s) x 3 (t) a
1
p 1
9. (a) 2 4 (b) 22 3
(c) 100 (d) 4 (e) 11 2n
1 p 1 1
y
(f) a 3k (g) a 2 (h) x a (i) t 8 (j) k k
3 3 9
10. (a) T. (b) F. x 2 (c) T. (d) F. x 2 (e) F. y a (f) F. a 2
1 7
(g) T. (h) F. x 2 (i) T. (j) F. (x + 1) 2 (k) T.

48
11. (a) 9 (b) 3 24 = 9
³¡ ¢ 1 ´ 12 ¡4¢1 p
1 16 1 1
(c) 1 + 9
2
= p
3
+ 3
2
=p
3
+ p23 = p3 = 3
3
32
1
12. (a) 3 3 is the larger.
1 1 1 1
Let x = 2 2 and y = 3 3 . Now x 6 = (2 2 )6 = 8 and y 6 = (3 3 )6 = 32 = 9.
x 6 < y 6 and we know x and y are both positive numbers, so x < y.
[The importance of x and y both being positive is demonstrated by the example
x = 1, y = −2.]
¡1¢1
2
(b) 2
is the larger.
1
¡1¢1 12 1 ¡1¢1 1
2 3
2 = 1 = 1 . Similarly 3 = 1 .
1
22 22
1
33 1 1 1 1
Both 3 3 and 2 2 are positive numbers and 3 3 > 2 2 , so 1 < 1 .
33 22
¡1¢1 ¡1¢1
3 2
That is 3 < 2 .
[If we divide by a negative number within an inequality, its direction will be reversed.]
1 1 2
13. The length of each edge is 6 3 , so each side has an area (6 3 )2 = 6 3 .
2 5
Total SA = 6.6 3 = 6 3 .
1
14. (a) Let x be the length of each side. Then V = x 3 , so x = V 3 .
Total surface area = 6x 2 = 6V 32 .
(b) Let r be the radius.
¡ A ¢ 12
(i) Since A = πr 2 , r = π
¡ A ¢ 12 1 1 1 1
(ii) Circumference = 2π = 2πA 2 π− 2 = 2π 2 A 2
π
¡ ¢1 1 1
(c) Let y be the length of each side. A2 = y 2 . So y = A2 2 = 2− 2 A 2 .

2.7 Factorising the base

Look at these problems. In all these examples we’ll be using the product rule
(ab)n = a n b n and the power rule (a m )n = a mn .
p
4
27
Example 1 Express p in the form 3n .
3
Solution
There is a clue in the question, which points us in the right direction: 27 = 33 ,
and that is our starting point.
p
4
p
4 3 3
27 3 34 1
p = p = 1 = 34 .
3 3 32
It was factorising that did the trick.
q
729
Example 2 Evaluate 625
.
Solution
Recognising that both bases are exact powers:
¶1 1 1
(36 ) 2 33
r µ
729 729 2 729 2 27
= = 1
= 1
= 2
=
625 625 625 2 (54 ) 2 5 25

QQ
1. Write these in index form as simply as possible:

p p p p
3
(i) 9 3 (ii) 128 (iii) 8 × 16
p
5 ( a)3
(iv) p
3 (v) a
25
2. Find the values of:

3 2 p
(i) (810000) 4 (ii) (512) 3 (iii) 44100.
What if a base doesn’t happen to be the power of a single number? Or has an

algebraic component?
1 5 1 5 1 4 11 1 1
19
1. (i) 32 × 3 2 = 3 2 (ii) (25 ) 2 = 2 2 (iii) 2 2 × 2 3 = 2 6 (iv) 5 3 (v) a 2
3 2 1
2. (i) (34 × 104 ) 4 = 27000 (ii) (29 ) 3 = 26 = 64 (iii) (9 × 49 × 100) 2 = 210.

2
Example 3 Evaluate 64000 3 .
Solution
The base 64000 is not the power of a single number but it can be factorised.
So let’s see if its factors can.
2 2 2
64000 3 = (64 × 1000) 3 = (26 × 103 ) 3
2 2
= (26 ) 3 × (103 ) 3
= 24 × 102 = 1600.
Notice that we didn’t need to find the prime20 factorisation of the base; any
factorisation that works is fine.
Let’s look at a nice algebraic example.
4x+1
Example 4 Simplify the expression
82x
Solution
In Example 1 we were able to write one base in terms of the other (27 = 33 ).
We could do that here, but a simpler approach would be to write both of them
as a power of 2:
4x+1 = (22 )x+1 = 22x+2 and 82x = (23 )2x = 26x
Now we can simplify easily:
4x+1 22x+2
= 6x = 22−4x .
82x 2
This technique can give rise to some interesting equations too.
2 3
20
It’s perfectly fine to do so: 64000 = 29 53 , so (6400) 3 = (29 53 ) 2 = 26 52 = 1600.

Example 5 Solve the equation 1000x = 0.01.
Solution
Both sides can be expressed as powers of 10:
1000x = (103 )x = 103x and 0.01 = 10−2 .
So 2
3x = −2 ⇒ x = − .
3
Example 6 Solve the equation 91−2x = 27x .
Solution
Using the same technique
91−2x = 33−6x and 27x = 33x
so 1
3 − 6x = 3x ⇒ x = .
3

QQ
1. Simplify these:
q
3
a a 27 y−1 169
(i) 900 2 (ii) 81 × 27 (iii) 9 y (iv) 256
2. Solve these equations:
1
(i) 8x = 2 (ii) 7 y = 49 (iii) 16x = 41−x
1
(iv) 9a = 3 (v) 125x = 0.2
13 1 1
21
1. (i) 2700 (ii) 37a (iii) 3 y−3 (iv) 2. (i) x = (ii) y = −2 (iii) x =
16 3 3
(iv) x = − 12 (v) x = − 31

Irrational indices
We now know how to deal with rational indices, i.e. ones that can be
expressed as a fraction. The fraction can be anything, of course. For example
5 p
9
7 9 = ( 7)5
85 p 85
100
230.85 = 23 100 = ( 23)
1
640.333....... = 64 3 = 4
You may have been wondering if it’s possible to have an irrational22 index.
For example p
7 3 or 10−π ?
If you want to know, turn to page A-10 in Appendix A (or click on the page
number).
22
An irrational number is one that can’t be written as a fraction (using integers).

Exercises 7
1. By rewriting bases, simplify these.
511
(a) 23 × 45 (b) 25 (c) 1007 × 104 (d ) 3x × 9
(e) 2 y × 82y ( f ) 36a × 6 (g ) 42t × 8t (h) 25x+1 × 5x
49b ×73 163x (9a)3 4k

(i ) (j) (k) (l )
7b 8x (3a)4 23k
25x 9 y+1 94a
(m) (n) (o)
54x 27 y−1 32a ×27a
125 y ×5
(p)
52y
2. Simplify these as far as you can.
y N ×y 2N +1 3t ×32t (2x)a ×x 4a
(a) (b) (c)
y 27t −1 4x 2a
(100x)3 ×x 2 (32x y)3

(d ) (e) (27y)4 × 9y 2 ( f )
(10x 2 )2 x 2 y×16x y 2
3. By re-writing bases, simplify these as far as possible.

1 5 1 1 2 5 3 1
(a) 8 4 × 2 4 (b) 100 3 × 10 3 (c) 9 3 × 24 3 (d ) 10 2 × 40 2
1 4 1
3 1 3 82 7 3 ×16 3 1 3
(e) 2 × 12 × 3
2 4 4 (f ) 1 3 (g ) 1 (h) 48 2 × 6 2
27 2 ×6 2 14 3
1 1 p
10 3 p 18
84 ×
(i ) 2 (j)
25 3 3
4. Write each of these as a single term.

1 3 p p 1 p
(a) 20 2 + 5 2 (b) 27 − 3 (c) 2(72) 2 + 3 8
1 1 p 1 p 5 3
(d ) 16 3 + 2 3 (e) 3 12 − 6(3 2 ) ( f ) 2(2 2 − 2 2 )
5 1
(g ) 3 2 − 2(3 2 )
5. Write 3a+2 − 3a in the form k 3a , where k is a constant.
6. Show that 2x+1 2 y = 2x 2 y+1 .

Solutions 7
511
1. (a) 23 × 210 = 213 (b) 52
= 59 (c) 1014 × 104 = 1018
(d) 3x × 32 = 3x+2 (e) 2 y × 26y = 27y (f) 62a+1
(g) 24t × 23t = 25t (h) 52x+2 × 5x = 53x+2 (i) 72b+3−b = 7b+3
36 ×a 3 9 22k
(j) 212x−3x = 29x (k) = (l) = 2−k
34 ×a 4 a 23k
52x 32y+2
(m) = 5−2x (n) = 3(2y+2)−(3y−3) = 35−y
54x 33y−3
(o) 38a−(2a+3a) = 33a (p) 52y+1−2y = 5
2a x 5a
2. (a) y 3N (b) 33t −3(t −1) = 33 = 27 (c) = 2a−2 x 3a
22 x 2a
103 x 5 215 x 3 y 3
(d) = 10x (e) 312 y 4 × 32 y 2 = 314 y 6 (f) 4 3 3 = 211
102 x 4 2 x y
1 5 8
3. (a) (23 ) 4 × 2 4 = 2 4 = 4
1 1 3
(b) (102 ) 3 × 10 3 = 10 3 = 10
1 1
or, by the bracket rule, (100 × 10) 3 = 1000 3 = 10
2 5 4 5 5
(c) (32 ) 3 × (3 × 23 ) 3 = 3 3 + 3 × (23 ) 3 = 33 × 25 = 864
3 1 1
(d) 10 2 × 10 2 × 4 2 = 200
3 1 1 3
(e) 2 2 × (22 ) 4 × 3 4 × 3 4 = 22 × 3 = 12
3 4 4
22 1 7 3 ×2 3
(f) 3 3 3 = 27 (g) 1 1 = 14
3 2 ×2 2 ×3 2 2 3 ×7 3
1 3 7 1 p
(h) (3 × 24 ) 2 × (3 × 2) 2 = 32 × 2 2 = 9 × 23 × 2 2 = 72 2
1 1 1 p
3 3 1
2 3 ×5 3 2 5 1
= 25 or 2 4 ×2 2 ×3
3
(i) 4 5
(j) 1 = 24 32
53 32
1 1 1 p p p
4. (a) 4 2 5 2 + 5 × 5 2 = 2 5 + 5 5 = 7 5
1 1 1 p p p
(b) 9 2 3 2 − 3 2 = 3 3 − 3 = 2 3
1 1 1 1 p
(c) 2 × 36 2 × 2 2 + 3 × 4 2 × 2 2 = 18 2
1 1 1 1 p
3
(d) 8 3 2 3 + 2 3 = 3 3 or 2
1 1 1
(e) 3 × 4 2 3 2 − 6 × 3 2 = 0
1 5 3
(f) 2 2 (2 2 − 2 2 ) = 23 − 22 = 4
1 1 p
(g) 32 3 2 − 2 × 3 2 = 7 3
5. 3a+2 = 3a × 32 = 9 × 3a ⇒ 3a+2 − 3a = 8.3a .
6. 2x+1 = 2(2x ). Similarly 2 y+1 = 2(2 y ). So each side is equal to 2(2x )(2 y ).

Common mistakes with indices

You have now met all the index rules, and hopefully they make sense to you.
But if you keep making mistakes, don’t be too hard on yourself: they are not
intuitive; its only when you’ve used them a lot that applying them will be
instinctive. The best way to avoid these mistakes is to study the common
ones: if it looks wrong to you, why is it wrong, and how would you correct it?
At the end is a list of questions for you to test yourself.
We mentioned one of the commonest errors: trying to apply the bracket rule
(ab)n = a n b n
to addition or subtraction. Beware:
(a + b)n 6= a n + b n
(a − b)n 6= a n − b n
For example:
(x + 3)2 6= x 2 + 32 , (y − 2)2 6= y 2 − 22 .
Putting x = 1 and y = 3 quickly reveals the error:
(1 + 3)2 = 16 6= 1 + 9 and (3 − 2)2 = 1 6= 9 − 4.

To see what’s going on, and why it doesn’t work, let’s look at the algebra:
(a + b)2 = (a + b)(a + b) = a 2 +2ab + b 2 6= a 2 + b 2

We get an extra 2ab term.
We can also demonstrate this geometrically:
a b
The area of the whole square is (a + b)2 .
a
It’s also equal to a 2 + ab + ba + b 2 .
In the case of (a − b)2 , not only do we get a −2ab term, the sign of b 2 is wrong
too:
(a − b)2 = (a − b)(a − b) = a 2 −2ab +b 2 6= a 2 − b 2
Misapplying the rule with when n = 12 is also common, for example

p p
x 2 + 9 6= x + 3, x 2 − 1 6= x − 1
If either of them “looks" as if they ought to be equal, put x = 4 into it.
Some of the other mistakes students make are illustrated below.

(35 )2 6= 37 24 × 35 6= 69
24 + 25 6= 29 (x + 2)3 6= x 3 + 8
3(2a)3 6= 6a 3 (1 − x)2 6= 1 − x 2
Study these. Can you see why they are wrong?

Brackets can trip us up too. Remember an index only applies to what it is
attached to. It is particularly easy to forget this with negative indices.
3 1 1
3a −2 = not or
a2 3a 2 (3a)2
4 1
4(x + 1)−1 = x+1
not 4(x+1)
p2 1
p 2 q −1 = q
not p2q
If all these mistakes leap out of the page at you, then you have a good
understanding of the rules. If they don’t, spend some time mulling over
them. There are more examples on pages 2-2, 2-18 and 2-26. It helps to write
down your own examples—until the rules start to look intuitively right.
Until then, the only answer is to slavishly follow the rules.

Summary of the rules of indices
a m × a n = a m+n
am
= a m−n
an
(ab)n = a n b n
¡ a ¢n a n
= n
b b
(a m )n = a mn
1
a −n =
an
m p p
a n = ( n a)m or n a m

Self test of common mistakes with indices.

Go through the list of statements below. Correct the ones that are false by
rewriting the right hand side. Unlike exam questions, these have been
designed to try and trip you up.
1. (a) (52 )4 = 56 (b) (10p )2 = 102p (c) (4y)3 = 4y 3
(d) (112 )3 = 115 (e) 72 × 73 = 75 (f) 22 × 54 = 106
(g) 34 × 54 = 154 (h) (a + 3)2 = a 2 + 9 (i) (t + 1)2 = 1 + 2t + t 2

³ ´−1 y 1
5 1
(j) y
=5 (k) 34 = 3 4 (l) (7−2 )−3 = 76
4
(m) 43 × 43 = 49 (n) (x − y)2 = x 2 − y 2 (o) x 3 = 4x −3
2
(p) a x × a x = a x (q) 5 y × 5−y = 1 (r) 23 + 24 = 27
1 1
(s) 2x −3 = 2x 3 (t) 10x = 10−x (u) (3a 2 )2 = 9a 4
³ 2 ´2
2 3 5 a4 a 1
(v) (5y ) = 125y (w) y 2 = y
(x) 6−2 = − 36
1 1 1 1
2. (a) 2(y + 4)−1 = 2(y+4) (b) p = 3− 2 (c) (4y)−2 = 4y 2
3
11 1 1 3
(d) (1 − x)− 2 = p (e) ppq = (pq) 2 (f) 100 2 = 150
1−x
¡ a ¢− 1 q
5 1 1
(g) 6 −2
= −36 (h) 5 2
= a
(i) (2x)− 3 = (2x)3
1
(j) (y x ) y = y x y (k) a 2 b −1 = a 2 b

Answers to self-test.
1. (a) F. 58 (b) T. (c) F. 64y 3
(d) F. 116 (e) T. (f) F. Can’t be simplified.
(g) T. (h) F. a 2 + 6x + 9 (i) T.
(j) T. (k) F. 3−4 (l) T.
(m) F. 46 (n) F. x 2 − 2x y + y 2 (o) T.
(p) F. a 2x (q) T. (r) F. Can’t be simplified.

2
(s) F. x3
(t) T. (u) T.
1
(v) F. 125y 6 (w) T. (x) F. 36
2 1
2. (a) F. y+4
(b) T. (c) F. 16y 2
(d) T.
1 1
(e) F. (pq)− 2 (f ) F. 1000 (g) F. 36
(h) T.
1 a2
(i) F. 1 (j) T. (k) F. b
(2x) 3

2.8 Surds
2.8.1 What is a surd?
Sometimes when we take the root of a number the result happens to be a
whole number, or some other rational23 number. For example
r
p 4 2 p
25 = 5 or = or 0.01 = 0.1
9 3
When the answer isn’t rational (and in general it won’t be) we call the result a
surd. Some examples of surds would be
r
p 7 p
3
p
2, 5, and 0.1.
11
So if a surd is an irrational number, what’s the difference? It’s a historical
thing. The word surd originates from the study of roots (square roots, cube
roots, fourth roots etc), so it’s only used if there is a root involved. So
although π is irrational it’s not a surd.
In the language of sets, surds are a subset of the irrational numbers.
For us the study of surds will be all about learning to manipulate expressions
involving square roots24 .
23
A fraction in which the numerator and denominator are both integers. See page 1-5
24 p
A term you will occasionally see is radical: it refers to the square root sign itself (i.e. ).
For example we might say we are going to square an expression to get rid of the radical.

And the point is?
Why would we take the trouble to develop specialist skills just for
manipulating square roots? Why not just get a calculator out? And if it’s
algebraic we could use fractional indices—we already know how to
manipulate those.
It’s true, we could always use fractional indices in place of roots. However it’s
often quicker and easier to use roots. And it’s those kinds of manipulations
we are going to be studying here.
The reason we try to avoid using a calculator is because mathematicians
want to be able to distinguish between an exact value and an approximate
one—and prefer to use an exact value if possible.
If we work with numbers in surd form we are preserving exact values.
p
Suppose we do a calculation and get the result x = 2, that is the exact
answer. It’s not a very useful answer if you are an engineer or a cabinet
maker; you would probably prefer something like x = 1.41. If you need
greater precision, you can go to more decimal places, say 1.4142. But it still
won’t be exact. From a mathematician’s perspective, any approximation is
losing something, however many decimal places we include.
Notice we say “greater precision", not “more accurate".

In Maths we use the word “accurate" to mean error-free, so “more accurate"
means “fewer errors". The word “precise" is better in this context.

This little calculation illustrates the value of maintaining exactness, using

surds. Look at this example:
To find the area of an isosceles right-angled triangle with a hypotenuse of

2cm.
We let the length of each of the other sides be x.
C
By Pythagoras’ Theorem
x 2 + x 2 = 22
2x 2 = 4
p 2 x
x = 2
or x = 1.4142 (4 DP)
(Discarding the negative square root.)

A x B
The area of the triangle ABC is 12 x 2 .

First we’ll use the decimal approximation for x:
(≈ means “is approximately equal to”)
1
Area ≈ × (1.4142)2 = 0.99998082 cm2 .
2
We may suspect that the exact answer is in fact 1, but we can’t be sure of that
just because it looks like it.
p
Using the exact value of x, the area = 21 × ( 2)2 = 1 cm2 .
p
Writing the number as x = 2 keeps it exact, and can even make the
calculations slightly less onerous. It also avoids introducing rounding errors,
which may be compounded in subsequent calculations.

Surds in trigonometry
The triangle we just considered we can think of as a square cut in half, so its
angles are 45◦ , 45◦ and 90◦ . We can use such a triangle to find an exact form
for the trig ratios of 45◦ .
The trig ratios depend on the shape of the triangle only, not its dimensions,
so we may as wellp choose a triangle in which x = 1. Now by Pythagoras the
hypoteneuse is 2.
Using SOHCAHTOA25
1
sin 45◦ = p
2
1 p
cos 45◦ = p 2 1
2
tan 45◦ = 1
45◦
1
Another very important triangle—one you will come across frequently—is a

30◦ , 60◦ , 90◦ triangle. We can think of this one as an equilateral triangle cut in
half along a line of symmetry. See the diagram below.
It’s ratios we are going to be interested in, so any dimensions will do. Since
we are going to be halving one of the sides,
p we make each one 2 units, then
the triangle has sides of length 1, 2 and 3 units.
p
Pythagoras tells us that R N = 3, so using SOHCAHTOA
R
T
p
T
3
30◦
T
sin 60◦ = T
2 T
T
1 2 p T
cos 60◦ = 3 T
2 T
p T
tan 60◦ = 3 T
T
◦
60
T
T
P 1 N Q
Write down for yourself the equivalent list for 30◦ .

These ratios (in their surd form) are so important that eventually you will
want to memorise them.
25 Opposi t e Ad j acent Opposi t e

sin = H y pot eneuse ; cos = H y pot eneuse ; tan = Ad j acent .

How tiny errors can grow

We have justified the use of exact values largely on the basis of some kind of
mathematical “purity". Look at this example
The problem is to find the value of
1 p
y= when x= 10.
(x − 3.15)2
p
If you evaluate this using your calculator (inserting 10 into the calculation),
you will find that the answer (to 3 SF) is
1
y= p = 6, 630.
( 10 − 3.15)2
If we make the same, 3 SF approximation for x at the outset, i.e. x = 3.16, we
have
1
y= = 10, 000.
(0.01)2
This is something like a 50% error!
We have picked an artificial example to make the point; the error is rarely
going to be this big, but the principle remains the same: use the exact value if
you can.
Precision, surds and exams

There is one final—and rather compelling—reason to become expert at
dealing with surds. An exam question may say something like: "Leave your
answer in surd form".
Of course as a budding mathematician you will find yourself preferring to
use an exact value anyway, and this will extend well beyond the use of surds.
For example you will prefer to write 31 rather than 0.3333 . . ., and π rather
than 3.1416 . . .. It’s all part of your transmogrification.
Where an answer can’t be expressed exactly, you will generally be told the level
of precision to use; if not, assume 3 significant figures (SF). You’ll find more
about SF and DP precision on page A-17.

2.8.2 Multiplying surds

We start with a definition of what a square root26 is:
p p
a × a = a.
p
If we are multiplying a number of square roots together, each pair of a’s will
give an a. So for example
p p p p p p
a × a × a × a × a × a = a3
p p p p p p
a × a × a × a × a = a2 a
From this (or from the index rules) we can see that
p
( 2)12 = 26 = 64
p p
( x)9 = x 4 x
p
( 10)2k = 10k
(If we raise a square root to an even power, the result contains no square roots;
if we raise it to an odd power, a single root remains.
p p
From the basic definition of a square root a × a = a, we can multiply out
and simplify expressions like these:
p p p
5( 5 + 2) = 5 + 2 5
p p p p
3( 3 − 1) + 3( 3 + 1) = 4 3
p p p p p
( 2 + 3)( 2 − 1) = 2 + 3 2 − 2 − 3 = 2 2 − 1
p p p p
(2 6 + 1)( 6 − 3) = 12 − 5 6 − 3 = 9−5 6
Check these for yourself. In the last two examples it is just like multiplying
out any pair of brackets: each term in the left hand one is multiplied by each
term in the right hand one. (You may be familiar with the “happy face” rule
for doing this quickly. If not go to page 3-10 in Chapter 3.)
1 1
26
In index notation: a 2 × a 2 = a 1 .

p p
Suppose we have something like ( 2−1)3 or 2−1)4 ? All we do is build up to
them, step by step27 .
p p p p p
( 2 − 1)2 = ( 2 − 1)( 2 − 1) = 2 − 2 2 + 1 = 3 − 2 2
p p p p
∴ ( 2 − 1)3 = ( 2 − 1)(3 − 2 2) = 5 2 − 7
p p p p
∴ ( 2 − 1)4 = ( 2 − 1)(5 2 − 7) = 17 − 12 2
p p
We could also obtain ( 2 − 1)4 by squaring 3 − 2 2. Can you see why?
Here are two more examples involving pairs of brackets:

p p p p p p
( 11 − 7)( 11 + 7) = ( 11)2 − ( 7)2 = 4
p p
( 2 + 1)( 2 − 1) = 2 − 1 = 1
You may recognise in these the ubiquitous difference of two squares pattern:
A
(X + Y )(X − Y ) = X 2 − Y 2
!
A
A
A
We shall pick this idea up again and use it to great effect a little later.
27
This gets rather tedious if the index is higher than about 4.
You’ll learn a more more direct (and very ingenious) method later.

QQ

p p p p p p
(a) 3(4 + 3) − 2 (b) (3 5 + 1)( 5 + 2) (c) ( 6 + 1)( 6 − 1)
p p p
(d ) ( 3 + 1)2 (e) ( 3)8 ( f ) ( 5)7
p p
(g ) ( y)10 (h) ( 2)17
28
p p p
(a) 4 3 + 1 (b) 17 + 3 5 (c) 5 (d) 2(1 + 3)
p p p
(e)34 = 81 (f ) 53 5 = 125 5 (g) y 5 (h) 256 2

2.8.3 Decomposing surds

We shall first see how decomposition can help when multiplying or dividing
surds, then we shall see how it can be used when adding or subtracting them.
But first what do we mean by decomposing surds? It’s best explained with a
few simple illustrations. Suppose we have the surds
p
98.
Although it is a genuine surd29 , it can be simplified, because 98 = 49 × 2.

Using the index rule
1
(ab)n = a n b n with n =
2
we have 1 1 1
(ab) 2 = a 2 b 2
i.e.
p p p
ab = a b
So p p p p p
98 = 49 × 2 = 49 2 = 7 2.
Sometimes decomposing surds throws up surprises. Look at this example:

p p
48 × 75.
We know we can split up each square root:
p p p p p p
16 × 3 × 25 × 3 = 16 × 3 × 25 × 3
p p
= 4 3 × 5 3 = 60
This simple answer is not be quite so unexpected if we think of it like this:

wrapped up within the 48 is a factor 42 and wrapped up within the 75 is a
factor 52 , so we can see at the outset that some significant simplifying is
going to be possible.
We could, of course, do the same simplification using the index notation:
p p 1 1
48 × 75 = (24 × 3) 2 × (52 × 3) 2
1 1
= 22 × 3 2 × 5 × 3 2
= 4 × 3 × 5 = 60
although we probably wouldn’t—it’s much easier to use the square root

approach.
29
i.e. it can’t be written as a rational number

Here is another example:

p p p
10 × 105 × 14.
In this case we’ll break each number down into its prime factors:
p p p p p p p p p p
2×5× 3×5×7× 2×7 = 2× 5× 3× 5× 7× 2× 7
p p
which, after collecting pairs of roots, becomes 2 × 5 × 7 × 3 = 70 3.
There are some cases of embedded squares that crop up a lot. You’ll find it
useful to familiarise yourself with them:
p p
8 = 2 2
p p
27 = 3 3
p p
75 = 5 5
p p
1000 = 10 10

We can use the same approach when dividing surds. The index rule tells us
that
³ a ´ 1 a 12
2
= 1
b b2
or
q p
a pa
=
b b
A simple application of this principle you will probably have seen before,
something like: r p
4 4 2
=p = .
9 9 3
This is consistent with what we mean by a square root, because
2 2 4
× = .
3 3 9
Here is a more interesting example. Neither numerator or denominator in

the fraction p
18
p ,
50
is a perfect square, although looking at their factors we can see that they
both contain embedded squares (18 = 2 × 9 and 50 = 2 × 25).
q p
a a
We can use
b
= p . First we apply it in reverse, and then forwards:
b
p r r p
18 18 9 9 3
p = = =p = .
50 50 25 25 5
p p p
In fact we could simplify the fraction using ab = a × b:
p p p p
18 9 × ZZ2 9 3
p =p p =p = .
50 25 × ZZ 2 25 5
An example from the world of decimals:

r p
p 16 16 4
0.16 = =p = = 0.4.
100 100 10

Of course expression involving surds, although they may simplify, don’t

necessarily simplify to the same extent. For example:
p p p p p p
3 21
3 × 3
7 3
p = p p p =p
42 7 23
2
Notice that there is only one termpin the numerator;

p the × sign may make it
look a bit like two. We can cancel 7 and 3.
Finally a very similar, but slightly trickier example:
p p p1 p1 p p
3 + 21 3 +
3 7 1+ 7
p = p p p = p
42 7 2 3 14
p really does contain two terms, and only one of

In this case the numerator
them contains a factor of 7.
There is nothing new here but the presence of square roots in a fraction just
adds another level of difficulty.

Now we are going to see how decomposing square roots can sometimes help
when adding or subtracting surds. In each case
the key is to spot the presence of a square embedded within the surd.
Some examples
p p p p p
3 5 + 2 20 = 3 5 + 2 4 5
p p
= 3 5+4 5
p
= 7 5
p p p p
27 − 12 = 32 × 3 − 22 × 3
p p
= 3 3−2 3
p
= 3
p p p p p p
8(3 2 + 50) = 2 2(3 2 + 5 2)
= 12 + 20 = 32
There are often several ways we could do the simplification. For example in
the last example we could multiply the surds rather than decompose them:
p p p p p p p
8(3 2 + 50) = 3 8 2 + 8 50
p p
= 3 16 + 400
= 12 + 20 = 32
Everything we have done in this section has been based on the principles
(derived from index rules) that:
p p p
ab =
a× b
q p A
a pa !
A
A
=
b b A

QQ
Simplify these expressions

p p p p p
(a) 32 (b) 45 20 (c) 12 15
q
9
q
1000
p p
(d ) 16 (e) 320 (f ) 150 + 54
p p p p p p
27+2 3
(g ) 3 8 − 2 (h) 72 − 2 18 (i ) p
75
p p p p p
30
(a) 4 2 (b) 30 (c) 6 5 (d) 3
4 (e) 5 4 2 (f ) 8 6 (g) 5 2 (h) 0 (i) 1

2.8.4 Rationalising surds

Look at this expression p
1 12
p + .
3 3
When adding fractions our first thought is to putpthe numerator over a
common denominator. In this case that would be 3 3. However we can do
slightly better than that. To see how, let’s first convert the basic definition of
p p
a square root: a × a = a, into two versions that includes division:
p
a p 1 a
p = a or p = .
a a a
With a = 3, the second version gives us

p
1 3
p = .
3 3
Now both fractions in our sum have the same rational denominator, i.e. 3.
We call this process rationalising the surd. It means writing

a fraction so that it doesn’t have any (square) roots in the
denominator.
Proceeding with the addition:

p p p p p
3 12 3 2 3 3A 3 p
+ = + = = 3
3 3 3 3 3A
Rationalising the surd is particularly effective in this instance, although the

way we did it was hardly intuitive. Let’s see how we normally do the
rationalising.

Take the fraction:

6
p .
2
The first approach to rationalising it (and the one we normallypuse) is to
multiply top and bottom by the surd we want to get rid of (i.e. the 2):
p p
6 6 2 6 2 p
p = p p = =3 2
2 2 2 2
Another way to look at this

p ispwould be to take a factor of 2 out of the
numerator and write it as 2 × 2, like this:
p p
3×2 3 × 2 × 2 p
p = p = 3 2
2 2

As with the previous example, rationalising has eliminated fractions

altogether. This won’t always be the case, of course. Some more examples:
p
10
p 2×5
p 2 5
= =
3 5 3 5 3
p p
p7 7
p p 2 7 2
= =
2 2× 2 2
p p p p p p
9p 2 9 2 6 3 12 3 4 3 p
= = = = 3 3
6 6 2 2
QQ
1. Rationalise these surds
8 2 21
(a) p (b) p (c) p
2 8 7
p
10 12 4 p
(d) p p (e) p (f) p + 2
2 5 3 2 2
2. Simplify and rationalise this expression (we hope you enjoy this one)
Ã p !
p 2 3
( 6 + 2) 3 − p .
2
p p p p p
31 2 sqr t 3
1. (a) 4 2 (b) 2 (c) 3 7 (d) 3 (e)2 2 (f ) 3 2
p
2. 6

p
So it seems that rationalising is a pretty simple process. To get rid of a a
p
from the denominator, multiply top and bottom by a. Well not quite. Look
at this fraction:
6
p
3−1
p
If we multiply top and bottom by 3 we get
p p
6 3 6 3
p p = p .
3( 3 − 1) 3 − 3
It hasn’t rationalised it, in fact we’ve now got square roots in the numerator
and the denominator.
The impasse is overcome by using the difference of two squares pattern:
p p p
( 3 − 1)( 3 + 1) = ( 3)2 − 12 = 2.
p p
Multiplying the top and bottom by 3 + 1 (rather than 3) we have
p p
6 6( 3 + 1) 6( 3 + 1) p
p = p p = = 3( 3 + 1)
3−1 ( 3 − 1)( 3 + 1) 2
It succeeds in rationalising it (eliminating any roots from the denominator).

p p
We say that 3 + 1 is the conjugate surd of 3 − 1, or—because each is the
conjugate of the other—we say they form a conjugate pair.
A few examples of conjugate pairs. Notice that there will only ever be two
terms but both may contain square roots:
p p
3+ 2 3− 2
p p
3 5−1 3 5+1
p p
8+ 3 8− 3
p p p p
3 2+ 7 3 2− 7
For the difference of two squares to work, the pair must be identical except
for the sign.

Another example. To rationalise

p
3 7+1
p .
3+ 7
p
The conjugate surd is 3 − 7, and
p p
(3 + 7)(3 − 7) = 9 − 7 = 2
p
so, multiplying top and bottom by 3 − 7, we have
p p p
(3 7 + 1)(3 − 7) 8 7 − 18 p
= = 4 7 − 9.
2 2
And an example in which there are two square roots in the denominator:
p
2
p p .
3 2+ 7
p p
Multiplying top and bottom by the conjugate surd 3 2 − 7, we have
p p p p p
2(3 2 − 7) 6 − 14 6 − 14
p p p p = =
(3 2 + 7)(3 2 − 7) 18 − 7 11

It may have crossed your mind that we might be able to get rid of square
roots from a denominator by squaring top and bottom? There are two very
good reasons why this would not work.
Firstly, squaring a fraction alters its value! Multiplying top and bottom by the
same number doesn’t—it’s the principle of equivalent fractions.
Secondly, even if squaring were appropriate, there’s another problem.
Suppose we wanted to square the fraction
3
p p
2+ 5
an
¡ a ¢n
We know from the index rule b = b n that
µ ¶2
3 9
p p = p p
2+ 5 ( 2 + 5)2
p p
but ( 2 + 5)2 6= 2 + 5. We have a bracket to multiply out:
p p p p p p p
( 2 + 5)( 2 + 5) = 2 + 2 2 5 + 5 = 7 + 2 10.
It’s a timely reminder that the bracket rule for indices applies to
multiplication and division, but not to addition or subtraction:
(a + b)n 6= a n + b n
(a − b)n 6= a n − b n .
We end this section with a few more examples of rationalising surds that bring
out different aspects.

Example 7
p p p
3−1 ( 3 − 1)2 3−2 3+1 p
p = p = = 2− 3
3 + 1 ( 3)2 − 12 2
p
If you are tempted to consider cancelling 3, remember we can only cancel
if we can find something that divides into every term in the numerator and
every term in the denominator. Only then do the numerator and
denominator share a common factor.
Example 8
p p p p p p p p
6 6( 6 − 3) 6− 2 3 3 6−3 2 p
p p = = = = 2− 2
6+ 3 6−3 3 3
p p p
Seeing the 2 and the 3 “wrapped up" in the 6 is the key here. Another
way to do this would be to write:
p p p p p
6 2 3 2ZZ
3
p p =p p p =p p
6+ 3 2 3 + 3 ZZ 3( 2 + 1)
giving p p p
2 2( 2 − 1) p
p = = 2− 2
2+1 1
Example 9
p p p p
1 1 3+1 3−1 3 + 1 − ( 3 − 1) 2
p −p = − = = = 1
3−1 3+1 2 2 2 2
Notice how introducing a bracket helps us deal correctly with the signs—we
need to subtract the whole of the right hand fraction.
How beautifully some surds simplify!

QQ
Rationalise these expressions

p
5 6 2 21 3
(a) p (b) p (c) p (d ) p p (e) p
2+ 3 6−2 2+1 2 3− 5 5+ 6
If you are intrigued and want to see some wonderful, more challenging
examples see page A-25 in Appendix A (or click on the page number).
p p p p p p
32
(a) 5(2 − 3) (b) 6( 6−4
6+2)
= 3( 6 + 2) (c) 2( 2 − 1) = 2 − 2
p p p p p p
21(2 3+ 5)
(d) 12−5 = 3(2 3 + 5) (e) 3(5− 6)
25−6 =
3(5− 6)
19

A little irrational exploration

In this chapter we looked in some depth at indices, and then at surds. What if
we were to combine these ideas? This thought came to me when I stumbled
across a mention of the number
p p2
2
in a book. Yes it is perhaps sad that I would be reading a Maths book rather
than a good novel.
p p p2
We know 2 is an irrational number. We wouldn’t be surprised if 2 were
as well, but that’s not the same as knowing it is for sure. So I Googled it.
Apparently mathematicians couldn’t say either way, until 1934, when
Alexandr Gelfond and Theodor Scheider proved that it is indeed irrational.
(Actually they proved something more general but that’s another story.)
We are into the realms of extremely esoteric stuff here, but with what we
already know, we can prove that an irrational number raised to the power of
another irrational number could be rational.
To do it we are going to look at two numbers:
µ p ¶p2
p p2 p 2
2 and 2
and we are going to show that one of these numbers is definitely rational.
The reasoning goes like this:
p p p2
We know 2 is irrational so if 2 is rational we have proved the result.
(Actually we know it isn’t but we are trying to prove it without using a fancy
p p2
theorem.) OK, so the only alternative is that 2 is irrational.
But—and here comes the good bit
¶p2
p p2 p p p p
µ
2 = ( 2) 2× 2 = ( 2)2 = 2
which is rational. We have discovered an irrational number raised to the

power of another irrational number, the result of which is rational. Neat, eh?
We end with a complicated looking expression for you to simplify. Don’t be

daunted: the answer is lovely:
p
p p  2
1+ 2
 2 p  .
p 1− 2
2
See the footnote33 .
33
You’ll know if you’ve got it: the answer is an integer.

Exercises 8
1. By collecting and/or combining square roots, simplify these as far as

possible.
p p p p
(a) ( 3)3 3 (b) ( 2)10 (c) ( 5)7
p p p p p
(d) ( 8)3 (e) ( 8)3 2 (f ) ( 3)5 + 2 3
p p p p 18
(g) 4( 2)3 − 2 (h) 8( 2)6 − ( 2)12 (i) p 4
( 3)
p
100
p 8 12 3
(j) (k) p 3 p (l) p p
2( 5)3 ( 2) +3 2 6 3−( 3)3
p p 3
3( 5)
(m) p
15
2. Simplify each of these as far as possible.

p p p p p p
(a) 3( 3 + 2) (b) 4 7 + 2(1 − 7) (c) 3 11 − (1 − 11)
p p p p 6
(d) 3 2(2 + 2) (e) 2 5(3 5 − 4) − 29 (f ) p
6
20 p p p p
(g) p (h) 6 − 3(1 + 3) (i) (6 − 3)(1 + 3)
10
p 12
p 25 p p
(j) 6 + p (k) 14 5− p (l) ( 2 + 3)( 2 − 1)
6 5
p p p p
(m) (2 3 + 1)(3 − 3) (n) (4 2 + 3)2 (o) ( 6 − 1)2
p p p
2 p3+9
(p) (1 + 3)2 (q) (1 + 3)3 (r)
3
p p p p
10+4
p 5 1
(s) (t) p (4 8 + 8) (u) (2 5 + p1 )( 5 + 10)
2 5 8 5
1 p p p 2 p
(v) ( p +4)2 (w) (3 + 2)2 + (3 − 2)2 (x) (3 + 2) − (3 − 2)2
2
(Can you see a quick method for part(x)?)
3. By extracting square roots wherever possible, simplify these.

p p p
(a) 8 (b) 20 (c) 200
p
(d) p3 (e) 21
p (f ) 256
3 7
p p p p p
4p18
(g) (h) 12 + 5 3 (i) 7 2 + 18
2
p p p p p p
(j) p175 (k) 50 − 32 (l) 2 27 + 12 − 75
28
p p p p p p p
(m) 3 × 48 (n) 40 × 90 (o) 2 × 14 × 7
p p p p p p p
15×
p 5
(p) 18 × 8 (q) (r) 22 × 6 × 33
2 3
p p p
(s) p 10 (t) p32 (u) p p24
250 72 3× 128
q
9
q
100
p
(v) 16 (w) 49 (x) 0.64
p
(y) p0.9 (z) p1
1.6 0.04

4. Rationalise each of these surds, and simplify your answer.

p p
8 21 12
(a) p (b) p (c) p
2 7 2 6
p p p
15 16 3 4 2
(d) p (e) p (f) p
10 6 8
p
1 5 8 7
(g) p (h) p (i) p
3 3 2 7 2
p
14p 16 3 6
(j) (k) p (l) p
3+ 2 5−1 6+1
p p
6 3 6p 5−1
(m) p (n) p (o) p
3 3−5 4 2−2 3 5+1
p p
2+p3 2 3+3 20 p
(p) (q) p (r) p −3 8
2− 3 2 3−3 8
p
4 5−1 p 28 1
(s) p (t) 7+ p (u) p +1
5+2 14 2+1
p p
(v)
2− 3
1p
+ 2+ 3 (w) 2− 3 − 2+ 3 (x) p5−p2 + p5+2p2
1p 1p 1p 2
5. Find a simpler way to write each of these expressions.

p p
p a+a 4 x−x
(a) ( y + 2)2 (b) p
a
(c) p
x
p p p p
(d) (3 x + 1)2 (e) ( p2 − k)2 (f) ( p − 1)2 − 2 p
k
p 3 p p p p
(g) x( x + 3) − x 2 (h) (2 y + 1)( y − 2) (i) 4 t − (2 + t )2
p p p p p
(j) ( x + p1x )2 (k) ( x + 3)2 + ( x − 3)2 (l) ( x + 3)2 − ( x − 3)2
6. Show that
p p p p
2+1 2− 3 8+ 2
(a) p
2+ 3
−p2−1
= 0 (b) p p = 3
8− 2
p p p p
r r
2−p3 3+p5 1
(c) = 2− 3 (d) = 2
(3 + 5)
2+ 3 3− 5
rp
p p 1− p 1 p
3+1
(e) p6−2 = 3− 2 (f ) 1 = 3
6+2 p −1
3−1
p
7. A sphere of radius 3 cm is inscribed in a cylinder (as shown in the
diagram). [Inscribed here means touching at all surfaces.]
(a) Show
p that the volume of the sphere is
4 3 π m and find the volume of the
3
cylinder.
[Leave your answer in terms of π.] p
3 cm -
(b) Show also that the surface area of
the sphere is the same as the
curved surface area of the cylinder.

8. (a) Find the exact area of an equilateral triangle of side 2 units.

(b) Write down the area of a regular hexagon of side 2 units.
(c) The hexagon is inscribed in a circle. Find the area between the
hexagon and the circle?
9. Find the hypoteneuse of a right-angled triangle whose other sides are

of length
r r
x 1 x 1
+p and −p .
2 2 2 2
p
10. The diagram shows a circle of radius a surrounded by an annulus (or
1
ring) of width pa .
(a) Show that the area of the annulus is
µ ¶
1
π 2+ .
a
(b) Hence show that, as a gets bigger, the area of the annulus gets
smaller.
p p1
a a
--
259 p
11. (a) Evaluate ( π − 1) on your calculator.
200
Explain why its exact value cannot be 1.
p
5
(b) Explain why 69, 343, 950 cannot be exactly 37.
p p p
12. (a) Expand (2 + 3)2 . Hence find, in surd form 7 + 4 3.
p p
(b) Use the same method to find, in surd form 3 + 2 2.
13. Find the values of the positive integers a and b, if

p p p
(a + 3)(b − 3) = 7 + 3 3.
14. Given that p

a 3−1
p =
a 3+2 2
find the value of a.
15. Find the exact length of the hypoteneuse in a 30◦ , 60◦ , 90◦ triangle that
has a total perimeter of 20 cm.
(This problem is solved approximately on page 6-16. The exact trig
ratios you will need for sin, cos and tan are derived on page 12-7.)
p
2 2 1
16. Rationalise (a) p p , and (b) p p p .
3+ 2−1 2+ 3+ 5

Solutions 8
p
1. (a) 9 (b) 32 (c) 125 5
p p p p p p p
(d) 8 8 or 16 2 (e) 16 2 2 = 32 (f ) 9 3 + 2 3 = 11 3
p p p
(g) 8 2 − 2 = 7 2 (h) 64 − 64 = 0 (i) 2
p p p
50 10 8 8 4 2 12
3
(j) p = p = 2 5 (k) p p = p = 5
(l) p= 4
p
5 5 5 2 2+3 2 5 2 63−33
p p
3×5× 5
(m) p p = 5
5× 3
p p p
2. (a) 3 + 2 3 (b) 2 7 + 2 (c) 4 11 − 1
p p p p
(d) 6 2 + 6 or 6( 2 + 1) (e) 1 − 8 5 (f) 6
p p p
(g) 2×10
p = 2 10 (h) 3 − 3 (i) 3 + 5 3
10
p p p p p p p
(j) 6 + 2 6 = 3 6 (k) 14 5 − 5 5 = 9 5 (l) 2 2 − 1
p p p
(m) 5 3 − 3 (n) 41 + 24 2 (o) 7 − 2 6
p p p p
(p) 4 + 2 3 or 2(2 + 3) (q) 2(5 + 3 3) (r) 2 + p9 = 2 + 3 3
3
p p p
(s) 5 + 2 (t) 4 + 8 (u) 11 + 22 5
33 p p p
(v) 2 +4 2 (w) 22 (x) 6 × 2 2 = 12 2
[In part (x) the quickest method is to use the difference of two squares.]
p p p p
3. (a) 2 2 (b) 2 5 (c) 10 2 (d) 3
p p
(e) 3 7 (f ) 16 (g) 12 (h) 7 3
p 5 p p
(i) 10 2 (j) 2 (k) 2 (l) 3 3
(m) 12 (n) 60 (o) 14 (p) 12
p
4p2
(q) 5
2 (r) 66 (s) 1
5 (t)
6 2
= 32
1 3 10
(u) 4 (v) 4 (w) 7
q p q
9 1
(x)
64
100
= 8
10
= 0.8 (y) p
16
= 3
4
(z) p 4
= 100
4
= 5
100
p p p p
2
4. (a) 4 2 (b) 3 (c) 2 or 21 2
q p p p p
(d)
15
= p3 = 6 16 3
(e) p p = 8 2 (f) 2
10 2 2 3 2
p p p
3 5 7
(g) 9 (h) 14 (i) 4 14
p p p p p
3 6( 6−1) 3(6− 6)
(j) 2( 3 − 2) (k) 4( 5 + 1) (l) 5
= 5
p p p p p
3 6(2 2+ 3) 3− 5
(m) 3(3 3 + 5) (n) 10
(o) 2
p p p
(p) 7 + 4 3 (q) 7 + 4 3 (r) − 12 8
p p p p
(s) 22 − 9 5 (t) 7 + 2 14 (u) 2
p p
2 10
(v) 4 (w) 2 3 (x) 3

p p p p
5. (a) y + 4 y + 4 (b) 1 + a (c) 4 − x (d) 9x + 6 x + 1
4 p 3 p
(e) k
−4+k (f ) p + 1 (g) 3x because x x = x 2 (h) 2y − 3 y − 2
p p
(i) −4 − t (j) x + 2 + x1 (k) 2(x + 9) (l) (2 x)(6) = 12 x
[Part (l) by the difference of two squares.]
p p p p
( 2+1)( 2−1)−(2− 3)(2+ 3)
6. (a) p p . The numerator is 0.
(2+ 3)( 2−1)
p p
(b) We could rationalise it as it stands. A quicker way is to notice that 8 = 2 2,
p p
2 2+ 2 2+1
so the l.h.s. could be written p p = 2−1 .
2 2− 2
In the next two questions we rationalise the fraction.
q p p
2
(c) (2−4−33) = 2 − 3
q p p
2
(d) (3+9−55) = 3+p 5
4
(e) If we adopt the same approach for this one we get

qp p p p p
( 6−2)2 6−2
p p6 − p2 = 3 − 2.
6−4
= =
2 2 2
p p p p p
There is a slightly quicker, and neater, way. Since 6 = 3 2, and 2 = 2 2 we can divide
p
top and bottom by 2, to give
rp p p p p p
( 3− 2)2
p3−p2 and the fraction p3−p2 rationalises to .
3+ 2 3+ 2 1
p p p
1 ( 3−1) 3− 3 1 3−1
(f) 1 − p = 1 − 2 = 2 . Similarly p = 2 .
3+1 p 3−1
3− 3
The entire fraction is therefore p . Once again we can rationalise this in the usual way,
p p 3−1
3( 3−1) p
or write it as p = 3.
3−1
In parts (b), (e) and (f) we saw how it made things a bit easier to do
some decomposing of surds.
4 4 p 4 p p
7. (a) Volume of sphere is 3 π r 3 = 3 π ( 3)3 = 3 π 3 3 = 4 3 π cm 3 .
p p p p
Height of cylinder is 2 3, so volume is π ( 3)2 × 2 3 = 6 3 π cm 3
(b) Surface area of the sphere is 4πr 2 = 12π cm 2 .

Curved surface area of the cylinder is circumference × height
p p
= 2π 3 × 2 3 = 12π cm 2 .
p p
8. (a) Area= 21 × 2 × 3 = 3 square units. [See section 7.4.1.]
(b) The hexagon can be thought of as 6 equilateral triangles pushed together,

p
so its area is = 6 3 square units.
p
(c) The radius of the circle is 2 units, so the area is π 22 − 6 3
p
= 2(π − 2 3) square units.
9. Let hypoteneuse be h, then

³q ´2 ³q ´2
x p1 x p1
h2 = 2
+ + 2
−
2 2
µ q ¶ µ q ¶
x Z x 1 1 x Z x 1 1
= 2+ Z 2 Z2 + 2 + 2 − 2 = x + 1.
p Zp +
2 Z2
Z p Z
Hypoteneuse is x + 1.

³p ´2 p 1
(a) Area of outer circle is π a + p1a = π a + 2 + a1 , because a× p = 1.
¡ ¢
10.
a
p 2 1
Area of inner circle is π( a) = πa. The difference is π 2 + a .
¡ ¢
1
(b) As a gets bigger, a gets smaller, so the area gets smaller.
[The inner radius gets bigger but the ring gets narrower—to such an extent that the net effect
is a reduction in the area of the ring.]
11. (a) 1.0003277 (to 8 s.f.)

259 p
Suppose we assume for the moment that 200 ( π − 1) = 1.
p
This would mean that π = 200 259
+ 1, which is rational.
But if a number is rational its square is also rational, which would imply that π is rational.
259 p
We know this is not the case, so our assumption that 200 ( π − 1) = 1 must have been wrong.
p
(b) If 5 69, 343, 950 were equal to 37 exactly, then 375 would equal 69, 343, 950.
This can’t be so, because the result of multiplying a number ending in 7 by itself
—any number of times—can never be a number ending in a 0.
This kind of reasoning is given a fancy name: proof by contradiction.
p p p p p
12. (a) (2 + 3)2 = 7 + 4 3, ∴ 7 + 4 3 = 2 + 3.
p p p p p
(b) By inspection (1 + 2)2 = 3 + 2 2 ∴ 3 + 2 2 = 1 + 2.
13. (a) a = 2, b = 5.
p p p p
(a + 3)(b − 3) = (ab − 3) + (b − a) 3 = 7 + 3 3
[We want two positive integers whose product is 10 and difference is 3.]
14. (a) a = 2.
p p p p
2a = ( 3 − 1)(a 3 + 2) = 3a − 3a + 2 3 − 2.
p p
Collecting terms: a( 3 − 1) = 2( 3 − 1).
p
15. (a) 20
3 (3 − 3).
p p
b 23 + b 12 + b = 20, which simplifies to b(3 + 3) = 40.
40p .
Then rationalise: b =
3+ 3
p p
16. (a) 3 − 2 + 1.
We use the standard approach but need two shots at it. Thinking of the denominator as
p p p p
3 + ( 2 − 1), we multiply top and bottom by 3 − ( 2 − 1).
p p p
p 2
p 2 2( 3−( 2−1)) p p
The new denominator is 3 − ( 2 − 1) = 2 2, giving p = 3 − 2 + 1.
2 2
p p p p p p p
6 1
(b) 12 ( 2 + 3 − 5) or 12 (2 3 + 3 2 − 30)
p p p
Treating the denominator as ( 2 + 3) + 5, we have
p p p p p p p p p p
( p2+ p3)− 5 ( 2+ p3)− 5 6( 2+ 3− 5)
= =
2
( 2+ 3) −5 2 6 12
You may have chosen to group the terms in the denominator differently, for example as
p p p p p p
2 + ( 3 + 5). Multiplying top and bottom by 2 − ( 3 + 5), the new denominator is
p
−6 − 2 15. Grouping in this way has not benefited from the fact that 2 + 3 = 5,
and the second stage of rationalising requires a bit more work:
p p p p p p p
p p p
2− 3− p 5 = ( 2− 3− 5)(3− 15) = 1 (2 3 + 3 2 − 30).
−2(3+ 15) −2(−6) 12


• The basic rules of indices:
am
¡ a ¢n a n
m n m+n m−n n n n
a ×a = a , = a , (ab) = a b , = n
an b b
1 m p p
(a m )n = a mn , a −n = n , a n = a m = ( n a)m . a 0 = 1
n
a
1
• Applying associativity and commutativity e.g. 16a 3 × 2a = 8a 2
5 1 1 1
• Negative fractional indices e.g. 8− 3 = 5 = 25 = 32
83
¡ 16 ¢− 1 ¡ 25 ¢ 1 p
25 25 2
• Simplifying by combining rules e.g. 25 =p =4 = 16
16
p 1 3 p 5
• Converting from root to index form e.g. 3 3 = 31 ×3 2 = 3 2 , ( 2)5 = 2 3
3
1 1 1
• Factorising the base e.g. 64000 3 = (26 ) 3 × (103 ) 3 = 22 × 10 = 40
1
• Equations e.g. 7x = 49 ⇒ x = −2, and 271−x = 92x ⇒ 33−3x = 34x ⇒
3
x =7
p p p p
6 6 3
• Surds e.g. p = 3 = 2, (4 2 + 1)( 2 − 3) = 5( 2 + 1)
3
p p p p
2 3−1 (2 3−1)(p 3−1) 7−3 3
Rationalising the denominator e.g. p = p = 2
3+1 ( 3+1)( 3−1)
p p p p p p
Decomposing e.g. 75 × 48 = 3 × 25 × 3 × 16 = 3 × 5 × 4 = 60
• Common errors e.g. 23 + 25 6= 27 , (52 )4 6= 56 , (x + 2)2 6= x 2 + 4, 23 × 34 6= 67

1 1
p 2 q −1 6= p 2 q , 4(x + 1)−2 6= 4(x+1)2

Chapter 3
BRACKETS AND FACTORISING
Overview
If you don’t need this chapter and are looking for solving quadratic equations,
go straight to Chapter 7.
3.1 Expressions, terms and coefficients. Terminology.
3.2 Multiplying out brackets. For example
3pq − p(1 − 2q), 4(x + 2)(x − 3), (2x + y)2 , (a + b − 2c)3
A proof of Pythagoras’ Theorem.
3.3 Polynomials. Adding subtracting and multiplying them. For example
To find 2g (x) − f (x), where f (x) = x 2 − 2 and g (x) = 2 + 3x + x 4 .
Finding the coefficient of x 4 in the polynomial (x 3 + x 2 + 1)(2x 2 − 5)2 .
3.4 Factorising and the Highest Common Factor in algebraic expressions.
3.5 One-bracket factorising, i.e. taking out a common factor. The factor
could be a bracket and we may need to form the bracket. For example
8x 2 y + 16x y 2 − 12x 4 , 4(1 − 2x) − (1 − 2x)2 , 2x y − 3y − 4x 2 + 6x
3.6 Two-bracket factorising.
Quadratics of the form ax 2 + bx + c. Various cases. For example:
8x 2 − 18x, x 2 − 8x + 15, x 2 − 3x − 10, 7x 2 − 29x + 4, 2x 2 − 7x y − 9y 2
The difference of two squares. For example 4x 2 − 9, 5 − 80y 2
Perfect squares. For example x 2 + 6x + 9 = (x + 3)2 , 4x 2 − 4x + 1 = (2x − 1)2
Disguised quadratics. For example
p
x 3 + 2x 2 + x, x 4 + 4x 2 + 3, x + 6 x − 7
3-1
CHAPTER 3. BRACKETS AND FACTORISING
3.1 Expressions, Terms and Coefficients

Take something like 2x 2 − 3x + 5. We call this an expression and it consists of
three terms added together: an x 2 term (2x 2 ), an x term (−3x) and a
constant term (+5).
We don’t think of an expression as a series of additions or

subtractions: all the terms are added, but some of them may
be negative.
Within each term there is typically some multiplication going on, although
not always.
A term will usually be of the form

constant part× algebraic part
in which case we call the constant part the coefficient. In the example above
the coefficient of x 2 is 2, the coefficient of x is −3 and the 5 is called the
constant coefficient (or constant term).
To take another example, the expression 5y x 2 + y 3 has two terms, the y x 2

term has a coefficient 5 and a the y 3 term a coefficient 1.
Actually the notion of a coefficient is slightly more subtle than this. Here is a
variation on the quadratic expression we started with:
2x 2 + ax − 5.
In this case we would not describe the second term as an ax term with
coefficient 1. We would say it is still an x term with coefficient a. This is
because the variable is clearly x, and a is a constant (a fixed number), even if
we don’t know what that number is yet.
To help distinguish variables from constants you will find that variables are
often labelled using the letters towards the end of the alphabet e.g.
x, y, z, t , u, v etc, and constants using letters near the beginning, e.g. a, b, c, d ,
although this is not a hard and fast rule. When we run out of letters
mathematicians tend to call on the letters of the Greek alphabet e.g. α, β, γ, δ
etc, pronounced alpha, beta, gamma, delta1 . Greek letters usually stand for
constants.
1
A clue to where the English word alphabet comes from.

Combining like terms

When we are simplifying expressions, we can combine together terms for
which the algebraic parts are identical. Such terms we call like terms (terms
that are like each other in the sense that only their coefficients differ).
To take a trivial example, we can see that combining the terms in 3a + 2a

makes sense (think 3 apples plus 2 apples), whereas trying to combine
3a + 2b doesn’t (3 apples plus 2 bananas?). The terms 3a and 2a are like
terms, whereas 3a and 2b are not, and neither are 3a and 2a 2 . To be like
terms the algebraic parts need to be identical 2 .
Each of the expressions below have like terms, which can be combined:
9x + 4x + x = 14x
y 3 + 7y 3 = 8y 3
10pq 2 − 4q 2 p = 6pq 2
In the last example, we have like terms because p × q × q = q × q × p (the

multiplication can be done in any order).
In the examples below there are no like terms, so we can’t combine any of
them
2x 2 + 3x
2p 2 q + 7pq 2
a 2 − 3ab + 4b 2
y4 + y3
QQ
Simplify these expressions by combining terms where possible (the answers

are in the footnote3
(a) x 3 + 2x 2 + 5x 2 − x 3 (b) 1 + y + y 2 + y 3 (c) 6pq − 2q p
(d ) 12 x 2 + 12 x 2 − x 2 (e) a 2 b + 2ab − b 3 ( f ) 2p 2 qr 2 + 3r 2 p 2 q
(g ) 3x 4 y + x 2 y − x y 4 + 7y x 2
2
In fact it’s a bit more subtle than this. To see why, turn to Annex A page A-11
(or click on the page number).
3
(a) 7x 2 (b) None possible. (c) 4pq (d) 0 (e) None possible. (f) 5p 2 qr 2
4 2 4
(g) 3x y + 8x y − x y

3.2 Multiplying out brackets

You will have met the idea of multiplying out brackets, sometimes referred to
as expanding brackets. At A Level it will be important for you to understand
why we do what we do. Look at this expression
4(x + 2) − 5(3 − 2x).
Within it we see more than one occurrence of x and also some numbers. It
would be good to combine like terms, but as it stands the like terms are
“wrapped up" within the brackets; to release them we need to write the
expression without brackets. Expanding the first bracket is straight forward:
4(x + 2) = 4x + 8.
We can think of this as two different ways to describe the area of a 4 by (x + 2)

rectangle.
x 2
For the second bracket, we use the box principle we met in Chapter 1 (see
page 1-45), which says we should see every sign as being attached to the
number that follows it. So in this case we don’t think of the expression as
“subtract 5(3 − 2x) from 4(x + 2)"; instead we see the minus sign in the
middle as part of the directed number −5, which must be multiplied by
everything inside the (3 − 2x) bracket.
4(x + 2) − 5(3 − 2x) = 4x + 8 −5 ( +3 −2x )

= 4x +8 −15 +10x
= 14x − 7
We’ve put the boxes in here to underline the thinking process. In practice you
won’t need to, not once the process has become second nature.

So far so good. Now let’s see how the box principle applies to an example like
6a − (2a − 5).
This looks like a common or garden subtraction? The simplest way to tackle
this is to insert an imaginary −1 in front of the bracket. Then we can apply
the box principle:
6a − (2a − 5) = 6a −1 ( 2a −5 )
= 6a −2a +5
= 4a + 5
Tackling it this way means we sidestep the issue of which are additions and
which are subtractions.
Notice that −1 × −5 = +5. If you need a reminder about what happens when
we multiply directed numbers together see page 1-47.
Another way to understand the last example is to think about what the
bracket means: it tells us we need to subtract the 2a and subtract the −5. But
subtracting −5 means −− 5 = +5. So
−(2a − 5) = −2a + 5
The box principle is particularly helpful in a more complicated looking

example like this (notice the imaginary 1 in front of the last bracket):
−6y + 2y(4 − y) − (1 + y 2 ) = −6y +2y ( +4 −y ) −1 ( +1 +y 2 )

= −6y + 8y − 2y 2 − 1 − y 2
= −3y 2 + 2y − 1
Look at the start of this expression: it appears to begin with −6y + 2y, which
would combine into −4y. The reason we can’t do this is the BIDMAS (or
BODMAS)4 rule, which tells us we must do any multiplication or division
before any addition or subtraction. That means any brackets have to be
multiplied out before we can start collecting terms; only after we have
multiplied out +2y(4 − y) can we look for like terms to combine them with.
4
B=Brackets, I=Indices (O=Of ), D=Divide, M=Multiply, A=Add, S=Subtract.
For example: 7(23 − 5). We must work out what’s in the Bracket before Multiplying
by 7, and in the bracket we must apply the Index 3 in 23 before Subtracting the 5.
The answer is therefore 7(8 − 5) = 21.
1
Since “Of" means multiply, as in 4 of 12 = 14 × 12
1 = 3, we can dispense with
BODMAS.

Some more examples:
5q(p − 1) + 3p − p(7 − 2q) = 5q p − 5q + 3p − 7p + 2pq

= 7pq − 5q − 4p
ab 3 + 2ab(5a 2 − 6b 2 ) = ab 3 + 10a 3 b − 12ab 3

= 10a 3 b − 11ab 3
x 2 (x 2 + y 2 ) − (x 4 − y 4 ) = x 4 + x 2 y 2 − 1(x 4 − y 4 )
= x4 + x2 y 2 − x4 + y 4
= x2 y 2 + y 4
QQ
Try these examples for yourself (the answers are in the footnote5 ).
(a) 6x + 7(2 + x) (b) 3(2y − 1) + 5(2 − y) (c) −9x − 2(x − 1)

(d) −2p(p + 1) − (1 − 2p 2 )
5
(a) 13x + 14 (b) y + 7 (c) −11x + 2 (d) −2p − 1.

Now we come to the case of two brackets multiplied6 together. This means
everything in the left hand bracket has to be multiplied by everything in the
right hand bracket. For example
(x + 3)(x + 2)
A simple diagram helps us to see how we can write the expression without
brackets. We can think of the the problem as that of finding the area of a
rectangle measuring x + 3 by x + 2. This breaks down into four smaller
rectangles (whose areas are shown in circles).

3 3x 6

x x2 2x

x 2
This shows us that
(x + 3)(x + 2) = x 2 + 3x + 2x + 6 = x 2 + 5x + 6
In practice we just need to make sure we multiply every term in one bracket
by every term in the other, which we could set out like this:
(x + 3)(x + 2) = x(x + 2) + 3(x + 2) = x 2 + 2x + 3x + 6 = x 2 + 5x + 6
Here are two more interesting examples. They follow exactly the same
princple but in each case there is an added layer of complexity. In the second
example we have to be very vigilant about the signs. To ensure we subtract
all the terms that arise when we multiply the two brackets, we have
introduced an extra pair of brackets (we’ve used square ones).
(3p 2 + q)(4q − p 2 ) = 3p(4p 2 − q) + q(4p 2 − q) = 12p 3 − 3p 2 q + 4p 2 q − q 2

= 12p 3 + p 2 q − q 2
b(2a + b) − (2a − b)2 = 2ab + b 2 − (2a − b)(2a − b)

= 2ab + b 2 − [2a(2a − b) − b(2a − b)]
= 2ab + b 2 − 1[4a 2 − 2ab − 2ab + b 2 ]
= 2ab + b 2 − 4a 2 + 2ab + 2ab − b 2
= 6ab − 4a 2
= 2a(3b − 2a)
6
In algebra being adjacent means “multiply", which is why we write 3y to mean 3 × y.

We are not confined to two terms in a bracket, of course. We can have as many
as we like as long as we multiply all the terms in one by all the terms in the
other. For example:
(4a + b)(3a − b + 5) = 4a(3a − b + 5) + b(3a − b + 5)

= 12a 2 − 4ab + 20a + 3ab − b 2 + 5b
= 12a 2 − b 2 − ab + 20a + 5b
(x + 2y − 3z)2 = (x + 2y − 3z)(x + 2y − 3z)

= x(x + 2y − 3z) + 2y(x + 2y − 3z) − 3z(x + 2y − 3z)
= x 2 + 2x y − 3xz + 2y x + 2y 2 − 3z y − 3zx − 2z y + 9z 2
= x 2 + 2y 2 + 9z 2 + 4x y − 5z y − 6zx

QQ
Try these examples for yourself (the answers are in the footnote7 ).
(a) (x + 4)(x + 7) (b) (3x + 5)(2x − 1) (c) (5x + y)2

(d)(a + 2b − 4)(3a − b + 2)
7
(a) x 2 + 11x + 28 (b) 6x 2 + 7x − 5 (c) 25x 2 + 10y + y 2
(d) 3a 2 − 2b 2 − 12a + 4b − 8 .

The happy face rule

You may have met the “happy" face rule for multiplying when there are two
terms in each bracket. Taking one term from the left bracket and one from
the right bracket gives 4 pairs of terms—each of these pairs must be
multiplied, and the resulting terms added together.
It is common to do the two top multiplications and then do the two bottom
ones, although of course it doesn’t matter which order we do them in. The
important thing is not to forget any of them.
× ×
(x + 3)(x + 2) = x 2 + 3x + 2x + 6 = x 2 + 5x + 6
×
×
Where there are negative numbers remember the box principle:
(x -6 )(x + 4) = x 2 − 24 − 6x + 4x = x 2 − 2x − 24
and
(4x -3 )(5x -2 ) = 20x 2 + 6 − 15x − 8x = 20x 2 − 23x + 6
We only use the happy face method when there are two terms in each
bracket. Repeate examples (a) and (b) on the previous page using this
method:
(a) (x + 4)(x + 7) (b) (3x + 5)(2x − 1).

The happy face method speeds things up a bit. And it can be used whenever
we have two brackets each containing two terms—whatever the terms are, for
example
(a + 3b)(4a + b) = 4a 2 + 13ab + 3b 2 and (p 2 − 2q)(p 2 + 2q) = p 4 − 4q 2
We have
(a + 3b)(4a + b) = 4a 2 + 3b 2 + 12ab + ab = 4a 2 + 13a + 3b 2

and
(p 2 -2q )(p 2 + 2q) = p 4 − 4q 2 + 2p 2 q − 2q p 2 = p 4 − 4q 2
Sometimes the expression in the bracket will be reversed, like in this

example
(6 − 5x)(x + 1).
In this case you can also think of it as
(−5x + 6)(x + 1) = −5x 2 + x + 6.
For the product of two brackets, we multiply all possible pairs of terms,
taking one from each bracket. Then we add together all the answers.
We never multiply terms from the same bracket.

So far so good.
The next step is to multiply brackets in your head,

i.e. without writing down the intermediate steps.
Why is it important to be able to do this? Well, because unless you can do

this you will find the reverse process—factorising into two brackets8 —well
nigh impossible.
QQ
Multiply these expressions out and collect the terms completely in your head.
(a) (x + 5)(x + 2) (b) (x + 3)(x − 7) (c) (3x + 1)(4x + 2)

(d) (8 − x)(2x − 3) (e) (x − 1)2
The exercises at the end of this section will provide you with more practice at
this.
8
See Section 6.
9
(a) x 2 + 7x + 10 (b) x 2 − 4x − 21 (c) 12x 2 + 10x + 2
(d) −2x 2 + 19x − 24 (e) x 2 − 2x + 1

When the multiplying out is part of a bigger simplification, there can be a lot
to think about and plenty of scope for making mistakes. Take an expression
like this, where minus signs abound (this is definitely not one to attempt
completely in your head!)
−(4x − 7)(x − 1) + 2x − 6x(x + 3).
Don’t forget the order of simplifying needs to follow the BIDMAS rule, and
for that reason we can’t combine the 2x with the 6x (because that would
mean doing a subtraction before doing all the multiplying).
We’ll start with −(4x − 7)(x − 1). One approach10 would be to expand (4x −
7)(x − 1) and multiply the answer by −1,
i.e. −(4x 2 − 11x + 7) = −4x 2 + 11x − 7 .
Combining this with the rest of the expression:
(−4x + 7)(x − 1) + 2x − 6x(x + 3) = −4x 2 + 11x − 7 + 2x − 6x 2 − 18x

= −10x 2 − 5x + 7
The thing to notice is that none of the steps here were difficult: each one you
can do in your head quite reliably, but mistakes can creep in simply because
of the number of steps—the scenario you are most likely to meet in an exam.
QQ
Try simplifying this expression. The answer is in the footnote11 .
−7a + (a + 3)(2 − a) − a(1 − 2a).
10
Or we could regard it as (−1)(4x − 7)(x − 1), and multiply the −1 by the first bracket, to
give (−4x + 7)(x − 1).
11 2
a − 9a + 6 .

Here is an example which hides a trap.
10(1 − x) − (x + 3)2 .
It might be tempting to write
10 − 10x − x 2 + 6x + 9.
Can you see what it should be? The answer is in the footnote12 .
QQ
Try simplifying the following juicy expression. See if you can get it right first
time.
7y + y(y − 2) − 3(1 + y)(2y − 1)
If your first attempt is wrong, see if you can diagnose the problem for yourself.
Get used to the idea that your first answer will sometimes be wrong: it’s par for
the course.
12
The whole of the expression (x + 3)2 is to be subtracted, so first we need to introduce
a bracket: 10 − 10x − (x 2 + 6x + 9) = 10 − 10x − x 2 − 6x − 9 = 1 − 16x − x 2 .
13
Answer: −5y 2 + 2y + 3
First expand (1 + y)(2y − 1) = (y + 1)(2y − 1) = 2y 2 + y − 1, then
7y + y(y − 2) − 3(1 + y)(2y − 1) = 7y + y 2 − 2y − 3(2y 2 + y − 1)

= 7y + y 2 − 2y − 6y 2 − 3y + 3
= −5y 2 + 2y + 3

We end this section with a problem that can be solved simply by multiplying
out some brackets.
Show that, if x is real
(2x−3)2 +6(2x+1) ≥ 15.
First, don’t be put off by the words “if x is real". For us it simply means “for
any value of x" (see Section 1.2). Multiplying out the brackets on the left hand
side:
(2x − 3)2 + 6(2x + 1) = 4x 2 − 12x + 9 + 12x + 6
= 4x 2 + 15
The reason is now clear: 4x 2 ≥ 0, so 4x 2 + 15 ≥ 15.
This problem illustrates an important principle about doing Maths: if you

can’t see how to proceed, see if you can write what’s in front of you in a
different way. My Maths teacher at school used to say “suck it and see". In
other words try out different ways of writing it; see if it helps.

A proof of Pythagoras’ Theorem

We end this section with a neat proof of Pythagoras’ Theorem. It relies on no
more than a clever picture and the ability to multiply out brackets.
Here is any right angled triangle. Call the lengths of the sides a. b and
hypoteneuse c. We want to prove that a 2 + b 2 = c 2 .
XXX
c
XXX
b XXX
XX
X
a
We start by drawing an (a + b) by (a + b) square, with points marked on each

side dividing it into two segments of lengths a and b. Joining these four
points as shown forms a smaller square inside the larger one. We can see
from the diagram above that small square has sides of length c. This smaller
square plus the four right-angled triangles combine to form the larger
square.
b X a
XXX
XXX b
c XXXX
X

a c

c a

X
c
XX XXX
b XXX
XX
X
a b
Now the area of the larger square is (a + b)2 and the area of the smaller
square is c 2 . The four triangles each have an area 21 ab, so all four have a total
area of 2ab.
By writing the area of the larger square in two different ways, we have
(a + b)2 = c 2 + 2ab
Now expanding the bracket
a 2 + + b 2 = c 2 + 2ab
2ab

a2 + b2 = c 2
We have proved Pythagoras’ Theorem.

Exercises 9
1. Multiply out the brackets and simplify where possible.

(a) 5(x + 3) + 2(y − 4) (b) 7y − 4(1 + y)
(c) 3(2c − 1) − (2 − 3c) + c (d) 3x(2 + y) + 4x y
(e) 5(x − 1) − 6(x + 2) + (4x − 3) (f) x 3 − (1 − x 3 )
(g) 2p(q − p) − q(2p − q) (h) −(1 − k) + k 2 − k(1 + k)
(i) 4ab(2a + 3b) (j) x y + x(1 − 2y)
(k) t (t 2 + 1) + 3t (1 − t ) − t 2 (l) 2zx(4z 2 + x)
(m) pq 2 (2p − q) + 4q 2 (p 2 + 2pq) (n) y 2 − (1 + 3y 2 )
(o) (a 2 + b 2 ) − (a 2 − b 2 ) (p) x y(3y − x) − y 2 (x − 2y + 1)
(q) x(1 + x + x 2 ) − (1 + x + x 2 ) (r) (x − 1)(1 + x + x 2 · · · + x n )
2. Multiply these out in your head.
(a) (x + 8)(x + 2) (b) (x − 8)(x − 2) (c) (x + 1)(x + 3)
(d) (x + 7)(x − 2) (e) (x − 4)(x + 10) (f ) (y + 1)(y − 9)
(g) (y − 1)(y + 9) (h) (2x + 3)(x − 5) (i) (4x + 1)(3x + 2)
(j) (2z − 3)(z + 7) (k) (5x − 6)(2x − 1) (l) (4a − 3)(4a + 3)
(m) (8x − 11)(3x + 2) (n) (5t − 12)(5t + 12)
3. In each case multiply out the brackets and simplify your answer.
(a) a(3a − 1)(2a + 1) (b) (5x + 3)(3x − 4)
(c) (1 − 2y)(y + 6) (d) (2ab + 1)(2ab − 1)
(e) (x + 5y)2 (f ) (a − 3b 2 )2
(g) (3y + 1)2 (h) (3y + 1)3
(i) (10t + 1)(10t − 1) (j) (x + 2)(x − 3)(x + 1)
(k) 2x(2x + 1)2 (l) (y − 1)(1 + y + y 2 + y 3 )
(m) (1 − 21 t )(1 + 12 t ) (n) (1 − x + x 2 )(1 + x + x 2 )
(o) (x 2 + 2x + 2)2 (p) (x 2 + 2x + 2)(x 2 − 2x + 2)
(q) (x + 5)(x + 2) − (1 + x 2 ) (r) (y + 1)2 + (3y + 2)
(s) 2(a − 3)2 − 1 (t) (x 2 − 1)(x 2 + 1) − x 2
(u) (2 + x) − (x + 1)(2 − x) (v) (a + 2b)2 − (a − 2b)2
(w) (x + y + x)2 − (x 2 + y 2 + z 2 ) (x) (p + 1)(p + q + 1)2
4. In each case multiply out the brackets.
(a) (10a + 1)(10b − 1) (b) 3(3x + 3 y ) (c) (2x + 1)2
1 1
(d) (4 y − 1)(4 y + 1) (e) (x 2 + 1)(1 − x 2 ) (f) x −2 (x 2 + 1)2
5. The hypoteneuse of a right-angled triangle is of length a + b. One of the
other sides is of length a − b. How long is the third side?
6. (a) A farmer has a rectangular field that is x m long an y m wide (x > y).
The crop he grows in it is proving very successful so he decides to make it 100
m longer and 50 m wider. By how much does this increase its area?
(b) Would it be better to make it 50 m longer and 100 m wider?

7. Show that
(a) 2(a + b)2 − (2a + b)2 + (2a 2 − b 2 ) = 0
(b) (2x + y + z)2 + (x + 2y + z)2 + (x + y + 2z)2 = 5(x + y + z)2 + x 2 + y 2 + z 2
by writing 2x + y + z = x + (x + y + z), etc.
(c) (2x + y + z)2 + (x + 2y + z)2 + (x + y + 2z)2 can also be written
6(x 2 + y 2 + z 2 ) + 10(x y + y z + zx).
(d) (a + b − c)2 + (b + c − a)2 + (c + a − b)2 = 4(a 2 + b 2 + c 2 ) − (a + b + c)2
[Hint in footnote14 .]
8. Expand (3x + 1)(x + 1). By choosing a value for x, use the identity to
factorise 30, 401 completely.
p
9. Expand (2x + 1)2 . Use the resulting identity to find 4, 004, 001 without
using a calculator.
10. Using the identity x 2 − 1 ≡ (x + 1)(x − 1), factorise 999, 999 completely.
11. N = (1 + x 2 )(2 + x 4 ). Multiply the bracket and hence express N as a
decimal when x = 10−2 .
12. A garden is exactly square in shape, each side being x m in length.
Around the edge of the garden is a path y m wide. Find the area covered by
the path.
x
y

- x
13. Find in its simplest form an expression for the surface area of a solid
rectangular cuboid15 whose sides are of length x, x + 1 and x + 2.
14.∗ A function is defined by f (x) = 2x(x − 1).
(a) Show that f (x + 1) = 2x(x + 1).
(b) Find and simplify an expression for (i) f (x) + f (x + 1),
and (ii) f (3x) − 3 f (x).
14
You may find it helpful to write a + b − c = (a + b + c) − 2c, etc.
15
A rectangular cuboid is like a cube except the faces are rectangles rather than
squares.

Solutions 9
1. (a) 5x + 2y + 7 (b) 3y − 4 (c) 10c − 5

(d) 6x + 7x y (e) 3x − 20 (f) 2x 3 − 1
(g) −2p 2 + q 2 (h) −1 (i) 8a 2 b + 12ab 2
(j) x − x y (k) t 3 − 4t 2 + 4t (l) 8z 3 x + 2zx 2
(m) 6p 2 q 2 + 7pq 3 (n) −1 − y 2 (o) 2b 2
(p) 2x y 2 − x 2 y + 2y 3 − y 2 (q) x 3 − 1 (r) x n+1 − 1
2. (a) x 2 + 10x + 16 (b) x 2 − 10x + 16 (c) x 2 + 4x + 3

(d) x 2 + 5x − 14 (e) x 2 + 6x − 40 (f ) y 2 − 8y − 9
(g) y 2 + 8y − 9 (h) 2x 2 − 7x − 15 (i) 12x 2 + 11x + 2
(j) 2z 2 + 11z − 21 (k) 10x 2 − 17x + 6 (l) 16a 2 − 9
(m) 24x 2 − 17x − 22 (n) 25t 2 − 144
3. (a) 6a 3 − 5a 2 − a (b) 15x 2 − 11x − 12

(c) 6 − 11y − 2y 2 (d) 4a 2 b 2 − 1
(e) x 2 + 10x y + 25 (f) a 2 − 6ab 2 + 9b 4
(g) 9y 2 + 6y + 1 (h) 27y 2 + 27y + 9y + 1
(i) 100t 2 − 1 (j) x 3 − 7x − 6
(k) 8x 3 + 8x 2 + 2x (l) y 4 − 1
2
(m) 1 − 14 t 2 or 1 − t4 (n) 1 + x 2 + x 4
(o) x 4 + 4x 3 + 8x 2 + 8x + 4 (p) x 4 + 4
(q) 7x + 9 (r) y 2 + 5y + 3
(s) 2a 2 − 12x + 17 (t) −1
2
(u) x (v) 8ab
(w) 2y z + 2xz + 2x y (x) p 2 + pq + 2p + q + 1
4. (a) 10a+b − 10b + 10a − 1 (b) 3x+1 + 3 y+1 (c) 22x + 2x+1 + 1
(d) 42y − 1 (e) 1 − x (f) x 2 + 2 + x −2
5. Let the length of the unknown side be x. By Pythagoras
x 2 = (a + b)2 − (a − b)2 = a 2 + 2ab + b 2 − (a a − 2ab + b 2 ) = 4ab.
p
Length of the third side is 2 ab.
6. (a) 50(2y + x + 100) m 2 .

(b) Yes, if the only criterion is the increase in area. It would increase it
by 50(y + 2x + 100), and since x > y that is a bigger increase.
7. (b) (2x + y +z)2 = [(x + y +z)+x]2 = (x + y +z)2 +2x(x + y +z)+x 2 . From this
we can see that (x +2y +z)2 = (x + y +z)2 +2y(x + y +z)+ y 2 , and (x + y +2z)2 =
(x + y + z)2 + 2z(x + y + z) + z 2 .
So the whole expression is 3(x + y + z)2 + 2(x + y + z)(x + y + z) + x 2 + y 2 + z 2 .
(c) Derive it from the answer to part (b) by expanding (x + y + z)2 .
(d) (a + b − c)2 = [(a + b + c) − 2c]2 = (a + b + c)2 − 4c(a + b + c) + 4c 2 .
By symmetry we can see that the whole expression is
3(a + b + c)2 − 4(a + b + c)(a + b + c) + 4(a 2 + b 2 + c 2 ). The result follows.

8. (3x + 1)(x + 1) = 3x 2 + 4x + 1. Putting x = 100 we have 30, 401 = 301 × 101,

both of which are prime.
9. (2x +1)2 = 4x 2 +4x +1. Substituting x = 1, 000, we have 2, 0012 = 4, 004, 001,
p
so 4, 004, 001 = 2, 001.
10. Putting x = 1, 000 we have 999, 999 = 1001 × 999. 1001 = 7 × 11 × 13 and
999 = 3 × 3 × 3 × 37. The complete prime factorisation is 33 × 7 × 11 × 13 × 37.
11. N = 2 + 2x 2 + x 4 + x 6 , so
N = 2 + 0.0002 + 0.00000001 + 0.000000000001 = 2.000200010001.
12. We can find the area covered by the path either by breaking the area into
rectangles, or by subtracting the area of the inner square from the area of the
outer square. Using the second method:
x 2 − (x − 2y)2 = x 2 − (x 2 − 4x y + 4y 2 ) = 4x y − 4y 2 or 4y(x − y).
13. Surface area is 2[x(x + 1) + (x + 1)(x + 2) + x(x + 2)], which simplifies to

2(3x 2 + 6x + 2).
14. (a) f (x + 1) = 2(x + 1)(x + 1 − 1).
(b)(i) 2x[(x − 1) + (x + 1)] = 4x 2 .
(ii) 6x(3x − 1) − 3.2x(x − 1) = 18x 2 − 6x − 6x 2 + 6x = 12x 2 .

3.3 Polynomials
Definition
All these expressions we call polynomials:
1 2
2x 4 + 7x − 11, 9 − x 3, (x + 4)2 , t − 3.
2
To be a polynomial expression (in the variable x), it must be possible to write
it down using a finite 16 number of terms, each of which is of the form ax n ,
where a is a constant and each index n is a non-negative integer (i.e. positive
or zero). Including the case n = 0 is important because without it we couldn’t
have constant terms, like −11 = −11x 0 . The variable need not be x of course.
The expression (x + 4)2 qualifies because the bracket can be expanded to
give x 2 + 8x + 16, which ticks all the boxes. This we recognise as a quadratic
expression.
The order of a polynomial is the highest power of the variable

that it contains—after any brackets have been multiplied out.
A linear expression like 5x + 2 is a polynomial or order 1, a quadratic is a

polynomial of order 2, one of order 3 we call a cubic, one of order 4 we call a
quartic. Beyond that we normally just say “polynomial of order 5",
“polynomial of order 6", etc. What is a polynomial of order zero (the answer
is in the footnote17 )?
Some more example of polynomials (with their order):
6x 3 − 4x 2 + 11x + 1 (order 3)
y2 − y6 (order 6)
5 2 4
1− 3x −x (order 4)
1 + x + x + x + . . . x n (where n is a positive integer)
2 3
(order n)
16
Later you will come across expressions containing an infinite number of such terms.
17
A polynomial of order zero is a constant.

These are also polynomials:
2x(x 3 + 3) = 2x 5 + 6x (order 4)
2 2
(7x − 2) = 49x − 28x + 4 (order 2)
x 4 −6x 2 x 4 6x 2 x 3 1 3
= − = − 3x = x − 3x (order 3)
2x 2x 2x 2 2
(1 + 2x)5 = 1 + 10x − 40x 2 + 80x 3 − 80x 4 + 32x 5 (order 5)

You can probably see how all these have been derived—except the last one.
We could arrive at the answer by multiply all five brackets together:
(1 + 2x)(1 + 2x)(1 + 2x)(1 + 2x)(1 + 2x).
This would be very tedious. Fortunately you will learn a much simpler way
to do it.
These expressions are not polynomials, because it’s impossible to write any
of them as a collection of terms of the form ax n (where each n is a
non-negative integer):
1
5x 2 + 3− x 2 = 6x 2 + 3 − x −2 (the last power is negative)
p 1
4 x + x3 = 4x − x 3
2 (the first power is not an integer)
These aren’t polynomials either:
1
p
2
x +1
, 2x 2 − 3x + 7, x 3 sin x, 2x
because we don’t have a way of writing any of them as a sum of separate

terms, let alone ones of the form ax n .
Finally, this is not a polynomial
1 + x + x2 + x3 + x4 + . . .
because the dots at the end mean the list of terms goes on for ever. We say it
has an infinite number of terms.

QQ
Which of these expressions is a polynomial? If it is, state its order. (The

answers are in the footnote18 .)
1 x 3 +5x 2
(a) 12 x − x 7 (b) 3 + 2x 2 (c) y(y 2 − 5) (d ) x2
x 3 +5x 1
(e) x2
( f ) x 3 +3 (g ) (x + 1)(x − 9) (h) (2x − 3)2
p 1 p p
(i ) 1 + 3 x ( j ) x2
(k) − 7 (l ) ( x − 1)( x + 1)
(a) o(7). (b) No. [= 3 + 21 x −2 ] (c) o(3). (d ) o(1). [x+5]

−1
18 (e) No. [= x + 5x ] ( f ) No. (g ) o(2). (h) o(2).
1
(i ) No. [= 1 + 3x ]2 ( j ) No. [= x −2 ] (k) o(0). [= −7x 0 ] (l ) o(1). [x − 1]
3.3.1 Adding and subtracting polynomials

Since every polynomial is a series of terms of the form ax n , adding and
subtracting them is merely a matter of collecting like terms.
Example 1
Add the two polynomials 3 − x 2 and x 2 + 6x − 11.
Solution
(3 − x 2 ) + (x 2 + 6x − 11) = 3 − x 2 + x 2 + 6x − 11
= 6x − 8
The brackets are not necessary in this case.
Exam questions will often adopt the function notation when describing
polynomials (see Section 1.7 on page 1-31), as in the examples below.
Example 2.
The polynomials f (x) and g (x) are defined by
f (x) = x + 3 and g (x) = 1 − 2x − x 2 .
Find in its simplest form the polynomial f (x) − 4g (x).
Solution
f (x) − 4g (x) = x + 3 − 4(1 − 2x − x 2 )
= x + 3 − 4 + 8x − 4x 2
= −4x 2 + 9x − 1
Notice how important it was to put brackets round the 1 − 2x − x 2 . It’s not a
bad idea to default to putting in brackets as a matter of habit.
QQ
Try this one for yourself.
Using the functions f (x) and g (x) as defined in Example 2, find and simplify
the expressions
2 f (x) + g (x) and f (x) − g (x).
19
2 f (x) + g (x) = −x 2 + 7.
f (x) − g (x) = x + 3 − (1 − 2x − x 2 ) = x + 3 − 1 + 2x + x 2 = x 2 + 3x + 2

Here is a slightly harder example.
Example 3.
Suppose two polynomials are defined by:
p(x) = x 2 − 2 and q(x) = 2 − 3x 2 + x 4
(a) Evaluate p(3) and q(−2).
(b) Solve the equation p(x) = 23.
(c) Show that x = −1 is a solution of the equation q(x) = 0.
(d) Write in its simplest form the polynomial p(x)q(x).
Solution
(a) p(3) = 32 − 2 = 7 and q(−2) = 2 − 3(−2)2 + (−2)4 = 2 − 12 + 16 = 8.
(b) The equation is x 2 − 2 = 23, or x 2 = 25, so there are two solutions: x = 5

and x = −5.
(c) q(−1) = 2 − 3(−1)2 + (−1)4 = 2 − 3 + 1 = 0.
(d) p(x)q(x) = (x 2 − 2)(x 4 − 3x 2 + 2)

= x 2 (x 4 − 3x 2 + 2) − 2(x 4 − 3x 2 + 2)
= x 6 − 3x 4 + 2x 2 − 2x 4 + 6x 2 − 4
= x 6 − 5x 4 + 8x 2 − 4

Sometimes its helpful to set out a polynomial addition or subtraction in

tabular or column form. For example to subtract x 3 + x 2 − 10x − 5 from
1 − x − 6x 2 + 3x 3 we write one above the other like this:
3x 3 − 6x 2 − x + 1
3 2
x +x − 10x − 5
and subtract each power of x in turn. It’s conventional—but by no means
essential—to put them in order of descending 20 powers of x. Here is the
answer; see if you agree with it.
2x 3 − 7x 2 + 9x + 6.
The problem consists of simple enough steps, but the prevalence of negative
numbers means it is remarkably easy to make mistakes, for example
−x − (−10x) = +9x.
From an examination point of view, a slip-up in a question like this can be
very costly. The whole subtraction is likely to be worth 2 marks, every error
carrying a 1 mark penalty. So one mistake and you lose half the available
marks; two mistakes and you lose all of them—even though you got a lot
right. This may seem harsh, but it is a very effective way of getting students
to prize accuracy!
Lining up powers of x above each other may mean leaving gaps. For example
to subtract 2x 3 − 7x + 5 from 1 − x 2 :
−x 2 +1
3
2x − 7x + 5
Do this subtraction for yourself (the answer is in the footnote21 ).
20
There in an arithmetic parallel: the place value notation we use in our arabic number
system assumes each column corresponds to a power of 10, written in descending
powers: 2734 = 2(103 ) + 7(102 ) + 3(10) + 4.
21
2x 3 − x 2 + 7x − 4.

3.3.2 Multiplying polynomials

We already know how to multiply polynomials. For example:
(x − 3)(5x 2 − x + 2) = x(5x 2 − x + 2) − 3(5x 2 − x + 2)
= 5x 3 − x 2 + 2x − 15x 2 + 3x − 6
= 5x 3 − 16x 2 + 5x − 6
Use this method to expand
(2x 2 + x − 4)(3x 2 − 5x − 2)
Remember we need to multiply every term in the first bracket by every term
in the second, so the calculation starts
(2x 2 + x − 4)(3x 2 − 5x − 2) = 2x 2 (3x 2 − 5x − 2) + x(3x 2 − 5x − 2) − 4(3x 2 − 5x − 2)
If both polynomials are long ones, a neat method is to use a table. Take the
example
(x 3 + 2x 2 − x − 5)(x 2 − 7x − 1).
It doesn’t matter which of the polynomials we put across the top, and which
down the side:
× −1 −7x x2
−5 +5 +35x −5x 2
−x +x +7x 2 −x 3
2x 2 −2x 2 −14x 3 +2x 4
x3 −x 3 −7x 4 +x 5
The example highlighted in red boxes shows how we complete the table: e.g.
multiplying 2x 2 × −7x = −14x 3 . When we have done this for every pair of
terms (one from each polynomial), we collect up all the terms in the body of
the table. This gives:
x 5 − 5x 4 − 16x 3 + 36x + 5.
Notice that all the terms involving the same23 power of x will lie along a
single (forward sloping) diagonal. For example one such diagonal holds all
the x 3 terms: −x 3 , −14x 3 and −x 3 , giving −16x 3 . Not only does this make it
easy to collect the terms, it also provides an error-check: if the powers of x
are not identical along one of these diagonals you know you have made a
mistake.
22
6x 4 − 7x 3 − 21x 2 + 18x + 8.
23
This works provided we enter the initial two polynomials in strictly ascending or strictly
descending powers of x. In our example we have chosen ascending order.

If you wish you can fill in the table like this:
× −1 −7x x2
−5 +5 +35x −5x 2
−x +x +7x 2 −x 3
2x 2 −2x 2 −14x 3 +2x 4
x3 −x 3 −7x 4 +x 5
5 +36x −16x 3 −5x 4 +x 5
Each column in the body of the table now contains terms in the same power
of x.
Use the tabular approach to multiply (2x 2 + x − 4)(3x 2 − 5x − 2). The answer
is in the footnote at the bottom of the previous page.
If you want to know about dividing polynomials this is in Chapter 4:

Algebraic Fractions & Dividing Polynomials. See page 4-41.

An elegant approach to multiplying polynomials

There is an other approach. It is the most direct, the quickest and it gives the
most insight into what is going on. But it requires a little more thought. You
can skip this method if you wish and go straight to the Exercises at the end of
this chapter (missing out Questions 6 and 7); it won’t impair your ability to
do C1 or C2 exam questions.
We’ll illustrate the approach with the product:
(2x 2 + x − 4)(3x 2 − 5x − 2).
Rather than do all the multiplications and then collect terms, as we have
done before, we look at it the other way round. We ask ourselves: “where
does each coefficients in the answer come from?".
The highest order term (the x 4 term) is very easy, it can only come from
multiplying 2x 2 by 3x 2 , so it’s 6x 4 .
Moving on to the x 3 term, this can come from two products: multiplying the
2x 2 in the left hand bracket by the −5x term in the right hand bracket, and
multiplying the x in the left bracket by the 3x 2 in the right hand bracket.
Summarising:
x 3 coefficient: (2)(−5) + (1)(3) = −3
Now the x 2 term. This arises from three products: (2x 2 )(−2), (x)(−5x) and
(−4)(3x 2 ). The x 2 coefficient is therefore −21.
Continuing this process for all the coefficients and listing all our results:
x 4 coefficient: (2)(3) = +6
x 3 coefficient: (2)(−5) + (1)(3) = −7
2
x coefficient: (2)(−2) + (1)(−5) + (−4)(3) = −21
1
x coefficient: (1)(−2) + (−4)(−5) = +18
x 0 coefficient: (−4)(−2) = +8
The product of the two polynomials is 6x 4 − 7x 3 − 21x 2 + 18x + 8. This has
taken a long time to explain, but is undoubtedly the quickest to perform
because you can do most of the intermediate working in your head. Try it on
this example: (x 2 + 3x − 2)(4x 2 − x − 6). (The answer is in the footnote24 .)
This method is particularly helpful if we only need to find some of the
coefficients. The following example demonstrates this.
Although you won’t need to adopt this approach in any of the problems you
meet in Core 1 or Core 2, you may want to. The next example illustrates why.
24
4x 4 + 11x 3 − 17x 2 − 16x + 12.

Example 4
Find the coefficient of x 4 in the polynomial
(x 3 + x 2 + 1)(2x 2 − 5)2
Solution
Expanding the right hand bracket using the happy face rule:
(2x 2 − 5)(2x 2 − 5) = 4x 4 − 20x 2 + 25, leading to the product
(x 3 + x 2 + 1)(4x 4 − 20x 2 + 25)
from which we get two x 4 terms: (x 2 ) × (−20x 2 ) and (1) × (4x 4 ), so the answer
is −16.

Exercises 10
1. Establish which of these is a polynomial. If it is, express it in its simplest
form.
³ ´
(a) 4x x − 12 (b) 4x x − x1 (c) 4x x − x12
¡ ¢ ¡ ¢
p ³p 1
´ 1
(d ) 4 x x − x p (e) (x 2 + 1)3 ( f ) (x 2 + 1) 2
¢2 ¡ ¢2 1 1
(g ) x + x1 − x − x1 (h) 3 (12x 2 − 1) (i ) 3x (12x 2 − 1)
¡
2. Use the tabular method to perform these subtractions. (Write each

polynomial in descending powers of x, and keep equal powers of x in
columns.) Check your answers by collecting terms.
(a) Subtract x 3 − 6x 2 − x + 7 from 3x 3 + x 2 − 6x + 2.

(b) Subtract x 2 + x − 6 from x 2 − 1.
(c) Subtract 4x 2 − x from 1 + 7x.
3. Use the tabular method to multiply these polynomials.
(a) 3x + 2 × x 2 + 7x − 3
(b) 2x 2 − x + 5 × 1 + x − 3x 2
(c) x 3 + 4x − 1 × x −2
4. Repeat Question 4,
(a) by multiplying brackets and collecting terms.

(b) by the direct method (i.e. writing the answer straight down, any
intermediate calculations having been done in your head).
5. The polynomial f (x) = 2x 3 + x 2 − 4x + 1
(a) Evaluate f (3).

(b) Show that f (4) + f (−4) = 34.
(c) Find the value of f 21 .
¡ ¢
(d) Show that x = 1 is a solution of the equation f (x) = 0.

(e) Write down the polynomial 3 f (x).
(f) The polynomial g (x) is defined by g (x) = 3 f (x) + k, where k is a
constant. Find k if g (0) = 0.
6. If f (x) = 5x 2 − x + 7, and g (x) = 2 − x 3 , write each of these in its simplest

form, in decreasing powers of x.
(a) f (x) + g (x) (b) f (x) − g (x) (c) g (x) − 2 f (x)
(d ) f (x)g (x) (e) [ f (x)]2 ( f ) f (x)[ f (x) + g (x)]
7. The polynomials h(x) and k(x) are defined by

h(x) = 1 + x 2 and k(x) = 1 + x + x 2 + x 3 .
(a) Show that k(x) = (1 + x)h(x).
(b) By trial and error find a solution of the equation k(x) = 0.
8. Show that (x + 2)3 = x 3 + 6x 2 + 12x + 8.

Hence solve the equation
x 3 + 6x 2 + 12x − 19 = 0.

Solutions 10
1. (a) Yes. 4x 2 − 2x (b) Yes. 4x 2 − 4 (c) No. [4x 2 − 4x −1 ]

p
(d) Yes. 4x − 4 (e) Yes. x 4 + 2x 2 + 1 (f ) No. [ x 2 + 1 6= x + 1]
1
(g) Yes. 2x 2 + 4 (h) Yes. 4x 2 − 2 (i) No. [4x − 31 x −1 ]
2. (a) 3x 3 + x 2 − 6x + 2 (b) x 2 −1 (c) +7x + 12

x 3 − 6x 2 − x + 7 2
x +x −6 2
4x − x
2x 3 + 7x 2 − 5x − 5 −x + 5 −4x 2 + 8x + 1
3. (a) × x 2 + 7x −3 (b) × − 3x 2 + x +1
+3x 3x 3 + 21x 2 − 9x +2x 2 − 6x 4 + 2x 3 + 2x 2
+2 2x 2 + 14x − 6 −x + 3x 3 − x 2 − x
+5 − 15x 2 + 5x + 5
Ans: 3x 3 + 23x 2 − 5x − 6 Ans: −6x 4 + 5x 3 − 14x 2 + 4x + 5
(c) × x3 + 4x −1
+x x4 + 4x 2 − x
−2 − 2x 3 − 8x + 2
Ans: x 4 − 2x 3 + 4x 2 − 9x + 2
5. (a) 52 (b) 34. [ f (4) = 129, f (−4) = −95]

(c) − 12 (d) f (1) = 0
(e) 6x 3 + 3x 2 − 12x + 3 (f ) g (0) = 3 + k, so k = −3
6. (a) −x 3 + 5x 2 − x + 9 (b) x 3 + 5x 2 − x + 5
(c) −x 3 − 10x 2 + 2x − 12 (d) −5x 5 + x 4 − 7x 3 + 10x 2 − 2x − 14
(e) 25x 4 − 10x 3 + 71x 2 − 14x + 49
(f) −5x 5 + 26x 4 − 17x 3 + 81x 2 − 16x + 35
7. (a) (1 + x)h(x) = (1 + x)(1 + x 2 ) = 1 + x + x 2 + x 3 = k(x)

(b) The equation is 1 + x 2 = 2x, or x 2 − 2x + 1 = (x − 1)(x − 1) = 0
There is one (repeated) root: x = 1.
(c) Since all the coefficients are > 0, the solution must be < 0.
By trial and error x = −1 fits the equation so this is a root.
(d) For all real x, x 2 ≥ 0, so h(x) = x 2 + 1 ≥ 1. h(x) can never be 0.
8. Expand the brackets by any method to give x 3 + 6x 2 + 12x + 8.

This suggests we write the equation: x 3 + 6x 2 + 12x + 8 = 27,
which is (x + 2)3 = 27. So x + 2 = 3 and the only root is x = 1.

3.4 Factorising & Highest Common Factor

Factorising
You will do a lot of factorising. But what exactly is it?
Factorising is converting a number, or an algebraic expression,

from a sum of terms into a product of two or more expressions.
We first met factorising in the context of numbers. When we factorise the

number 132, we are converting the sum 100+30+2 into a product, for example
12 × 11:
132 = 12 × 11.
When we factorise 2x 2 − 10x, we are converting the sum of the terms 2x 2 and
−10x, into the product of the terms 2x and x − 5.
2x 2 − 10x = 2x(x − 5).
Highest Common Factor

So where does the Highest Common Factor25 (HCF) come in? You will have
met the term before: whenever you cancel a fraction into its simplest form
you are dividing numerator and denominator by the HCF of the two numbers.
The HCF of two numbers is the largest number that divides26 into both of
them.
Sometimes the HCF is quite easy to see, for example:
30 2
=
45 3
(the HCF is 15). If it’s not so obvious we’ll need to look more closely at the
factors of the two numbers. A systematic way of doing this is to find the prime
factorisation 27 of each of them. For example:
72 23 .32 2.3 6
= 2 = = .
420 2 .3.5.7 5.7 35
The HCF is the largest number we can divide into both the top and the
bottom, i.e. 22 .3 = 12. (Notice the use of the “dot" notation for
multiplication.)
25
In the USA they call it the Greatest Common Divider (GCD).
26
Used in this context we mean “divides without a remainder".
27
Breaking a number into factors and continuing the process until all the numbers are
prime numbers.
e.g. 132 = 2 × 66 = 2 × 2 × 33 = 22 × 3 × 11.

The notion of HCF is also relevant to factorising. Take this expression:
72x + 420y.
It has various factors, so we could write, for example
72x + 420y = 2(36x + 210y) or 3(24x + 140y).
To factorise the expression completely we want the HCF of the two

coefficients 72 and 420, which we know to be 12.
72x + 420y = 12(6x + 35y).
The algebraic parts of each term may also share a factor, and for a complete
factorisation it is once again the HCF we are interested in. For example
20b 3 − 15ab 2 = 5b 2 (4b − 3a)
because 5b 2 is the largest expression that divides into both 20b 3 and 15ab 2 .
All this takes us into the main theme for rest of this chapter: factorisation
algebraic expressions.

3.5 One-bracket factorising

The opposite to multiplying out brackets is factorising. We’re going to be
looking at the case of a single bracket here, what we call one-bracket
factorising. Later we’ll be looking at the two-bracket case. Some examples:
6x + 9 = 3(2x + 3)
y + 5y 2 − 2y = y(y 3 + 5y − 2)
4
16p 2 q + 12pq 2 = 4pq(4p + 3q)
In essence the process is a simple one: look at all the terms and see if there is
anything we can take out of every one of them—in other words do they have
any common factors? If we can take out everything they have in common in
one fell swoop, great. But it’s fine to do it step by step, for example
16p 2 q + 12pq 2 = 4(4p 2 q + 3pq 2 )

= 4p(4pq + 3q 2 )
= 4pq(4p + 3q)
At each stage we produce an expression which, when multiplied out gives us

what we started with. The most we can “take out" of every term is 4pq. This
is the Highest Common Factor (HCF) that we talked about in the previous
section.
Often the HCF will be obvious; if it isn’t it can be built up step by step like in
the example above. If there are a number of terms and the HCF is not
obvious we can use set theory to find it quickly and easily. For a description
of how this works see Appendix A, page A-26. It is something a
mathematician would find interesting; it is not a technique you need to
know at this stage.

Factorising means re-writing the whole expression as a product. There is

rarely anything to be gained by taking a factor from only some of the terms.
For example
2x 3 + 16x 2 − 3 = 2x 2 (x + 8) − 3
It’s rather like writing 35 as 2 × 17 + 1; it tells us nothing useful. Writing it as
5 × 7 could be.
So with one-bracket factorising, there is a simple underlying process: we
look at all the terms and take outside the bracket the largest factor that all
the terms have in common, i.e their HCF.
One-bracket factorising is not hard, but it is always easy to make a mistake

So we recommend strongly:
A
after factorising, multiply the result out in your head to check it.
!
A
A
A

Let’s look at some more interesting examples.
Example 5 Factorise 2x(1 + y) + x y + 7x.
Solution
To see exactly what we have, we’ll multiply out the bracket that’s already
there first:
2x(1 + y) + x y + 7x = 2x + 2x y + x y + 7x
= 3x y + 9x
= 3x(y + 3)
You might have spotted that every term has a factot x, suggesting a slightly
different approach:
x[2(1 + y) + y + 7] = x[2 + 2y + y + 7] = x[3y + 9] = 3x(y + 3).
Example 6 Factorise 10(x + 3) + 2(x + 3)2 .
Solution
We could multiply out the brackets and look for factors, but there is a much
better way.
The first term is 2 × 5 × (x + 3) and the second is 2 × (x + 3) × (x + 3).

We can take out28 2 and (x + 3) from each term, i.e. the HCF is 2(x + 3).
10(x + 3) + 2(x + 3)2 = 2 (x + 3) [5 + (x + 3)]

= 2 (x + 3) (x + 8)
We have put (x + 3) in a red box to highlight the fact that the whole bracket is
a factor.
Although we have ended up with two brackets, we still regard this as
one-bracket factorising, because the process was that of finding common
factors. It just so happens in this instance that one of the common factors
was itself a bracket.
Factorise these expressions. The answers are in the footnote29 .
(a) 6x 3 + 12x 2 y − 4x y 2 (b) 3(1 + 3x) + x + 2 (c) (x − 1)2 + 4(x − 1)
28
We use the phrase “take out" but it’s not anything to do with subtraction; it’s
division.
29
(a) 2x(3x 2 + 6x y − 2y 2 ) (b) 5(2x + 1) (c) (x − 1)(x + 3)

Example 7 Factorise y(2y − x) + 3(x − 2y).
Solution
“Taking out" a minus sign from the second bracket has the effect of reversing
the order of the subtraction. Then we have a common factor:
y(2y − x) + 3(x − 2y) = y(2y − x) −3(− x + 2y)

= (2y − x)(y − 3)
In the next two examples the common factor is not quite so obvious
Example 8 Factorise 2a(x + 3y) + 5x + 15y.
Solution
We can write the expression as
2a(x + 3y) + 5(x + 3y) = (x + 3y)(2a + 5)
So if there is no obvious factor, maybe it’s worth seeing if we can find one by
writing part of the expression in a different way.
Example 9
Factorise (a) 1 + x + x 2 + x 3 (b) 6x − 3y − 2x 2 + x y.
Solution
(a) 1 + x + x 2 + x 3 = (1 + x) + x 2 (1 + x)
= (1 + x)(1 + x 2 )
(b) 6x − 3y − 2x 2 + x y = 3(2x − y) − x(2x − y)

= (3 − x)(2x − y)
You may be wondering how you can be expected to spot such things. The
answer is that you should always be looking for factors. Take the last example:
the fact that 6 and 3 have a common factor of 3 will come to your mind. Then,
“seeing" 6x − 3y as 3(2x − y) in your mind’s eye, you may be inclined to check
for a factor of 3 or a (2x − y) elsewhere.
The key is to make a habit of looking out for factors.

QQ
(a) 1 + x 2 + 3x + 3x 3 (b) 5(x − y) + 3(y − x) (c) 4a − 8ab + 3a − 6ab
30
(a) (1 + x 2 )(1 + 3x) (b) 2(x − y) (c) (1 − 2b)(4a + 3b)

Finally a reminder. By its nature, factorising requires more thought and

mistakes can easily occur. Multiplying out on the other hand is easier and
you can usually do it in your head. So, remember:
A
After factorising, check your answer by multiplying it out.
!
A
A
A
Where next?
You will have met a rather different kind of factorising in the context of
quadratic expressions like this:
x 2 + 7x + 10 = (x + 2)(x + 5).
But the three terms have no factor common, nor there is an obvious way of
constructing one. How is this factorising?
It is factorising because we are converting a sum (of three terms) into a
product of the two expressions (x + 2) and (x + 5).
In the next section we shall explore this kind of two-bracket factorising.
Some of it may be familiar territory; some of it may not be.

Exercises 11
Factorise these expressions as far as possible.

1. (a) 3x y + 12y 2 (b) 8a 2 b + 2a 3
(c) 10p 2 q − 6pq 2 (d) 2x 4 + x 2 + 5x
(e) 6y 2 + 3 (f ) 12a 2 + 5ab
(g) 7 − 7t (h) 9x − 3x 3 + x 4
(i) 8x 2 y + 16x y 2 − 12x 4 (j) x 3 + 3x y 2 + 7y 3
(k) 5ab 3 − a (l) 4x + 1 + 6(x − 1)
(m) 1 − 3a + 5(a + 1) (n) y(y − 4) + 11y 2
(o) 4(1 − 2x) + (1 − 2x)2 (p) 4(1 − 2x) − (1 − 2x)2
(q) 8a + 6 + (4a + 3)2 (r) 3(x + 2)2 − 9x − 18
(s) (5x − 3)2 + 6 − 10x
2. (a) (y + 1)3 − (y + 1)2 (b) 2(a + b)2 − 3a 2 − 3ab

(c) 3x − 3y + 6 + (x − y + 2)2 (d) 3x − 3y + 6 − (x − y + 2)2
(e) 2(x − 4) + 3x − 12 (f ) 2x y + 5y + 4x + 10
(g) ab + 3b − a − 3 (h) 12y 2 + 2x y + 18y + 3x
(i) a 2 b + 2b − 2a 2 − 4 (j) 3x 2 − x y + 12x − 4y
(k) 2p 2 + 2pq + 4pr − p − q − 2r (l) 4(x + 1) + 5(x + 1)2 + (x + 1)3
(m) x(3x − 2) + y(2 − 3x) (n) 2a 2 (a − 4b) − 4b + a
(o) (5y + 1)(x − 3) + y(3 − x) (p) 5y 2 − y − 2x y + 10x y 2
(q) x(a − 2b − c) + 3(c + 2b − a) (r) 1 + 2x + x 2 + x 3 + 2x 4 + x 5
(s) x(x + 1)2 − (x + 2)(x 2 − 1)
3. (a) Add these numbers:

111
1110
11100
(b) By writing the sum as 111 + 111 × 10 + 111 × 100,

and factorising, show that 12, 321 = 1112 = 32 372 .
(c) Show that 1001 = 7 × 11 × 13, and 111, 111 = 111 × 1001.
(d) Find the complete prime factorisation of 12, 345, 654, 321.

Solutions 11
1. (a) 3y(x + 4y) (b) 2a 2 (4b + a) (c) 2pq(5p − 3q)

3 2
(d) x(2x + x + 5) (e) 3(2y + 1) (f ) a(12a + 5b)
(g) 7(1 − t ) (h) x(9 − 3x 2 + x 3 ) (i) 2x(4x y + 8y 2 − 6x 3 )
(j) Can’t be factorised. (k) a(5b 3 − 1) (l) 10x + 5 = 5(2x + 1)
2
(m) 2a + 4 = 2(a + 2) (n) 12y − 4y = 4y(3y − 1) (o) (1 − 2x)(5 − 2x)
(p) (1 − 2x)(3 + 2x) (q) (4a + 3)(4a + 5) (r) 3(x + 2)(x − 7)
(s) 5(5x − 3)(x − 1)
2. (a) y(y + 1)2 or (y + 1)2 y (b) (a + b)(2b − a)

(c) (x − y + 2)(x − y + 5) (d) (x − y + 2)(y − x + 1)
(e) 5(x − 4) (f ) (2x + 5)(y + 2)
(g) (a + 3)(b − 1) (h) (6y + x)(2y + 3)
2
(i) (a + 2)(b − 2) (j) (3x − y)(x + 4)
(k) (p + q + 2r )(2p − 1)
(l) (x + 1)(4 + 5(x + 1) + (x + 1)2 ) = (x + 1)(x 2 + 7x + 10)
= (x + 1)(x + 2)(x + 5)
(m) x(3x − 2) − y(3x − 2) = (3x − 2)(x − y)
2
(n) 2a (a − 4b) + (a − 4b) = (a − 4b)(2a 2 + 1)
(o) (5y + 1)(x − 3) − y(x − 3) = (5y + 1 − y)(x − 3)
= (4y + 1)(x − 3)
(p) y(5y − 1) − 2x y(1 − 5y) or (p) y[5y − 1 − 2x + 10x]
= y(5y − 1) + 2x y(5y − 1) = y[5y − 1 − 2x(1 − 5y)]
= (5y − 1)(y + 2x y) = y[5y − 1 + 2x(−1 + 5y)]
= y(5y − 1)(1 + 2x) = y(5y − 1)(1 + 2x)
(q) x(a − 2b − c) − 3(a − 2b − c) = (a − 2b − c)(x − 3)
2 2 2
(r) (1 + 2x + x ) + x (1 + 2x + x ) = (1 + 2x + x 2 )(1 + x 2 )
= (1 + x)2 (1 + x 2 )
(s) x(x + 1)2 − (x + 2)(x 2 − 1) = x(x + 1)2 − (x + 2)(x + 1)(x − 1)
= (x + 1)[x(x + 1) − (x + 2)(x − 1)]
= (x + 1)[x 2 + x − (x 2 + x − 2)]
= 2(x + 1)
3. (a) 12, 321. (b) 111(1 + 10 + 100) = 111 × 111 = (3 × 37)2 .
(c) 111(1000 + 1) = 111000 + 111.
(d) 12, 345, 654, 321 = 111, 111(1 + 10 + 100 + 1, 000 + 10, 000 + 100, 000).
= 111, 1112 = (111 × 1001)2 = 32 72 112 132 372 .

3.6 Two-bracket factorising: ax 2 + bx + c

If a = 0 we don’t have a quadratic so we’ll never need two brackets to factorise
it. So in all the cases that follow take it as read that a 6= 0. We start by assuming
that a is positive; later we see what to do if a is negative.
3.6.1 When c = 0
We can quickly dispense with this special case. Take the expression
x 2 + 5x.
It’s a quadratic because the highest power of x is 2, but it has no constant

term, so we can factorize it using a single bracket
x 2 + 5x = x(x + 5).
Similarly with the expression

6y − 3y 2 .
Here the variable is y, and the y 2 term is at the end.
6y − 3y 2 = 3y(2 − y).

3.6.2 When a = ±1
a =1
(We are assuming here that neither b nor c is zero. We deal with those special
cases elsewhere. )
There is no factor common to all the terms in the quadratic expression
x 2 + 5x + 6
so it doesn’t yield to one-bracket factorising, but we know it’s what we get

when we multiply out:
(x + 2)(x + 3).
This is two-bracket factorising. One-bracket and two-bracket factoring are
quite different processes, but they are both factorising because they are
converting a sum of terms into a product. When factorising an expression
remember to
check for both one-bracket and two-bracket factorising
If this is familiar territory for you, you’ll sometimes be able to spot the
factors. You may have done so in the example above. When you can’t, you’ll
need a systematic way of finding them. That’s what we’re going to look at
next, using the following example:
x 2 + 12x + 32 = (· · · + . . .)(. . . + . . .)
The "anatomy" of the expression on the left is that the x 2 term comes from
multiplying together the first term in each bracket. So no problem here: both
brackets must start with x.
x 2 + 12x + 32 = (x + . . .)(x + . . .)
Turning to the constant term 32, that comes from multiplying the two
numbers—where the dots are. So the dots must be replaced by two factors of
32. The options are 1 × 32, 2 × 16, 4 × 8. A quick check (either in your head or
on paper) tells us it’s the last pair that works:
x 2 + 12x + 32 = (x + 4)(x + 8).

Now consider this quadratic. It includes a minus sign:
x 2 − 8x + 12.
Once again each bracket must start with x and we are looking for factor pairs
for 12. This time the options are 1×12, 2×6, 3×4. To make sure we don’t miss
any it’s a good idea to write the list systematically, starting with the inevitable
factor of 1, then 2 etc. We stop as soon as the factors start repeating: if we
were to continue with the list: 4 × 3, 6 × 2, 12 × 1, we get the same three factor
pairs but written in reverse order 31 .
I have glossed over something: it’s not strictly true to say that the only factor
pairs of 12 are 1 × 12, 2 × 6 and 3 × 4. We would also get +12 if both numbers
were negative. Bearing that in mind we now have all the options. There are
six of them, all of which generate the correct x 2 term and the correct constant
term:
(x + 1)(x + 12) (x − 1)(x − 12)
(x + 2)(x + 6) (x − 2)(x − 6)
(x + 3)(x + 4) (x − 3)(x − 4)
It remains to find the one that gives the right x term. And you can now see
why it’s so useful to be able to work out the middle term in your head. The
correct factorisation is
x 2 − 8x + 12 = (x − 2)(x − 6)
Can you see that if we change both the signs in the brackets, we still get +12;
the only thing that changes is the sign of the x-term.
(x − 2)(x − 6) = x 2 − 8x + 12
(x + 2)(x + 6) = x 2 + 8x + 12
In these both these examples we needed to find a pair of factors of c whose
sum corresponded to the number in front of x (i.e. ignoring its sign).
To summarise, when the x 2 coefficient is 1
if the constant term is positive we are looking for the factors

whose sum matches the x-coefficient.
We say “matches" not “equals" because at this stage we ignore it’s sign.
31
You may be able to see that factors pairs we will start repeating when the first
p of
the two factors exceeds the square root of the constant term—in this case 10.
The logic is that beyond that point the second factor will be smaller than the first
and since they are still valid factor pairs we must be repeating ones we had earlier
in the list.

QQ
(a) x 2 + 8x + 12 (b) x 2 + 17x + 72 (c) x 2 − 10x + 24 (d) x 2 − 38x + 72

(e) x 2 + 12x.
32
(a) (x + 2)(x + 6) (b) (x + 8)(x + 9) (c) (x − 4)(x − 6) (d) (x − 2)(x − 36)
(e) x(x + 12).

What happens if the constant term is negative? Look at the quadratic
x 2 + x − 20.
The factor pairs for 20 are 1 × 20, 2 × 10 and 4 × 5. But since the constant term
is −20 one factor must be positive and the other negative. So the options are:
(x − 1)(x + 20) (x + 1)(x − 20)

(x − 2)(x + 10) (x + 2)(x − 10)
(x − 4)(x + 5) (x + 4)(x − 5)
A mental check through the x terms tells us that the correct answer is
x 2 + x − 20 = (x − 4)(x + 5)
If we swap over the two signs in the brackets on the right, the only thing that
changes on the left is the sign of the x-term.
x 2 − x − 20 = (x + 4)(x − 5).
What we see is that when c is negative, it’s the difference of its factors we are
looking for.
Summarising again, when the x 2 coefficient is 1.
if the constant term is negative we are looking for the factors

whose difference matches the x-coefficient.
QQ
(a) x 2 + 4x − 12 (b) x 2 − 3x − 18 (c) x 2 + 7x − 18 (d) x 2 − x + 20.
33
(a) (x − 2)(x + 6) (b) (x − 2)(x + 9) (c) (x + 4)(x − 5).

Let’s review where we have got to.
When a = 1
1. Make a list of the factor pairs of c, starting with 1 × c.

2. We can identify which factor pair is the right one—by looking at b.
Because b comes from
• the sum of the factors if c is positive
• the difference of the factors if c is negative
3. Check the two sign options:

• if c is positive, the signs must ++ or −−
• if c is negative, the signs must +− or −+
This simple method is particularly helpful if the constant terms has a lot of
factors, for example
x 2 − 9x − 36
the factor pairs are 1×36, 2×18, 3×12, 4×9, 6×6. The coefficient of x 2 is 1 and
we have a negative constant term, so we are looking for a factor pair whose
difference is 9. That leads us straight to the pair 3 and 12, and the two options
are (x − 3)(x + 12) and (x + 3)(x − 12)
A mental check of the x terms points to the second one.
a = −1
Before going on to look at quadratics for which a 6= 1, there is one special case
that we can deal with easily. That is when a = −1. For example
20 + x − x 2
All we need to do is write it as −(x 2 − x − 20).
Factorising x 2 − x − 20 = (x + 4)(x − 5), we have
20 + x − x 2 = − (x − 5)(x + 4)
which we may want to write as (5 − x)(x + 4).
QQ
Factorise the expression 15 + 2x − x 2 .

34
−(x + 3)(x − 5) or (x + 3)(5 − x).

3.6.3 When a 6= 1 or −1
(As before we assume here that neither b nor c is zero.)
If a 6= 1or −1 spotting the factors is seldom easy—and there are no simple
rules to help us.
That only leaves one alternative: systematically going through all the options.
Take this example:
3x 2 − 13x + 14.
As before, we first check to see if the three terms have any common factors. In
this case they don’t, so we go straight into the process of forming two brackets.
(. . . x ± . . .)(. . . x ± . . .)
The numbers in front of the two x’s must multiply to give 3 and the two
constant terms must multiply to give 14. The coefficient 3 can only arise
from the product of 1 × 3, and the coefficient 14 can arise in two different
ways: 1 × 14 and 2 × 7.
But there is a twist this time.
When the coefficient of x 2 is not 1, we need to include the factor

pairs for the constant term in reverse order.
In this example that means 1 × 14 and 14 × 1; 2 × 7 and 7 × 2. The reasons why

will become clear shortly. We only do this for c, we don’t do it for a as well35
We set out the options in rows and columns, as before. The rows correspond
to the factor pairs; the columns show the sign options.
In this case the constant term is positive so we need either two pluses or two
minuses. By the way you don’t have to put all the stuff in the columns at the
sides and along the top; we’ve put it in to show the structure of the table.
a + + − − c
1×3 (x + 1)(3x + 14) (x − 1)(3x − 14) 1 × 14
(x + 14)(3x + 1) (x − 14)(3x − 1) 14 × 1
(x + 2)(3x + 7) (x − 2)(3x − 7) 2×7
(x + 7)(3x + 2) (x − 7)(3x − 2) 7×2
Being systematic ensures no options are forgotten, or repeated. A quick

mental check of x-coefficients tells us that the factorisation is
3x 2 − 13x + 14 = (3x − 7)(x − 2).
35
Or we could make the choice the other way round, i.e. include the factor pairs for a in
reverse order, but not those for c. If we do it for both we will duplicate options.

Now we consider an example where there are rather more options:
6x 2 − 11x − 10
We first check to see if the three terms have any common factors. In this case
they don’t, so we go straight into the process of forming two brackets.
(. . . x ± . . .)(. . . x ± . . .)
The numbers in front of the two x’s must multiply to give 6 and the two
constant terms must multiply to give 10. So this time there are two sets of
factor pairs to consider: those that multiply to give 6 (i.e. 1 × 6, 2 × 3) and
those that multiply to give 10 (i.e. 1 × 10, 2 × 5).
As before we need to include the factor pairs for c written in reverse order,
i.e. 1 × 10, 10 × 1, 2 × 5 and 5 × 2. We’ll set out the options in the same way as
before: the different factor pairs for a and c running down the page and the
alternative signs options alongside them.
a + − − + c
1×6 (x + 1)(6x − 10) (x − 1)(6x + 10) 1 × 10
(x + 10)(6x − 1) (x − 10)(6x + 1) 10 × 1
(x + 2)(6x − 5) (x − 2)(6x + 5) 2×5
(x + 5)(6x − 2) (x − 5)(6x + 2) 5×2
2×3 (2x + 1)(3x − 10) (2x − 1)(3x + 10) 1 × 10
(2x + 10)(3x − 1) (2x − 10)(3x + 1) 10 × 1
(2x + 2)(3x − 5) (2x − 2)(3x + 5) 2×5
(2x + 5)(3x − 2) (2x − 5)(3x + 2) 5×2
We have sixteen candidates. They all generate the correct x 2 -term and the
correct constant term. At this stage we can go through them, checking the x
coefficients until we find −11. The answer is
6x 2 − 11x − 10 = (2x − 5)(3x + 2).
QQ
Use the tabular method to factorise this quadratic
21x 2 + 22x − 8.
When a and c have lots of factors the whole process can get very long-
winded. To an extent this is unavoidable, but in harder cases there is
often a neat way of dismissing some of the options by inspection i.e
just by looking at them. If you want to know how it works turn to page
A-12 in Appendix A (or click on the page number).
36
(3x + 4)(7x − 2). The table should contain 16 options.

When a < 0
We can deal with the case of a being negative in much the same way we dealt
with the special case a = −1. Take the example
8 − 22x − 21x 2 .
Rather than try to form brackets straight away, the simplest approach is to
first write the expression like this:
−(21x 2 + 22x − 8).
The expression inside the bracket is the one you factorised on the previous
page, so we have
−(3x + 4)(7x − 2).
This is the same as −(7x − 2)(3x + 4) which we may want to write as
(2 − 7x)(3x + 4).
QQ
Factorise this expression

6 − x − x 2.
The answer is the footnote37 .
37
−(x + 3)(x − 2) or (2 − x)(x + 3).

Look at this example:
24 − 3x − 21x 2 = −(21x 2 + 3x − 24).
The first thing we notice is that c has lots of factors, and because a 6= 1, the list
of options is starting to look very long—on the face of it 32. Very tedious!
Can you see what we could do to massively reduce this number?
The answer is in remembering that factorising can be of the one-bracket or
the two-bracket variety. In this case there is a common factor of 3:
−3(7x 2 + x − 8).
Finish this for yourself. You should find that there are now only 8 options to
consider. The answer is in the footnote38 .
We could have factorised the expression without taking out the factor 3, but
it would have involved looking at four times as many options. So remember:
A
always check for possible one-bracket factorising
!
A
A
before embarking on two-bracket factorising. A
38
−3(x − 1)(7x + 8) or 3(1 − x)(7x + 8).

3.6.4 The difference of two squares.

When b = 0 and c < 0 the pattern we get can be thought of as the difference of
two squares. You will find it cropping up all over the place. A simple example
would be:
x 2 − 9.
The x 2 coefficient is 1, we’ll use the method we arrived at in the previous
section. There is no x-term (b = 0) and the constant term is negative, so we
p zero. In other words the factors are
are looking for factors of 9 that differ by
the same, so they must be equal to the 9 and the factorisation is
x 2 − 9 = (x + 3)(x − 3).
The pattern can be expressed as:

A
2 2
!
A
X −Y = (X + Y )(X − Y ) A
A
The difference of two squares often appears in disguise. Here are some
examples. In the first one substitute: X = 3y and Y = 5.
9y 2 − 25 = (3y)2 − 52 = (3y + 5)(3y − 5)

1 − x2 = 12 − x 2 = (1 + x)(1 − x)
4a 2 − b 2 = (2a)2 − b 2 = (2a + b)(2a − b)
x2
¡ ¢2
= x5 − 22 = x5 + 2 x5 − 2
¡ ¢¡ ¢
25
−4
³ ´2 ³ ´³ ´
2 1 2 1 1 1
p − p2 = p − p = p+p p−p
Mull over these examples and try to absorb the underlying structure. That
will help you see the pattern when it occurs:
ä2 − ∆2 = ( ä − ∆ ) ( ä − ∆ )
You will be expected to spot when an expression can be thought of as the

difference of two squares.
QQ
Factorise each of these.

y2
(a) x 2 − 16 (b) p 2 − q 2 (c) 100x 2 − 1 (d) 9 − 4
39 y y
(a) (x + 4)(x − 4) (b) (p + q)(p − q) (c) (10x + 1)(10x − 1) (d) (3 + 2 )(3 − 2 )

The difference of two squares pattern can be in disguise in various other ways.
Look at this example:
49 − y 6 .
“Wrapped up" in the y 6 term is a square term: (y 3 )2 , so
49 − y 6 = 72 − (y 3 )2 = (7 + y 3 )(7 − y 3 ).
One way to look at this is to imagine replacing y 3 with (say) Y . Then y 6

becomes Y 2 and the pattern is clear:
49 − y 6 = 72 − Y 2 .
In this next example the fact that all the powers of x are odd could put us off
the scent:
x 5 − 16x.
The trick is to see that there is a common factor we can take out:
x 5 − 16x = x(x 4 − 16) = x(x 2 + 4)(x 2 − 4).
Did you notice too that we can go one step further in this case: we can
factorise the x 2 − 4, so
x 5 − 16x = x(x 2 + 4)(x + 2)(x − 2).
You may be wondering what happens if the constant term is negative but is not
an exact square.

Here is an example
x 2 − 5.
Is it possible to factorize this? Yes it is:
p p p
x 2 − 5 = x 2 − ( 5)2 = (x + 5)(x − 5)
and similarly p p p
3x 2 − 1 = ( 3x)2 − 12 = ( 3x + 1)( 3x − 1)
QQ
Factorise each of these expressions.
(a) x 4 − y 2 (b) x 6 − 25 (c) x 4 − 9x 2 (d) 3a 2 − 12 (e) y 2 − 2
40
(a) (x 2 + y)(x 2 − y)(b) (x 3 + 5)(x 3 − 5) (c) x 2 (x + 3)(x − 3)
p p
(d) 3(a + 2)(a − 2) (e) (y + 2)(y − 2).

You may have seen this pattern (the difference of two squares) used in
arithmetic calculations. For example to find the values of
(a) 20022 − 19982 and (b) 99.9 × 100.1
Recognising the pattern we have
(a) 20022 − 19982 = (2002 + 1998)(2002 − 1998) = 4000 × 4 = 16000
and using the relationship the other way round
(b) 99.9 × 100.1 = (100 − 0.1)(100 + 0.1) = 1002 − 0.12 = 10, 000 − 0.01 = 9, 999.99
Use the difference of squares to evaluate these:
(a) 1022 − 982 (b) 50.12 − 49.92 (c) 201 × 199 (d) 0.72 × 0.68
This is all you need for now, but there is a lot more juice in the difference of
two squares. For example as a simple way to simplify an expression like this
without multiplying out the brackets
(3x + y)2 − (3x − y)2
and, even more powerfully
(2x + 6y + a)2 − (2x + 6y − a)2 .
If you want see how, turn to page A-13 of Appendix A (or click on the page
number).
Summarising, the identity on which the difference of two squares is based:
X 2 − Y 2 = (X + Y )(X − Y )
41
(a) 200 × 4 = 800 (b) 100 × 0.2 = 20 (c) 2002 − 12 = 39, 999
(d) 0.72 − 0.22 = 0.49 − 0.04 = 0.45

3.6.5 When b = 0 and c > 0. The sum of two squares?

Suppose we have a quadratic with no x term and a positive constant term, for
example
x 2 + 9.
Could we still factorise it? Let’s see what happens if we try. As before a = 1,
so we can use the standard method. The x-coefficient is zero, so we must be
looking for equal factors of 9, but the constant term is positive so the only
options are:
(x + 3)(x + 3) or (x − 3)(x − 3).
Neither of these is correct: one generates a +6x term, the other a −6x term.
Consequently
x 2 + 9 does not factorise.
We will find the same with any42 quadratic in which b = 0 and c > 0.
There is no “sum of two squares" factorisation 43 .
42
With a > 0.
43
At least not using Real numbers. In Chapter 1 we briefly introduced the idea of
Complex numbers, and develop the ideas further in Chapter 14. If we allow
such numbers, then the sum of two squares can be factorised.

3.6.6 Perfect squares

Look at these expansions:
(X + Y )2 = (X + Y )(X + Y ) = X 2 + 2X Y + Y 2 A
!
A
(X − Y )2 = (X − Y )(X − Y ) = X 2 − 2X Y + Y 2 A
A
We say that an expression is a perfect squares when it can be written as
(something + something-else)2 or (something − something-else)2 .
Here are some examples. Check them by multiplying out the brackets (in your
head if you can).
y 2 + 10y + 25 = (y + 5)2
t 2 − 8x + 16 = (t − 4)2
a 2 + 6a + 9 = (a + 3)2
x 2 + 2x + 1 = (x + 1)2
x 2 − 2x + 1 = (x − 1)2
Look at the last two examples. Changing the sign on the right hand side from
+ to − only changes the middle term on the left hand side (from + to −).
We see this in the two identities at the top of the page. We derive the second
one by replacing Y with −Y . But (−Y )2 = Y 2 , so only the middle term
changes.

Un the same way that difference of two squares pattern can be disguised, so
can a perfect square. Take this expression:
9x 2 + 6x + 1.
This happens to fit the pattern if we regard 9x 2 as (3x)2 , because
(3x + 1)2 = (3x + 1)(3x + 1) = 9x 2 + 6x + 1.
Another way to look at it is to replace 3x with X (say), then
9x 2 + 6x + 1 = X 2 + 2X + 1 = (X + 1)2 = (3x + 1)2 .
Show that each of these expressions is a perfect square:

(a) 4x 2 + 20x + 25 (b) 36x 2 − 12x + 1 (c) 49y 2 + 28x + 4
(d) (x − 1)2 + 4x.
The answers are in the footnote44
44
(a) Putting X = 2x, X 2 + 10X + 25 = (X + 5)2 = (2x + 5)2 . (b) (6x − 1)2
(c) (7y + 2)2 (d) (x + 1)2

A word of caution. In each of the examples we have looked at, the clue to
the expression being a perfect squares is that it has 2 terms that are exact
squares, e.g. 9x 2 = (3x)2 and 25 = 52 . But that doesn’t automatically make the
expression a perfect square. For example
9x 2 + 20x + 25.
To be a perfect square, it would have to be (3x + 5)2 , making the middle term
30x. So it isn’t a perfect square.
It’s a small step from where we have reached to factorising an expression like
9p 2 − 24pq + 16q 2 .
Since 9p 2 = (3p)2 and 16q 2 = (4q)2 , we can write it
(3p − 4q)2 .
Check for yourself that the middle term is indeed −24pq.

As before you may prefer to think of it as replacing (say) 3p with P , and 4q
with Q, giving
P 2 − 2PQ +Q 2 = (P −Q)2 = (3p − 2q)2 .
QQ
Decide whether each of these expressions is a perfect square:

(a) 36x 2 + 24x + 1 (b) x 2 + 4x y + 4y 2 (c) 100a 2 + 60ab + 9b 2 .
We can’t move on without highlighting a very common error. We know that
X 2 + Y 2 6= (X + Y )2 .
But this doesn’t always stop students writing

q
x2 + y 2 as x + y.
We only have to take the example x = 3 and y = 4 to see that this is wrong.
Summarising, the identities on which perfect squares are based:
X 2 + 2X Y + Y 2 = (X + Y )2
X 2 − 2X Y + Y 2 = (X − Y )2
45
(a) No. (b) Yes. (x + 2y)2 (c) Yes. (10a + 3b)2

3.6.7 Disguised quadratics

Quadratic expressions may be in disguise. In fact they often will be. Most
examples fall into two categories: those requiring substitution and those
requiring one-bracket factorisation.
Starting with substitution, take this expression:
x 4 + 14x 2 + 49.
If we replace X for x 2 , we have
X 2 + 14X + 49 = (X + 7)2 = (x 2 + 7)2 .
This next example takes exactly the same idea a bit further:
p
x − 2 x − 3.
There doesn’t appear to be anything remotely like a square term; in fact we

p
have a square root term. But if we substitute46 X = x we can write x = X 2 ,
giving
p p
X 2 − 2X − 3 = (X − 3)(X + 1) = ( x − 3)( x + 1)
If you can see the structure clearly in your mind’s eye you could skip the
substitution stage and go straight to
p p p p
( x)2 − 2 x − 3 = ( x − 3)( x + 1),
p
thinking of the expression as a “quadratic in x".
Can you see that this next example is actually very similar?
2 1
x 3 − 2x 3 − 8.
The secret here is to write the terms using powers and roots47 :
p p
( 3 x)2 − 2 3 x − 8.
p
3
The quadratic structure is now clear. Putting X = x, we have
X 2 − 2X − 8 = (X − 4)(X + 2)
p p 1 1
and the answer is ( 3 x − 4)( 3 x + 2) or (x 3 − 4)(x 3 + 2).
46 p
Any variable name will do e.g. t = x or whatever.
47
If you want a reminder of how we do this turn to page 2-26 (or click on the page
number).

Moving onto the idea of taking out a common factor, look at this expression:
x 4 + 14x 3 + 49x 2 .
As it stands it is not a quadratic, but a (one-bracket) factorising gives
x 2 (x 2 + 14x + 49)
and a further (two-bracket) factorisation yields the answer: x 2 (x + 7)2 .
An example involving roots:

p p
x x − 2x − 3 x.
p p
Thinking of the middle term as −2 × x × x, we can take out a common
p
factor of x:
p p
x(x − 2 x − 3)
leaving (inside the bracket) one of the expressions we met on the previous
page. So the answer is
p p p
x( x − 3)( x + 1).
Finally an example involving fractional indices:

5 3 1
x 2 − 7x 2 + 12x 2 .
5 1 3
By the laws of indices (see page 2-26) we can write x 2 = x 2 × x 2 and x 2 =
1
x 2 × x, so we rewrite the expression as
1 1
x 2 (x 2 − 7x + 12) = x 2 (x − 3)(x − 4).
If this seems like a lot to take in, we’ll be re-capping when we cover quadratic
equations in Chapter 7.
QQ
Here are some disguised quadratics for you to factorise for yourself:
p
(a) x 8 + 2x 4 + 1 (b) y 5 − 6y 4 + 9y 3 (c) y + y −6
7 4 1
(d) a 4 − a 2 b 2 (e) x + 5x + x .
3 3 3
48 p p
(a) (x 4 + 1)2 (b) y 3 (y − 3)2 (c) ( y + 3)( y − 2) (d) a 2 (a + b)(a − b)

Algorithm for factorising the quadratic ax 2 + bx + c.

The flow diagram below summarises what we have discovered about
factorising different types of quadratics.
It is meant as a quick reference guide. “Common factors" means check for

any one-bracket factorising.
START
@
c =0 @ c 6= 0
@
@R
@
Common @
a =0 @ a 6= 0
factors @
@ R
@
Common @
factors
@
b=0 @
@b 6= 0
@
@
@
R
@
Common a <0 @ a >0
factors
@
R
@
c <0 @ c >0 Take out - Common
@ factor −1 factors
R
@
Difference No factors @
of two
@a 6= 1
@
a =1
squares @
@
R
@
Factor Factor
pairs of pairs of
coefft c a and c
@
c <0 @ c >0
?
@ List the
@
R
@ options
Difference Sum ?
of factors of factors Inspect
=b =b and
eliminate

Factors, roots and equations

Factorising is an essential first step in solving many kinds of equations. If the
factors are linear, they lead immediately to the solutions of the equation,
using the zero product principle we met in Chapter 1. Here is a quick
reminder.
zero.
? × ? =0
What could the numbers be?

It can’t work unless one of them numbers is zero. Or both could be zero. If
the product is in the form of two brackets multiplied together, perhaps the
result of a factorisation, we might have something like this
(x − 10)(x + 2) = 0 ⇒ x − 10 = 0 or x + 2 = 0
⇒ x = 10 or x = −2.
This will apply however many elements there are in the product, for example
x(x + 1)(x − 2)(x − 7) = 0.
Either x = 0, x + 1 = 0, x − 2 = 0 or x − 7 = 0.
The roots or solutions are x = −1, 0, 2 and 7.
This is the approach we will often use to find the roots of a quadratic or
polynomial equation. But rather than develop the ideas here, we will pick
them up again when we deal with quadratics and polynomials in Chapters 7
and 8.

Exercises 12
In questions 1 to 7 write each expressions as a product of factors. Where

possible check each answer by multiplying it out in your head.
1. (a) x 2 + 7x + 10 (b) x 2 + 7x + 6 (c) x 2 + 7x (d) x 2 + 5x + 6

(e) y 2 + 8y + 15 (f ) y 2 − 8y + 15 (g) y 2 − 8y (h) x 2 − 10x + 9
(i) x 2 + 6x + 9 (j) 6 − x − x 2 (k) x 2 + x (l) x 2 + 3x − 10
(m) 10 + 3x − x 2 (n) y 2 + 2y + 1 (o) x 2 − 8x + 12 (p) x 2 − 7x + 12
2. (a) 2x 2 + 11x + 5 (b) 3x 2 + 7x + 1 (c) 5x 2 + 7x + 2

(d) 7x 2 − 29x + 4 (e) 3y 2 + 8y + 4 (f ) 7p 2 − 3p − 4
(g) 6 + q − 2q 2 (h) 5y 2 + 18y − 8 (i) 2x 2 − 7x y − 9y 2
(j) 4t 2 + 23t − 6 (k) 6x 2 + 13x + 5 (l) 4x 2 − 11x + 7
(m) 4x 2 + 4x − 15 (n) 8y 2 + 13y − 6 (o) 3p 2 − 16pq − 12q 2
(p) 5x 2 + 15x + 10 (q) 6x 2 + 26x − 20 (r) 5 − 3x − x 2
(s) 4 + 7x − 2x 2 (t) 10 + 11y − 6y 2 (u) 8 − 22t − 6t 2
3. (a) y 2 + 2y − 63 (b) t 2 + 6t − 16 (c) w 2 − 15w − 16 (d) w 2 − 16

(e) w 2 − 16w (f) x 2 − 1 (g) p 2 − 100 (h) 9x 2 − 16
(i) y 4 − x 4 (j) 1 − 4t 2 (k) 3t − t 2 (l) 25a 2 − b 2
(m) 36 − k 6 (n) M 2 − 9N 2
4. (a) 3x 2 + 6x + x y + 2y (b) 10x y + 4x + 35y + 14

(c) 8ab + 4b 2 − 2ac − bc (d) 8ab − 2ac − bc + 4b 2
(e) 2x y + 3y − 2x 2 − 3x (f) 6x 4 + 5x 3 − 12x y − 10y
(g) 3a 2 + 2ac + 3ab + 2bc + 9a + 6c (h) x y + 2y − 4x 2 − 8
5. (a) x 3 + 7x 2 + 10x (b) x 3 + 10x (c) 4x 2 + 4x − 24

(d) 3x 3 − 18x 2 + 45x (e) 2x 2 − 50 (f ) 6y + 3y 4
(g) x 3 − x (h) p 4 + p 3 − 2p 2 (i) ax 2 − 6ax + 8a
(j) 2y x 2 − 12x y − 14y (k) k − kx 2 (l) x 6 − 9x 4
(m) 2x 2 − 9x 4
6. (a) x 2 + 6x + 9 (b) x 2 + 14x + 49 (c) y 2 − 4y + 4

(d) t 2 + 2t + 1 (e) x 2 + 6x (f ) p 2 − 2p + 1
(g) y 2 − 9 (h) 5q 2 + 20q + 20 (i) x 2 − 8x + 16
(j) x 2 − 16 (k) 2y 2 + 20y + 50 (l) 2y 2 + 20y
(m) 9x 2 + 6x + 1 (n) 1 − x 2 (o) 16p − 12p 2
(p) 4y 2 − 12y + 9 (q) 8k 2 − 24k + 18 (r) 8k 2 − 24k
(s) 2k 2 − 18 (t) 5x 2 − 20 (u) y 3 + 5y 2 + 6y
(v) 9y 2 − 4 (w) 16x 2 − 1 (x) 32x 3 − 2x
(y) 100 − y 2 (z) p 2 − q 2

7. (a) 3x 3 − 11x 2 − 4x (b) 4a 2 − b 2 (c) t 4 − 4

(d) 1 − p 2 q 2 (e) 8x 2 − 18 (f) 8x 2 − 18x − 18
(g) 8x 2 − 18x (h) 8x 4 − 18x 3 (i) x 6 − y 6
(j) 9y 2 − 4x 2 (k) t 3 − 9t (l) 2t 4 − 8t 2
(m) x 2 − 2 (n) 7 − y 2
8. Find the length of the hypotenuse of a right angled triangle in which the
lengths of the other two sides are 2t and 1 − t 2 .
9. Factorize these expressions completely.

(There are hints for some in footnotes49 .)
(a) x 4 + 7x 3 + 10x 2 (b) x 4 + 4x 2 + 3 (c) x 4 − 7x 2 + 10
(d) y 4 + y 2 − 2 (e) x 4 − 2x 2 − 15 (f ) 2x 4 − 7x 2 − 8
(g) 3y 4 + 7y 2 − 6 (h) 3x 4 − 27 (i) y 4 − 6y 2 + 9
(j) y 6 − 4 [Hint in footnote.] (k) p 4 + 2p 2 + 1 (l) x 6 + 7x 3 + 10
p p p
(m) x + 5 x + 6 [Hint in footnote.] (n) x + 6 x − 7 (o) 2y + 5 y + 3
(p) x 5 − 2x 3 − 3x (q) x 2 y 2 + 4x y + 3 (r) p 2 + 2px + x 2
(s) x 4 − 2x 2 y 2 − 8y 4 [Hint in footnote.] (t) x 2 + 7ax + 10a 2 (u) y 2 + 6t y − 7t 2
(v) a 2 − 8ax + 15x 2
10. Write each expression as a perfect square, e.g. x 2 + 6ax + 9a 2 = (x + 3a)2 .

(a) x(8 + x) + 16 (b) 36t 2 + 12t + 1 (c) x 2 + 2x y + y 2
(d) (x + 2)(x + 4) + 1 (e) a(a − 8) + 4(a + 1) (f ) 25 + x 2 (x 2 + 10)
1 p2 pq q2 1
(g) y 2 + y+ 4 (h) 9 + 6 + 4 (i) x 2 + 2 + x 2
1 9 6
(j) X 2 + Y 2 − X Y
11. Use the difference of two squares to factorise these expressions.

(a) (2x + 3)2 − (2x − 3)2 (b) (y 3 + 1)2 − (y 3 − 1)2
(c) (4s + t + 1)2 − (4s − t + 1)2 (d) (3a + b + 2c)2 − (3a + b − 2c)2
x2
¡ a ¢2 1 1 4q 2
(e) 9 −1 (f) b −4 (g) y 2 − 25 (h) 1− x 2 (i) p 2 −r 2
2 ¢2 y 2 ¡ y ¢2
(j) x + x1 − x − x1
¡ ¢ ¡ ¡ ¢
(k) 1 − x − 1 + x
12. Evaluate these in your head.
(a) 3022 − 2982 . (b) 20.12 × 19.92 .
13. Write down an expression for the area between two concentric circles
of radii 51.2 cm and 48.8 cm. Calculate the area in your head, leaving your
answer in terms of π.
14. Show that, for any even integer N
(N + 1)2 − (N − 1)2
is always a multiple of 8.
49 p
(j) Set y 3 = Y . (m) Set x = y. (s) First try factorizing x 4 − 2x 2 − 8.

15∗ . By writing z = x + y, show that the expression
A = x 2 + 2x y + y 2 − 4
can be written z 2 − 4. Hence or otherwise factorise the expression A.
16∗ . Expand
(x − 1)2 + (y − 1)2
Hence or otherwise show that, for all values of x and y,
x 2 + y 2 ≥ 2(x + y − 1)
[There is a hint in the footnote50 .]

Verify this result for yourself by substituting values for x and y. For what
values of x and y would x 2 + y 2 = 2(x + y − 1)?
17∗ . Write a 2 − 2ab + b 2 as a perfect square.

Hence or otherwise show that a 2 − 2ab + 2b 2 ≥ 0 for all values of a and b.
[Same hint as for Question 12.]
18∗ . If x − 2 is a factor of x 2 + px + 8, find the value of p.
19∗ . Given that x 3 + 6x 2 + ax − 2 ≡ (x + 1)(x 2 + 5x + b), find a and b.
20∗ . Find the value of λ if x + 4 is factor of x 2 + x + λ.
21∗ . What is the value of the constant k if 2x + 1 is a factor of 6x 2 − x + k?
22∗ . Show that, whatever the values of x and y,
x 2 + y 2 ≥ 2x y.

q
23∗ . Simplify the expression x 1 − x8 + 16
x2
.
24. Show that the sum of the cubes of three consecutive integers is a multiple
of 3.
25. Show that for any integer n, n(n 2 − 1) is dividable by 6.
50
Hint: Any number squared ≥ 0.
51
Expand (x − y)2 and use the previous hint.

Solutions 12
1. (a) (x + 2)(x + 5) (b) (x + 1)(x + 6) (c) x(x + 7)

(d) (x + 2)(x + 3) (e) (y + 3)(y + 5) (f) (y − 3)(y − 5)
(g) y(y − 8) (h) (x − 1)(x − 9) (i) (x + 3)(x + 3) = (x + 3)2
(j) (2 − x)(x + 3) (k) x(x + 1) (l) (x + 5)(x − 2)
(m) (5 − x)(x + 2) (n) (y + 1)2 (o) (x − 2)(x − 6)
(p) (x − 4)(x − 3)
2. (a) (x + 5)(2x + 1) (b) (x + 2)(3x + 1) (c) (x + 1)(5x + 2)

(d) (7x − 1)(x − 4) (e) (3y + 2)(y + 2) (f) (7p + 4)(p − 1)
(g) (2q + 3)(2 − q) (h) (y + 4)(5y − 2) (i) (2x − 9y)(x + y)
(j) (4t − 1)(t + 6) (k) (3p + 2q)(p − 6q) (l) 5(x + 1)(x + 2)
(m) −(x + 5)(x − 2) or −(x − 2)(x + 5) or (2 − x)(x + 5)
(n) 2(3x − 2)(x + 5) (o) −(x − 4)(2x + 1) or (4 − x)(2x + 1)
(p) (5 − 2y)(3y + 2) (q) 2(t + 4)(1 − 3t )
3. (a) (y + 9)(y − 7) (b) (t + 8)(t − 2) (c) (w − 16)(w + 1)

(d) (w + 4)(w − 4) (e) w(w − 16) (f) (x + 1)(x − 1)
(g) (p + 10)(p − 10) (h) (3x + 4)(3x − 4) (i) (y 2 + x 2 )(y 2 − x 2 )
(j) (1 + 2t )(1 − 2t ) (k) t (3 − t ) (l) (5a + b)(5a − b)
(m) (6 + k 3 )(6 − k 3 ) (n) (M + 3N )(M − 3N )
4. (a) (3x + y)(x + 2) (b) (5y + 2)(2x + 7) (c) (2a + b)(4b − c)

2
(d) (2a + b)(4b − c) (e) (2x + 3)(y − 1) (f ) (6x + 5)(x 3 − 2y)
(g) (3a + 2c)(a + b + 3) (h) (x 2 + 2)(y − 4)
5. (a) x(x + 2)(x + 5) (b) x(x 2 + 10) (c) 4(x + 3)(x − 2)

(d) 3x(x − 1)(x − 5) (e) 2(x + 4)(x − 5) (f ) 3y(2 + y 3 )
(g) x(x + 1)(x − 1) (h) p 2 (p + 2)(p − 1) (i) a(x − 2)(x − 4)
(j) 2y(x + 1)(x − 7) (k) k(1 + x)(1 − x) (l) x 6 − 9x 4
(m) 2x 2 − 18x 4
6. (a) (x + 3)(x + 3) = (x + 3)2 (b) (x + 7)(x + 7) = (x + 7)2

(c) (y − 2)2 (d) (t + 1)2
(e) x(x + 6) (f) (p − 1)2
(g) (y + 3)(y − 3) (h) 5(q + 2)(q + 2) = 5(q + 2)2
(i) (x − 4)2 (j) (x + 4)(x − 4)
(k) 2(y + 5)2 (l) 2y(y + 10)
2
(m) (3x + 1) (n) (1 + x)(1 − x)
(o) 4p(3p − 4) (p) (2y + 3)2
(q) 2(2k − 3)2 (r) 8k(k − 3)
(s) 2(k + 3)(k − 3) (t) 5(x + 2)(x − 2)
(u) y(y + 2)(y + 3) (v) (3y + 2)(3y − 2)
(w) (4x + 1)(4x − 1) (x) 2x(4x + 1)(4x − 1)
(y) (10 + y)(10 − y) (z) (p + q)(p − q)

7. (a) x(3x + 1)(x − 4) (b) (2a + b)(2a − b) (c) (t 2 + 2)(t 2 − 2)

(d) (1 + pq)(1 − pq) (e) 2(2x + 3)(2x − 3) (f ) 2(4x − 3)(x + 3)
(g) 2x(4x − 9) (h) 2x 3 (4x − 9) (i) (x 3 + y 3 )(x 3 − y 3 )
(j) (3x + 2y)(3x − 2y) (k) t (t + 3)(t − 3) (l) 2t 2 (t + 2)(t − 2)
p p p p
(m) (x + 2)(x − 2) (n) ( 7 + y)( 7 − y)
8. Hypoteneuse is 1 + t 2 . If the hypotenuse is x, then by Pythagoras,

x 2 = (1 − t 2 )2 + (2t )2
= 1 − 2t 2 + t 4 + 4t 2 = 1 + 2t 2 + t 4
= (1 + t 2 )2
9. (a) x 2 (x + 2)(x + 5) (b) (x 2 + 3)(x 2 + 1) (c) (x 2 − 2)(x 2 − 5)

(d) (y 2 + 2)(y 2 − 1) (e) (x 2 − 5)(x 2 + 3) (f ) (2x 2 + 1)(x 2 − 4)
(g) (3y 2 − 2)(y 2 + 3) (h) 3(x 2 + 3)(x 2 − 3) (i) (y 2 + 3)2
p p
(j) (y 3 + 2)(y 3 − 2) (k) (p 2 + 1)2 (l) ( x + 2)( x + 3)
p p p p
(m) ( x + 7)( x − 1) (n) (2 y + 3)( y + 1) (o) (x 3 + 2)(x 3 + 5)
(p) x(x 2 + 1)(x 2 − 3) (q) (x y + 1)(x y + 3) (r) (x 2 + 2y 2 )(x 2 − 4y 2 )
(s) (p + x)2 (t) (x + 2a)(x + 5a) (u) (y + 7t )(y − t )
(v) (a − 3x)(a − 5x)
10. (a) (x + 4)2 (b) (6t + 1)2 (c) (x + y)2 (d) (x + 3)2
¢2 ¡ p q ¢2
(e) (a − 2)2 (f) (x 2 + 5)2 (g) y + 21
¡
(h) 3 + 2
¢2 ¢2
(i) x + x1 (j) X1 + Y3
¡ ¡
11. (a) 24x (b) 4y 3 (c) 4t (4s + 1) (d) 8c(3a + b)

¡x ¢¡x ¢ ¡a ¢¡a ¡ ¢ 1
¢¡ 1
¢
(e) 3 +1 3
−1 (f) bb
−+1
1 (g) y + 5
y − 5
³ ´ ³ ´
¡ 1
¢¡ 1
¢ 2p 2p
(h) 1 + x 1 − x (i) q + r q
−r
³ ´
¡2¢ 2y 4y
(j) (2x) x = 4 (k) − x (2) = − x
12. (a) 2400. (b) 399.99.

By the difference of two squares: (a) = 600 × 4 and (b) = 202 − 0.12 .
13. 240π cm2 . Area = (51.22 − 48.82 )π = (100)(2.4)π cm2 .
14. N is divisible by 2 and
(N + 1)2 − (N − 1)2 = [(N + 1) + (N − 1)][(N + 1) − (N − 1)]

= [2N ][2] = 4N
so the (N + 1)2 − (N − 1)2 is divisible by 8.

15. z 2 = x 2 + 2x y + y 2 ,
so A = z 2 − 4 = (z + 2)(z − 2) = (x + y + 2)(x + y − 2).
16. Let K = (x − 1)2 + (y − 1)2 = x 2 + y 2 − 2x − 2y + 2.

Since (x − 1)2 and (y − 1)2 are both ≥ 0 (for all values of x and y), K ≥ 0.

It follows that x 2 + y 2 ≥ 2x + 2y − 2.
17. a 2 − 2ab + b 2 ≡ (a − b)2 ≥ 0 (for all a and b).
18. x 2 + px + 8 ≡ (x − 2)(x + ?).

Comparing x 2 terms the coefficient of x in the second bracket must be 1.
Comparing constant terms, the second bracket must be (x − 4), so p = −6.
19. Comparing constant terms b = −2, and we have
x 3 + 6x 2 + ax − 2 ≡ (x + 1)(x 2 + 5x − 2)
Comparing x terms, a = 3.
20. Since the product of the brackets must start x 2 +. . ., the factorisation must
look like
x 2 + x + λ ≡ (x + 4)(x + ?)
To get an x term with coefficient +1 on the right hand side the factors must
be (x + 4)(x − 3). λ = −12.
21. Comparing x 2 coefficients, 6x 2 − x + k ≡ (2x + 1)(3x + ?). Comparing x

coefficients the factors must be (2x + 1)(3x − 2). k = −2.
22. (x − y)2 ≥ 0, so x 2 − 2x y + y 2 ≥ 0, or x 2 + y 2 ≥ 2x y.
x 2 −8x+16
¡ x−4 ¢2
23. Expression under the root is = , so the answer is x − 4.
x2 x
24. We could take the integers to be n, n + 1 and n + 2. The working is a bit

simpler if we take them to be n − 1, n and n + 1.
(a + b)3 = (a + b)(a 2 + 2ab + b 2 ) = a 3 + 3a 2 b + 3ab 2 + b 3 ,
so (n − 1)3 + n 3 + (n + 1)3 = (n 3 + 3n 2 + 3n + 1) + n 3 + (n 3 − 3n 2 + 3n − 1) =
3n(n 2 + 2)
25. n(n 2 − 1) = (n − 1)n(n + 1). These three consecutive integers must contain
at least one. Even number and at least one multiple of 3.


• Terminology: expression, term, coefficient. Combining terms
e.g. 5x 2 y + 6x y + y 2 x − x y(y + 2x) = 3x y 2 + 6x y = 3x y(y + 2)
• Multiplying out brackets e.g. 3pq −p(1−2q), (2x + y)2 , (a +b)(a −2b)2 .
A proof of Pythagoras’ Theorem.
• Adding, subtracting and multiplying polynomials e.g. finding 2g (x) −

f (x), where f (x) = x 2 − 2 and g (x) = 2 + 3x + x 4 , and the coefficient of
x 4 in the polynomial (x 3 + x 2 + 1)(2x 2 − 5)2 .
• Factorising and highest common factor e.g. 20b 3 − 15ab 2 = 5b 2 (4b −

3a).
• One-bracket factorising, i.e. taking out a common factor. The factor

could be a bracket e.g. 4(1 − 2x) − (1 − 2x)2 = (1 − 2x)(3 + 2x),
2x y − 3y − 4x 2 + 6x = (2x − 3)(y − 2x)
• Two bracket factorising.

Quadratics of the form ax 2 + bx + c. Various cases e.g.
8x 2 −18x, x 2 −8x +15, x 2 −3x −10, 7x 2 −29x +4, 2x 2 −7x y −9y 2
• The difference of two squares e.g. 4x 2 − 9 = (2x + 3)(2x − 3),

5 − 80y 2 = 5(1 + 4y)(1 − 4y), (2a + 1)2 − (a − 3)2 = (3a − 2)(a + 4)
• Perfect squares e.g. x 2 + 6x + 9 = (x + 3)2 , 4x 2 − 4x + 1 = (2x − 1)2

2x 2 − 16x + 32 = 2(x − 4)2
• Disguised quadratics e.g. x 3 + 2x 2 + x = x(x + 1)2 ,

p p p
x 4 − x 2 + 2 = (x 2 + 1)(x 2 − 2), x + 6 x − 7 = ( x + 7)( x − 1)

Chapter 4
ALGEBRAIC DIVISION AND

FACTOR THEOREM
Overview
4.1 Introduction. From arithmetic fractions to algebraic fractions.
4.2 A notation issue. This clarifies a small but important notation issue:
that fractions like
2 2a 1 x+1
3
a and 3 are equivalent, as are 2 (x + 1) and 2 .
4.3 Cancelling algebraic fractions. For example
3a 2 b 18y 3 x 2 +5x+6 x−1
, , ,
9b 3 6x y−9y 2 4x+12 x 2 −1
Common mistakes. One of them is illustrated with a “proof" that 2 = 1.
4.4 Multiplying algebraic fractions. This builds on 4.2. For example
1 3 a 1 2 1−a 1 y−1 ¢2
2a 2 × a , x + x1
¡
9x × 4, × b ,
4b 2 y 2 −y
× y ,
2
with a reminder that 3× 3 6= 69 , and that 8× 3x
2
= 81 × 3x
2
= 12x.
4.5 Dividing algebraic fractions: the “hammer and boot" rule.
3 x2
1
x y 4 2 x−3x+4
, 4 , ,
2 2+ x3 1
4 x−1
y2
4.6 Adding/subtracting algebraic fractions. For example
4 3a 2 2 x(x−1) x−3 4
a
+ 2, 3
x + 6 , x
+ x+1 .
4.7 Algebraic fractions with brackets.(Optional.)
4.8 Dividing polynomials with a linear divisor. For example

x 2 +2x−8 3x 3 +11x 2 −7x−29
x+4
, x+2
4.9 Factor Theorem. For example finding the factors of x 3 + 6x 2 − 9x − 14.
4-1
CHAPTER 4. ALGEBRAIC DIVISION AND FACTOR THEOREM
4.1 Introduction to algebraic fractions

At this level of Maths we will occasionally use the ÷ sign for division, but we
will almost always indicate division by means of a fraction. And since those
fractions will increasingly be algebraic rather than arithmetic, it is hard to
overstate the importance of being completely comfortable with algebraic
fractions.
It may surprise you to know that every Maths student I have ever taught has
at some point had problems with fractions—including those who have gone
on to study Maths degrees at the top universities. So if it applies to you too,
you are in good company.
What am I saying? That students can never do simple fraction calculations?

Not at all. Rather that when they do, they rarely understand what’s really
going on, so when they meet algebraic fractions, they can come unstuck.
They may be very happy with
5 3 5 3
× or −
6 10 6 10
but less so with something like
5 x 5 x
× or − .
2x 10 2x 10
If the algebraic ones don’t look any harder to you than the arithmetic ones,
great: you can probably skip a lot of this chapter. If they do, it’s time for a
quick trip back to basics. All will then become clear. (If you are not sure see
answers in the footnote1 at the bottom of the page.)
But first an important notation issue that arises in the world of algebraic
fractions.
15 3 x x 25−x 2
6 × 10 = 41 , 5
6
3
− 10 = 15 , 5
2x × 10 = 1
4,
5
2x − 10 = 10x .

4.2 An important notation issue

The division x ÷ 3 we could write either as
x 1
or x
3 3
x
We can check this by giving x a value, say x = 12. Then 3
= 12 ÷ 3 = 4, which
is the same as 31 × 12, i.e. 13 of 12.
x 1 1 x x
Better still, if we write x as we can see why: x = × = .
1 3 3 1 3
Generalising this:
1
dividing by n is the same as multiplying by .
n
Similarly:
2 2 a 2a
a= × = .
3 3 1 3
a
The point of writing a as is to make it crystal clear that the a is not “on the
1
line"; it’s very definitely in the numerator.

Some more examples:
1 1 (x + 1) x + 1
(x + 1) = × =
2 2 1 2
and
3 3 (2y 2 ) 3 × 2y 2 6y 2
(2y 2 ) = × = =
5 5 1 5 5
The key is to make it clear what’s in the numerator and what’s in the
denominator, so when it come to combining fractions we know what to do.
For example
2 a 2a a 2a + a 3a
a+ = + = = = a,
3 3 3 3 3 3
Or µ ¶
2 1 2 1
a+ a = a + = a
3 3 3 3
(taking out the common factor of a).
This next example brings out two key issues. The first is the usefulness of
2y
regarding the 2y as 1 . Then a reminder that to add or subtract fractions
they must have identical denominators:
5 2y 5y 14y 5y 9y
2y − y = − = − = .
7 1 7 7 7 7
2y
We have multiplied the top and the bottom of 1
by 7 (using the principle of
equivalent fractions) .
The idea of writing every term as a fraction will crop up a lot. For example
3x 1 2 x2 a 5a
3x = , x = , 5 =
1 2 2 b b
makes crystal clear what belongs in a numerator and what belongs in a
denominator—essential when adding or subtracting expressions involving
fractions.
QQ
Simplify these expressions

3y 2
(a) 91 x + 29 x (b) 5
− 15 y (c) 4x 2 + 2x3 (d) 34 x − x.
x y 14x 2
2
(a) 3 or 13 x (b) 5 or 51 y (c) 3 (d) − x4 or − 14 x.

4.3 Cancelling algebraic fractions

When cancelling fractions we need to find factors shared by the numerator
and denominator (the Highest Common Factor). Sometimes this is obvious;
if it isn’t we can always find it by factorising, for example:
39 3 ×13
3
= = .
1001 7 × 11 ×
13
77
Turning to an algebraic example, the principle is the same: examine the

factors and see what cancels. For example
15ab 2 3 × 5 × a × b ×
b 3a
3
= = .
20b 4 × 5 × b × b 4b
b ×
If it’s immediately clear that the HCF is 5b 2 , we don’t need the middle step.

Now look at this fraction

15a + 10
.
5
How would the cancelling work in this case?
It’s not uncommon to see students give the answer:
3a + 10 or 15a + 2.
The mistake in each case was to divide the denominator—but only part of
it—by 5.
Since we are looking for common factors it makes sense to factorise:
15a + 10 5 (3a + 2)
= = 3a + 2.
5 5
Alternatively3 we could divide every term in the numerator and every term in
the denominator by 5:
3a 2
15a + 10 15a
+ 10

= = 3a + 2
5 5
1
This amounts to the same thing as taking out a factor of 5 from top and
bottom. The important things is that the whole of the top and the whole of
the bottom must be divided.
What about the prospect of cancelling this fraction?

3x + 4 2a + a 2
and .
6 1+a
We can’t cancel either of them. Although there are various factors common
to top and bottom, neither contains a factor that divides into every term.
3
Or we could split it into two fractions added together, and cancel each one separately:
15a + 10 15a 10
= + = 3a + 2.
5 5 5
We are just applying the reverse process to adding fractions. Obviously this method
doesn’t work if it’s the denominator that contains more than one term.

In this rather more interesting example every term has a factor 3y:
6y 2
18y 3 18y
3
6y 2
=
=
6x y − 9y 2 y −
6x
2
9y 2x − 3y
2x 3y
or we could cancel by factorising first:
18y 3 . 6y 2
3y 6y 2
= =
6x y − 9y 2 3y(2x
− 3y) 2x − 3y
We can turn this into a simple rule: to cancel a fraction with more than one
term in its numerator or its denominator,
either
• factorise, and then cancel, or A
!
A
• cancel by dividing into every term. A
A
QQ
Cancel these fractions as far as possible:

6x 2 y 8x 16ab+20b 2 x 2 +x y
(a) (b) (c) (d) .
9x y 3 4+12x 2ab−6b 2 x 2 −x y
4 2x 2x 2(a+5b) x+y
(a) (b) (c) (d) .
3y 2 1+3x a−3b x−y

We won’t always have a choice of methods. Look at this example:
2x 2 + 6x
.
3x + 9
There’s nothing that divides into every term, so factorising is the only way to
go:
2x(x + 3) 2xX(xX+X3)
X 2x
= X = .
3(x + 3) 3(xX+X3)
X 3
In this case the common factor was a bracket.
In this next example

4a 2 b − 2ab 2
2ab 2 − b 3
we could divide every term (top and bottom) by b. But as you will see we still
need to factorise, so dividing by b can just as well come later:
4a 2 b − 2ab 2 2ab (2a

X−Xb) 2aSb 2a
X
X
2 3
= 2
= = .
2ab − b b (2a − b) b
XXX
X bA2
The natural next step is to include the possibility of two-bracket factorising.

You will need this for Core 2, but probably not Core 1. Two examples:
x 2 + 5x + 6 (x + 2)X
(xX+XX x +2
3)
= =
4x + 12 4(xX
X +X3)
X 4
y −1 (yX
X −X1)
X 1
2
= = .
y − 1 (y + 1)(yX
X −XX y +1
1)
QQ
Cancel these fractions as far as possible. (Be careful: some may not cancel at
all):
10x 4 x 2 +y 2 x 2 +x y x+4 x 2 −9
(a) (b) (c) (d) (e) .
3x 3 y−2x 4 x+y x+y 2
x +5x+4 x−3
5 10x 2 x(x+y)
(a) (b) No cancelling possible (c) =x
3x y−2x 2 x+y
x+4 1 (x+3)(x−3)
(d)
(x+1)(x+4)
= x+1 (e)
x−3
= x + 3.

Exercises 13
1. In each case combine the terms into a single fraction.

a
(a) 5 + 2a
5
1
(b) 5 a+ 5 a
2 1
(c) 2 x+ 2 x
1
3 1 1 1 7x
(d) 4 p− 4 p (e) 6 (a + 1)+ 6 (a + 1) (f ) 2 − 12 x
k+1
(g) x+ 3 x
5
(h) 2 − 31 (k + 1) (i) 2
5x
− x4
2 a y2 y2 1 3
(j) 3 a− 6 (k) 2 + 8
(l) 2 x− 4 x
1 1 1
(m) 2 x y+ 3 x y+ 4 x y
2. In each case find the prime factors of the numerator and the
denominator, and hence cancel as far as possible.
128 28 51 121 35
(a) 144 (b) 180 (c) 340 (d) 1001 (e) 98
3. Cancel these fractions as far as possible

3a 2 b 7 pq 2 r 3
(a) 9b 3 (b) 63x (c) p 3 q 2 r
15x 2 y 2 4x 2 +10x y 2a 2 −6a
(d) 20y z (e) 2
(f) 4a
20x+3y 3y 2 +12y x 14pq
(g) 30
(h) 3y 2
(i) 21p−7q
a 2 +b 2 4x 2 +8x y t 2 +3t
(j) 2ab (k) x+2y
(l) t 2 +t
4. Cancel these fractions as far as possible.

3x 2 −9x x 2 −4 2(x+1)2
(a) x−3
(b) x+2 (c) x 2 −1
x 2 +7x+10 2y−y 2 x 2 +8x+15

(d) x 2 −x−6 (e) 4−y 2 (f) x+5
y 2 −6y+9 2x 3 +10x 2 a 2 +2ab+b 2
(g) 2y−6
(h) x+5
(i) a+b
x−1 x 4 −1 k(4−y 2 )
(j) 7x 2 −3x−4 (k) x 2 −1 (l) 2+y
x 3 +x 2 −6x x 2 −1 x 2 −1
(m) x 2 −2x
(n) x 2 −x (o) (x−1)2
x 2 +1 12x 2 +4x y+9x+3y x(x+2)−4x−8

(p) (x+1)2 +(x−1)2 (q) 2(3x+y)
(r) x+2
x(x+1)−3x+8 (a+b)2 −1 1+x+x 2 +x 3
(s) x+2
(t) a+b−1 (u) 1+x
4y 2 −9 2a 2 +28a+98
(v) 2y 2 −3y−9 (w) a 2 +6a−7
5. Simplify these fractions by cancelling.

(3x−6)2 (4x+2)2 4 (a y+a)3 (t 2 −t )2
(a) 3
(b) 10x+6 (c) (2y−10)2 (d) 2a
(e) t 2 (t 2 −1)

Solutions 13
3a 3a 3 1 p
1. (a) 5 (b) 5 or 5 a (c) x (d) 2 p or 2
3(k+1)
1
(e) 3 (a + 1) or 3
a+1
(f) 3x
8
(g) 3 x or 3
8x
(h) 6
− 2(k+1)
6
= k+1
6
9x a 5y 2 1
(i) 4 (j) 2 (k) 8 (l) − 2 x
13
(m) 12 x y
8 7 3 11 5
2. (a) 9 (b) 45 (c) 20 (d) 91 (e) 14
a 1 r2 3x 2 y
3. (a) 3b (b) 9x (c) p 2 (d) 4z (e) x(2x + 5y)
a−3 y+4x 2pq
(f) 2 (g) None possible. (h) y
(i) 3p−q (j) None possible.
t +3
(k) 4x (l) t +1
2(x+1)
4. (a) 3x (b) x − 2 (c) x−1
x+5 y
(d) x−3 (e) 2+y (f ) x + 3
y−3
(g) 2 (h) 2x 2 (i) a + b
1 (x 2 −1)(x 2 +1)
(j) 7x+4 (k) x 2 −1
= x2 + 1 (l) k(2 − y)
x(x+3)(x−2) (x+1)(x−1) x+1 (x+1)(x−1) x+1
(m) x(x−2) = x + 3 (n) x(x−1) = x (o) (x−1)2
= x−1
x 2 +1 1 (3x+y)(4x+3) 4x+3 (x+2)(x−4)
(p) 2x 2 +2 = 2 (q) 2(3x+y)
= 2 (r) x+2
= x −4
x 2 −2x+8 (a+b+1)(a+b−1) (1+x)(1+x 2 )
(s) x+2
= x −4 (t) a+b−1
= a +b +1 (u) 1+x
= 1 + x2
(2y+3)(2y−3) 2y−3 2(a+7)2 2(a+7)
(v) (2y+3)(y−3) = y−3 (w) (a+7)(a−1) = a−1
32 (x−2)2 22 (2x+1)2 2(2x+1)2 4 1

5. (a) 3
= 3(x − 2)2 (b) 2(5x+3) = 5x+3 (c) 22 (y−5)2 = (y−5)2
a 3 (y+1)3 a 2 (y+1)3 t 2 (t −1)2 t −1
(d) 2a
= 2
(e) t 2 (t +1)(t −1) = t +1

4.4 Multiplying algebraic fractions

What does it mean to multiply two fractions? For example
1 6
× .
3 7
6
The fraction we can regard as 6 portions of a cake that has been divided
7
into 7 pieces. So since in arithmetic × and “of" are equivalent, we can think
1 6 2
of this as of . In other words 2 pieces, or .
3 7 7
1 6 2
× = .
3 7 7
A more direct way to arrive at the answer is the familiar rule:
to multiply fractions together
multiply all the numerators;

multiply all the denominators.
We have
1 6 6
× =
3 7 21
2
which we can simplify, by dividing top and bottom by 3, to give .
7
The most direct way to get to this answer is to cancel right at the start:
2
1 6A 2
× = .
3A 7 7
1
Who says we can do this? The following bit of mathematical “sleight of

hand" explains why:
2
1 6 1 × 6 6 × 1 6A 1
× = = = × .
3 7 3 × 7 3 × 7 3A 7
1
What this shows is that within a multiplication of fractions
we can cancel anything on the top with anything on the bottom.
By the way, students sometimes think they need to find a common

denominator when multiplying fractions. They don’t. They are confusing it
with adding or subtracting fractions.

Make it a practice to do as much cancelling as you can before multiplying.

The next example illustrates why this is such a good idea:
1 3
7 12 3 7A 3 9 12
Z
Z
× × = × × = .
8 5 14 8A 5 Z14
Z 20
2 2
This is a great deal easier than cancelling after multiplying:
7 × 12 × 3 252 126 63 9
= = = = .
8 × 5 × 14 560 280 140 20
Remember, the cancelling must involve one number from the top and one
from the bottom—so not the 12 and the 3 for example.
Mixed fractions need to be first converted6 to proper (“top-heavy”) fractions.

For example
2 1 12 10
2 ×3 = × = 8.
5 3 5 3
2×5 + 2
6
For example 2 25 = 21 + 25 = 10
+ 52 = 12
2 . Or more directly
5 5
whol e number × d enomi nat or + numer at or
This can be encapsulated into a simple rule:
d enomi nat or
3 5×4 + 3 23
Another example: 5 4 = = .
4 4

Moving on to an algebraic example:

2 3
4 9a 2 4A × 9A × aA2
× = = 6a
3a 2 3A ×Za × 2A
1 1 1
Here too, it’s often easier to cancel before doing any multiplying:
2 3
2
4 9a 4A 9A aA2
× = × = 6a
3a 2 3AZa 2A
Another example:
1
2
3x yx 3A x S y x2 x3
× = × =
5y 2 12 5y A2 Z 12
Z 20y
4

In section 4.1 we saw that it may not be immediately obvious what is in the
numerator and what is in the denominator:
3 1 3x 1 x
x× = × =
4 6 4 6 8
8 1 3 8 a2 4a
3× × a2 = × × =
15ab 2 1 15ab 2 5b
Satisfy yourself that these are correct.
In this next example (where the dots stand for all the fractions in between)
we don’t have the option of multiplying before cancelling:
1 2 3 4 5 N −1
× × × × ×......×
2 3 4 5 6 N
There are lots of pairs that cancel: the pair of 2’s, the pair of 3’s, the pair of 4’s
etc, running up to a pair of N − 1’s (there must be an N − 1 on the bottom of
the penultimate fraction). This just leaves a 1 at the start and an N at the end:
the answer is
1
.
N
If this isn’t completely clear to you, try it with a particular number e.g. N = 8,
and you should see how it works.
QQ
Simplify these as far as possible:

5 9 14 9x 2 2y 1 b2
(a) 7 × 10 × 15 (b) 8y 2 × 3x (c) 6 ×3a × a 2
2 3 4 5 99
(d) 3 × 4 × 5 × 6 × . . . . . . × 100 .
7 3 3x b2 1
(a) (b) (c) (d) .
5 4y 2a 50

We end this section with one of the commonest mistakes students make with
fractions. It also leads on to a very simple but useful little principle. First the
mistake:
2 24
12 × 6= .
3 36
The confusion is with equivalent fractions, where were we are multiplying
the top and the bottom by 12, leaving the fraction the same size—the same
number expressed in a different way:
12 × 2 24
= .
12 × 3 36
No, what we are trying to do here is make the fraction twelve times bigger:
4
2 Z 12
Z 2
12 × = × = 8.
3 1 3A
1
12
As soon as we see the 12 in front as , the mistake is avoided.
1
Some more examples, this time involving algebraic fractions:
3x 8 3x
8× = × = 12x
2 1 2
1 3p 2 1 p
3p 2 × = × =
9p 1 9p 3
As you get more used to expressions like this, you’ll be able to do some of the
cancelling in your head.
Now for an important special case.

We often find ourselves multiplying by the denominator of the fraction. For

example
1
2 3A 2
3× = × = 2
3 1 3A
1
a result we can generalise by saying
if we multiply a fraction by its denominator

the denominator disappears.
Or we could express it like this:
somet hi ng
× = somet hi ng .

The right hand is, in effect, on the top of the fraction , so it cancels with
1
the left hand .
Some more examples:
2a 3
×5 = 2a 3
5
11
a× = 11
a
2x + y
× 6x y = 2x + y
6x y
y
(y − 1) × = y
y −1

A key point to take away from this whole section is:
when multiplying fractions treat every element as a fraction, A
!
A
by inserting a denominator of 1 if necessary. A
A
This makes it crystal clear what’s on the top and what’s on the bottom.
QQ
Simplify all these in your head:

7 5x 9x 13 x+y
(a) 30× 15 (b) 4× 8 (c) 7× 7 (d) y × y (e) 3x × 3x .
8 5x
(a) 14 (b) (c) 9x (d) 13 (e) x + y.
2

Exercises 14
1. In each case multiply the fractions, cancelling where possible.

2 3 7 2 3 7 5a 2a
(a) 5 × 4 × 9 (b) 5 × 4 × 9 (c) 12 × 3
1 3 2 14 3 2
(d) 9 x × 4 (e) 7 × 15 a (f) 8 x × 9 y
6 2 a 1 1 2 3
(g) x × 3 x (h) 4b × 2 b 2 (i) 2 x × 3 x × 4 x
2x 2 1 2x 2 4x
(j) 3y × 6x y (k) 3y × 6x y (l) 7 × 7
2 x+1 6
(m) 5 × 5 y (n) 3 × 3 (o) y × y
4 1−a 3
(p) 5t 2 × 5t 2 (q) 2a × 2a (r) k 2 × 4k 2
2. Multiply the fractions, factorising and cancelling where possible.

16a+4b 6a t 2 +6t t +1
(a) 3a
× 4a+b (b) t +1 × t
x 2 −6x 3x+6 1 y−1

(c) x+2 × 2x−12 (d) y 2 −y × y
3x 1
(e) x+1 × (x + 1)2 (f ) (2 − y) × y−2
5b
(g) 6a × 3a 2 (h) (y 2 − y) ÷ (y 2 + y)

Solutions 14
7 1 5a 2 x 1
1. (a) 30 (b) 21 (c) 18 (d) 12 or 12
x
4a 4 1 ab
(e) 15 or 15
a (f) 12 y (g) 4 (h) 8
x3 x
(i) 4 (j) 9y 2 (k) 4x 3 (l) 4x
(m) 2y (n) x + 1 (o) 6 (p) 4

3
(q) 1 − a (r) 4
3x 1
2. (a) 8 (b) t+6 (c) 2 (d) y 2
5ab y−1
(e) 3x(x + 1) (f ) −1 (g) 2 (h) y+1

4.5 Dividing algebraic fractions:

hammer and boot rule
What about divsions involving fractions?
Let’s jump straight in with a hard one. What do we mean by this division?
1 1 3 12
3 ÷1 or .
2 6 1 16
We are asking how many 1 16 ’s must we add together to make 3 21 ?
The answer is 3 (satisfy yourself that this is right).
Of course this is an artificial problem constructed to illustrate a point. How

do we divide any two fractions?
At the start of the chapter we saw that
n 1
dividing by n = 1 is the same as multiplying by .
n
This principle carries across to any number:
a
dividing by the fraction b
b
is the same as multiplying by a
This is the familiar rule: to divide by a fraction we turn it upside down and
multiply.
4 2 4 3 12 6
÷ = × = = .
5 3 5 2 10 5
If it’s a mixed fraction we’ll need to convert9 it into an improper
(“top-heavy") fraction first:
1 1 7 7 7 6
3 ÷ 1 = ÷ = × = 3.
2 6 2 6 2 7
Some algebraic examples:

2a a 2 2a 9 6
÷ = × 2=
3 9 3 a a
4 4y y 2 4y 1 4
y ÷ y2 = ÷ = × 2=
3 3 1 3 y 3y
3×2 + 1
9
For example 3 12 = 31 + 12 = 62 + 21 = 72 . Or more directly
2
whol e number × d enomi nat or + numer at or
This can be encapsulated into a simple rule:
d enomi nat or
3 5×4 + 3 23
Another example: 5 4 = = .
4 4

There is a more direct technique„ which is usually quicker and easier. I call
it the hammer and boot rule. It’s for whenever we have a fraction within a
fraction, like in the example
3 12
.
1 16
We’ll explain the hammer and boot technique using arithmetic fractions, but
of course it applies equally to algebraic ones. We start with this example:
3
5 3 3 1 3
= ÷2 = × =
2 5 5 2 5×2
We’ve deliberately not finished the calculation because it’s the pattern we are
after, not the answer. What has happened? We imagine hitting the top
fraction 53 with a hammer, and think of it pushing the 5 down to join (and
multiply) the 2 underneath, like this:
3
5 3 3
↓ = =
2 5 × 2 10
Here are some more examples:
1 3 a 5
3 1 1 3
8 1 a
7 2x 5
↓ = = , ↓ = = , ↓ = , ↓ = .
2 3×2 6 9 8 × 9 24 b 7b 3x 6x 2
1
Can you see that the first answer makes sense? (Half of 3
is 16 .) It’s not so easy
to see why the others are correct.

Now let’s see what happens if the fraction within the fraction is on the bottom.
As before:
2 5 2 7 2×7
5
= 2÷ = × =
7
7 1 5 5
In this case we imagine booting the 7 up onto the top.
-2 14
5 = 15
7
Some more examples
-4 -2
1 = 12
1
= 12 6 = 2×7
6
= 73
3 7
- 2y 8y 3 - 5x
3 = 3 x = 10x
x
= 10
4y 2 2
When we have one fraction divided by another, we hammer and kick:

2 x 1
7 2 × 5 10 2 4x a 3
3
= = , 1
= = 2x, 2a
=
5
7 × 3 21 4
2 3
2a 2
QQ
Simplify these for yourself (the answers are in the footnote10 .)

2 4 a
3 7 5 y
(a) (b) (c) (d) 1
5 2 3 4
2 ab 1
6x 7 6 x
(e) 3 (f) 3 (g) a (h) 1
x 5 5 x3
10 10 2 a 10 5b
(a) (b) (c) (d) 4y (e) 2x 2 (f) (g) (h) x 2
3 7 15 21 6

Get into the habit of seeing the hammer and boot in action whenever you
find yourself with a fraction within a fraction. After a while you will be
simplifying and doing any easy cancelling in a single step, in your head. For
example take this expression
10x 2 y
3
2x
y
You may be able to see that, having applied the hammer and boot rule, the
2 will divide into the 10, the 3 multiplies with the 2 and there is an x that
cancels, leaving the fraction
5x y
6
Doing a simplification in your head carries risks of course, but if you can do
it reliably it is preferable. Why? you may ask. It is not because it saves a
minute or two; it’s because the simplifying the fraction will be just one step
in a bigger problem, and having to work it out on paper inevitably sidetracks
you from thinking about the wider problem. It’s like a car driver who can’t
change gear without consciously thinking about it; she can’t also maintain
the required focus when negotiating complicated traffic conditions.
Some more examples. They look a little more complex but the hammer and
boot principle applies in exactly the same way:
(2x + 1)2 3y(2x + 1)2

4
=
3y
4
5x 5x(y − 1)
x = = 5(y − 1).
y−1
x
In the last example we had to introduce a bracket to make sure the whole of
1 − y was multiplied by 3x. If we don’t put in the bracket we’ll find ourselves
5x y−1
writing x
.
To apply the rules we must be clear what is in a numerator and what is in a

denominator. That means we may need to modify the notation first (either
explicitly or in your head). For example:
1 x+1
9
(x + 1) 9 3(x + 1) x + 1
1
= x = = .
3
x 3 9x 3x

Exercises 15
Simplify each expression by writing it as a single fraction in its simplest form.

4 1 5
9 2 8 7 12
1. (a) (b) (c) (d) 8 ÷3 (e) 11 ÷6
3 5 20
2 5 4 1 1
(f) 3 (g) 1 (h)2 (i) 3 (j)9
4 4 3 5 2
2 1 3 5 1
5 3 4 8 4
(k) 3 (l) 3 (m) 1 (n) 15 (o) 1
7 8 2 16 5
3 a 1 x2
x 2 y y 1
2. (a) (b) (c) (d) (e) ÷b
2 5 2y 4x ab
5p 2
q 3 2y 5b 8
(f) (g) a (h) x (i) 10 (j)2
10p 2 y b x
2x a y2
1 3t 3 3 2
(k) a (l)9t 2
(m) y (n) 2 (o) 3y
b 2 5 b 5
1 4 3
3x 3y ax 2
(p) 2 (q) 3 (r) a
x y2 2
π p 1
p
2 πr 2 x 4 y
3. (a) (b) r (c) (d)
3 3 p1 1
x y
2 2
( 12 )(− 21 ) ( 52 )( 23 )( 12 ) x−1 x−1
(e) (f) (g) 1 (h) 1
32 3×2
(x−1)2 x 2 −1
4. In this question p = 12 , q = 27 , r = 4 and s = − 56 .

p q 1 1
(a) q (b) 2 (c) q 2 (d) p 2 + q 2 (e) pq
rs r +1 p− r1 1 1
(f) 4q (g) s (h) (i) 1+ s (j)
p+ r1 p+ q1
6( 32 ) 1 ( 12 )( a1 ) 6x −3
5. (a) (b) (c) (d)
4 3( x6 ) 1 1
4a 2 2x
³ ´ ³ ´2
xy 2 5p x+1 3
( ) 3 2 2y 2x
(e) x2
(f) q 2 (g) 3 (h) 1
6 ( )
2 x+1 24x

Solutions 15
4 1 1 7 2
1. (a) 27 (b) 10 (c) 32 (d) 24 (e) 11
8 5 2
(f) 3 (g) 20 (h) 6 (i) 3 (j) 9
14 8 3 2 5
(k) 15 (l) 9 (m) 2 (n) 3 (o) 4
3 a 1 x 1
2. (a) 2x (b) 10 (c) 2y 2 (d) 4y (e) ab 2
p 6 2y 2 b2
(f) 2q (g) a (h) x (i) 2 (j) 4x
b 2 10x ab 5y
(k) a (l) 3t (m) 3y (n) 2 (o) 6
3 4y 6
(p) 2 (q) 9 (r) x 2
p
π y
3. (a) 6 (b) 3πr (c) x (d) 4
1 5
(e) − 36 (f) 16 (g) 2(x − 1) (h) 2(x + 1).
7 1 4 25
4. (a) 4 (b) 7
(c) 49
(d) 2 (e) 7
35 3 1 1
(f) − 12 (g) −6 (h) 5 (i) − 5 (j) 4
2 12
5. (a) 1 (b) x (c) 2a (d) x 2
10p (x+1)3
(e) 2y 2 (f) q 2 (g) 12y 2 (h) 3(23x ).

4.6 Adding and subtracting algebraic fractions

At the start of the chapter we said that we need to understand how adding
or subtracting fractions works, otherwise we can’t deal with examples that
involve algebra, like
5 x
− .
2x 10
If you have no trouble with this, you can skip this section. (You might want
to do the examples at the end of the section just to check.)
We are going to start by putting the cat among the pigeons with a story.
It’s the end of their holiday, Cain and Pete have enjoyed their final meal and
they want to tip the waitress out of the foreign currency they have
left—which isn’t much. Cain has $3 left, and contributes $1, i.e. 13 of what he
has left; Peter has $5 left and chips in with $3, i.e. 35 of what he has left. They
will be giving the waitress a total of $4 out of the $8 they have between them,
or 48 of the total they have left. In other words
“ "
1 3 4 1
+ = =
3 5 8 2
3
Except that this can’t be right, because the result is less than !
5
The flaw is that they are fractions of different amounts. Fractions can only be
added or subtracted if they represent fractions of the same whole11 .
11
NB. This does not apply when we’re multiplying or dividing fractions.

So how do we deal with different denominators—or rather, why does the

method we have learned work? Although we are working with arithmetic
examples, their purpose is to clarify the principles that we can then apply to
algebraic fractions.
Suppose we divide a cake into 8 equal sized portions. The size of each
portion is 18 of the whole cake and we’ll take 5 of these portions. These
represent a fraction (or proportion) equal to 58 of one whole cake.
We then cut up a second cake, identical to the first, and add a quarter of this
cake to what we have taken from the first one. Because the whole is the same
size in each case, it makes sense to describe the total amount of cake as the
result of the addition
5 1
+ .
8 4
The question is: how does it makes sense to add these fractions? (As to
opposed to: what do we do?)
If we cut the 14 in half, to give 2 pieces each the size of an 81 of a whole cake,
we can combine these with the other 5 pieces that are the same size:
5 2 7
+ = .
8 8 8
7
The total is 8 of a whole cake.
Vital though it is to understand what is going one when we add fractions, we

obviously don’t want to have to go through a thought process like his every time!

Arithmetically what we’ve done is make the two denominators the same, using
the notion of equivalent fractions. The key point is that
to add or subtract fractions all the denominators must be the same
Whatever the context, fractions that have the same denominator represent
portion sizes that are the same, and can therefore be added or subtracted.
Pictorially:
5
8
+
1
4
=
7
8
Now we apply this to fractions with algebraic numerators:
x 3x 2x 3x 2x + 3x 5x 5
+ = + = = or x
2 4 4 4 4 4 4
2a 1 2a a 6a 5a a 1
− a= − = − = or a.
5 3 5 3 15 15 15 15
Before going any further, a reminder about a couple of issues, including the
notation issue we talked about on page 4-2.

An expression like
1 1
(y − 3) + (2y − 1)
2 5
can be written
y − 3 2y − 1
+ ,
2 5
which we can combine into single fraction in the same way12 :
5(y − 3) 2(2y + 1) 5y − 15 + 4y + 2
+ =
10 10 10
7y − 13
=
10
Notice the brackets we need to introduce in this case.
If one of the terms is not a fraction, we apply the usual trick of giving it a
denominator of 1:
x −2 3x x − 2
3x + = +
4 1 4
12x + x − 2
=
4
13x − 2
=
4
12
By using equivalent fractions to give them both the same denominator.

With algebraic fractions there is scope for sign errors in a way that there isn’t
with arithmetic ones. Suppose the two fractions we just added had been
subtracted instead; we start off as before:
x − 2 12x x − 2
3x − = −
4 4 4
The thing to notice is that at some point we will be subtracting x − 2. But it is
the whole of it we need to subtract, so it makes sense to put a bracket in:
12x − (x − 2) 12x − x + 2 11x + 2

= = .
4 4 4
As a general rule, it never hurts to enclose something in brackets, but it can
be positively dangerous not to when adding or subtracting algebraic
fractions.
Here is a slightly more complicated example:
2 2 x(x − 1)
x + .
3 6
Writing each expression as a fraction, and giving them both the same
denominator:
4x 2 x(x − 1) 4x 2 + x(x − 1)
+ = .
6 6 6
Now simplifying the numerator:
4x 2 + x 2 − x 5x 2 − x x(5x − 1)
= =
6 6 6
We have written the numerator in factorised form. In general
it’s better to leave an answer in factorised form where possible.

We can combine the fractions provided they have the same denominator. It
doesn’t matter what that is. In the last example we could have put them both
over 18, for example. We used 6 because that’s smallest denominator we
could have chosen, i.e. the lowest common denominator (LCD).
It’s not always obvious what the lowest common denominator is. We are
looking for the smallest number each denominator will divide 13 into, in
other words their Lowest Common Multiple. In this example
a 2a 1
+ −
6 9 4
the LCM is 36, because it’s the smallest number that 6, 9 and 4 all divide into,
so we have
6a 8a 9 14a − 9
+ − =
36 36 36 36
QQ
Combine these fractions. Simplify your answers as far as possible.
x
(a) 5 + 3x
10
(b) 2y− 3 y
1 7x
(c) 12 + x+6
4
7x
(d) 12 − x+6
4
13
In this context we mean divide exactly.
14 x 5y 5x+9 2x−9
(a) (b) (c) (d)
2 3 6 6

In this next example one of the denominators is algebraic. The process is

exactly the same; the only difference is that it tends to give rise to more juicy
algebra.
4 3a
+
a 2
In this case the lowest common denominator is 2a (the smallest number that
4
a and 2 divide into), so multiplying the top and bottom of by 2, and the top
a
3a
and bottom of by a, we have
2
8 3a 2 8 + 3a 2
+ = .
2a 2a 2a
In exams at this level15 , you’ll probably find you will quickly spot what the
lowest common denominator must be—from your understanding of what it
means.
15
For trickier examples (you will meet these later on), there is a neat systematic way of
deriving the LCM, using the concept of the union of sets (see Appendix A page A-26 or click
on the page number).

In this next example finding the lowest common denominator is not the
problem; it’s the manipulation that follows that is more demanding:
x +6 4
− .
x x −1
The numbers x and x + 1 do not have a common factor, so their LCM16 is

x(x + 1).
(x + 6)(x − 1) 4x x 2 + 5x − 6 − 4x
− =
x(x − 1) x(x − 1) x(x − 1)
x2 + x − 6 (x + 3)(x − 2)
= =
x(x − 1) x(x − 1)
Notice we had to expand (x − 3)(x + 1) so we could combine it with the 3x.
But there is no reason to expand the x(x + 1).
When writing down an answer it’s a good idea to do any factorising that is
possible. In this case it happens that there is a quadratic in the numerator
which factorises.
QQ
Combine these fractions. Simplify your answers as far as possible.
2 5 y x+4 x
(a) x + 3x (b) 4 + 4y 1
(c) x−2 + 5 (d) x−3 − x
1
We end this chapter by bringing together what we learned about adding and
subtracting algebraic fractions with the hammer and boot technique we saw
earlier.
16
If this isn’t completely clear, think of an arithmetic parallel, like 3 and 4. They have no
common factor so their LCM must be 3×4. Compare this with, say, 6 and 10. They have a
common factor of 2 so the LCM is 3 × 2 × 5 = 30.
x y 2 +16 (x+3)(x−1) x2 (x−3) 2
17
(a)
2
(b)
4y
(c)
5(x+2)
(d)
x−3
− x(x−3) = xx(x−3)
−x+3

More algebraic fractions within a fraction

In the following example the expression in the numerator has a fraction in
just one of the terms:
5
3
x +2
4x
Before applying the hammer, we convert the expression in the numerator into
a single fraction: 5x
3
+ 2 = 5x+6
3
, so we have
5x+6
3 5x + 6
=
4x 12x
Here is an even more interesting example:

1 2
3 (t + 5t + 4)
1
6
t + 23
t 2 +5t +4
3 6(t 2 + 5t + 4) 2X
(t X
+X4)(t
X + 1)
= t +4
= = = 2(t + 1).
6
3(t + 4) (t X
X +X4)
X
You will have an opportunity to practise examples like this in the exercises.
The principle is: whenever you have fractions within fractions
write the numerator and denominator as a single fraction
QQ
Multiply out these expressions, simplifying your answers as far as possible.

2 1
(a) x(x+ x ) (b) (1 + y 3 )(4y − y32 ) (c) ( p
2x
+ 1)2 + ( 2p1 x − 1)2
1− 12 a
(d)
1− 14 a 2
This is all you need for now, but if you want to go a bit deeper with algebraic
fractions, and see a lovely example from a C4 exam paper, turn to page A-15 in
Appendix A (or click on the page number).
3 1 2(2−a) 2(2−a) 2
18
(a) x 2 + 2 (b) 4y 4 + y− 3
y
(c)
x
+2 (d)
4−a 2
= (2+a)(2−a) = 2+a

4.7 More algebraic fractions

Often it’s when we put a few techniques together that things get interesting.
We are going to see some examples here.
We are wanting to multiply out the brackets; what’s new is that now there are
fractions involved. (Remember that every expression can be turned into a
fraction by giving it a denominator of 1.)
³ x´ 3x x 3x 2
3x 2 + = 6x + . = 6x +
2 1 2 2
x3 1
µ ¶
1
3
x 4 − 2 = 4x 3 − . 2 = 4x 3 − x.
x 1 x
Now some examples in which two brackets are multiplied together (we can
use the happy face rule19 ):
1 a 4a 2
µ ¶µ ¶
1 2 1 2
4a − a+ 2 = 4a 2 − + 2
−
a a a1 1 a a a2
8 2
= 4a 2 − 1 + − 3
a a
µ ¶2 µ ¶µ ¶
and 1 1 1
x+ = x+ x+
x x x
x 1 1
= x2 + 2 + 2
1x x
1
= x2 + 2 + 2
x
We could also do these two using negative indices to describe the fractions20 :
(4a − a −1 )(a + 2a −2 ) = 4a 2 − a 0 + 8a −1 − 2a −3
1 1
= 4a 2 − 1 + 8 − 2 3
a a
8 2
= 4a 2 − 1 + − 3
a a
and (x + x −1 )(x + x −1 ) = x 2 + 2x 0 + x −2
1
= x2 + 2 + 2
x
19
See section 3.2 on page 3-4
20
Negative indices are described on page 2-18, and fractional indices on page 2-26.

Let’s do a nice juicy one involving square roots:

µ ¶2 µ ¶µ ¶
1 1 1 1 2
p +1 = p +1 p +1 = + p + 1.
3 x 3 x 3 x 9x 3 x
Or using fractional indices:

µ ¶µ ¶ µ ¶µ ¶
1 1 1 1 1 −1 1 −1
p +1 p +1 = x +1
2 x +1
2
3 x 3 x 3 3
1 −1 2 − 1
= x + x 2 + 1.
9 3
1 1 2 1
= . + . +1
9 x 3 x 12
The results are the same of course; the root notation is probably the easier
route. When manipulating expressions it is normally easier to use powers
and roots rather than index form.

Exercises 16
1. Combine each of these into a single fraction in its simplest form.
(a) 1 1
2 (x + 3) + 4 (x − 1) (b)
2x−1
5
+ x+1
2
(c) 3a + 13 (6a − 1) (d) 1
3
y + 2y − 5
1−x
(e) x− 4 (f) 2x − 32 (x + 1) − 12 (x − 1)
x(x−1)
(g) 2
+ 14 x 2 (h) 83 (a − 1) + 21 (a + 1) + 18 (a − 1)
2. In each case find the LCD and combine into a single fraction.
4 2 2 1 3b a
(a) +3 (b) a+ (c) − 2b
x 5 a a
6 1 3x 1 1+x 1
(d) x
+ x+2 (e) x+2
− x+1 (f) x2
+ 2x
t −1 t x+1 x−1
(g) − (h) − x+1
t t +1 x−1
3. Perform these substitutions, simplifying the results as far as possible.

5x−y
(a) K = 3 , where x = 13 and y = − 43
a+3b
(b) x = 4a−b , where a = 21 and b = 56
1
(c) y = 3 (p 2 − q 2 ), where p = 25 and q = − 15
(d) T = (3x + y)2 , where y = 12 x
(e) X = 2a 2 − ab + 3b 2 , where b = − 32 a
6y
(f) z = 2x+y , where y = x
2a+b
(g) y = 3 , where b = − 12 a
y 2 −x 2
(h) T = y 2 +x 2 , where y = 34 x
4. Simplify each expression by writing it as a single fraction in its simplest

form.
2+ x3 5y 1
(a) (b) y (c) a 5
4 2
−1 2 +8a
4b b
3 −2
1 1
2 y− 4 x
x+ 34
(d) (e) 3y (f)
2
2
x+ 13
x 2 + 12 x 3a− 12 x 2 −4
(g) (h) (i)
4x+2 a+ 14 1 2
2
x −x
1 2 3 1 2 7
5
x + 5 x−2 (x+1)2 3
y − 3 y+ 38
(j) 1 (k) 3 2 (l) 1
5
x+1 4
(x −1) 2
y−4
5. Simplify these expressions.

x− x1 1 1
x − 1+x
2 1
1−x + x
(a) x+1 (b) x (c)
x
1− 1+x x+2+ x1

6. In each case multiply out the brackets and simplify as far as possible.
³ ´
(a) y 1 + 1y (b) 6x 2 1 − x3
¡ ¢ ¡1 ¢
(c) 2ab ab + 4 − ab
³ ´
y¢
(d) x 2 2y + x3y (e) 5a b − b2 (f) xy 3 + x
¡ ¢ ¡
¢2 ¢³ ´ ³ ´2
(g) 1 + x2 (h) y − 3 1 + 1y 3
¡ ¡
(i) a2
+b
¢2 ¡p ¢ ¡p
(j) 3 + x2 (1 + x) (k) x + x1 x + x1 x − x1
¡ ¢ ¡ ¢
(l)
³ ´³ ´ ³ ´¡
(m) p1 − 1 p1 + 3 (n) x + x12 x 2 − x1 (o) 1 + x1 x − 1 + x1
¢ ¡ ¢¡ ¢
t t
(p) 1 − x1 x + 1 + x1
¡ ¢ ¡ ¢
7. Repeat Question 6, tackling each expression in a different way.

(i) Re-write the algebraic components of ¡any fractions within the
2 3 2
expression in index form. For example 6x 1 − x = 6x (1 − 3x −1 ).
¢
(ii) Simplify the expression, leaving your answer in index form.

(ii) Check that it is equivalent to the answer you got in Question 6.
8. Choose any number and cube it. Subtract the number you started with.
Divide the result by one more than the number you started with. Now
add back the number you started with and take the square root of the
result. What is the answer?
9. ∗ Simplify these expressions. (There is a hint in the footnote21 .)

1+a −2 x −1 +y −1 p −2 −1
(a) (b) (c)
a x+y p −2 +1
1−x 1−4x −2 1− x2 − 3
x2
(d) −4 (e) (f)
2x 2+x (1+ x1 )2
1 1 3 1 2 1
x 2 −x − 2 (1+y) 2 −(1+y)− 2 a 3 −a − 3
(g) 1 (h) 1 (i) 5 1
x2 (1+y)− 2 a 3 −a − 3
21
Convert each of the terms with a negative index into fraction.

Solutions 16
3x+5 3x+5 15a−1 7y−15

1. (a) (b) (c) (d)
4 4 3 3
5x−1 x(x−2)
(e) (f ) −1 (g) (h) a
4 4
2
12+2x 2a +5 6b 2 −a 2 7x+12
2. (a) (b) (c) (d)
3x 5a 2ab x(x+2)
3x 2 +2x−2 x 2 +2x+2
(e)
x(x+2)
(f)
2x 2
(g) − t (t1+1) (h)
4x
(x+1)(x−1)
5
+4 3
3. (a) 1. Because 3 3 3 = 3 .
1 5
18 2+2 3
(b) 7 . Because = .
2− 56 7
6
1 41¡ 1
¢ 1¡ 3 ¢
(c) 25 . Because 3 25 − 25 = 3 25
.
49x 2 ¢ 7x
Because 3x + x2 = 2 .
¡
(d) 4
.
28a 2 2a 2 2
(e) 9
. Because X = 2a 2 + 3 + 4a9 .
(f) 2.
3a
a
(g) 2 . Because y = 32 .
7 7x 2 25x 2
(h) 25 . Because y 2 − x 2 = 9 and y 2 + x 2 = 9 .
6+x 10y
4. (a) 12 (b) y−2
8 5b
(c) 9a (d) 12
2y−x 3x+4
(e) 3y (f) 3x+1
6a−1
x 2 2(6a−1)
(g) 4 (h) 4a+1 = 4a+1
4
2(x+2)
(x−2)
2(x+2) (x+5)(x−2)
(i) x =
(x−2) x
(j) x+5
= x −2
4(x+1)2 4(x+1) 2(y 2 −7y−8) 2

(k) 3
(x+1)(x−1)
= 3(x−1)
(l) 3(y−8)
= 3 (y + 1)
x 2 −1
x 2 −1 x (x+1)(x−1)
5. (a) x − 1. Numerator: x
. Hence fraction is x+1 = x+1
.
x
1
1 1 1 x(x+1)
(b) x . Numerator: x(x+1)
; denominator: x+1 . Hence 1 .
x+1
1 x+1 x 2 +2x+1 (x+1)2
(c) 1−x 2 . Numerator: x(1−x) ; denominator: x = x
.
x+1
x(1−x) x (x+1)

Hence (x+1)2
=
2
.
x(1−x)(x+1)
x

6. (a) y + 1 (b) 6x 2 − 18x (c) 2 + 8ab − 2a 2 b 2

3x 10a 3x
(d) 2x 2 y+ y (e) 5ab− b (f) y +1
4 4 3 9 6b
(g) 1+ x + x 2 (h) y − 2− y (i) a 4 + a 2 +b 2
2 1 1
(j) 5+ x +3x (k) x 2 + 2+ x 2 (l) x− x 2
1
(m) t + p2t −3 (n) x 3 − x13 1
(o) x+ x 2
1
(p) x− x 2
7. (a) y + 1 (b) 6x 2 − 18x (c) 2 + 8ab − 2a 2 b 2
(d) 2x 2 y + 3x y −1 (e) 5ab − 10ab −1 (f) 3x y −1 + 1
(g) 1 + 4x −1 + 4x −2 (h) y − 2 − 3y −1 (i) 9a −4 + 6ba −2 + b 2
(j) 5 + 2x −1 + 3x (k) x 2 + 2 + x −2 (l) x − x −2

1
(m) t −1 + 2t − 2 − 3 (n) x 3 − x −3 (o) x + x −2
(p) x − x −2
8. The answer is the number you started with.

xX
(x+1)(x−1)
XX
x → x 3 → x 3 − x = x(x 2 − 1) = x(x + 1)(x − 1) → = x2 − x
Xx+1
XX
→ x 2 → x.
a 2 +1
8. (a) . [a −2 = a12 . Multiply top and bottom by a 2 .]
a3
1 x+y
(b) . [Num is x1 + 1y = x y . Then hammer.]
xy
1−p 2 x 4 (1−x)
(c) 2 (d) . [Denominator is x24 .]
1+p 2
x−2 2
(e) . [Num is x 4−4 = (x+2)(x−2) ]
4 4
x−3 2 (x−3)(x+1)
¡ x+1 ¢2 (x+1)2
(f) . [Num is x −2x−3 2 = 2 . Den is = x 2 .]
x+1 x x x
x+1 1 1
(g) or 1+ . [Multiply top and bottom by x 2 .]
x x
3 1
(1+y) 2 − 1
(1+y) 2 (1+y)2 −1
(h) 1 = = y 2 + 2y
1
1
(1+y) 2
1
[Multiply top and bottom by (1 + y) 2 .]
2 1
a3− 1
a3 a−1 1
(i) 5 = = [Because a 2 − 1 = (a + 1)(a − 1)]
a 3 − 11 a 2 −1 a+1
a3

4.8 Dividing polynomials

In Chapter 3 we multiplied polynomials together. If you want a reminder see
page 3-27.
In Section 3 of this chapter we found we could sometimes divide polynomials
by cancelling. For example:
x 2 + 2x − 8 (x − 2)X
(xX 4)
+X
X
= = x −2
x +4 x+
X XX4
and
4x 2 − 1 (2x
X+X1)(2x − 1) 2x − 1
X
.
X
= =
4x + 2 2(2x +
X1) 2
XX
X
Be careful not to cancel inappropriately. For example we can’t cancel the two
4’s in the second example. If this is not entirely clear to you see page 4-5.
What if there is no cancelling possible and we have no choice but to divide?

We will only be considering the case when the denominator is a linear
expression. But first a review of how arithmetic long division works.

Terminology and layout

We are going to use the same method and layout as with arithmetic long
division. Although this may be very familiar, it is important to understand
what is going on, so that we can carry across the same thinking into the
realm of polynomials. Remember: we are delving in order to clarify the
process, not to find the answer!
1 3 9 1 ← QUOTIENT
DIVISOR → 7 9 7 4 2 ← DIVIDEND
− 7
?
2 7
− 2 1
?
6 4
− 6 3
?
1 2
− 7
5 ← REMAINDER
The task is to find the number (the quotient) which when multiplied by 7 (the
divisor) gives 9, 742 (the dividend).
We start with the thousands column and generate as much of the 9, 000 as
we can by picking the number which, when multipled by 7, gives the largest
number that doesn’t exceed 9. That number is obviously 1, generating 7, 000
of the 9, 000 we need.
That leaves 2, 000 still to generate, so we must treat this as 20 hundreds. This is
combined with the 7 hundred in the dividend to give 27 hundreds. Repeating
the process 3 × 7 = 21 (meaning 300 × 7 = 2100) generates 21 hundreds from
our target of 27 hundreds. This leaves 6 hundreds, which we must treat at 60
tens and combine with the 4 tens to make 64 tens, and so on. At the end of
the process there is 5 left over, the remainder.
One way to express what we have found is to say
8, 270 2
= 2, 756 +
3 3
or
8, 270 = 3 × 2, 756 + 2.
These forms tend to be more helpful than using the notion of a remainder.
You may have met the term dividend in the context of investments. There
it describes the part of a company’s annual profit that is divided among its
shareholders.

We are going to use a similar layout to
divide 3x 3 + 11x 2 − 7x − 29 by x + 2.
In the numerical example we kept the 1000’s in one column, the 100’s in the
next column, and so on. Here we keep all the x 3 s in one column, all the x 2 s
in the next column, and so on.
In the arithmetic example we started by trying to generate 9, 000. Here we
start by generating the term with the highest power of x, i.e. the 3x 3 . We
need a term in the quotient which, when multiplied by x + 2, will generate
3x 3 . That has to be 3x 2 , which we put in the x 2 column. Then as before, we
multiply the 3x 3 by the x + 2 to give 3x 3 + 6x 2 .
Putting 3x 2 in the quotient has generated 3x 3 + 6x 2 towards the dividend
3x 3 + 11x 2 − 7x − 29.
3x 2
x +2 3x 3 + 11x 2 − 7x − 29
3x 3 + 6x 2
5x 2
We can now tick off the 3x 3 term and turn our attention to generating the
11x 2 term. We have already generated 6x 2 so we need a further 5x 2 . The
mechanics of this is that we subtract 3x 3 + 6x 2 from 3x 3 + 11x.
To generate this 5x 2 we need to put 5x in the quotient. But when we do this
we will also generate an x term, so in anticipation we “bring down” the −7x.
Then multiplying +5x by x + 2 we have 5x 2 + 10x. Subtracting this from
5x 2 − 7x gives −17x.
3x 2 + 5x
x +2 3x 3 + 11x 2 − 7x − 29
3x 3 + 6x 2
?
2
5x − 7x
5x 2 + 10x
−17x
We can tick off the 11x 2 term and move on to the −7x term. The process is
the same as before and we are left with a remainder of 5. Notice that
−29 − (−34) = −29 + 34 = +5.
3x 2 + 5x − 17
x +2 3x 3 + 11x 2 − 7x − 29
3x 3 + 6x 2
5x 2 − 7x
5x 2 + 10x
?
−17x − 29
−17x − 34
+5
You will simply be asked to perform the division, stating what the reminder
is (if any).

But what does this mean? It is not like arithmetic division where the
meaning is clear. For example 20 ÷ 3 = 6 reminder 2.
To say x + 2 divides 3x 2 + 5x − 17 times into 3x 3 + 11x 2 − 7x − 29, with a
remainder 5, may not be particularly helpful.
In the arithmetic example we could write the result like this:
20 2
= 6+
3 3
or 20 = 3 × 6 + 2
In the same way we can express the result of the algebraic division as
3x 3 + 11x 2 − 7x − 29 5
= 3x 2 + 5x − 17 +
x +2 x +2
or 3x 3 + 11x 2 − 7x − 29 = (x + 2)(3x 2 + 5x − 17) + 5.
QQ
1. Complete this division
2x 3
4x − 3 8x 4 + 6x 3 − 13x 2 + 23x − 12
8x 4 − 6x 3
+ 12x 3 − 13x 2
2. Divide 4x 3 + 22x 2 + 13x + 7 by x + 5.

3. Divide 3x 3 − 4x 2 − 6x + 5 by x + 1.
4. Divide 6x 4 − 7x 3 − 13x 2 + 19x − 11 by 2x − 3.
The answer to Question 1 is over the page. The other answers are in the
footnote22 .
22
2. 4x 2 + 2x + 3 rem 10 3. 3x 2 − 7x + 1 rem 4 4. 3x 3 + x 2 − 5x + 2 rem −5.

Answer to Question 1.
2x 3 + 3x 2 − x + 5
4x − 3 8x 4 + 6x 3 − 13x 2 + 23x − 12
8x 4 − 6x 3
+ 12x 3 − 13x 2
+ 12x 3 − 9x 2
− 4x 2 + 23x
− 4x 2 + 3x
+ 20x − 12
+ 20x − 15
+ 3
i.e. 2x 3 + 3x 2 − x + 5 remainder 3.

Alternative method for polynomial division
There is another, slicker method. The underlying process is the same for
both methods: constructing each coefficient in turn, starting with the
highest power. With this method we do it by inspection, working out in our
head what each coefficient must be. It is the quicker method but it is
inevitably more prone to errors, so always check your answer by multiplying
out at the end.
We start by writing
3x 3 + 11x − 7x − 29 = (x + 2)( )+?

To generate the 3x 3 term we need to start the second bracket with 3x 2 .
3x 3 + 11x − 7x − 29 = (x + 2)(3x 2 )+?

But multiplying 3x 2 by x + 2 will also generate +6x 2 . We need 11x 2 altogether
so we have to produce an extra 5x 2 . This we can do by adding +5x to the
bracket:
3x 3 + 11x − 7x − 29 = (x + 2)(3x 2 + 5x )+?
Can you see that multiplying what we have so far gives 3x 3 + 11x 2 + 10x? We
have dealt with the x 3 and x 2 terms and in the process generated 10x. We
need a total of −7x so we must produce an extra −17x. We can do this with a
−17 in the bracket:
3x 3 + 11x − 7x − 29 = (x + 2)(3x 2 + 5x − 17) + ?
Multiplying this out gives us 3x 3 +11x 2 −7x −34. We are almost there. We just
need a +5 to make the constant term −29. We can’t correct this from inside
the bracket; +5 is the remainder.
3x 3 + 11x − 7x − 29 = (x + 2)(3x 2 + 5x − 17) + 5
3x 3 + 11x 2 − 7x − 29 5
or = 3x 2 + 5x − 17 + .
x +2 x +2
The first method is more “mechanical”. It is less prone to error, too, because
we do less in our heads. The advantage with this second method is that we
can see what is going on; it is more obvious why we are doing what we do.
Repeat the four questions on page 4-44 using the inspection method.

Example 1
Divide the polynomial 2x 4 + 10x 3 + x 2 + 2x − 15 by x + 5.
Solution
2x 3 + x − 3
x +5 2x 4 + 10x 3 + x 2 + 2x − 15
2x 4 + 10x 3
+ x 2 + 2x
x 2 + 5x
− 3x − 15
− 3x − 15
This differs from the example we have just looked at in two ways. First the 2x 3
in the quotient happens to generate the x 3 we need as well as the x 4 term. We
therefore go straight to the x 2 terms, and bring down the 2x term as well,
anticipating that we will also be generating an x term.
Although it’s not strictly necessary, it is much easier if we to keep the integrity
of the columns: all the x 4 s under each other, all the x 3 s under each other
etc. This means leaving a space in the quotient where an x 2 term would have
gone.
The other difference is that there is no remainder, i.e.
x + 5 is a factor of 2x 4 + 10x 3 + x 2 + 2x − 15.
We can write
2x 4 + 10x 3 + x 2 + 2x − 15
= 2x 3 + x − 3
x +5
or 2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)(2x 3 + x − 3).

Example 1 (Version using the alternative method)
(a) Show that x + 5 is a factor of the polynomial 2x 4 + 10x 3 + x 2 + 2x − 15.

(b) Hence, or otherwise, show that
x − 1 is a also factor of 2x 4 + 10x 3 + x 2 + 2x − 15.
Solution
(a) We won’t assume it is a factor; at this stage we will allow for there to be a
remainder.
2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)( )+?

To generate the 2x 4 term we need to start the second bracket with 2x 3 .
2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)(2x 3 )+?

This deals with the x 4 term. But multiplying 2x 3 by x + 5 will also generate
+10x 3 . This happens to be exactly right, so we can move on to generating the
x 2 term; for this we need an x in the right hand bracket:
2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)(2x 3 + x )+?
This deals with the x 2 term. But multiplying x by x + 5 also generates 5x. We
need a total of 2x so we must produce an extra −3x. We can do this with a −3
in the right hand bracket:
2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)(2x 3 + x − 3) + ?
Multiplying −3 by x + 5 deals with the x term, and also produces the correct
constant term −15, as we expected. There is no remainder.
2x 4 + 10x 3 + x 2 + 2x − 15 = (x + 5)(2x 3 + x − 3)
2x 4 + 10x 3 + x 2 + 2x − 15
or = 2x 3 + x − 3.
x +5
(b) We could check that x +1 is also a factor by repeating the process. There is
a slightly quicker way. We want to show that x + 1 is a factor of (x + 5)(2x 3 +
x −3). We know x +1 is not a factor of x +5, so it must be factor of (2x 3 +x −3).
So yes, we can repeat the process but with this cubic. We leave you to check
this:
2x 3 + x − 3 = (x + 1)(2x 2 + 2x + 3).
Later we shall learn a simple rule for checking whether a linear expression is a
factor of a polynomial—without having to do the division.

Example 2
(a) Divide 8x 3 − 10x 2 − 17x + 21 by 2x − 3.

(b) By choosing a suitable value for x, use the result of part (a) to simplify the
fraction
64a 3 − 40a 2 − 34a + 21
.
4a − 3
Solution
(a)
4x 2 + x − 7
2x − 3 8x 3 − 10x 2 − 17x + 21
8x 3 − 12x 2
+ 2x 2 − 17x
2x 2 − 3x
− 14x + 21
− 14x + 21
We need to be careful with the subtractions. For example
−10x 2 − (−12x 2 ) = +2x 2 .
(b) We have found that
8x 3 − 10x 2 − 17x + 21
= 4x 2 + x − 7.
2x − 3
Putting x = 2a.
8(2a)3 − 10(2a)2 − 17(2a) + 21

= 4(2a)2 + (2a) − 7
2(2a) − 3
64a 3 − 40a 2 − 34a + 21
so = 16a 2 + 2a − 7
4a − 3

Missing terms
Divisions can be a little trickier if there are terms missing. In the next
example there is no x term in the dividend (its coefficient is zero).
dividing 2x 3 + 7x 2 + 2 by 2x − 1.
The process is no different, but we strongly advise that you keep the different
powers of x in strict columns. In this case it means leaving a space between
7x 2 and −2 in the dividend.
Check through each step to make sure you understand it. Or try it for
yourself first.
x 2 + 4x + 2
2x − 1 2x 3 + 7x 2 −2
2x 3 − x 2
+ 8x 2
+ 8x 2 − 4x
+ 4x − 2
+ 4x − 2
In the next example we find a term missing in the quotient:
dividing 4x 4 − 5x 3 − 6x 2 − 6x − 12 by 4x + 3.
x 3 − 2x 2 − 4
4x + 3 4x 4 − 5x 3 − 6x 2 − 16x − 12
4x 4 + 3x 3
− 8x 3 − 6x 2
− 8x 3 − 6x 2
? ?
− 16x − 12
− 16x − 12
At the second subtraction the difference is zero. That tells us that the x 2 term
is correct and we don’t need to generate any more. We can therefore move
straight on to the −16x term in the dividend. So we bring that down, and
bring down the −12 at the same time.
Whenever this happens during a long division, i.e. a subtraction yeilds zero,
we bring down the next two terms.

QQ
Divide
(a) 4x 3 − 11x + 5 by 2x − 1
(b) 3x 4 + 7x 2 − x − 2 by x + 2
(c) 6x 3 + x − 7 by x − 1
(d) x 4 + x 3 − 12x 2 − 8x + 20 by x − 3.
(e) 6x 4 + 4x 3 − 9x 2 − 3x + 2 by 3x + 2
(f) x 4 − x 2 by x − 1
23
(a) 2x 2 + x − 5 (b) 3x 3 − 6x 2 + 5x − 1 (c) 6x 2 + 6x − 7
(d) x 3 +4x 2 −8 rem 4 (e) 2x 3 −3x +1 (f ) x 2 (x +1) [x 4 −x 2 = x 2 (x 2 −1) = x 2 (x +1)(x −1)]

Example 3
The function f (x) is defined by

f (x) = 2x 3 + 3x 2 + 2x + 6.
Express f (x) in the form
f (x) = (x + 1)g (x) + r
where g (x) is a quadratic and r is a constant.
Solution
This looks harder than it is because the question is expressed in a very formal
way. We can see what’s going on if we divide both sides by x + 1:
f (x) r
= g (x) + .
x +1 x +1
In other words when we divide f (x) by x + 1 the quotient is g (x) and the
remainder is r . So let’s do the division:
2x 2 + x + 1
x +1 2x 3 + 3x 2 + 2x + 6
2x 3 + 2x 2
x 2 + 2x
x2 + x
x +6
x +1
+5
2
i.e. g (x) = 2x + x + 1 and r = 5.
f (x) = (x + 1)(2x 2 + x + 1) + 5.

Example 4
The polynomial f (x) is defined by
f (x) = 2x 3 − x 2 − 25x − 12.
(a) Show that x + 3 is a factor of f (x).

(b) Use the result of part (a) to factorise f (x) completely.
(c) Hence find the roots of the equation f (x) = 0.
Solution
(a) & (b). Using the Inspection Method, f (x) can be written
f (x) = (x + 3)(2x 2 − 7x − 4)
= (x + 3)(2x + 1)(x − 4)
(c) f (x) = 0 when (x + 3)(2x + 1)(x − 4) = 0, i.e. when x + 3 = 0, 2x + 1 = 0 or

x − 4 = 0.
1
The roots (i.e. the solution) are therefore x = −3, − 2 and 4.

Example 5
N is an integer.
(i) Explain why N (N − 1) is an even number.
(ii) Divide N 3 + 1 by N + 1.
N 3 +1
(iii) Hence or otherwise show that is always an odd integer.
N +1
(iv) By writing 8, 001 in the form N 3 + 1, factorise 8, 001 completely.
Solution
(i) One of the integers N and N − 1 will be an even number, so their product
will be even.
(ii)
N2 − N + 1
N +1 N3 +1
N3 + N2
− N2
− N2 − N
+N + 1
+N + 1
Notice that there is nothing to “bring down” until the last step when we bring
down the +1.
N 3 +1
(iii) = N 2 − N + 1 = N (N − 1) + 1.
N +1
We know from part (i) that N (N − 1) is an even number so N (N − 1) + 1 is an
odd number.
(iv) 8, 001 = 203 + 1.
We also know that N 3 + 1 = (N + 1)(N 2 − N + 1), so factorising we have24
203 + 1 = 21 × (202 − 20 + 1) = 21 × 381
= 3 × 7 × 3 × 127
Answer: 8, 001 = 32 × 7 × 127.
24
There is a rule that says n is divisible by 3 if the sum of n ’s digits is divisible by 3. In this
case 3 + 8 + 1 = 12, which is a divisible by 3.
p
Then to check for factors of 127 we only need to check for factors up to 127, i.e. we can stop
at 11. We can also speed up the checking process by realising that if, for example, 3 is not a
factor, 6 or 9 can’t be factors either. Similarly no even number can be a factor is n is not even.

4.9 The Factor Theorem

If a quadratic has factors we know how to find them—by inspection. In some
special case we can factorise higher power polynomials e.g.
x 3 + x 2 − 6x = x(x 2 + x − 6) = x(x + 3)(x − 2)
x 4 − 16x 2 = x 2 (x 2 − 16) = x 2 (x + 4)(x − 4)
x 4 − 16 = (x 2 + 4)(x 2 − 4) = (x + 2)(x − 2)(x 2 + 4).
What about a polynomial like
f (x) = x 3 − 2x 2 − 5x + 6
where no such technique can be applied?
There is no easy way to find the factors of a polynomial, but—if they
exist—the Factor Theorem will sometimes allow us to find them.
We’ll explain how using the function f (x). With a little bit of multiplying you
will see that
f (x) = (x − 1)(x 2 − x − 6) = x 3 − 2x 2 − 5x + 6.
Writing it this way shows that
f (1) = 0 and x − 1 is a factor.

You may want to put x = 1 into x 3 − 2x 2 − 5x + 6, although we know the
answer must be zero because if x = 1, the bracket (x − 1) is zero.
Since x 2 − x − 6 = (x − 3)(x + 2), we can easily show that f (x) could also be
written
f (x) = (x − 3)(x 2 + x − 2)
and deduce that
f (3) = 0 and x − 3 is a factor.

Similarly
f (−2) = 0 and x + 2 is a factor.
In general, these two statements are equivalent:
(x − a) is a factor of f (x) and f (a) = 0
If one of these statements is true, so is the other one.
We can express this slightly more succinctly, in the single statement

A
(x − a) is a factor of f (x) ⇐⇒ f (a) = 0.
!
A
A
A
The symbol ⇐⇒ we read as “implies and is implied by”, or “if and only if”.
We’ve shown there is a link between the factors of a polynomial function f (x)
and the roots of the equation f (x) = 0. This is the Factor Theorem.

Returning to the function f (x), having one factor of f (x) we can easily
complete the factorisation:
f (x) = x 3 − 2x 2 − 5x + 6
= (x − 1)(x 2 − x − 6)
= (x − 1)(x − 3)(x + 2)
In this example we had a “foot in the door”; we knew one of the factors up
front. Suppose we don’t?
Here is another example:
f (x) = x 3 + 2x 2 − 11x − 12.
We can use the Factor Theorem to hunt for factors. This may seem rather hit
and miss. Of course it is, but in an exam you wouldn’t be expected to look
further than these factors:
(x ± 1) (x ± 2) (x ± 3),
in addition to factors like x, x 2 etc. So let’s see what we can find.

Notice when substituting in values, we don’t need to complete the
evaluation if it’s obvious the answer isn’t going to be zero:
f (1) = 1 + 2 − 11 − 12 6= 0 ⇒ x − 1 is not a factor

f (2) = 8 + 8 − 22 − 12 6= 0 ⇒ x − 2 is not a factor
f (3) = 27 + 18 − 33 − 12 = 0 ⇒ x − 3 is a factor
so
f (x) = (x − 3)(x 2 + 5x + 4)
= (x − 3)(x + 1)(x + 4).
Our strategy was: try x = 1, x = 2, x = 3, x = −1 etc, until we find a factor.

We could have tried: x = ±1, x = ±2, etc:
f (1) = 1 + 2 − 11 − 12 6= 0 ⇒ x − 1 is not a factor
f (−1) = −1 + 2 + 11 − 12 = 0 ⇒ x + 1 is a factor
so
f (x) = (x + 1)(x 2 + x − 12)
= (x + 1)(x + 4)(x − 3).
Obviously we don’t know in advance which strategy will turn out to be
quicker. Since f (−1) is normally quicker to evaluate that f (2), we may
choose to opt for the latter strategy.

QQ
Use the Factor Theorem to show that, for the polynomial

f (x) = x 3 + 6x 2 − 9x − 14
(i) x − 2 is a factor (ii) x + 1 is a factor (iii) x − 1 is not a factor.
Factorise f (x) completely.
25
(i) f (2) = 8 + 24 − 18 − 14 = 0 (ii) f (−1) = −1 + 6 + 9 − 14 = 0
(iii) f (1) = 1 + 6 − 9 − 14 6= 0
(x − 2)(x + 1)(x + 7)
We can divide f (x) by x − 2 or x + 1 and proceed in the usual way. But since we have
two factors, we can identify the third by inspection:
f (x) = x 3 + 6x 2 − 9x − 14 = (x − 2)(x + 1)(ax + b)
The coeff of x 3 is 1, so a must be 1. The constant coeff is −14 so b must be 7.

Finding factors by inspection

We don’t mean spotting that a polynomial has a certain factor by doing the
factorising in our head! That is not going to happen. We spot the factor by an
indirect route.
The Factor Theorem tells us that finding a factor is equivalent to finding a
root. For example:
f (x) = x 4 + 7x − 8.
We may spot that the sum of the coefficients is zero, so f (1) = 1 + 7 − 8 = 0.
x − 1 is a factor.
In the very similar polynomial
g (x) = x 4 − 7x − 8
the coefficients 1, −7 and 8 don’t sum to zero, but if we put x = −1 the terms
do:
g (−1) = (−1)4 + 7(−1) − 8 = 1 + 7 − 8 = 0.
x + 1 is a factor of g (x).
By the same reasoning we can see that, if n is a positive integer, the

polynomial
p(x) = x n − 1
has a factor x − 1.
QQ
1. If x + 1 is a factor of
q(x) = x n + 1, where n is a positive integer
what can we say about n?
2. By inspection, identify a factor of the function

x 3 − x − 6.
26
1. n is odd.
2. x − 2. If we call the function f (x), f (2) = 0

Example 6
Factorise completely these polynomials

(a) f (x) = 2x 3 + x 2 − 13x + 6
(b) g (x) = x 4 + 2x 3 − 16x 2 − 2x + 15.
Solution
(a) Checking values:

f (x) = 2x 3 + x 2 − 13x + 6
f (1) = 2 + 1 − 13 + 6 6= 0 ⇒ x − 1 is not a factor
f (2) = 16 + 4 − 26 + 6 = 0 ⇒ x − 2 is a factor
so
f (x) = (x − 2)(2x 2 + 5x − 3)
= (x − 2)(2x − 1)(x + 3).
We used the Inspection Method to complete the factorisation. In part (b) we’ll
use the Long Division Method.

(b) Checking values:

g (x) = x 4 + 2x 3 − 16x 2 − 2x + 15
g (1) = 1 + 2 − 16 − 2 + 15 =0 ⇒ x − 1 is a factor
so we have
x 3 + 3x 2 − 13x − 15
x −1 x 4 + 2x 3 − 16x 2 − 2x + 15
x4 − x3
+ 3x 3 − 16x 2
+ 3x 3 − 3x 2
− 13x 2 − 2x
− 13x 2 + 13x
− 15x + 15
− 15x + 15
g (x) = (x − 1)(x 3 + 3x 2 − 13x − 15)

= (x − 1)h(x)
Now we need to find a factor of the cubic h(x). That will give us a quadratic
and we can complete the factorisation from there.
We can show, for example, that h(3) = 0, so one factor of h(x) is x − 3, and
dividing:
4x 2 + x − 7
x −3 8x 3 − 10x 2 − 17x + 21
8x 3 − 12x 2
+ 2x 2 − 17x
+ 2x 2 − 3x
− 14x + 21
− 14x + 21
h(x) = x 3 + 3x 2 − 13x − 15
= (x − 3)(x 2 + 6x + 5)
= (x − 3)(x + 1)(x + 5)
so
g (x) = (x − 1)(x − 3)(x + 1)(x + 5)

QQ
Factorise these polynomial functions completely.
(i) f (x) = x 3 − 4x 2 + x + 6
(ii) f (x) = 2x 3 + 3x 2 − 17x − 30
(iii) f (x) = 3x 3 − 19x 2 − 15x + 7
(iv) f (x) = x 4 + 7x 3 + 10x 2
27
(i) (x − 2)(x + 1)(x − 3) (ii) (x − 3)(x + 2)(2x + 5) (iii) (x + 1)(x − 7)(3x − 1)
(iv) x 2 (x + 2)(x + 5)

Factors of the form ax + b

If a polynomial function f (x) has a factor ax + b we must be able to write it
as
f (x) = (ax + b) × another polynomial,
so f (− ba )= 0.
For example we saw (in Example 5) that
f (x) = 2x 3 + x 2 − 13x + 6 = (2x − 1)(x − 2)(x + 3).

1
Since 2x − 1 is a factor of f (x) we should find that f ( 2 )= 0. Let’s check that.
µ ¶ µ ¶3 µ ¶2 µ ¶
1 1 1 1
f = 2 + − 13 +6
2 2 2 2
1 1 13
= + − +6
4 4 2
= 0
In general, the two statements are equivalent:
f − ba = 0
¡ ¢
(ax + b) is a factor of f (x) and
If one of these statements is true, so is the other one.

We can express this slightly more succinctly, in the single statement
f − ba = 0.
¡ ¢
(ax + b) is a factor of f (x) ⇐⇒

Example 7
(a) Show that 2x + 3 is a factor of the function f (x) = 6x 3 − 13x 2 − 41x − 12.
(b) Write f (x) as the product of linear factors.
(c) Solve the equation
6x 3 − 13x 2 − 41x − 12 = 0.
Solution
3
(a) When substituting x =− 2 into f (x) we recommend you break done the
simplifying into modest steps:
µ ¶ µ ¶3 µ ¶2 µ ¶
3 3 3 3
f − = 6 − − 13 − − 41 − − 12
2 2 2 2
µ ¶ µ ¶ µ ¶
27 9 3
= −6 − 13 + 41 − 12
8 4 2
1 1 13
= + − +6
4 4 2
81 117 123
= − − + − 12
4 4 2
−198 + 246 − 48
=
4
= 0
So 2x + 3 is a factor of f (x).
(c)Dividing:
3x 2 − 11x − 4
2x + 3 6x 3 − 13x 2 − 41x − 12
6x 3 + 9x 2
− 22x 2 − 41x
− 22x 2 − 33x
− 8x − 12
− 8x − 12
so f (x) = 6x 3 − 13x 2 − 41x − 12

= (2x + 3)(3x 2 − 11x + 4)
= (2x + 3)(3x + 1)(x − 4).
(c) If you want a quick reminder about solving equations, see Chapter 1 page
1-21
(2x + 3)(3x + 1)(x − 4) = 0

⇒ 2x + 3 = 0 or 3x + 1 = 0 or x −4 = 0
⇒ x = − 32 or x = − 13 or x = 4

Example 8
1
(i) Find the values of a and b if f (1) = f ( 3 )= 0, where
f (x) = ax 3 + 7x 2 + bx + 5.
(ii) For these values of a and b, factorise f (x) completely.
Solution
(i)
f (1) = a + 7 + b + 5 = 0
⇒ a + b = −12
and
µ ¶ µ ¶3 µ ¶2 µ ¶
1 1 1 1
f = a +7 +b +5
3 3 3 3
a 7 b
= + + +5
27 9 3
a + 21 + 9b + 135
= =0
27
a + 21 + 9b + 135 = 0
⇒ a + 9b = −156
We have found two simultaneous equations for a and b:

a + b = −12
a + 9b = −156
Subtracting gives 8b = −144.

The solutions are a = 6, b = −18, so
f (x) = 6x 3 + 7x 2 − 18x + 5.

(ii) The factor theorem tells us that f (x) has two factors x − 1 and 3x − 1 , i.e.
f (x) = (x − 1)(3x − 1) × a linear expression.
We would normally divide to complete the factorisation, and we can do so
here, but since we already have two of the three factors there is a more direct
route.
Suppose the linear expression is c x + d , then
f (x) = (x − 1)(3x − 1)(c x + d )

= 6x 2 + 7x 2 − 18x + 5
We don’t need to multiply out the three brackets but we can see what some
of the terms would be if we did. The coefficient of x 3 would be 3c. But we
know it is 6, so c = 2.
Similarly the constant coefficient would be +d , so d = 5.
f (x) = (x − 1)(3x − 1)(2x + 5).

Example 9
Solve the equation 2x 3 − 9x 2 + 4x + 15 = 0.
Solution
We are not given any clues. (Actually there is a formula for solving cubic
equations just like there is for quadratic equations, but it is complicated so
almost never used.)
This is just one step on from what we did in Example 6. We’ll find a factor of
the expression using the Factor Theorem. That will allow us to factorise it
completely and give us the roots.
We’ll call the function f (x).
f (x) = 2x 3 − 9x 2 + 4x + 15
f (1) = 2 − 9 + 4 + 15 6= 0 ⇒ x − 1 is not a factor
f (2) = 16 − 36 + 8 + 15 6= 0 ⇒ x − 2 is not a factor
f (3) = 54 − 81 + 12 + 15 = 0 ⇒ x − 3 is a factor
so
f (x) = (x − 3)(2x 2 − 3x − 5)
= (x − 3)(x + 1)(2x − 5).
We used the Inspection Method to complete the factorisation. Now
(x − 3)(x + 1)(2x − 5) = 0
⇒ x −3 = 0 or x +1 = 0 or 2x − 5 = 0
5
⇒ x = 3 or x = −1 or x = 2
.

Exercises 17
1. Find, by inspection, a solution to the equation

3x 3 − x 2 − x − 1 = 0.
2. Find, by inspection, a solution to the equation x 4 = x + 14.
3. Find the value of a for which x = 2 is a solution of the equation

x 3 + ax 2 − 7 = 0.
4. The function f (x) = x 3 + 3x 2 + 5x − 9.
(a) Find the value of (i) f (2) (ii) f (0) (iii) f (−1) (iv) f (−3).
(b) By inspection, write down a value of k for which f (k) = 0.
5. Two of the roots of the equation

2x 3 + px 2 + 4x + q = 0.
are x = 1 and x = −2. Find the values of p and q.
6. Find the constants A, B and C in the following identities.

(a) (x + 3)(Ax + B ) = x 2 + 7x + 12
(b) (x + 1)(Ax + B ) = 2x 2 + 5x + 3
(c) (x + 4)(Ax + B ) = 3x 2 +C x + 8
(d) (x − 1)(Ax + B ) = x 2 + 3x +C
(e) (x − 2)(Ax + B ) = 3x 2 − 7x + 2
(f) (2x + 1)(Ax + B ) = 8x 2 +C x + 3
(g) (3x + A)(x + B ) = 3x 2 + B x + 10
(h) (x 2 + 3)(Ax + B ) = 2x 3 − x 2 + 6x +C
(i) (4x 2 + 1)(Ax 2 + B ) = 8x 4 − 2x 2 +C
7. Given that (ax + 2)(x 2 + bx + c) = 3x 3 + 8x 2 + x − 2, write down the
values of a and c. Hence find b.
8. Determine A, B and C in the following identities.
[Find them in any order.]
(a) (x + 2)(Ax 2 + B x +C ) = x 3 + 7x 2 + 12x + 4
(b) (x + 1)(Ax 2 + B x +C ) = 2x 3 + 3x 2 + 5x + 4
(c) (x − 1)(Ax 2 + B x +C ) = 3x 3 + x 2 − 5x + 1
(d) (x + 3)(Ax 2 + B x +C ) = x 3 − 3x 2 − 16x + 6
(e) (2x + 1)(Ax 2 + B x +C ) = 6x 3 + x 2 − 7x − 5
9. In this question use the Long Division Method described in the text.
(a) Divide x 3 + 5x 2 + 7x + 3 by x + 3
(b) Divide x 3 + 10x 2 + 19x + 6 by x + 2
(c) Divide x 3 + 3x 2 − 12x + 4 by x − 2
(d) Divide 2x 3 + x 2 − 7x + 4 by x − 1
(e) Divide 3x 3 + 8x 2 + 19x + 10 by 3x + 2
(f) Divide 2x 3 − 15x − 9 by x − 3
(g) Divide x 3 − 11x + 20 by x + 4

(h) Divide x 4 + 2x 3 + 4x − 40 by x − 2
(i) Divide 4x 4 − x 2 + 6x + 3 by 2x + 1
10. Find the values of a, b and c if

(ax + 3)(x 3 + bx 2 + x + c) = 4x 4 − 5x 3 − 2x 2 + 14x + 9.
11. By factorising simplify these expressions

x 2 +5x+6 x 3 +x 2 −6x x 2 +x−2
(a) (b) (c)
x+2 x+3 x 2 +7x+10
4x 2 −64 (x+3)2 x 2 −1
(d) (e) 2 (f )
x−4 x −9 x 2 −x−2
12. Use any method to simplify these expressions
x 3 +9x 2 +15x+2 x 4 −x 3 −22x+12 x 3 −11x+20
(i) (ii) (iii)
x+2 x−3 x+4
4x 4 −x 2 +6x+3 2x 3 −15x−9 x 3 −8
(iv) (v) (vi)
2x+1 x−3 x−2
13. If x − 2 is a factor of x 3 + 3x 2 + ax − 12 find a.
14. Find a and b if x − 1 and x + 3 are factors of 2x 3 + ax 2 + bx − 3.
15. x + 2 and x − 4 are factors of f (x) = 3x 3 + px 2 − 22x + q. Find the values

of p and q, and hence factorise f (x) completely.
16. Use the Factor Theorem to factorise completely each of these

polynomials:
(a) x 3 + 6x 2 + 5x (b) x 3 + 6x 2 + 5x − 12
(c) 2x 3 + 3x 2 − 23x − 12 (d) 2x 3 − 18x
(e) 6x 3 + 13x 2 + 2x − 5 (f ) 8x 3 − 42x 2 + 67x − 30
(g) x 3 − 3x 3 + 3x − 9 (h) x 3 − 1
(i) 1 + x + x 2 + x 3 (j) 3x 6 − 48x 2 .
17. Show that 3x − 2 is a factor of the function
3x 3 + 10x 2 − 44x + 24.

Express the function as a product of linear factors.
18. Find the three roots of the equation28
x 3 − 8x 2 + 19x − 14 = 0.
p
Give the two non-integer roots in the form p ± q, where p and q are
integers.
19. Show that x + 1 and x − 1 are both factors of the polynomial
x 6 + 2x 4 − x 2 − 2.
Hence or otherwise factorise the polynomial completely.
20. Show that x 3 + 4x 2 + 4x + 3 = (x 2 + x + 1)(x + 3).

By selecting a suitable value for x, find the prime factorisation of 1, 443.
28
If you want a reminder about solving quadratic equation using the formula see Chapter
1 page 1-21.

21. Show that x − 1 is a factor of x 4 − 1

Without using a calculator show that 63 is a factor of 224 − 1 (i.e.
16, 777, 215).
22. If m is a positive integer, show that x 2 − 1 is a factor of x 2m − 1.
23. The functions f (x) and g (x) are defined by
f (x) = x 2 + x − 2 and g (x) = x 2 − 4x + 3.

In each part of this question look for a quick method.
(a) Find the coefficient of x 3 in the polynomial f (x)g (x).
(b) Show that (x − 3) f (x) = (x + 2)g (x).
f (13) 3
(c) Hence, or otherwise, show that = .
g (13) 2
24. The functions p(x) and q(x) are defined by
p(x) = x 2 − x and q(x) = x 2 + x.

(a) Find the values of p(1)q(1), q p(1) and pq(1).
(b) Show that p(x)q(x) = x 2 (x 2 − 1).
(c) Find in its simplest form the polynomial [q(x)]2 − [p(x)]2 .
p(x) 2
(d) Show that = 1−
q(x) x+1
1
1+ x−1 x+1
25. Show that the expression 1 simplifies to x−1
1− x+1
26. Show if k is an odd integer, N k + 1 is divisible by N + 1.
27. Use the Factor Theorem to factorise x 3 − x −6. Hence, or otherwise find
all the roots of the equation x 6 − x 2 − 6 = 0.

Solutions 17
1. x =1
2. x =2
1
3. a = − 4 [8 + 4a − 7 = 0]
4. (a) (i) 21 (ii) −9 (iii) −12 (iv) −24

(b) a = 1
5. p = 10, q = −16. [p + q = −6, 4p + q = 24]
6. (a) A = 1, B = 4 (b) A = 2, B = 3 (c) A = 3, B = 2, C = 14

(d) A = 1, B = 4, C = −4 (e) A = 3, B = −1 (f ) A = 4, B = 3, C = 10
(g) A = 2, B = 17 (h) A = 2, B = −1, C = −3 (i) A = 2, B = −1, C = −1
7. a = 3, c = −1, b = 2 [a and b by inspection from x 3 and constant coefficients]
8. It is easiest to find A and C first, and use these to find B .

(a) A = 1, B = 5, C = 2 (b) A = 2, B = 1, C = 4 (c) A = 3, B = 4, C = −1
(d) A = 1, B = −6, C = 2 (e) A = 3, B = −1, C = −5
9. (a) x 2 + 2x + 1 (b) x 2 + 8x + 3 (c) x 2 + 5x − 2

(d) 2x 2 + 3x − 4 (e) x 2 + 2x + 5 (f) 2x 2 + 6x + 3
(g) x 2 − 4x + 5 (h) x 3 + 4x 2 + 8x + 20 (i) 2x 3 − x 2 + 3
10. a = 4, b = −2, c = 3. [a and c by inspection, hence b]

£ (x+2)(x+3) ¤ £ x(x+3)(x−2) ¤
11. (a) x + 3 x+2
(b) x(x − 2) x+3
h i
(x−1) (x−1)(x+2)
£ 4(x+4)(x−4) ¤
(c) (x+5) (x+5)(x+2)
(d) 4(x + 4) x−4
h i h i
(x+3) (x+3)3 (x−1) (x+1)(x−1)
(e) (x−3) (x+3)(x−3)
(f ) (x−2) (x+1)(x−2)
12. (i) x 2 + 7x + 1 (ii) x 3 + 2x 2 + 6x − 4 (iii) x 2 − 4x + 5

(iv) 2x 3 − x 2 + 3 (v) 2x 2 + 6x + 3 (vi) x 2 + 2x + 4
13. a = −4 [Let cubic be f (x) then f (2) = 8 + 12 + 2a − 12 = 0]
14. a = 5, b = −4 [Let cubic be f (x) then f (1) = 2 + a + b − 3 = 0 and

−54 + 9a − 3b − 3 = 0 ⇒ a + b = 1 and 3a − b = 19]
15. p = −7, q = 8, (x + 2)(x − 4)(3x − 1) [ f (−2) = f (4) = 0

⇒ 4p + q = −20 and 16p + q = −104]
16. (a) x(x + 1)(x + 5) (b) (x − 1)(x + 3)(x + 4)

(c) (x − 3)(x + 4)(2x + 1) (d) 2x(x + 3)(x − 3)
(e) (x + 1)(3x + 5)(2x − 1) (f ) (x − 2)(2x − 5)(4x − 3)
(g) (x − 3)(x 2 + 3) (h) (x − 1)(x 2 + x + 1)
(i) (1 + x)(1 + x 2 )
17. f ( 23 ) = 89 + 40
9
− 88
3
+ 24 = 48−264+216
9
=0
(3x − 2)(x − 2)(x + 6)

p p
18. x = 2, 3 + 2 and 3 − 2 [Equation is (x − 2)(x 2 − 6x + 7) = 0]
19. All powers of x are even so let x 2 = X giving f (X ) = X 3 − 2X 2 − X − 2.

f (1) = 0 so X − 1 divides (exactly) into f (X )
⇒ x 2 − 1 = (x + 1)(x − 1) divides into f (x)
⇒ x + 1 and x − 1 both divide into f (x)
Dividing gives f (X ) = (X − 1)(X 2 + 3X + 2) = (X − 1)(X + 1)(X + 2)
so complete factorisation is (x + 1)(x − 1)(x 2 + 1)(x 2 + 2).
Alternatively, we could divide f (x) by (say) x + 1 to give x 5 − x 4 + 3x 3 − 3x 2 + 2x − 2

Writing the quintic as x 4 (x − 1) + 3x 2 (x − 1) + 2(x − 1) = (x − 1)(x 4 + 3x 2 + 2)
we have the complete factorisation (x + 1)(x − 1)(x 2 + 1)(x 2 + 2)
20. Multiply out the right hand side.

Let x = 10, then 1443 = 111 × 13 = 3 × 37 × 13
21. If f (x) = x 4 − 1, f (1) = 0 so x − 1 is a factor.

Let x = 64 = 26 , then 63 is a factor of (26 )4 − 1 = 224 − 1.
22. Let x 2 = X . Factor Theorem tells us that X − 1 is a factor of X m − 1

so x 2 − 1 is a factor of (x 2 )m − 1 = x 2m − 1.
23. −3 [Coefft of x 3 is −4 + 1]
f (x) = (x + 2)(x − 1), g (x) = (x − 3)(x − 1) [Each side is (x + 2)(x − 1)(x − 3)]
In part (b) let x = 13 ⇒ 10 f (13) = 15g (15).
24. (a) 0, 0, 2 [p(1)q(1) = 0 × 2 = 0, q p(1) = q(p(1)) = q(0) = 0, pq(1) = p(q(1)) = p(2) = 2]

(b) p(x)q(x) = x(x − 1)x(x + 1) = x 2 (x 2 − 1)
(c) 4x 3 [By difference of two squares (q(x) + p(x))(q(x) − p(x)) = 2x 2 .2x]
p(x) x−1 x+1−2 x+1 2 2

(d) = = = − = 1−
q(x) x+1 x+1 x+1 x+1 x+1
A slightly easier way would be to start on the right hand side:
2 x+1 2 x+1−2 x−1
1− = − = =
x+1 x+1 x+1 x+1 x+1
x x
25. Numerator simplifies to x−1 and the denominator to x+1 .
The result follows from the hammer and boot principles.
26. If f (N ) = N k + 1, f (−1) = (−1)k + 1 = 0, so N + 1 is a factor.
27. If f (x) = x 3 − x − 6, f (2) = 0. The function factorises to (x − 2)(x 2 + 2x + 3).

By substitution or otherwise x 6 − x 2 − 6 = (x 2 + 2)(x 4 + 2x 2 + 3).
x 2 + 2 = 0 has no roots and the discriminant of x 4 + 2x 2 + 3 < 0. So the
equation has no roots.

Proof of the Factor Theorem

Suppose x − a is a factor of the function f (x), then we can write
f (x) = (x − a)g (x)
where g (x) is another function (which we could find by dividing f (x) by
x − a).
Substituting x = a into both sides
f (a) = 0 × g (a) = 0.
More generally, if the factor is ax + b, we can write
f (x) = (ax + b)h(x)
where h(x) is another function (which we could find by dividing f (x) by
ax + b).
b
Substituting x =− a into both sides
b b
µ ¶ µ ¶
f − = 0 × h − = 0.
a a
Strictly speaking we haven’t quite finished. We’ve only shown the theorems
hold in one direction. We haven’t shown, for example, that if f (a) = 0, x − a
must be a factor of f (x).
To fill in this last piece of the jigsaw we can use proof by contradiction.
We shall assume for a moment that f (a) = 0 but x − a is not a factor of f (x).
If x − a is not factor we get a (non-zero) remainder when we divide f (x) by
x − a, i.e.
f (x) = (x − a)g (x) + r where r 6= 0.
But that means f (a) 6= 0, which we know to be false.

Exercises 18
Some of these questions involve solving equations. If you need a reminder

about this see Chapter 1 page 1-21).
1. For the function f (x) = 2x 3 − x 2 − 6x + 3, find
(i) f (0) (ii) f (1) (iii) f (−1) (iv) f (2) (v) f (−3)
2. Use the Remainder Theorem to find the remainder when

(a) x 3 + x 2 + 7x − 1 is divided by x − 1
(b) x 3 + x 2 + 7x − 1 is divided by x + 1
(c) x 3 + 2x 2 − 4x + 9 is divided by x − 2
(d) x 3 − 3x 2 − 6x + 8 is divided by x − 3
(e) 4x 3 − 5x − 2 is divided by 2x + 1
(f) 8x 3 − 2x 2 + 3 is divided by 2x − 3
3. Find the value of a if f (x) = x 3 + ax 2 − 8x + 6 has remainder 10 when

divided by x − 2.
4. Find the value of b if f (x) = x 3 + 3x 2 + bx − 2 has remainder 14 when

divided by x + 2.
5. Find the value of c if f (x) = c x 3 + x 2 + 5x + 3c is divisible by x + 1.
6. When the expression x 2 + ax + b is divided by x − 1 the remainder is 5.

When the same expression is divided by x + 1 the remainder is 11. Find
the values of a and b.
7. The remainder when x 2 + px − 2 is divided by x − 2 is four times the

remainder when the same expression is divided by x −1. Find the value
of p.
8. The function f (x) = 2x 2 + ax + b has a factor x − 4, and when f (x) is

divided by x −2 the remainder is −6. Find the values of a and b, and the
other factor of f (x).
9. x − 2 is a factor of x 3 + ax + 2a. Find the value of a, and the remainder

when the same expression is divided by x + 2.
10. Given that x 2 − 1 is a factor of x 3 + ax 2 + bx − 3, find the values of a and

b.
11. The function f (x) = 4x 3 + ax 2 + bx − 1 has a factor 2x − 1. When f (x) is

divided by x + 1 the remainder is 3.
(i) FInd two equations in a and b.
(ii) Solve these to find a and b.
12. When f (x) = 4x 3 + ax 2 + x + b is divided by x − 1 the remainder is 10.

When f (x) is divided by x + 2 the remainder is −5.
(i) Find two equations in a and b.
(iii) Hence show that f 21 = 0.
¡ ¢
(iv) Use this result to find all the linear factors f (x).

13. When f (x) = 3x 3 + ax 2 + bx + 2 is divided by x − 2 the remainder is 20.

When f (x) is divided by x + 1 the remainder is 8.
(i) Find two equations in a and b.
(iii) Hence show that f 31 = 0.
¡ ¢
(iv) Use this result to solve the equation f (x) = 0.
14. (i) Find the linear factors of the function f (x) = x 3 − 3x 2 − x + 3.

(ii) Hence solve the equation f (x) = 0.
(iii) Write the function in the form
f (x) = (x − 2)g (x) + r
where the constant r and the function g (x) are to be determined.
15. Express 6x 3 + 5x 2 − 3x − 2 in the form (x + 1)(ax 2 + bx + c), stating the

values of a, b and c.
Hence solve the equation 6x 3 + 5x 2 − 3x − 2 = 0
16. (a) Show that

x 2 − 2x − 8 x − 4
= .
x2 − 4 x −2
(b) Factorise the expression f (x) = x 3 − 3x 2 − 6x + 8.
(c) Hence, or otherwise, simplify
x 3 − 3x 2 − 6x + 8
.
x2 − 4
17. (i) Show that 3x − 1 is a factor of f (x) = 3x 3 − 4x 2 − 5x + 2.

(ii) Hence, or otherwise, simplify the fraction
3x 3 − 4x 2 − 5x + 2
.
3x 2 + 2x − 1
18. (a) Show that 2x − 1 is a factor of f (x) = 2x 3 − 5x 2 − 6x + 4.

equation f (x) = 0, giving the two non-integer solutions in
(b) Solve the p
the form a ± b, where a and b are integers.
19. (i) Show that x = 1 is a solution of the equation

x2 1
+ = 1.
x + 1 3x − 1
(ii) Write the equation in the form of a polynomial equation.
(iii) By factorising, solve the equation.
20. When f (x) is divided by x − 3 the remainder is 4. Write down the value
of the constant k if
f (x) k
= x2 + 2 + .
x −3 x −3
Find the values of a, b and c if
f (x) = x 3 + ax 2 + bx 2 + c.

21. Write down a solution of the equation

(x + 1)(3 − x) = 2x 2 (x + 1).
Find the other solutions of the equation.
22. (i) Write down one root of the equation
x 3 = 8 − 9x 2 .
p
(ii) Find the other two solutions, expressing them in the form a ± b 6,
where a and b are integers.
23. (a) Show that when x 3 − 5x 2 − 4x + 18 is divided by x − 2 the remainder

is −2.
(b) Hence, or otherwise, write the expression
x 3 − 5x 2 − 4x + 20
as a product of linear factors.
24. (i) Show that x − a is a factor of the function

f (x) = x 2 + 2ax − 3a 2 ,
where a is a constant.
(ii) Write down the value of k for which
f (x) = (x − a)(x + ka).
(iii) Hence find, in terms of a, the roots of the equation f (x) = 0.
25. (a) Show that x − k is a factor of the function

f (x) = x 3 + (3 − k)x 2 − 2kx − k 2 ,
where k is a constant.
(b) Write f (x) in the form
f (x) = (x − k)g (x),
where g (x) is a quadratic whose coefficients may include k.
(c) Find the remainder when f (x) is divided by x − (k + 1).

Solutions 18
1. (i) 3 (ii) −2 (iii) 6 (iv) 3 (v) −42

51
2. (a) 8 (b) −8 (c) 17 (d) −10 (e) 0 (f ) 2
3. a =3
4. b = −6
5. c = 2 [ f (−1) = −c + 1 − 5 + 3c = 0]
6. a = −3, b = 7 [1 + a + b = 5, 1 − a + b = 11]
7. p =3 [ f (2) = 4 f (1) ⇒ 2 + 2p = 4(−1 + p)]
8. a = −9, b = 4. 2x − 1 [ f (4) = 32 + 4a + b = 0, f (2) = 8 + 2a + b = −6]
9. a = −2. Rem −8 [ f (2) = 8 + 2a + 2a = 0, f (−2) = −8 + 4 − 4]
10. a = 3, b = −1 [x − 1 and x + 1 are factors. ⇒ a + b = 2 and a − b = 4]
11. (i) a + 2b = 2 (ii) a = 6, b = −2 [ 48 + a4 + b2 − 1 = 0, −4 + a − b − 1 = 3]
a−b =8
12. (i) a + b = 5 (ii) a = 8, b = −3 [ f (1) = 10, f (−2) = −5]
4a + b = 29
(iii) f 12 = 48 + 2 + 12 − 3 = 0
¡ ¢
(iv) 2x − 1, 2x + 3 and x + 1. [ f (x) = 4x 3 + 8x 2 + x − 3 = (2x − 1)(2x 2 + 5x + 3)]
13. (i) 2a + b = −3 (ii) a = 2, b = −7 [ f (2) = 20, f (−1) = 8]

a−b =9
(iii) f 13 = 19 + 92 − 73 + 2 = 0
¡ ¢
(iv) x = 31 , 1 or −2. [ f (x) = (3x − 1)(x 2 + x − 2) = (3x − 1)(x − 1)(x + 2)]

14. (i) x − 1, x + 1 and x − 3 [ f (1) = 0 ⇒ (x − 1) is a factor. f (x) = (x − 1)(x 2 − 2x − 3)]
(ii) x = 1, −1 or 3 [x − 1 = 0, x + 1 = 0 or x − 3 = 0]
(iii) r = −3, g (x) = x 2 − x − 3 [ f (2) = −3, so f (x) + 3 = x 3 − 3x 2 − x + 6 = (x − 2)g (x)]
15. a = 6, b = −1 and c = −2. [By inspection a = 6 and c = −2 hence b]

2
x = −1, 3
or − 12 . 2
[6x − x − 2 = (3x − 2)(2x + 1)]
(x+2)(x−4)
16. (a) Expression is (x+2)(x−2)
(b) (x − 1)(x + 2)(x − 4) [ f (1) = 0 ⇒ x − 1 is a factor. Then divide and factorise.]
(x−1)(x−4)
(c) (x−2)
[Numerator is (x − 1)(x 2 − 2x − 8)]
17. f ( 13 ) = 0 [ 91 − 94 − 53 + 2]
(3x−1)(x+1)(x−2)
x −2 [Numerator is (3x − 1)(x 2 − x − 2). Factorising: (3x−1)(x+1)
]
18. f ( 21 ) = 0 [ 82 − 54 − 26 + 4]
1
p p
2 , 1 + 5, 1 − 5 [ f (x) = (2x − 1)(x 2 − 2x − 4) = (2x − 1)((x − 1)2 − 5)]
1
19. (i) Substituting x = 1, left hand side is = 2 + 12 .
(ii) x 3 − 3x 2 + 4x + 5 = 0 [x 2 (3x − 1) + (x + 1) = (x + 1)(3x − 1)]
(iii) x = 1 (repeated), x = − 23 [(x − 1) is a factor, hence (x − 1)(3x + 2)(x − 1)]
20. k =4
a = −3, b = 2, c = −2 [ f (x) = (x − 3)(x 2 + 2) + 4]

21. x =1 [It satisfies the equation.]

x = −1, x = − 32 [(x + 1)(2x 2 − (3 − x)) = 0, so 2x 2 + x − 3 = (x + 1)(2x + 3) = 0]
22. (i) x = −1
p
(ii) −4 ± 2 6 [x + 1 is a factor of x 3 + 9x 2 − 8 = (x + 1)(x 2 + 8x − 8)]
23. (a) Call the function f (x), then f (2) = −2.

(b) (x − 2)(x + 2)(x − 5) [ f (x) = (x − 2)g (x) − 2, so f (x) + 2 has a factor x − 2]
24. (i) f (a) = a 2 + 2a 2 − 3a 2 = 0

(ii) k = 3 [The constant terms must be −3a 2 ]
(iii) x = a, x = −3a [ f (x) = (x − a)(x + 3a) ⇒ x − a = 0 or x + 3a = 0]
25. (a) f (k) = k 3 + (3 − k)k 2 − 2k 2 − k 2 = 0

(b) By inspection f (x) = (x − k)(x 2 + 3x + k).
(c) k 2 + 6k + 4 [From (b), f (k + 1) = (k + 1 − k)(k 2 + 2k + 1 + 3k + 3 + k)]

Revision Exercise 1
Because fluency with algebraic fractions is so important, and it has been a

long chapter, here are some revision questions.
In each case write the expression as a single fraction, simplifying where
possible.
5 1 9 7 5x 2 6y 3 3a+6b 10a
1. (a) 7 × 3 × 10 × 11 (b) 2y × x
(c) 5a
× a+2b
y 2 −4y 3y+3 5x 2
(d) y+1 × 2y−8 (e) 11× 11 (f) a× a
3
5 3y
(g) x 2 × 4x 2 (h) 4x ×2x 2 (i) 27
7 3
4 9 16
(j) 3 (k) 2 (l) 9
5 5 4
5
2y 3 6a
(m) 3x (n) y 4 (o) 4
a
1 x 13x 5 1
(p) x (q) 7 + 7 (r) 9 y − 9 y
y
1 7a 2 a+2 a−1
(s) 2 a − 2 (t) y + 5 y (u) 4 + 3
3−2a
(v) 23 (a + 1) + 61 (2a − 1) (w) 6p − 21 (4p + 3) (x) a − 11
4p 2 +8pq t −4
2. (a) p+2q
(b) t 2 −3t −4
x x 2 −2x+1 x 2 +1
(c) x−1 × 2x 2
(d) x 2 −1
x+2 x 2 +x 1
(e) x+1 × x 2 +6x+8 (f) x 3 +x ×(3x 2 + 3)
(g)
4a−16
8a
× a(a+1)
a 2 −16
2
(h) (y − 1)× y 2 −2y+1
1+ a x
(i) 3 2 (j) 1
3 x−1
7 2y− 21
(k) a 5a (l)
4+ 8 y+ 43
x 2 + 13 x
(m) 2+6x (n) 4x + 12 (x − 1) − 21 (3x + 1)
y(y+2) 1 2
(o) 5
+ 10 y (p) 32 (t + 4) − 61 (t + 4) + 13 (t − 1)
3 4 y
(q) x + 5 (r) 4x − 3x
y
1 2 2y 3
(s) x+1 − x (t) y−1 − y+1
2x−1 3x
(u) x
− 2x+1

Solutions to Revision Exercise 1

3 3y
1. (a) 22 (b) 15x y 2 (c) 6 (d) 2 (e) 5x
5 3x y 3 20
(f) 2 (g) 4 (h) 2 (i) 14 (j) 3
35 1 5 2 3a 2
(k) 18 (l) 12 (m) 3x (n) y (o) 2
y 4 7y
(p) x (q) 2x (r) 9 y (s) −3a (t) 5
7a+2 2a+1 8p−3 13a−3
(u) 12 (v) 2
(w) 2
(x) 11
1 x−1
2. (a) 4p (b) t +1 (c) 2x
x 3
(d) No simplification. (e) x+4 (f) x
a+1 a+1 2 2+a
(g) 2a+8 = 2(a+4) (h) y−1 (i) 6
3x 8 8y−2 2(4y−1)
(j) x−3 (k) a (l) 4y+3 = 4y+3
x 3y 2 +4y y(3y+4)
(m) 6 (n) 3x − 1 (o) 10
= 10
5x+10 5(x+2) 15−4x y 2 −12x 2
(p) 6
= 6 (q) 5x
(r) 4x y
(x+2) 2y−3 x 2 −1
(s) − x(x+1) (t) y−1 (u) x(2x+1)


2 2a 1 x+1
• Important notation issue e.g. 3 a = 3 , and 2 (x + 1) = 2
3a 2 b a2 18y 3 3y 2
• Cancelling algebraic fractions e.g. 9b 3 = 3b 2 , 6x y−9y 2 = x−3y , where
we can cancel every term in numerator and denominator, or we may
x 2 +5x+6 x+2 x−1 1
need to factorise e.g. 4x+12 = 4 , x 2 −1 = x+1 .
• Multiplying algebraic fractions, making sure we understand what is in

a numerator and what in a denominator
x y−1
1 1
e.g. 91 x × 34 = 12 , y 2 −yy
×
= y2
2 6 3x
with a reminder that 3× 3 6= 9 , and that 8× 2 = 81 × 3x
2
= 12x.
• Dividing algebraic fractions: the “hammer and boot" rule

3 x2
x 3 y y x2 4 4x
e.g. 2 = 2x , 4 = 4 , = 2x+3 ,
2+ x3
y2
• Adding/subtracting algebraic fractions

4 3a 8+3a 2 2 2 x(x−1) x(5x−1) x−3 4 (x+3)(x−1)
e.g. a + 2 = 2a
, 3x + 6 = 6
, x + x+1 = x(x+1) .
• Dividing polynomial by a linear factor, by factorising

x 2 +2x−8 (x+4)(x−2)
e.g. x+4
= x+4
= x +2
or using algebraic long division, with or without remainder
3x 3 +11x 2 −7x−29
e.g. x+2
• Factor Theorem, finding an initial factor by inspection

e.g. f (x) = x 3 + 6x 2 − 9x − 14. f (−1) = 0 so x + 1 is a factor.
Then divide and factorise if possible:
f (x) = (x + 1)(x 2 + 5x − 14) = (x + 1)(x + 2)(x − 7)
Factorising with a given factor e.g. 2x − 3, so that f ( 32 ) = 0

Chapter 5
COORDINATE GEOMETRY
Overview
5.1 Introduction. Representing algebraic relationships pictorially.
5.2 Formulae for the equation of a straight line, given

(a) gradient m and y-intercept c: y = mx + c
(b) gradient and one point (x 1 , y 1 ): y − y 1 = m(x − x 1 )

y−y 1 y 2 −y 1
(c) two points on the line (x 1 , y 1 ) and (x 2 , y 2 ): =
x−x 1 x 2 −x 1
x y
(d) x-intercept a and y-intercept b: + = 1
a b
Perpendicular lines: if their gradients are m 1 and m 2 , then m 1 m 2 = −1.
5.3 A head-scratching puzzle. (Optional.)

5.4 Distances between points.
Distance between (x 1 , y 1 ) and (x 2 , y 2 ):
d 2 = (x 1 − x 2 )2 + (y 1 − y 2 )2
5.5 Formula for the mid-point of the line between (x 1 , y 1 ) and (x 2 , y 2 ):

µ ¶
1 1
(x 1 + x 2 ), (y 1 + y 2 )
2 2
Lines within triangles: medians, altitudes, perpendicular bisectors.
And the points where they meet: centroid, orthocentre, circumcentre.
5.6 Circles. Theorems and properties. (The equation of a circle is covered in
Chapter 7, Section 12.)
5-1
CHAPTER 5. COORDINATE GEOMETRY
5.1 Introduction
Mathematics in real life often boils down to understanding the relationships
between quantities we can measure. To take a trivial example, suppose two
physical measurements can each vary but the two must always add up to 3.
If one distance is x and the other y, we can summarise this information by
writing the beautifully concise statement:
x + y = 3, where x ≥ 0 and y ≥ 0.
We need algebra. But algebra lives in the abstract world, a world that the
human brain is ill-equipped to deal with. It’s hardly surprising therefore that
when René Descartes1 cooked up a simple way to express algebraic
relationships pictorially—what we now refer to as coordinate geometry, or
Cartesian geometry—mathematics took huge leaps forward and developed
in whole new areas.
This is not new to you; every time you draw a graph with axes, and plot
points on it you are using coordinate geometry. We’ll just be going deeper.
The equation x + y = 3, is a linear relationship, because its equation takes

the form
ax + b y + c = 0
where a and b are constants (in this case a = b = 1 and c = −3). These are also
linear relationships:
1 y −1
2x − 5y = 7, y = 1 − x, x = 0.3y, = 2,
2 x
because they can all be written in the form ax + b y + c = 0.
Examples of equations that can’t be written in this form, i.e. non-linear
relationships, would be:
sin x
y 2 = x + 1, x y = 3, y= p
x
1
Descartes was a French philosopher born at the very end of the 16th century.

If a relationship is linear its Cartesian representation will be a straight line,

hence the name.
Examples of points on the line

x+y =3
line would be (3,0), (0,3), (1,2), (4,-1).
Since x and y can take any values that sum to 3, the corresponding line
stretches to infinity in both directions. We can only ever draw a small part of
it, of course.
y y
6 6
5 5
@ @
@ 4 @ 4
@ @
3
@ 3 • (0, 3)
@
x
@ @
+
2 @ 2 @• (1, 2)
y
=
@ @
3
1 @ 1 @
@ @
@ - x @• (3, 0) - x
−2 −1 0 1 2 3@ 4 5 −2 −1 0 1 2 3@ 4 5
−1 @ −1 @• (4, −1)
@ @
y

6
Another example of a straight 7
y = 2x + 1

line would be 6

y = 2x + 1 5 •(2, 5)

Here the relationship is that 4 3
•( , 4)
every y value is 1 more than 2

3
twice the corresponding x
value. 2

The graph opposite shows 1 (0, 1)

the line that represents this - x
equation. Check for yourself −2 −1 0 1 2 3 4 5
−1
that the points labelled satisfy
this relationship. (− 32 , −2) • −2

−3

If the equation happens to include only one variable, its line will be parallel
to one of the coordinate axes (see the graph below).
On the blue vertical line all the points have an x-coordinate of 3, so its
equation is x = 3. Similarly the red horizontal line has equation y = 4. The
two lines cross where x = 3 and y = 4, i.e. at (3, 4).
y
6
8 x =0 x =3
7
6
y =4
5
(1, 4) (3, 4)
? 4 • •
3
2
y =0 (3, 1)
1 •
? - x
−2 −1 0 1 KAA2 3 4 5
A
A
I
@
@ A
@ A
@ A
@ A
@ A y = 0 is the x-axis
@
@ x = 0 is the y-axis
In this chapter we confine ourselves to linear relationships. In future chapters

we learn a number of powerful techniques for exploring non-linear ones.

Coordinate geometry in three dimension

Before we embark on the meat of the chapter, relationships could contain
more than two variables: there could be 3, 4 or indeed any number. If an
equation in two variables needs a 2-dimensional graph to represent it, one in
3 variables needs 3 dimensions etc., so we will briefly see how we might
represent graphically an equation in three variables if you wish.
We are going to consider the equation x + y + z = 3.

A physical 3-dimensional model is the best way to get your head round this.
Since this is rarely an option, we shall need to become comfortable with
using a 2-D picture to represent 3 dimensions.
Try to “see" it like this: think of the x-axis as coming out of the paper, with
the y- and z-axes lying flat on the paper. Or if you find it easier, think of the
x- and y-axes lying flat on the paper and the z-axis rising up out of it. It
doesn’t matter how you do it as long as you can find a way of visualising that
3rd dimension in some way.
The three points: (3, 0, 0), (0, 3, 0) and (0, 0, 3) all satisfy the equation. We have
marked them off in the diagram and joined each pair with a line. Imagine
the three lines being made of rigid wire. Now think of resting an infinitely
large (and of course infinitely thin) sheet of card against them. The card
would be pierced by all three axes. We call this a plane, and the coordinates
of every point in the plane satisfy the equation x + y + z = 3.
Some other points lying in the plane (and whose coordinates therefore
satisfy the relationship) would be (0, 2, 1), (1.5, 0, 1.5) and (1, 1, 1). Can you
visualise the positions of these points in 3-dimensional space on the
diagram? The first point has x-coordinate equal to zero so it is no distance in
the x direction. That means it lies in the y-z plane (a vertical sheet of
cardboard containing the y-axis and the z-axis). This plane has equation
x = 0, because every point in this plane has an x-coordinate equal to 0.
z
6
@
@
@
@
@ (0, 0, 3)
@
@
@
@
@
(0,3,0) -

@
@ y
@

@
@

(3, 0, 0)

x

In later modules you’ll be studying coordinate geometry in three

dimensions—and if you go on to do Maths at degree level in n-dimensions.
But for now we are going to stick to equations in two variables, represented
by straight lines in a 2-dimensional plane—more formally known as a
2-dimensional Cartesian space.

5.2 Linear equations and straight lines

So what’s the link between the equation and the line—between the algebra
and the geometry. If we have a line how can we work out what its equation is,
and vice versa? The answer is the subject of this section. What we shall see is
that
there are two key features of any straight line:
gradient and y-intercept.
That is to say how steep the line is and where it crosses the y-axis. If we know
these, we can write down what its equation is; conversely, its equation will tell
us what the line looks like.
5.2.1 Gradient
You will have met the notion of gradient before, but it’s worth a refresher
because it is so central to the study of both linear and non-linear2
relationships.
We’ll start with the graph of y = 2x + 1. To find its gradient we draw a right
angled triangle using any two points on the line (see the diagram below). We
imagine moving along the line, and see how far this takes us in the x and y
directions:
the gradient is how far up we go for every unit we move from left to right.
y Imagine moving from A to B .

6 To get there we need to move 1
8
unit to the right (we always go
7 •
C from left to right).
6
+1
6
Then we need to move up 2 units

2x
5 to arrive at B . This makes the

y=

gradient 2.
4
We can choose any two point

3 B
• -
6 on the line, but it’s obviously
2 easiest to use points that have

1 -•
integer coordinates. We could, for
A example, go from B to C , which
- x
−1 0 1 2 3 4 5 involves moving 2 to the right and
−1 4 up; for every unit we have moved

from left to right we have moved 2
up.
Try it for yourself using the journey from A to C.
2
It was the middle of the 17th Century when in England Isaac Newton, and Gottfried
Leibniz working independently in Germany, made the breakthrough that allowed
them to calculate the gradient at any point on a curve, and thus the field of Calculus
was born.

Put another way if we divide the distance moved “up" (the y-axis) by the
distance moved to the “right" (the x-axis) we get the gradient—which leads
us to a simple formula.
For the line between (x 1 , y 1 ) and (x 2 , y 2 ):

•

increase in y-coordinate
(x 2 , y 2 )
gradient =
increase in x-coordinate

y2 − y1
=
x2 − x1

•
(x 1 , y 1 )


Let’s see how this works for some pairs of points, starting with the line y =
2x + 1, for which we know the gradient to be 2.
y y
6 6
8 8

7 • 7
•
(3, 7) (5, 7)

6 6

5 5

4 4
•
(1, 4)

3 • 3
(1, 3)

2 2

1 1

- x - x
−1 0 1 2 3 4 5 −1 0 1 2 3 4 5
−1 −1

y 2 −y 1 7−3 4 y 2 −y 1 7−4 3
grad = = = = 2 grad = = =
x 2 −x 1 3−1 2 x 2 −x 1 5−1 4
y y
A
A
6 A• 7 6
6
(−1, 7) A
5 A6
A
4 • 5A
(3, 4) A
3 4 A
A
2 3 A
A
1 2 A
A
- x 1 A
−1 0 1 2 3 4 5 A
−1 A -
−2 −1 0 1 2 A3 4
−2 • −1 A• x
(1, −2) (3, −1) A
−3 −2 A
A

y 2 −y 1 4−− 2 6 y 2 −y 1 −1−7 8
grad = = = = 3 grad = = = = −2
x 2 −x 1 3−1 2 x 2 −x 1 3−− 1 4
It doesn’t matter which point we designate to be (x 1 , y 1 ) and which (x 2 , y 2 ).

Swapping round (3, −1) and (−1, 7) in the last example we still get the same
gradient:
7−− 1 8
= = −2.
−1−3 −4

The gradient in the last example came out negative. The line is precisely as
steep as the first one but it slopes “backwards". If we are using the formula
to work out the gradient, the formula will take care of whether the gradient is
positive or negative.
But what if we are using the triangle approach? How does that give us a
negative gradient? The answer is that in the last example, in moving along
the line (remember we always move from left to right) we go down not up.
Put another way, the movement in the y-direction is negative. We can think
of the gradient as “the change in y divided by the change in x”, provided we
remember to make it negative if the line is sloping backwards.
Below we find the gradients again using the triangle method—and add an
extra line for luck.

(3, 7) (5, 7) •
•
6
6

+3

+4 (1, 4) •

-

(1, 3) +4
• -
+2

+4 +3 3
grad = = 2 grad = =
+2 +4 4
A
+4
A
A• -
e
(−1, 7) A e
+5
A e• -
A (−2, 6) e
A e
A e
A e
A -8 e -6
A e
A e
A e
A e
A e
A e
A e•?
A (3, 0) e
A•? e
e
(3, −1) A
A
A
−8 −6 6
grad = = −2 grad = =−
+4 +5 5

To sum up: if we move downwards as we go from left to right along the line,
we translate this decrease in y as a negative increase in y, so the formula
stands:
A
y 2 −y 1
increase in y-coordinate
!
A
gradient = = A
increase in x-coordinate x 2 −x 1 A
Sometimes you will want to find a gradient using the formula and
sometimes using the triangle approach. This formula captures both.

The gradient of a line parallel to one of the axes

How does this formula apply to a line that’s parallel to one of the axes?
First if it is “horizontal", i.e. parallel to the x-axis, y 2 = y 1 , the numerator in
the formula will be zero, so a horizontal line has gradient zero.
If the line is “vertical", i.e. parallel to the y-axis, x 2 = x 1 , the denominator will
be zero, so its gradient is infinite. We met ∞ on page 1-9.
A
for a line parallel to the x-axis, the gradient = 0
!
A
A
for a line parallel to the y-axis, the gradient = ∞ A
horizontal lines:
y gradient = 0
6

- x
y vertical lines:
6 gradient = ∞

- x
NB: The x-axis is the line y = 0. The y-axis is the line x = 0.

QQ
1. In each case find the gradient of the line joining the two points.
(a) (b)
(4, 9)
•
(5, 13) •
(2, 1) • (2, 6) •
(c) (d )
•
(−1, 2)
(4, 7) •
(−2, 4) •
•
(4, −3)
2. Find the gradient of each line using the endpoints marked A and B .
(a) (b)
6 B• 6
PP•
A PPP
P
PP B
P• P
A• - -
(c) (d )

6 6

B•
T
T

T

T•

AT

T

T

- -
A•
T
T B

T•

T
T
3 7 1 1 3 5
1.(a) 4 (b) (c) (d) −1 2.(a) 1 (b) − (c) (d) −
3 2 3 2 3

5.2.2 The equation of a straight line: y = mx + c

We are going to be drawing lots of straight lines, and later on lots of curves.
But we won’t be plotting them on graph paper because we won’t be doing
any measuring.
For us the purpose of the drawing will be to show the key features in pictorial
form: the overall shape; whether it cross the axes (and if so where); how the
gradient changes, and so on. We call such a diagram a sketch.
The word sketch is a slightly unfortunate term to use in this context, because
the word might suggest either an artistic rendering, or something that is
rough and ready. It means neither. A sketch is a diagram, not drawn to scale,
that makes the key features clear. It’s a good idea to use a pencil when
drawing a sketch because often we want to amend our first attempt.
By the end of this section you’ll be able to sketch a line just by looking at its
equation, and starting with a sketch, work out quite easily what its equation
is.

Our starting point is the line y = x (drawn using a thick line in the diagram),
whose gradient is 1. Saying the equation is y = x is shorthand for saying that
at every point on the line “the y-coordinate = the x-coordinate".
Suppose we were to move every point on this line up 4 units. So for example
(2, 2) becomes (2, 6), and so on. Rather than being equal to it, the
y-coordinate of every point on the new line is now 4 units bigger then its
x-coordinate. In other words its equation is y = x + 4.
One particularly important point on a line is the y-intercept, which is where
it crosses the y-axis. For the line y = x it was (0, 0) For this new line it is (0, 4).
In exactly the same way we can see that the lowest line on the diagram has
equation y = x − 2. (Starting with the line y = x we have moved every point
down 2 units). For this line the y-intercept is at (0, −2).
Each of the lines has equation y = x + c, where c is the value of its
y-intercept.
6
y 4
6
+
x
=
+4
y
−2
x
+4 ?
=
y
2
−
x
=
y
- x
0
−2

Now we are going to look at lots of lines with different gradients, but this
time all passing through the origin. Use some of the the points marked to
satisfy yourself that their equations and their gradients are correct.
grad = 4

grad = −3 y
grad = 2
grad = 1
B 6
B 8 •
B
B 7
y=
B
x
y =4
B
−3x
•B 6
2x

y=
B
B 5
grad = 21
x
B
=

4 • •
y
B
B
1 x
•B 3 y =2
B

B
B 2 • • •
Q Q B
Q B1 • •

Q B
Q
QB - x
0 Q 1 2 3 4 5 6 7 8

−2 −1 B Q
B Q
−1
B Q
Q
B Q
−2 Q•
Q y=
B Q
Q −2
B
−3 B Q 3x
B Q
−4 Q
B •Q
grad = − 32
B Q

We can see that each line has equation y = mx, where m is its gradient.

Which brings us to this crucial result: if the equation of a straight line is
y = mx + c, its gradient is m and its y-intercept is c.
Take the example of the line below. We can see that its gradient is 2, and that
its y-intercept is +3.
The equation must therefore be y = 2x + 3.
y

6 We can easily verify that
8
this is correct. The point
7
(1, 5) lies on the line because

6 its coordinates satisy the

equation, as does (0, 3) of
5 •(1, 5)
course.
4

3 •

2

1

- x
−1 0 1 2 3 4 5

−1
y
L
Another example: to find the equation L 5 6
of the line in the graph opposite. L
L4
The y-intercept is 3, so c = 3. To find L
3 L•
The gradient, we look for two points on L
the line both of whose coordinates are 2 L
L
integers: (0, 3) is one; another is (2, −2). 1 L
To move between these points from left
L
L - x
to right we need to go along 2 and down −1 0 1 L 2 3 4 5
5 −1 L
5, so its gradient is − 2 . L
5 −2 L•
The equation is therefore y = − 2 x + 3. L
−3 L
L
−4 L
If we can know the gradient of a line and its y-intercept, we can write
down its equation.

Conversely if we have the equation of a line, in the form y = mx + c, we

can construct what its graph looks like.
y
We’re going to construct the line
whose equation is 5
6
y = 3x − 2. 4
3
All we need is two points on the
line. We already have one: the 2
y-intercept (0, −2). And since the
1 •
gradient is 3 we can find a second
- x
one by moving from it 1 units to
−1 0 1 2 3 4 5
the right and 3 units up. That −1
takes us to (1, 1).
−2 •
The two points are marked on the
−3
graph, and they tell us where the
line sits.
y
Let’s look at a harder example: the
equation
6
5
2 4•
y = 4 − 3 x.
or 3
2
y = − 3 x + 4.
2 •
As before we already have one 1
point: (0, 4). - x
2
Since the gradient is − 3 we can −1 0 1 2 3 4 5
find a second one by moving from −1
it 3 units to the right and 2 units −2
down. That takes us to (3, 2).
−3
The two points marked give us our
line.
What we have seen in this section can be summed up by:
gradient
↓
y = mx +c
↑
y-intercept

QQ
1. By identifying its gradient and intercept, write down the equation of each
of these lines, in the form y = mx + c. (The grid lines are at 1 unit intervals.)
(a) 6 (b)
6

- -

(c) (d )
6 6
HH
HH
HH

H
HH
HH

- -

2. From the gradient and y-intercept of these straight lines, sketch them.
(a) y = x − 3 (b) y = 3x − 2 (c) y = 32 x + 1 (d) y = − 12 x + 5
4 2 1
1. (a) y = x + 3 (b) y = 2x − 1 (c) y = x + 1 (d) y = − x + 4
3 2
2. For example (a)(0, −3), (3, 0) (b)(0, −2), (1, 1) (b)(0, 1), (3, 3) (b)(0, 5), (2, 4).

We said that any linear relationship can be represented by a straight line. If

the relationship happens to be expressed in the form y = mx + c, we know
immediately that the gradient of the line must be m and its y-intercept c. If
the equation is written in a different form we’ll need to do a bit of
rearranging to extract that information.
Since it means writing the equation in the form y = . . ., that amounts to
rearranging it as a formula, making y the subject. (If you need a reminder
about this see page 6-22.)
Example 1
For each of these lines find its gradient and y-intercept.

(a) 2y − 4x = 3 (b) x = 3y + 7 (c) 0.6x + 0.25y = 1.
Solution
(a) The equation needs rearranging into the form y = mx + c.
(+4x) 2y = 4x + 3
3
(÷2) y = 2x +
2
3
The gradient m = 2 and the y-intercept c = 2 .
(b) Rearranging
3y = x − 7
1 7
y = x−
3 3
1 7
m =3, c = −3.
(c) Rearranging
0.25y = −0.6x + 1
(×4) y = −2.4x + 4
12
m = −2.4 or − 5 , c = 4.
Now we can move on to the kinds of problems you could meet in an exam, each
of which can be solved using y = mx + c.

Example 2
Find the equation of the line that is parallel to the line 2y + 5x = 3 which
passes through the point (0, −4).
Solution
5 3
The equation 2y + 5x = 3 can be written y = − 2 x + 2 .
The line we are looking for has the same gradient, so its equation must be
5
y = − 2 x + c, for some value of c.
5
But we know its y-intercept is −4, so the equation is y = − 2 x − 4.
The answer is correct, but we’d normally tidy it up a bit. We can get rid of the
fractions (in this case by multiplying by 2). It’s also conventional to try and
avoid expressions beginning with a minus sign. So we might write it as
2y = −5x − 8
2y + 5x + 8 = 0
It’s worth a quick check that x = 0, y = −4 satisfies this equation.

Example 3
The line 5y = 3x + k passes through the point (−1, 3). Find the value of k and
hence determine where the lines crosses both axes.
Solution
x = −1, y = 3 satisfy the equation, so 15 = −3 + k, i.e. k = 18.
The equation of the line is 5y = 3x + 18.
We could rearrange it into the form y = mx + c, but since we are after both
intercepts, it’s easier to substitute x = 0, and then y = 0 into the equation.
18
The y-intercept is y = (0, 5 ) and the x-intercept (−6, 0).
Example 4
Find the value of p for which the two lines are parallel.
px + 3y = 1
y = 2x + 4
Solution
p 1 p
Rearranging the first equation: y = − 3 x + 3 , so its gradient is − 3 .
p
But it is parallel to the second line, whose gradient is 2, therefore − 3 = 2, or
p = −6.
QQ
(a) Find the gradient m and the y-intercept c of the line 3y − 2x + 12 = 0.

(b) Write down the equation of the line parallel to y = 2 + 5x that passes
through the point (0, −3).
(c) the point (−1, 4) lies on the line y = ax + 1. Find the value of a.
5 2
(a) m = , c = −4 (b) Gradient is 5, so y = 5x − 3
3
(c) Substituting x = −1 and y = 4 gives 4 = −a + 1 ⇒ a = −3.

5.2.3 Methods for finding the equation of a line

You will often have to find the equation of a straight line, given some
information about it. That information is unlikely to be its gradient and
y-intercept. (If it was you could simply write the equation down.)
In this section we’ll be looking at what to do if we don’t have this information
but we do have sufficient to work out what the equation must be.
We start with an example in which we know the gradient of the line, and we
have the coordinates of a point through which the line passes—a point not
on the y-axis.
There are two possible approaches. The first is to use the form y = mx + c,
and here are two examples.
Example 5.
Find the equation of the line with gradient 3 passing through the point (4,1).
Solution.
Using the form y = mx + c, since the gradient is 3, it must be possible to write
the equation as y = 3x + c, for some value of c.
To find c, we know that the coordinates of (4, 1) must satisfy the equation so
1 = 12 + c or c = −11
The equation is y = 3x − 11.
Before showing you the second method we’re going to develop a new
formula.

Example 6
1
What is the equation of the line through (−4, 3) with gradient 2 ?
Solution
The equation must look like this:
1
y = x + c.
2
for some value of the constant c. Substituting x = −4 and y = 3 into the
equation we have
1
3 = × (−4) + c
2
3 = −2 + c ⇒ c = 5
The equation of the line is

1
y = x + 5.
2
This is the correct answer, but since it is always possible to get rid of fractions
from an equation, we generally do so. Multiplying through by 2 we have:
2y = x + 10.
This is the first method. But it relies on us knowing the gradient up front. This
won’t always be the case. Very often what we know is two points the line
passes through. The easiest way to deal with this case is using the formula
that we are now going to develop.

Starting with two points that lie on the line, we want a formula that tells us
the equation of the line.
Suppose the two points are A(2, 1) and B (5, 7) on the graph below.
Using the formula

y2 − y1 y
gradient = ,
x2 − x1 6
the gradient of the line AB is B •

7−1 6 (5, 7)
= = 2.
5−2 3 P •(x, y)

Now here is the clever bit. This
formula will work for any pair
of points on the line, so we

imagine any point at all—we’ll
call it P and give it coordinates

(x, y). In the diagram we have A•
put P on the line but for the (2, 1)
- x
moment imagine it could be 0
anywhere.

The reasoning goes like this: if the gradient of AP is also 2, then P is on the
line; if the gradient is not 2, P is not on the line.
That means the statement “P is on the line" can be expressed
mathematically by turning the statement “the gradient of AP = 2" into
algebra using the coordinates of P .
Using the formula for the gradient of a

line, but instead replacing x 2 with x and
P (x, y) • y 2 with y, the gradient of AP is:

y −1
x −2

so P lies on the line if x and y are
connected by the relationship:

y −1
= 2.
A(2, 1) x −2
•

In other words the equation of the line is

y − 1 = 2(x − 2)

which simplifies to y = 2x − 3.

We used the gradient formula to write down the gradient of AP ; we could

also use the triangle approach.
P is y units above the x-axis and

P (x, y) • A is 1 units above the x-axis, so to

6 get from A to P we need to go up

a distance y − 1 units.
Similarly to get from A to P we

y −1
need to go along a distance x − 2

units.
The gradient of AP is therefore

A(2, 1)
• -
y −1
x −2
x −2

as before.
We can turn this into a formula that works for any two point (x 1 , y 1 ) and
(x 2 , y 2 ). The gradient of the line between the two points is
y2 − y1
x2 − x1
and the equation of the line passing through the two points(x 1 , y 1 ) and
(x 2 , y 2 ) is
y − y1 y2 − y1
= .
x − x1 x2 − x1
Notice that in this formula (x 1 , y 1 ) has a privileged position: unlike (x 2 , y 2 ) it

appears on both sides of the equation. So is it important to have the points
in the right order? No it isn’t. We can verify this by swapping A and B in
the example above. (At first it looks as though we are going to get a different
answer—but it all comes out in the wash.)
y −7 1−7 −6
= = = 2
x −5 2−5 −3
y − 7 = 2(x − 5) = 2x − 10
y = 2x − 3

Earlier we saw how to construct the equation of a line given its gradient and
one point on it, using y = mx + c. At the time we also said there is an
alternative way to find the equation. It’s this.
Example 5. (Alternative method)
Find the equation of the line with gradient 3 passing through the point (4,1).
Solution.
We now know that the gradient of a line we can write as
y − y1 y −1
or in this case .
x − x1 x −4
But we already know gradient is 3, so we can write:
y −1
= 3 or y − 1 = 3(x − 4),
x −4
which simplifies to y = 3x − 11, the same result as before.
We can generalise this into saying that the
equation of a line through (x 1 , y 1 ) with gradient m is
y − y1
=m or y − y 1 = m(x − x 1 ).
x − x1
Often this is the quickest way to derive an equation.

Summarising all the results we have found:
Methods for finding the equation of a line.
Data Formula Method

m, c y = mx + c Write equation down.
Get rid of any fractions.
m, (x 1 , y 1 ) y = mx + c Substitute:
y 1 = mx 1 + c. Hence c
A
or and write equation down.
!
A
A
y − y 1 = m(x − x 1 ) Substitute and simplify. A
y−y 1 y 2 −y 1
(x 1 , y 1 ), (x 2 , y 2 ) = Substitute and simplify
x−x 1 x 2 −x 1
These formulae are the basis of many exam type questions.

QQ
In each case find the equation, given either two points on the line, or one
point and the gradient. Express your answer in a form that doesn’t involve
any fractions.
1 3
(a) (0, −4); gradient = − 3 (b) (1, −2); gradient = 2
(c) (−1, 3) and (6, 5) (d) (− 25 , 1); gradient = −2
(e) (k, 0) and (0, k).

For part (e) you can use the formula or you may want to draw a sketch. The
answers are in the footnote6 . Yours may be in a slightly different form.
6
(a) 3y + x + 12 = 0 (b) 2y = 3x − 7 (c) 7y = 2x + 23 (d) 5y + 10x = 1
y k
(e) = = −1 ⇒ x + y = k.
x−k −k

Before we move on, a word about diagrams. It is not necessary to draw a

sketch for every question involving coordinate geometry. The two main
advantages of doing so are to help our understanding—a diagram tells us
things that words may not, and as an error check—a sketch often tells us
roughly what answer to expect.
Example 7.
Find where the line joining the points P (−1, −2) and Q(3, 1) cut the
coordinate axes.
Solution.
y The equation of PQ is
5
6 y − (−2) 1 − (−2)
=
x − (−1) 3 − (−1)
4
y +2 3
3 =
x +1 4
2
4(y + 2) = 3(x + 1)

1 • Q(3, 1)
4y + 5 = 3x

- x 5
−2 −1 0 1 2 3 4 5 When x = 0, y = − 4 ,
−1
5
P (−1, −2) −2
•
when y = 0, x = 3 .

−3 The line cuts the axes at
(0, − 54 ) and ( 53 , 0).
From the sketch these are the kinds of answers we would expect. It doesn’t
tell us they are correct, but had we made a mistake it might have told us.

In this next example we are going to use a completely different approach: no

formulae, just simple reasoning about the relationships between points.
Example 8.
ABC D is a parallelogram. The coordinates of A, B and C are (−1, 3), (−3, −2)
and (1, 0) respectively. Find the coordinates of D.
Solution.
Since ABC D is a
parallelogram AB //C D
and AD//BC , and the
parallel sides are also the
y same length. We can use
6 D this to work out where D
5 •
is, starting either from A or
4 from C .

A(−1, 3) • 3 To get from B to A we must

2 move 2 units to the right and
5 units up.
1

C (1, 0) The same must therefore
• - x
apply from C to D,
0
−3 −2 −1 1 2 3 4 5
−1
so D is (3, 5).

•
−2 Check this answer for
B (−3, −2)
yourself by finding D again,
but this time starting from
A.
Notice that we didn’t need to find the equation of any of the lines. We didn’t
even use the notion of gradient—not explicitly at least. It was the idea of a
translation, or a vector, that solved the problem. The concept of vectors is
simple but very powerful. You will meet them again later.

y
5.2.4 The equation of a straight line x + = 1
a b
We can do everything we need to with straight lines using the form of

equation y = mx + c. Sometimes it’s slightly easier to use a different form.
Quite a common one is

x y
+ = 1. y
a b
6
When y = 0, x = a, and
when x = 0, y = b. aa
aa
So the line goes through baa
aa
(a, 0) and (0, b). aa
aa
a
These points are the
a aa x
aa -
x-intercept and the y- 0 aa
intercept.
It means we can write down the equation of a line if we know where it

intersects the axes.
For example the lines below (notice the negative sign in the right hand one).
x y x y
+ =1 and − = 1.
3 6 4 2
y
AA 6 y
A 6
A (0, 6)
A
A
A
A
-
x
A
A
(4, 0)
A
A
A (0, −2)

A (3, 0)

- x

A
AA
Conversely we can easily sketch the line if we have the equation in this form.

For example take the line
2x + 5y = 8.
To get it into the form we need, there

y
has to be a 1 on the right hand side,
so we divide every term by 8: 6
2x 5y
+ =1 (0, 1.6)
8 8
aa
aa
Now the coefficients of x and y in
aa
each numerator must also be 1, so aa
aa
we divide top and bottom of each aa
a - x
fraction by the appropriate number:
aa
0
aa
(4, 0) aa
x y
+ =1
4 1.6
and the intercepts are now clear.
QQ
Write down the x- and y-intercepts for these lines. The answers are in the
footnote7 .
x 2y x y y
(a) 5 + 7
=1 (b) 2 − 8=1 (c) 3x + 5y = 30 (d) 3 − 7x
2
+1 = 0.
7
(a) x = 5, y = 3.5 (b) x = 2, y = −8 (c) x = 10, y = 6
7x y
(d)
2
− 3 = 1 ⇒ x = 27 , y = −3

This next example is for anyone who is finding it all too easy so far. You are
free to skip it if you wish, or come back to it later
Example 9.
x y
A line has equation + = 1 (k 6= 0)
2k k
(a) Show that its gradient is independent of the value of k.

(b) Find the area enclosed by the line and the two coordinate axes.
(c) Hence, or otherwise, find the perpendicular distance from the origin to
the line.
Solution 8 .
y
(a) The intercepts are (2k, 0) and (0, k). 6
A quick sketch shows us that the HA
H
k HH
gradient of the line is t
H
HH
H
k 1 d HH
− =− H
2k 2
HH
H
HHB -
(negative because the line slopes 2k HH x

O
backwards).
Notice we don’t necessarily need to have numerical values to sketch a graph.
(b) The area of triangle O AB is 21 OB.O A = k 2 .
(c) The area is also equal to 21 AB.d = k 2 , where d is the perpendicular

distance OP from the origin to the line.
p
By Pythagoras AB 2 = k 2 + 4k 2 = 5k 2 , i.e. AB = 5 k.
1p 2k
Hence 5 k.d = k 2 , therefore d = p .
2 5
When a question says “hence or otherwise" it is almost always easier to use

the “hence" route. But you may not see how to do it this way. An “otherwise"
approach for finding d in this case would be to use trigonometry (or similar
triangles, which amounts to the same thing).
A Since the angle B is in two triangles:

HH
P BOP and BO A, we can write down
tH
H
what the cosine of B in two ways:
H
k
HH
H
d HH OP O A
sin B = = .
H
HH OB AB
H
H d k 2k
∴ 2k = p , giving d = p .
H
O B
5k 5
8
To see how the solution to part (a) ties up with the y = mx + c approach, we can
rearrange the equation:
(×2k) x + 2y = 2k
2y = 2k − x
x
y = − +k
2
from which we can read off the gradient − 12 and y-intercept (0, k).

5.2.5 Perpendicular lines

There is a simple relationship between the gradients of two lines that are
perpendicular to each other. Look at the diagram. If we rotate the red
triangle OPQ through 90◦ anti-clockwise about O, we get the blue triangle
OP 0Q 0 .
The effect of the rotation is
to make each side in the new
Q L triangle perpendicular to the
corresponding side in the
gra
L
original triangle:
d=
L
L
−2
5
L
L !! ! OP 0 ⊥ OP, OQ 0 ⊥ OQ.
0
!L t
P ! !
2 !! L
But more importantly (for our
d = 5
! L
gra !!!
!
L purposes)
! ! L
!
P 0Q 0 ⊥ PQ,
! L
!!
• L
Q0 O P
so let’s compare their gradients:
5 2
The gradient of PQ is − and the gradient of P 0Q 0 is .
2 5
The product of these gradients is − 52 × 52 = −1, a result that applies to any pair
of perpendicular lines:
if the gradients of two perpendicular lines are m 1 and m 2 , then A
!
A
m 1 m 2 = −1. A
A
gradient gradient
Another way to look at it is this: to get from one of l 1 of l 2
gradient to the other, all we have to do is turn it 7 4
−
upside down and change the sign. 4 7
1
3 −
3
The table opposite gives examples of the gradients 1
of pairs of perpendicular lines (l 1 and l 2 ). −9
9
The rule works both ways, of course, because if we −0.8 1.25
repeat the process we get back to the gradient we 1
m −
started with. m
a b
−
b a

QQ
Try these example for yourself.

1. In each case the gradient of a line is given. Find the gradient of a line
perpendicular to it. The answers are in the footnote9 .
1 3 a 1
(a) 4 (b) −1 (c) − 5 (d) 0.5 (e) 2 (f) k−1
2. For each pair of lines, say if they are parallel, perpendicular or neither.
(a) 5x + y = 6, 10x + 2y = 13 (b) y = x + 7, x + y = 7

(c) y = 3x + 1, 3y + x = 4 (d) 6x + 7y = 1, 6x = 7y + 3
9 5 2
1.(a) −4 (b) 1 (c) (d) −2 (e) − (f) −(k − 1) = 1 − k
3 a
2.(a) parallel (b) perpendicular (c) perpendicular (d) neither

Example 10
Find the equation of the line perpendicular to 2y = 4 − 3x that passes
through the point (−5, 0).
Solution
3
In y = mx + c form, the equation of the line 2y = 4 − 3x is: y = − x + 2.
2
2
The gradient of the perpendicular line is therefore 3 and its equation is
y −0 2
=
x − (−5) 3
3y = 2x + 10
Example 11
The point P (4, 7) lies on the circumference of a circle, and at P the tangent to
the circle has gradient equal to −2. The centre of the circle C lies on the line
x = −3. Find the coordinates of C .
Solution
The tangent to a circle and the radius through the same point on its
circumference are at right angles. So the line through P with gradient 12 is a
diameter (i.e. it passes through the centre of the circle). Its equation is
y −7 1
=
x −4 2
2(y − 7) = x − 4
2y − 14 = x − 4
2y = x + 10.
The centre also lies on the line x = −3, so substituting this into the equation
7
of the line gives y = 2 .
The centre of the circle is at (−3, 72 ).
The last three examples bring us up to the standard of the most demanding
exam questions you are likely to meet.

Example 12
4x + k y = 3 (5.1)
x − 9k y + 1 = 0 (5.2)
are two perpendicular lines. Find the possible values for k.

Solution
4 3
The first equation can be written: y = − x + ,
k k
1 1
and the second written: y = x + .
9k 9k
The lines are perpendicular, so the product of their gradients must be −1,
4 1
i.e. − . = −1, or k 2 = 49 . The possible values for k are 23 and − 32 .
k 9k

Example 13
Taken in order anticlockwise, A, B , C and D are the vertices of a square. The
coordinates of A and B are (2, −1) and (4, 3) respectively.
(a) Show that AD passes through the origin and C lies on the y-axis.
(b) Where does the line DB cross the x-axis?
Solution
(a) Gradient of AB is 2, and gradient of y

1
AO is − 2 , so AO ⊥ AB . 6
C?
But since ABC D is a square, AD must
also be ⊥ AB , so O lies on AD. B (4, 3)
•

What about the position of C ? Knowing
D?
that BC is the same length as AB , and

perpendicular to it, we can construct - x
O

the coordinates C from the coordinates •
of B , using the vector approach we A(2, −1)

adopted in Example 8:
To get from A to B we move 2 to the right and 4 up, so to get

from B to C , or from A to D, we must move 4 to the left and 2 up. The
coordinates of C must therefore be (4 − 4, 3 + 2), i.e. (0, 5).
Similarly the coordinates of D must be (2 − 4, −1 + 2), i.e. (−2, 1).
(b) Now we know where D is, we can use the formulae to find the equation of
DB . Its gradient DB is:
3−1 1 y−1 1
= , so its equation is =
4+2 3 x+2 3
i.e. 3(y − 1) = x + 2, which simplifies to 3y = x + 5.

Substituting y = 0 into the equation we find where DB crosses the x-axis i.e.
at x = −5.

Example 14
The vertices of a triangle are A(1, 0), B (6, 0) and C (4, 4). The altitude 10 that
passes through C meets the x-axis at N , and the altitude through B meets
AC at M . Find where the altitudes meet.
Solution
C N is the line x = 4.
4
y The gradient of AC is 3 , and
B M ⊥ AC so the equation of
6 C (4, 4) B M is
A
A y 3
A = −
M A x −6 4
ZZ
P or 4y = −3x + 18.
A
Z A
Z A
Z A P lies on this line x = 4.
A AB - x
Z
Z So we have
0 1 N 6 3
4y = 6 or y = .
2
Incidentally, we can verify that the latitude from A also passes through P .
3
The gradient of AP is 2
3 = 12 ; the gradient of BC is − 24 = −2.
AP is perpendicular to BC , so the altitude through A passes through P .
QtQ orthocentre

The line through the vertex of a triangle

Q

aa QQ altitude
t
a )
10
perpendicular to the opposite side is called an aa

Q

altitude. The three altitudes of any triangle all
aa Q
t aa
meet at the same point, which is known as the
Q

a
Q
orthocentre of the triangle.

5.2.6 Intersecting lines

In all these examples, finding where two lines cross was easy because one of
them was parallel to one of the axes. Suppose the two lines were:
x + 3y = 15
y = x +1
To find where these cross we shall have to use one of the methods of solving
linear simultaneous equations in Section 6.5 of Chapter 5 (see page 6-41).
The answer is x = 3, y = 4 (verify that these satisfy both equations).
To interpret this y
graphically, let’s write 6
PP
the first equation as 5 PP
P
(3,
P•P
P4)
x y
+ =1
PP
P
15 5 PP
P
1 PP
which intersects the P PP - x
axes at (15, 0) and −1 15P
(0, 5).
The second line is easy to sketch and the point on the graph where they
cross, (3, 4), looks about right.

We’ve been able to solve all the examples we’ve seen so far. Does that mean
any pair can be solved? The answer is no, because the lines they represent
could be parallel, in which case there is no point that lies on both of them.
Although it’s not immediately obvious, the equations
x+y =4 and 2y = 3 − 2x
correspond to a pair of parallel lines. Let’s see what happens if we try to find
where they cross.
Substituting y = 4 − x from the first
equation into the second, we have
2(4 − x) = 3 − 2x y
8 − 2x = 3 − 2x
@ 6
@
4
@
which seems to imply that 8 = 3!! This @
nonsensical conclusion tells us the @
@
lines must be parallel.
x
+
@
y
@
=
@
Writing the second equation as @
4
@
1.5
@
x @
3 +
@
y @
x+y = , =
@
3
@
2 @ 2
@
@ -
x
0 1.5@ 4@
we can see that both lines have @ @@
@
gradient −1.

QQ
Try these example for yourself.
1. In each case the gradient of a line is given. Find the gradient of a line
perpendicular to it. (The answers are in the footnote11 )
1 3 a 1
(a) 4 (b) −1 (c) − 5 (d) 0.5 (e) 2 (f) k−1
2. For each pair of lines, say if they are parallel, perpendicular or neither.
(a) 5x + y = 6, 10x + 2y = 13 (b) y = x + 7, x + y = 7

(c) y = 3x + 1, 3y + x = 4 (d) 6x + 7y = 1, 6x = 7y + 3
11 5 2
1.(a) −4 (b) 1 (c) (d) −2 (e) − (f) −(k − 1) = 1 − k
3 a
2.(a) parallel (b) perpendicular (c) perpendicular (d) neither

5.3 An extraordinary puzzle

It is often said that seeing is believing. We start with some simple shapes.
Now we assemble them together, in two different ways. In the first case we
get a triangle whose base is 13 units and height 5 units.
It appears that we get the same in the second case, except there is a square
missing. What’s going on?
The shapes haven’t been doctored in any way; there is a perfectly sensible
explanation.
The footnote12 contains a clue to the conundrum.
12
There is a good reason for the puzzle to be placed in this chapter.

5.4 Distances between points and lines

You will not need to know how to find the distance between a point and a line,
or the distance between two parallel lines. But for completeness there is some
optional material that briefly introduces these ideas if you are interested.
5.4.1 The distance between two points

To calculate the distance between two points, we start by finding the
difference between their y-coordinates and the difference between their
x-coordinates, just as we do when finding a gradient. An example:
y y
6 6
B (4, 6) (4, 6)
6 6

4
6

?
6 A(1, 2) (1, 2)
-
2 3
- x - x
-
1
-
4
Imagine moving from A to B . In the x direction that means moving from a

point 1 unit to the right of the y-axis to a point 4 units to the right, i.e. a
movement of 3 units in the x direction. Similarly in the y direction, the
movement is from a point 2 units up from from the y-axis to a point 6 units
up, i.e. a movement of 4 units in the x direction. Then by Pythagoras:
AB 2 = 32 + 42 = 25, ∴ AB = 5.
We can turn this into a general formula.
y y
6 6
B (x 2 , y 2 ) B
6 6

y2 − y1
y2

?
6 A(x 1 , y 1 ) A
-
y1 x2 − x1
- x - x
-
x1
-
x2
By Pythagoras: AB 2 = (x 2 − x 1 )2 + (y 2 − y 1 )2

A reminder
Before doing an example, a reminder of something we met in Chapter 2.
When using Pythagoras we will often be finding the square root of a number,
and it will pay off to get used to writing square roots in their simplest form.
p 1
For example 20 = 20 2 . We can write this
1 1 1 1 1
20 2 = (4 × 5) 2 = 4 2 × 5 2 = 2 × 5 2 .
Or, in root notation
p p p p p
20 = 4 × 5 = 4 × 5 = 2 5.
One other point. Try to avoid using a calculator—whether it be dealing with
square roots or subtracting negative numbers; do as much as you can in
your head. You will get to know how far you can go with this by the number
of mistakes you make.
QQ
In each case find the distance between the two points, expressing your
answer as simply as possible.
(a) (2, 0) and (5, 4) (b) (−2, 1) and (3, 13) (c) (0, 9) and (2, 3)
(d) (5, −11) and (−2, 13) (e) (1, 1 − a) and (a + 1, 1)
Before we leave this topic, an example of the kind of question you could be set
in an exam.
13
p p p p
(a) 5. (b) 13. (c) 40 = 2 10. (d) 25 (e) 2a 2 = a 2.

Example 15
The vertices of a triangle are P (−4, 1), Q(4, 7) and R(−1, −3).
(a) Find the lengths of PQ, QR and RP .
(b) Hence, or otherwise, show that the triangle is right-angled.
Solution
(a) Let’s start with a quick sketch (it doesn’t need to be accurate). Since we are
only interested in distances and angles we don’t even need to include axes.
Q(4, 7)
•

P (−4,1)
•
@
@
@
@
@
@
@•
R(−1, −3)
Using the formula AB 2 = (x 2 − x 1 )2 + (y 2 − y 1 )2

PQ 2 = (4 + 4)2 + (7 − 1)2 = 100
QR 2 = (4 + 1)2 + (7 + 3)2 = 125
RP 2 = (−3 − 1)2 + (−1 + 4)2 = 25

p p
∴ PQ = 10, QR = 125 = 5 5 and RP = 5.
(b) A glance at the numbers tells us that
QR 2 = PQ 2 + RP 2
so PQR is a right-angled triangle with the 90◦ angle at P .
Alternatively, we could show P̂ = 90◦ , by showing that PQ ⊥ P R:
7−1 3 −3 − 1 4
gradient of PQ = = , gradient of P R = =− .
4+4 4 −1 + 4 3
The product of their gradients is −1, so the lines are perpendicular.
N.B. Had we not been asked to find the lengths (i.e. part (a)), the alternative
method definitely provides an easier solution to (b).
If you had wondered about finding the distance from a point to a line, you may
be intereted in the next couple of pages; if not, feel free to move straight to the
next section.

5.4.2 The distance between a point and a line

Strictly speaking it doesn’t really make sense to talk about “the distance
between a point and a line"—it depends on which point on the line we
choose. When we use the phrase we mean the shortest distance.
To get from the point P to

the line l , we could go to P •XXX
any point on the line, for L@ XXX
L@ XXX ! l
example A. L @ XX X!!
!
L @ !! D
The shortest distance is L @!!
@ !
along a line through P t LL !!C
!Y
perpendicular to l , i.e. the

! !!B H HH
!
distance from P to B . !!
A Foot of the perpendicular
The point B we refer to as the foot of the perpendicular from P to l .
Example 16
Find the distance between the point M (2, 8) and the line 2y = x − 1.
Solution
Suppose the the foot of the perpendicular from M on to the line is at N . The
distance from M to the line is the length of M N , which we will call d . To find
this we shall need the coordinates of N .
N is the point where the line M N

intersects the line 2y = x − 1, so we
must find the equation of M N .
y
The equation of the line
6 • M (2, 8)
2y = x − 1
A
A
A
A we can write as
dA
1 1
A y = x− ,
A
2 2
t
A
1
so its gradient is 2 . The gradient of
A
N

M N , therefore, is −2, and since it
-x
passes through (2, 8), its equation is

y − 8 = −2(x − 2)
y = 12 − 2x.
So N is at the intersection of these two lines:

2y = x − 1
y = 12 − 2x.
This is easily solved, to give x = 5, y = 2 (satisfy yourself this is correct).

Now we can find the distance M N : d 2 = (5 − 2)2 + (8 − 2)2 = 32 + 62 .

p
The
p exact p answer is 45, or since 45 contains a factor 9, we can express it as
9 × 5 = 3 5 units.
Sometimes we want a numerical value, i.e. 6.71 (3 SF).
In an exam, unless the question says otherwise, it is always better to give the
exact answer.
The distance between two parallel lines

As with a point and a line, the distance between two lines is the shortest
distance between them, that is along any of the lines perpendicular to both
of them,
The simplest case is where both lines are parallel to one of the axes— for
example the lines y = 3 and y = 7, which are clearly 4 units apart. You won’t
need any more than this at the moment.

5.5 The mid-point of a line

Or more properly the mid-point of a line segment, because we think of a
mathematical “line" as extending infinitely in both directions.
In the previous section we found a formula for the length of the segment;
Here we’ll be finding the coordinates of its mid-point. As before we consider
separately distances from the x-axis, and distances from the y-axis. AB is the
line in question (in red) and M is its midpoint.
y y
6 6
B (4, 6) (4, 6)
6

M

M • •
6
6

4
6 A(1, 2) (1, 2)
2
- x - x
- -
1 5
- 2
4
A is 2 units above x-axis and B is 6 units above, M must be 4 units

1 5
above—the average14 of 2 and 6. Similarly, in the x-direction M is 2 (1 + 4) = 2
units from the y-axis.
M is the point (2 21 , 4). We can turn this into a general formula.
y y
6 6
B (x 2 , y 2 ) (x 2 , y 2 )
6

M

M • •
6
y2
2 (y 1 + y 2 )
1

6 A(x 1 , y 1 ) (x 1 , y 1 )
y1
- x - x
- -
x1 1
- 2 (x 1 + x 2 )
x2
¡1
(x + x 2 ), 12 (y 1 + y 2 )
¢
The midpoint of (x 1 , y 1 ) and (x 2 , y 2 ) is 2 1
14
Strictly speaking we should say the mean or the mean average.

QQ
Try these examples. They start off quite simply but we have included some
algebraic examples, to make it a little harder.
For each pair of points, find the coordinates of the midpoint (the answers are
in the footnote15 ).
(a) (2, 5) and (6, 11) (b) (−3, −1) and (5, −11) (c) (0, −1) and (1, −2)
(d) ( 13 , − 12 ) and ( 73 , 23 ) (e) (a, 3b) and (5a, −b) (f) (2, 14 y) and (2x, y)
15
(a) (4, 8) (b) (1, −6) (c) ( 12 , − 32 ) (d) ( 43 , 21 ) (e) (3a, b) (f ) (x + 1, 85 y)

Example 17
The vertices of a triangle are P (2, −3), Q(10, 3) and R(0, 9).
Find the coordinates of the point where the medians16 of the triangle meet.
Solution
y
6 The midpoints of the sides are
at L(5, 6), M (1, 3) and N (6, 0).
R(0, 9)
We can choose any two
b
EJ b
EJ b
E J bb medians. MQ is an obvious
E J bL choice for one, as its equation
E bb
E
J
J
b
b is simple: y = 3.
G b Q(10, 3)
ME
E b
We’ll make R N the other: with
J
J

a gradient of − 32 its equation
E J
E
must be
J
J - x
E
N
3
E
E y = − x + 9.
E 2
P (2, −3)
There is no need to simplify this as we are going to solve it with y = 3.
3
3 = − x + 9 or x = 4.
2
The medians meet at G(4, 3).
When it comes to solving the simultaneous equations we won’t always have

a line parallel to one of the axes. In this example we could have chosen P L
(instead of MQ). You’ll see that its gradient is 3 so its equation is
y+3
= 3 which simplifies to y = 3x − 9.
x−2
Solve this with the equation of R N to show we get the same result as before.
Tackling the next example will be very illuminating; we’ll see how different
approaches are possible. They all arrive at the same result of course, but
you’ll be able to see what they each offer in terms of insight and
computational simplicity. (We all like things to be as simple as possible.)
C@ centroid
C @
C @
The line joining the vertex of a triangle to the midpoint aa C @

median
@
C
of the opposite side is called a median. The three
aa
C aa @
16 medians of any triangle meet at the same point—called C aa@

the centroid of the triangle.
C aa @

Example 18
The points A(−2, 6), B (1, 4) and C (µ, −2) lie on a straight line. Find the value
of µ.
Solution
We are going to solve it in four ways. This is not masochism. Seeing the same
problem from different viewpoints helps to deepen our understanding,
which boosts our confidence.
Method 1: Equating gradients.

Since the three points lie on a straight line, the gradient of AB is the same as
the gradient of BC , so we can write
4−6 −2 − 4
=
1+2 µ−1
−2 −6
=
3 µ−1
−2(µ − 1) = −18
µ = 10.
Method 2: Using the equation of the line.
The equation of the line (using the points A and B ) is

y −6 4−6 2
= = −
x +2 1+2 3
3(y − 6) = −2(x + 2)
3y + 2x = 14
But C lies on this line, so substituting in its coordinates (the coordinates
must satisfy the equation):
−6 + 2µ = 14 or µ = 10.
Method 3: Vector approach.
Drawing sketch is a pretty much essential with the last two methods.
•
Q
A(−2, 6)
−2
Q
Q
? -Q• B (1, 4)
Q
+3 Q
QQ
...
...
−6 ...
Q
Q
Q
Q
? -Q• C (µ, −2)
+?
In moving from A to B we go down 2 and along 3. (The gradient is − 23 .)

In moving from B to C we go down 6—three times as far as when we moved
from A to B .
The corresponding horizontal movement from B to C must therefore be 3 ×
3 = 9. C is 9 units to the right of B , that is µ = 10.
Hopefully this makes untuitive sense; in Chapter 13 we cover vectors in more

detail.

Method 4: Similar triangles.

Geometrically, the two triangles in the diagram are similar, the lower one
being an enlargement of the upper one, with scale factor 3 (think of the 2 unit
length being enlarged to become 6).
It follows that the 3 unit length becomes 9, making µ = 10.
It may seem like overkill to consider four different methods. It is certainly true
that you will only use one on a particular problem, but there are good reasons
to be familiar with a range of approaches.
First, different methods lend themselves more easily to different types of
problem: one approach may be quite tedious to apply and another very
simple. The second reason is that each one provides a slightly different
perspective on the underlying principles; understanding different approaches
gives you deeper insight into them, and that means you can answer harder
questions.

5.6 Circle Theorems

Before we begin a quick reminder of some geometrical theorems that you will
need.
Perpendicular bisector of a chord

@
@ Each of the diagrams opposite shows the chord
t
@
@• C of a circle and its perpendicular bisector (the
@ line through its midpoint, at right angles to it).
@ The theorem is that
@
@
the perpendicular bisector of a
chord passes through the centre
of the circle.
In the bottom circle we can see that it’s

• C obviously true, by symmetry. If we rotate this
t circle we get the diagram at the top—where
the result is less self-evident.
Circumcircles and circumcentres

The diagram opposite shows
the circle passing through all
the vertices of a triangle (there
will only be one circle that does
this). Such a circle we call the
circumcircle of the triangle, and HH
its centre the circumcentre. The HH
CH•
circumcentre of a triangle (the H H

centre of the circle that passes
H
H
H
H
through its vertices) may be H

outside the triangle, like in the one
opposite.
Visualise the third perpendicular
bisectors, also passing through C .

The angle in a semicircle

An important special case is when the
circumcentre of the triangle lies neither
inside nor outside but actually on one of
its sides. A

A
If this is the case then that side is a A
diameter of the circle, and we can say C A
t
• A
two things:

(a) the centre of the circle will be at its

midpoint;

(b) its opposite angle is a right-angle.
(You may recall the geometrical
theorem that says: the angle in a
semicircle is a right-angle.)
Tangent properties
B
t
@
@
@
@• C
@

D

A
The diagram shows a tangent from the point A

touching the circle at B .
C is the centre of the circle.
The radius through B is perpendicular to the tangent at B .
We can therefore use Pythagoras to find the length of the tangent AB if we
know the coordinates of A and C , and the radius of the circle.
The tangents from A to the circle are the same length: AB = AD.

Example 19 The vertices of a triangle are P (−2, −2), Q(7, 1) and R(5, 5). A
circle passes through all three vertices.
Find the coordinates of the centre of the circle, and its radius.
Solution.
The perpendicular bisectors of the sides meet at the centre of the circle.
First we draw a diagram showing the key lines. I hope you agree it is a
beautiful one.
y
6
R(5, 5)
AA
A
@
@ A
@ A•L(6, 3)
t
@ A
@ M(3, 3) A
2 2
@• A
t
Q(7, 1)
@•
A
@
C
@

- x

@
@

@
@

P (−2, −2)
The gradient of RQ is −2 and its midpoint L(6, 3), so its bisector is

1
y − 3 = (x − 6),
2
This simplifies to 2y = x (you may want to check this).
Similarly the gradient of P R is 1 and its midpoint (using the formula) is
µ ¶
3 3
M , ,
2 2
so its perpendicular bisector has gradient −1 and equation
µ ¶
3 3
y − = −1 x − ,
2 2
which simplifies17 to y = 3 − x.
Solving the equations
2y = x and y = 3−x
simultaneously tells us the centre is at C (2, 1).
For the radius of the circle we need to pick one of the vertices, say R, and
the radius r is the distance between (2, 1) and (5, 5).
That is r 2 = (5 − 2)2 + (5 − 1)2 = 25. The radius of the circle is 5.
You might have spotted that in this instance Q would have been a better
vertex to choose for finding the radius, as its y-coordinate happens to be
1—the same as C ’s. This confirms that the radius is 5.
17
(×2) 2y − 3 = −2 x − 23 = −2x + 3.
¡ ¢
Alternatively (and slightly more simply) y − 32 = −x + 23 ⇒ y = −x + 3

The final example in this section takes us into the real world.
Example 20
A solid cylinder rests with its axis horizontal on two stiff wires 40 cm apart
running parallel to its axis. The cylinder supports a horizontal sheet of wood
60 cm above the level of the wires. Find the radius of the cylinder.
Solution
The diagram is a cross section view. We have chosen to run the x-axis
through the level of the wires, and the y-axis through the centre of the
sphere. This puts the wires at (20, 0) and (−20, 0) and the sheet of wood
resting on the line y = 60.
To find the radius of the circle
y
we need to find where its centre
6 is. By symmetry it lies on the
y-axis.
60
B
B It also lies on the perpendicular

B
B bisector of any chord—we have
B chosen the one through (10, 30),
B whose gradient is 13 (satisfy
B

B (10, 30) yourself this is correct).
t
B
B
B Its equation is
B
B 1
B y − 30 = (x − 10),
BB 3
• •
-
x
−20 0 20 or 3y − x = 80,
which intersects the y-axis at (0, 80

3
).
80 100
The radius of the circle is therefore 60 − 3 = 3 cm.

Some truly amazing relationships
This is far beyond the A Level syllabus but is so extraordinary that I cannot
resist telling you about it. We have seen how, for any triangle, there are two
“special" points on each of the sides:
(a) the feet of the three altitudes from opposite vertices (L, M , N );
(b) the midpoints of the sides (P , Q, R).
Now a reminder about the three “centres" created using these points:
- the altitudes meet at the orthocentre (H );

- the medians (drawn from the midpoints) meet at the centroid (G);
- the perpendicular bisectors (also drawn from the midpoints) meet
at the circumcentre (C ).
C@
@
C @
C @
C @
C @ Q
R a

aa C @
@
@ aa C
@ M

a•C a G @
a @
@ C aa
t
a @
N HH
@ C aa@
t
•H
H H @
CC a@
a
@
HH @ P
HH
H @
HH@
H@ @
@
t H@ @
H
@
L @
@ Q
R H

t
t
HH @
H @
HH
•HC @
HH @
t @
@
P
We are about discover something amazing about the 6 points, L, M , N , P , Q

and R.
We start by marking all these points on a single triangle.
We know we can construct a circle passing through any three points, for
example L, M and N , or P , Q and R. Add an extra point and something
special is happening if all 4 points happen to lie on the same circle.
All six points that we have identified lie on a circle.
@
@
@
•M
@
@
N • @
R • •Q
@
@
@
@
@
@
• • @
L P

But we haven’t finished yet. In fact we have barely begun. Now we are going
to put in the altitudes and mark on the diagram the midpoints of the lines
that join H to the vertices.
As you can see, these further three points also lie on the same circle.
For this reason it is known as the 9-point circle18 .
@
@
@
• @• M
@
@
N •HH @
HH @ Q
R • HH •
@
HH @
H
HH @
H @
• •
HH @
H
HH @
HH@
H@
H@
• • H
L P
There is more. Look at the diagram below. The three centres of the triangle
H , G and C will always lie on a straight line; moreover G always divides this
line in the same ratio: HG : GC = 2 : 1.
@
@
@
• @• M
@
@
N • @
R •

• H @ Q
•
@ @
@• O @
@• G @
@
•
@@• C • @
@
@
@
• • @
L P
Finally, how is this line connected with the 9-point circle? Well, the centre of
the 9-point circle also lies on the line, and is in fact at the mid-point of H and
C . We’ve marked it O.
Isn’t that amazing.
18
It has many other names, mostly honouring the numerous mathematicians who
claim to have been the first to discover it.

Exercises 19
1. Find the gradient of each of these lines. Where possible work the
answer out in your head.
(a) y = 6x + 2 (b) y = 7 − 2x (c) 5y = 2x + 1
(d) x = 4y + 5 (e) 2y + 8x = 1 (f) 6x = 2 − 5y
x y
(g) 2x − 2y + 11 = 0 (h) 2x + 2y + 11 = 0 (i) =
3 2
x y x 2y
(j) 3(x + y) = 2 (k) − = 1 (l) + =1
5 10 9 3
2. Put these lines in ascending order, according to their gradients.
(a) y = 21 x (b) y = 1 − 2x
(c) 5y = 1 + 2x (d) 5y + 2x = 1
(e) x = 3y + 6 (f ) 5x + 3y = 2
3. These two lines are parallel:

6y − βx = 2
x = 3y + 1.
Find the value of β.
4. Find the equation of each line, given the information about the point(s)
is passes through and its gradient. Express the answer in a way that does
not include any fractions.
1
(a) Through (0, 1), gradient 4. (b) Through (0, 3), gradient − .
2
1
(c) Through (1, 5) and (4, 8). (d) Through (2, 4), gradient .
3
(e) Through (0, −2), parallel to the line y = 3x − 1.
(f) Through (−2, 0), parallel to the line y = 3x − 1.
(g) Through (2, 2) and (−1, −3).
(h) Through (−2, 7), perpendicular to the line 2x + 3y = 1.
5. Find the distance between each pair of points P 1 and P 2 . Write your
answer as simply as possible. (Do not use a calculator.)
P1 P2 P1 P2
(a) (1, 0) (4, 4) (g) (3, 2) (−2, 1)
(b) (−5, 2) (3, 8) (h) ( 41 , − 41 ) (− 34 , 34 )
(c) (− 12 , −1) (2, 5) (i) (−2, 1) (4, 4)
(d) (2, −12) (9, 12) (j) (30, −60) (80, 60)
p p
(e) (−0.1, 0) (−0.5, 0.3) (k) ( 11, 0) (0, 5)
p
(f) (1, 1) (5, 5) (l) ( 32 , 2) (1, 0)

6. Find the distance between each pair of points A and B . Write your
answer as simply as possible.
A B
(a) (a, 1) (a + 4, 4)
(b) (5p, p) (p, 4p)
(c) (k − 1, 2k − 1) (k + 1, 2k + 1)
(d) (t , 3t ) (6t , 15t )
(e) (q, −p) (−2q, 2p)
1
(f) −( λ , 0) ( λ3 , λ3 )
7. For each pair of points P and Q, find
(i) the coordinates of the midpoint of PQ,
(ii) the gradient of PQ, and
(iii) the equation of the perpendicular bisector of PQ.
In part (ii) your answer should be in a form that does not contain any
fractions.
P Q P Q
(a) (2, 3) (6, 7) (f) (0, 1) (3, 5)
(b) (4, 0) (−2, 2) (g) (2, −3) (6, −2)
(c) (−1, 2) (1, 6) (h) (−1, 1) (−2, 6)
(d) (−5, −4) (−1, 2) (i) (0.2, 0.1) (0.6, 0.3)
(e) (11, 2) (3, 6) (j) (3, 12 ) (1, 2)
8. In each of the examples below, A is one end of a line segment AB and

M is its midpoint. Find the coordinates of B .
A M B
(a) (2, 1) (4, 5)
(b) (7, −2) (8, −2)
(c) (−3, 12 ) (0, 0)
(d) (1, 1) (3, 32 )
(e) (a, b) (a + 1, b + 2)
9. The three points P (1, 3), Q(3, 7),R(6, k) lie on a straight line.
(i) Find the gradient of the line.
(ii) Write down an expression, in terms of k, for the gradient of QR.
(iii) Hence, or otherwise, find the value of k.
10. Find the perpendicular distance from the origin to the line
4y + 3x = 20.
11. Find the perpendicular distance

(a) from the point (6, 5) to the line x + y = 3,
(b) from the point (2, 6) to the line y = 12 x.

12. A triangle is formed by three points R(3, 1), S(−1, 1) and T (4, 6).
(i) Find the area of the triangle.
p
(ii) Show that the length of RT is 26.
(iii) Use your answer to part (ii) to
find the perpendicular distance from
the point S to the linea RT .
a
That is to say the line segment RT
extended infinitely in both directions.
y
6 T (4, 6)

S(−1, 1) R(3, 1)
-x
13. P and Q are the points (1, 0) and (3, 4) respectively. The lines through P
and Q perpendicular to PQ meet the line x = 0 at R and S.
Show that the length of RS is 5.
14. ABC D is a rectangle. The coordinates of three of its vertices are

A(−2, 1), B (4, 3) and C (5, 0). Find
(a) the position of the fourth vertex, and
(b) the area of the rectangle.
15. From the point C (7, 4), a straight line is drawn perpendicular to
x y
+ =1 y
4 8 A
meeting it at N . Find BAl
6
Al
A l
(a) the coordinates of N , A l
(b) the lengths of C N and AB , A l
l
A
l C (7, 4)
(c) the area of the triangle ABC , A
t
NA

(d) the length of AC , and A
(e) the perpendicular distance from A
- x
AAA
B on to the line AC .
16. Show the the triangle formed by the origin O and the points A(10, 0) and
B (6, 8) is isosceles.
(i) Write down the equation of the altitude 19 through B .
(ii) Find the the equation of the altitude from O to AB .
(ii) Find where the altitudes of the triangle meet20 .
17. Show that there is a circle that passes through the fours points (1, 0),
(5, 0), (5, 8) and (−1, 6).
18. Find the centroid of the triangle with vertices (0, 4), (2, −2) and (4, 10).
19
The line through B perpendicular to O A.
20
The orthocentre.

19. Find the centre of the circle that passes through (−2, 0), (0, 4) and (2, 6).
20. The two lines 2y + α x = 1 and 8y − α x = 5 are perpendicular. Find the

possible values for α.
21. Show that the triangle PQR is right angled, where P , Q and R are (0, −2),
(2, 2) and (0, 3) respectively.
Hence write down the coordinates of the centre of the circle that passes
through the vertices of the triangle.
22. X and Y are the points (1 + α, 0) and (1 − α, 2α).

(i) Find the midpoint of X Y .
p
(ii) Show that X Y = 2 2 α.
23. These three lines x−y = 1 (5.3)

x+y = 7 (5.4)
3y = 7x + 1 (5.5)
intersect to form a triangle. Show that the triangle is right angled and
hence, or otherwise, find its area.
q q−p p q+p
24. Find the distance between the points ( , p ) and ( p−q , p ).
p−q
25. The vertices of a triangle are A(2, −1), B (4, 3) and C (−2, 7). The
midpoint of AC is M and the midpoint of BC is N .
(i) Find the coordinates of M and N .
(ii) Show that M N is parallel to AB .
(iii) Show that AB = 2M N .
26. Find the centroid of the triangle joining the points P (−1, 2), Q(7, 0) and
R(9, 10).
27. The three points R, S and T lie on a straight line, with S. between R
and T . The coordinates of R and S are (1, 3) and (7, 7) respectively and
RS : ST = 1 : 2. Find
(a) the gradient of the line,

(b) the equation of the line, and
(c) the coordinates of T .
28. ABC D is a rectangle in which AB = 10 cm and AD = 6 cm. The line

through B perpendicular to AC intersects C D at T .
Show21 that DT : T C = 16 : 9.
D T
T "" "C
T"
" tT
6 cm "
"
T
" T
"
" T
"
" T
A 10 cm B
29. ABC D is a rectangle in which AB = a and AD = b. The line through B
perpendicular to AC intersects DC at T .
Show that DT : T C = a 2 − b 2 : b 2 .
21
Hint: use coordinate geometry with the origin at A.

30.∗ (i) The relationship

y = λx + (1 − λ)
describes a set of straight lines (one for each value of λ). Show that all
the lines pass through the point (1, 1).
(ii) Another set of lines is described by
y = µx + (3 + 2µ).
Find the coordinates of the point through which all these lines pass.
(iii) Find the equation of line that belongs to both sets.
31. A lamp shade is in the form of a right circular frustum22 . The diagram
shows a vertical cross-section, and gives its dimensions.
The shade will be fixed to a lamp
20 cm
- stand holding the light bulb. The

6 B
B designer has specified that the middle
B of the bulb should be at the centre of
B the sphere that touches both of the
60 cmB
B circular openings.
B
B Find the correct position for middle of

?
-
B the bulb.
80 cm
32. A sphere rests in a groove formed from two sheets of metal, one
vertical and the other at 45◦ to it. The diagram shows a vertical cross
section view.
The sheets meet at O, and the sphere touch

them at A and B , the point B being 4 cm above
O.
A (a) State the distance between B and the
vertical
p sheet and hence show that OB =
B 4 2 cm.
6
4 cm (b)
p Show that the radius of the sphere is 8 −
4 2 cm.
?
O (c) How far is the highest point on the sphere
above the level of O?
22
A frustum is a cone with the top “sawn off". The plane created by the “cut" must
be parallel to the plan containing the base. In this case the planes are both
horizontal. The axis of symmetry need not be perpendicular to these planes; the
fact that is a right circular frustum means that it is.

p
33. The vertices of a triangle are O(0, 0), H (4, 0) and K (2, 2 3). M is the
midpoint of OH . The perpendicular from M to H K meets the line at
N.
y
6 K
T
T
T
T
T
TN
t

"" T
"
" T- x
O M H
(a) Prove that ∆OH K is equilateral.
(b) Show that the equation of H K is
p
y = 3(4 − x).
(c) Find the coordinates of N .
(d) Hence, or otherwise, show that
the length of H N is 1.
34. Show that the distance between the points (1, −k) and (k 2 , k) is k 2 + 1.

Solutions 19
2 1
1. (a) 6 (b) −2 (c) 5 (d) 4 (e) −4 (f ) − 65
2
(g) 1 (h) −1 (i) 3
(j) −1 (k) 2 (l) − 23
2. (b), (f), (d), (e), (c), (a).
3. β = 2.
4. (a) y = 4x + 1 (b) 2y + x = 6 [y = − 12 x + 3]
(c) y = x + 4 (d) 3y = x + 10 [y − 4 = 13 (x − 2)]
(e) y = 3x − 2 (f) y = 3x + 6 [y − 0 = 3(x + 2)]
−3−2 5
(g) 3y = 5x − 4 [gradient is −1−2 = 3 ⇒ y − 2 = 53 (x − 2)]
(h) 2y = 3x + 20 [y − 7 = 23 (x + 2)]
13 5 169
5. (a) 5 (b) 10 (c) 2 [( 2 )2 +62 = 4 ]
p p p p
(d) 25 (e) 0.5 (f ) 4 2 [ 32 = 16 2]
p p p p p p
(g) 26 (h) 2 (i) 3 5 [ 45 = 9 5]
3
(j) 130 (k) 4 (l) 2
6. (a) 5 (b) 5p [(3p)2 + (−4p)2 = 25p 2 ]

p
(c) 2 2 (d) 13t
p 5 2 2 2
(e) 3p 2 [18p 2 = 2.9p 2 ] (f ) λ [ λ3 2 + λ4 2 = λ5 2 ]
7. (i) (ii) (iii) Working

(a) (4, 5) 1 y +x =9 y − 5 = −(x − 4)
(b) (1, 1) − 13 y = 3x − 2 y − 1 = 3(x − 1)
(c) (0, 4) 2 2y + x = 8 y = − 12 x + 4
3
(d) (−3, −1) 2 3y + 2x + 9 = 0 y + 1 = − 23 (x + 3)
(e) (7, 4) − 12 y = 2x − 10 y − 4 = 2(x − 7)
(f) ( 32 , 3) 4
3 8y + 6x = 33 y − 3 = − 34 (x − 32 ) = − 34 x + 89
(g) (4, − 52 ) 1
4
2y = 8x − 37 y + 52 = 4(x − 4)
(h) (− 32 , 27 ) −5 5y = x + 19 y − 72 = 15 (x + 23 ) = 51 x + 10
3
1
(i) (0.4, 0.2) 2 y + 2x = 1 y − 0.2 = −2(x − 0.4)
(j) (2, 54 ) − 34 12y = 16x − 17 y − 54 = 43 (x − 2) = 34 x − 83
8. (a) (6, 9) (b) (9, −2) (c) (3, − 21 ) (d) (5, 2) (e) (a + 2, b + 4)
k−7 k−7
9. (i) gradient = 42 = 2 (ii) 3
(iii) 3
=2 ⇒ k −7 = 6
⇒ k = 13.

10. Distance d = 4.
3
Gradient of the line 4y + 3x = 20 is − 4 .
4
Perpendicular line through the origin is y = 3 x or 3y = 4x.
Solving simultaneously (by any method), for example
12y + 9x = 60 and 12y = 16x ⇒ 16x + 9x = 60.
12 4 12 16
Hence x = 5 and y = 3 . 5 = 5 .
12 16 42 32 4 2 42 32 42
d 2 = ( 5 )2 + ( 5 )2 = 52 + 52 = 42 ( 52 + 52 ) = 42 .
p
11. (a) 4 2.
Gradient of the line is −1, perpendicular line from (6, 5) is
y = x + 1. The two lines intersect at x = 2, y = 1.
p p p
Distance between (6, 5) and (2, 1) is 42 + 42 = 16 2.
p
(b) 2 5.
Gradient of perpendicular line is −2, and its equation is
y = −2x + 10. The lines intersect at (2, 4).
p p
Distance is 22 + 42 = 20.
12. (i) Area = 12 × 4 × 5 = 10. [Height of T above SR is 5.]
p
(ii) RT = 26. [RT 2 = 12 + 52 .]
20 p
(iii) p or 3.92 (3 S.F.) [ 21 26 × h = 10.]
26
13. Gradient of PQ is 42 = 2.
y
Lines through P is:
6
y = − 12 (x − 1), i.e. 2y + x = 1, SHHH
which cuts the y-axis at (0, 21 ).
H
HH
Lines through Q is:
H t Q(3, 4)

y − 4 = − 12 (x − 3), i.e. 2y + x = 11,

which cuts the y-axis at (0, 11

2
).

The distance RS is therefore = 5.

[Can you find RS by trigonometry?] R HH
H t - x
O P (1, 0)
14. y
6
• B (4, 3)
B
B

A(−2,
1)
−3 B
• B
?-B•C-
B
(5, 0)
0 +1 x
−3
?- • D
+1
1
AB and BC are perpendicular as expected (gradients 3
and −3).
(a) D is at (−1, −2).
The simplest method is to think of moving from B to C : this involves
going down 3 and 1 to the right. The same must apply from A to D.
(b) Area of rectangle = 20.
p p
AB 2 = 62 + 22 , so AB = 2 10. BC 2 = 12 + 32 , so BC = 10.

15. (a) N is (3, 2).

Equation of AB can be written 2x + y = 8 (multiplying through by 8),
so its gradient is −2 and the gradient of C N is 12 .
Equation of C N is therefore y − 4 = 21 (x − 7) or 2y = x + 1.
Solving these two equations (e.g. by substitution) gives x = 3, y = 2.
p p
(b) C N = 2 5, AB = 4 5.
C N 2 = 42 + 22 . AB cuts the axes at x = 4 and y = 8,
so AB 2 = 42 + 82 .
p p
(c) Area = 20, because 12 × AB ×C N = 12 × 4 5 × 2 5.
(d) AC = 5 [AC 2 = 32 + 42 ].
(e) Distance = 8. Area = 20, so 21 × 5× height = 20.
16. OB 2 = 62 + 82 = 100, so
O A = OB = 10. ∆O AB is isosceles. y
(i) x = 6. 6
B (6, 8)
(ii) y = 12 x. AA
A
Since the triangle is isosceles with A• N
O A = OB , ON is a line of symmetry, A
A
where N is the midpoint of AB , A
i.e. (8, 4). A -x
O A(10, 0)
(iii) Orthocentre is at (6, 3).
We label the point P ,Q,R and S.

17.
y Our strategy will be to find the centre
6 R(5, 8) of a circle passing through three of the
•
points, and verify that the fourth point

S(−1, 6)
• lies on the same circle.
If we pick P ,Q and R, finding the
C (3, 4)
centre of its circumcircle is easy: the
perpendicular bisectors of PQ and QR
are x = 3 and y = 4, i.e. the centre is at
C (3, 4).
• • -x
P (1, 0) Q(5, 0) Then if the radius is r (= C P )
p
r 2 = 22 + 42 = 20; r = 2 5.
p
It remains to check that SC = 2 5 (or if you like that SC 2 = 20).
This is easily done: (3 + 1)2 + (4 − 6)2 = 16 + 4.
A slightly neater way to show that S lies on the circle is to use the
vector approach: to get from C to P , from C to Q or from C to R
we move 2 units parallel to one axis and 4 parallel to the other.
The same applies to S

18. The centroid (where the medians meet) y

is at (2, 4). 6
R(4, 10)
Labelling the vertices P , Q and R, and
the midpoints opposite them P 0 , Q 0 and R 0 ,

• Q 0 (2, 7)

the diagrama points to P P 0 and QQ 0 as

P (0, 4) •P 0 (4, 4)
the simplest medians to choose.
J
J
J• R 0 (1,1) - x
a
To see this the diagram has to be J
reasonably realistic. J Q(2, −2)
19. Centre is at (9, −3).

Line between (−2, 0) and (0, 4): midpoint is (−1, 2), gradient is 2.
⊥ bisector is y − 2 = − x1 (x + 1) or 2y + x = 3.
Line between (0, 4) and (2, 6): midpoint is (1, 5), gradient is 1.
⊥ bisector is y − 5 = −(x − 1) or y + x = 6.
Simultaneous solution of the two equations is x = 9, y = −3.
20. α = ±4.
α 1
2y + α x = 1 can be written y = − x+ ;
2 2
α 5
8y − α x = 5 can be written y = x+ .
8 8
α α
Since the lines are perpendicular, (− )( )= −1, or α2 = 16.
2 8
21. Centre is at (0, 21 ). y
Gradient of PQ is 2; gradient of QR is − 12 , 6
so PQ ⊥ QR. R(0, 3)
H
P R must therefore be a diameter of the HH
Q(2, 2)
circle through P , Q and R (angle in a

semicircle), so the centre is at its midpoint, -x
the y-coordinate of which is 12 (3 +− 2).

R(0, −2)
22. (i) (1, α).

(ii) X Y 2 = [(1 + α) − (1 − α)]2 + (2α − 0)2
= (2α)2 + (2α)2 = 8α2 .
23. Solving (6.13) and (6.14) simultaneously, their lines meet at A(4, 3).
Similarly the lines (6.13) and (6.15) meet at B (−1, −2), and
the lines(6.14) and (6.15) meet at B (2, 5),
7
Gradients of the sides of the triangle are AB : 1; BC : ; C A: −1.
3
AB ⊥ C A. Area of ∆ABC = 12 × AB ×C A.
p p
AB 2 = 52 + 52 = 50, AB = 5 2. C A 2 = 22 + 22 = 8, C A = 2 2.
p p
Area of the triangle = 21 × 5 2 × 2 2 = 10.
p
24. Distance = 5. (It is independent of p and q.)
p q 2 q+p q−p 2
Let the distance be L, then L 2 = ( − ) +( − )
p−q p−q p p
Taking each bracket in turn:
p q p−q q+p q−p q+p−(q−p) 2p
− = = 1, and − = = = 2.
p−q p−q p−q p p p p

25. (i) M (0, 3),N (1, 5). y

(ii) Gradients of M N and AB are both 2, 6
C (−2, 7)
so M N //AB . Q
A Q
(iii) Use the formula to find M N and AB , A Q • N (1, 5)
p p A QQ
giving M N = 5 and AB = 2 5. A Q
A•M (0, 3) Q B (4, 3)
Or (better) use a vector approach: to move A

from M to N , we go 1 along and 2 up. A
A - x
From A to B , we go 2 along and 4 up. A A(2, −1)
(We don’t need to find the lengths.)
26. Centroid is (5, 4).
If we label the midpoint opposite P as P 0 , etc, their coordinates
are P 0 (8, 5), Q 0 (4, 6), R 0 (3, 1).
Median P P 0 : gradient 13 , equation 3y = x + 7.
Median QQ 0 : gradient −2, equation y = 14 − 2x.
Median RR 0 : gradient 32 , equation 2y = 3x − 7.
Solving any pair simultaneously gives the position of the centroid.
2
27. (a) Gradient is 3 . (b) Equation is 3y = 2x + 7. (c) T is (19, 15).
To move from R to S we go
6 to the right and 4 up. •
T
The ratio RS : ST = 1 : 2,

so from S to T we must go •

S(7, 7)

12 to the right and 8 up. •
R(1, 3)
28. Let the distance be d , then

d 2 = (k 2 − 1)2 + (2k)2 = k 4 − 2k 2 + 1 + 4k 2 = k 4 + 2k 2 + 1
= (k 2 + 1)2 .


• Understanding the Cartesian plane linking algebra and geometry.
The coordinate axes as the lines x = 0 (y-axis) and y = 0 (x-axis).
• Formulae for the equation of a straight line, given

(a) gradient m and y-intercept c: y = mx + c
(b) gradient and one point (x 1 , y 1 ): y − y 1 = m(x − x 1 )

y−y 1 y 2 −y 1
(c) two points on the line (x 1 , y 1 ) and (x 2 , y 2 ): =
x−x 1 x 2 −x 1
x y
(d) x-intercept a and y-intercept b: + = 1
a b
• Perpendicular lines: if their gradients are m 1 and m 2 , then m 1 m 2 = −1.
1 2 3
e.g. 2 and − 2 , 3 and − 2 , − 41 and 4
• Distances between (x 1 , y 1 ) and (x 2 , y 2 ): d 2 = (x 1 − x 2 )2 + (y 1 − y 2 )2
• Mid-point of the line segment between (x 1 , y 1 ) and (x 2 , y 2 ):

¡1 1
¢
2 (x 1 + x 2 ), 2 (y 1 + y 2 )
• Lines within triangles:medians, altitudes, perpendicular bisectors.

And the points where they meet: centroid, orthocentre, circumcentre.
• Circles. Theorems and properties.

Chapter 6
LINEAR RELATIONSHIPS
Overview
6.1 Linear equations. For example
7x 1 25.9 3y−2 1
7y − 1 = 3(y + 5), 8
+ 12 = 3x, x
= 0.88, 8
= 10 y.
6.2 Formulating linear equations. Solving problems by converting textual

information into mathematical equations.
6.3 Changing the subject of a formula. The technique follows on naturally

from the method for solving linear equations. Formulae such as
1−t 2
q
1 1 x−2 4
y = 4 x+ 2 , y= 2 , V = 3 πr ,
3
T = k a+x
a−x
, x = 2+t 2
6.4 Linear inequalities. For example

3x − 1 < 5, 2(1 − 3a) ≤ a + 1.
6.5 Simultaneous linear equations. Solving pairs of equations in two
variables. For example
x−1
x + y +1 =0 2(p − q) =1 3
=y
and 2x + 3y =1 and 4p − 5 =q and y − 4x =7
6.6 Combining linear inequalities. One variable and two variable case.
For example In the two variable case, solving using a graph.
6.7 Linear modelling. Practical examples.
6-1
CHAPTER 6. LINEAR RELATIONSHIPS
6.1 Linear Equations

Wherever Maths is applied to real life problems linear relationships and
linear equations are ubiquitous.
To take one example, economists build highly complex mathematical

models, capturing the interactions between economic variables, mostly1
using linear equations. A whole branch of mathematics has been developed
to study these relationships, which when combines with some statistical
techniques comes under the heading of Econometrics.
A linear equation (in the variable x) is one involving x terms and constant
terms only, like the ones on the previous page. It can’t contain, for example,
p
terms in x 2 , x1 , x or sin x.
This section is about making sure you can to solve any linear equation you
come across. If you are wondering whether you can skip this section, look at
the Exercises at the end. If you’re comfortable with all of them, go straight to
Section 5.2.
To solve an equation we need the last line to read “x = something". There is a

standard procedure for getting to this, which always works. The first thing we
have to do is
move every term that includes the variable onto one side;
move everything else onto the other side.
How do we do this? By using the principle that
we can do whatever we like to the equation,

provided we do the same to both sides.
To be precise, we do this
by repeatedly applying inverse operations.
We’ll see how this works in practice—and clarify what we mean by an inverse
operation—using examples. Don’t be put off by the fact that the first
example is trivially simple—it’s easy to work the answer out in your head. It’s
purpose is to spell out what our strategy will be, which we can then apply to
any linear equation. After that the examples get progressively harder.
1
Non-linear relationships are sometimes considered.

Example 1. Solve the equation 3a + 7 = 19.
Solution
Think of the equation like a set of weighing scales that are in balance. On the
left we have three boxes each weighing a kg and a box weighing 7 kg. They
are in exact balance with a box weighing 19 kg on the right.
19
a a a 7
Since we want to end with the statement “a =something", we need to get any
terms involving a onto the left hand side and anything not involving a on
the other side. We can do this by taking 7 kg from both sides.
When we do this the scales will still be balance, looking like this:
12
a a a
∆
Mathematically we can describe what we’ve done like this. We have put in
the bracket on the left what we are doing to each side of the equation:
(−7) 3a + 7 − 7 = 19 − 7
3a = 12.
The answer is now obvious: a = 4. But as it won’t always be obvious, so we
need to articulate the process for getting there. The two sides will still be in
balance if we reduce what’s on each side by the same fraction. In this case we
divide each side by 3, leaving a kg on one side and 4 kg on the other.
(÷3) 3a ÷ 3 = 12 ÷ 3
a = 4.
a 4
∆

Using the scales analogy helps to get us thinking in the right way. In practice
we’ll simply apply the steps mechanically. The first step is to get rid of the +7
on the left hand side. We do this by applying the operation −7 to both sides of
the equation, because this is the inverse (or opposite) operation to +7, it
“unwinds" it. Put another way, adding +7 and then adding −7 is equivalent
to doing nothing.
Next we want to get rid of the 3 in front of the a. Subtracting 3 wouldn’t work
this time because our starting point was a × 3 not a + 3. The operation that
‘unwinds" ×3 is ÷3, so we divide both sides of the equation by 3.
Once we understand what’s going on, we can abandon the scales analogy
and write the complete solution:
3a + 7 = 19
(−7) 3a = 12
(÷3) a = 4.
Writing in the steps—the (−7) and the (÷3)—is not essential, but experience
suggests it is a very good idea until you are quite fluent with the method.
Example 2. Solve the equation 7y − 4 = 2(y + 8).
Solution
There is a y term on r.h.s. which we will want to combine with the one on the
l.h.s., but it is “wrapped up" in a bracket, so the first thing we need to do is
multiply out the bracket.
7y − 4 = 2y + 16
Let’s start with the numbers. On the l.h.s. we have a −4. The inverse operation
to this is +4, so adding 4 to both sides we have
(+4) 7y = 2y + 20
Now we want to collect the y terms on the l.h.s. For this we need to get rid of
the 2y on the right hand side: we do this by applying the operation −2y, i.e.
by subtracting 2y from both sides.
(−2y) 5y = 20
(÷5) y = 4
We end with a division as the last step (the inverse operation to multiplying
by 5).

4x
Example 3. Solve the equation = 2x − 5.
3
Solution
The fun starts when we introduce fractions. The good news is that
we can always turn an equation that contains fractions into one that
doesn’t, by multiplying both sides by some number.
In this case we obviously multiply by by 3, and recalling what we learnt in

section 4.4 on page 4-11
4x 3 4x
3× = × = 4x
3 1 3
which allows us to get rid of the fraction:

(×3) 4x = 3(2x − 5)
4x = 6x − 15
Notice we had to “unpack" the bracket (multiply it out) so we could separate

out the term that involves x and the one that doesn’t.
Now think of the expression 6x − 15 not as a subtraction but as two numbers
6x and −15 added together. (See section 1.10.2 on page1-45.) The natural
next step would be to subtract 6x from both sides. That would get all the x
terms on the l.h.s and leave the only number on the r.h.s. However, because
6x is bigger than 4x it’s easier to collect the x’s on the right and numbers on
the left.
4x = 6x − 15
(−4x) 0 = 2x − 15
(+15) 15 = 2x
15
(÷2) = x
2
A brief diversion: suppose you had collected the x terms on the left?

Beginning as before:
4x = 6x − 15
(−6x) − 2x = −15
−15 15
(÷− 2) x = =
−2 2
We ended up with a negative coefficient of x in our penultimate line. Here
we divided by that negative number to get the answer. Instead we could have
multiplied both sides by − 1:
−2x = −15
−
(× 1) 2x = 15 etc.
or we could have swapped sides to get rid of the minus signs:

−2x = −15
(+15) −2x + 15 = 0
(+2x) 15 = 2x etc.
Use whichever approach you prefer.

3x x
Example 4. Solve the equation = 2− .
10 15
Solution
We could, of course, multiply through by 10 and then by 15, or simply
multiply by 150. A better choice would be 30, the LCM of 10 and 15. Notice
that we need to multiply every term by 30, not just the fractions.
30 3x 30 x
(×30) × = 30 × 2 − ×
1 10 1 15
9x = 60 − 2x
To “move" the −2x term over to the left hand side, we apply the inverse
operation, i.e. we add 2x to both sides.
(+2x) 9x + 2x = 60 − 2x + 2x
11x = 60
60
(÷11) x =
11
QQ
Solve these equations:
x
(a) 8x = 5(x + 3) (b) 2x = 4 + 5 (c) 17 − 3y = 2(y + 1)
2x x
(d) − = 1.
3 4
So far all the fractions have had arithmetic denominators. It’s just as easy to
deal with algebraic ones.
2 20 12
(a) x = 5 (b) 8x = x + 20 ⇒ x = (c) y = 3 (d) 8x − 3x = 12 ⇒ x = .
7 5

5
Example 5. Solve the equation 4 = .
y
Solution
Multiplying both sides by the number y:
(×y) 4y = 5
5
(÷4) y = or 1.25
4
3 1
Example 6. Solve the equation +4 = .
2x x
Solution
The LCM of x and 2x (the smallest number they both divide into) is 2x, so
(×2x) 3 + 8x = 2
(−3) 8x = −1
1
(÷8) x = −
8
When there is more than one algebraic denominator in an equation, the LCM
is very often simply their product. Look at the next example.

4
Example 7. Solve the equation = 5.
1−3x
Solution
Multiplying both sides by 1 − 3x (and remembering to put it in a bracket on
the right hand side):
(1 − 3x) 4
XXX
= 5(1 − 3x)
X
×X
1 1−
X3xX
X
4 = 5(1 − 3x)
4 = 5 − 15x
1
which quickly leads to the solution x = 15 .
2x x+1
Example 8. Solve the equation = .
6x−5 3x
Solution
We start by multiplying both sides by 3x and by 6x − 5. This is sometimes
called “cross-multiplying".
It appears to give a quadratic equation:
6x 2 = (6x − 5)(x + 1)
but multiplying out the brackets we see that the x 2 terms cancel each other
out:
6x 2 = 6x 2 + x − 5
(−6x 2 ) 0 = x −5
(+5) 5 = x
The next example is the sort of equation we come across routinely in

trigonometry.

Example 9. Solve the equation
h sin 60◦ = 12.

Solution
First sin 60◦ = 0.8660 (to 4DP), so the equation is
h × 0.866 = 12.
This is no different from an equation like 3a = 12, except that the answer is
not obvious. In both cases we just need to
divide by the number in front of the variable.
In the case of the equation 3x = 12 it was the 3, in the current example it’s the
0.866.
12
(÷0.866) x= = 13.86 (4 SF)
0.866
Don’t be put off by the fact that, despite dividing, we get an answer that’s
bigger than 12. Can you see that we will always get an answer bigger than the
number we started with if we divide by a number less than 1? For example
had we divided 12 by 0.5, we would be asking how many halves are there in
12. The answer is 24.
This example reminds us of a slip that is very easy to make. It’s incorrectly
solving an equation like this:
6x = 2.
Can you see what we are getting at? (See the footnote3 )
3
Then answer is x = 13 , not x = 3. We divide by the number in front of the variable.

There are lots of ways to skin a cat (as they say). This applies to solving
equations: the order we do things in can be changed. But only up to a point.
Take this example:
4x = 3 − 5x
Since we are aiming to get to an equation starting x = . . ., we might be tempted
to begin by dividing by 4. This isn’t going to get us very far because it still
leaves an x term on the right hand side:
3 − 5x
x=
4
Happily there is an order which always works (see below). It’s the order we
have used in all the examples. (An algorithm is an ordered sequence of steps
for solving a problem—like the instructions for putting together a piece of
self-assembly furniture.)

Algorithm for solving a linear equation
1. If there are any fractions, to get rid of them by multiplying (by the
LCM).
2. Expand any brackets.
3. Collect terms, by adding or subtracting.
4. Divide by the number in front of the variable.
For ease of memorising the steps can be summarised as
• Fractions
• Brackets
A
!
A
• Collect Terms
A
A
• Divide
Notice that dividing is the last thing we do.

If a step isn’t needed, we just miss it out.
We shall be returning to this list later, because with one or two
embellishments it also serves as the basis for a fool-proof algorithm for
changing the subject of a formula.
QQ

2 2 1 4x+1
(a) = 5 (b) = 1− (c) = 3
x 3x x x−2
x x−2
(d) = (e) 0.125y = 2
x+4 x
4 2 5
(a) x = (b) x = (c) x = −7 (d) x 2 = x 2 + 2x − 8 ⇒ x = 4 (e) y = 16.
5 3

Precision
A word about precision before we go on. This is the number of decimal points
or significant figures we use when writing down a number. It is different from
accuracy: an inaccurate answer is deficient in some way. The distance from
the earth to the moon is 240, 000 miles, a figure that is perfectly good enough
for me, but not for anyone planning a lunar landing! The distance varies of
course as it’s orbit is elliptical, and it depends on exactly where we measure
from and to; the precision needed depends on the context.
For our purposes if an answer comes out as a fraction we generally leave it as
a fraction, rather than convert it to a decimal. And there’s rarely anything to
be gained by converting a top-heavy (improper) fraction to a mixed fraction.
If an exact answer isn’t possible the question will normally state the level of
precision required e.g. “give your answer correct 5 to 2 SF" or “to 1 DP". If the
question doesn’t specify a level of precision, use you common sense, but it’s
generally safe to give the answer correct to 3 SF.
However—and this is important—whatever the precision required for the
answer, don’t round until the end, otherwise you risk introducing A
!
A
rounding errors. How much extra precision to keep in? There is no A
simple rule, but 2 or 3 extra figures is usually sufficient6 . A
5
For a reminder of the meaning of SF and DP turn to page A-17 in Appendix A
(or click on the page number).
6
This tells an examiner that you recognise the need for extra precision in working.

6.2 Formulating linear equations

A common type of exam question is one that requires the formulation—and
then solution—of an equation 7 .
What’s usually involved is
finding two expressions for the same quantity—and making them equal
To do this we will often need to define a new variable.
Sometimes we only need to find one expression because we know the value
of one of them already.
Bear in mind this is what we are trying to do when looking at the following
examples.
Example 10
There are two alternative tariffs available for a mobile phone. Under tariff A
there is a fixed monthly charge of £5 plus 2p per minute for all calls. Under
tariff B there is no fixed change, but all calls cost 4.5p per minute. At what
level of monthly calls does it becomes cheaper to go for tariff A?
Solution
This is the same as saying at what level of usage is the cost the same on both
tariffs. We’ll call that level N minutes per month. Now we can write down a
formula for what the cost would be under tariff A, and what it would cost
under tariff B. The two formula will give the same result, i.e. the same cost,
when the two expressions are equal. We call this the break-even point.
Under tariff A the total monthly cost will be (in pence) 500 + 2N . Under B it
will be 4.5N . The two tariffs will cost the same if
500 + 2N = 4.5N
This is easily solved to give N = 200 minutes per month. Above this level A is
cheaper; below it B is cheaper.
7
Sometimes we need to form, and solve, more than one equation.
For now we’ll confine ourselves to a single equation.

Example 11
Find two positive numbers whose difference and quotient are both equal
to 3.
Solution
The quotient of two numbers is the result of dividing one by the other. At first
sight, it appears we have two numbers to find. But we can easily reduce it to
one by calling the smaller one x and the larger one x + 3— incorporating one
of the pieces of information about the two numbers right at the outset. (We
could of course assign them the other way round, i.e. make the two numbers
x and x − 3.)
Now since the quotient is a number > 1 and both numbers are positive 8 , the
larger of the two must be in the numerator of the division, i.e.
x +3
=3
x
This solves easily to give x = 32 and the two numbers are 3

2
and 92 . A quick
check verifies that this is correct.
Sometimes a problem can appear to be more difficult just because of the

numbers involved. Look at this next example.
8
You might like to try and solve the same problem but this time with both numbers
negative.
Hint: does it still make sense to divide the larger number by the smaller?

Example 12
The problem is to find length of the hypoteneuse in a 30◦ , 60◦ , 90◦ triangle
that has a total perimeter of 20 cm. We are looking for an answer that is
correct to 3 SF.
Solution
We start by defining a variable: the obvious one to choose is the thing we are
trying to find, the length of the hypoteneuse, b. As with Example 1 we find
two expressions that must be equal. In this case there are two ways of
expressing the perimeter: one as a formula based on b, the other the value
we know it must have i.e. 20. (We always drop any units during the course of
a calculation, and re-introduce them into our final answer.)
A

c

b

30◦

B a C
We have assumed9 that B = 30◦ and A = 60◦ , and adopt the usual
convention: BC = a and C A = B .
Using SOHCAHTOA
a b
= cos 30◦ and = sin 30◦
c c
a = c cos 30◦ = 0.86603c and b = c sin 30◦ = 0.5c
We now have two expressions for the perimeter:
0.86603c + 0.5c + c and 20
so we can write
0.86603c + 0.5c + c = 20
2.36603c = 20
20
c = = 8.45298
2.36603
The length of the hypoteneuse is 8.45 cm (3 SF).
For a juicier example, turn to Appendix A, page A-18 (or click on the page
number).
9
The result would be the same if we made B = 60◦ and A = 30◦ .

Summary
All these examples illustrates the three steps in formulating and solving an
equation:
1. Define a variable to stand for what we are trying to find.
2. Find two expressions involving that variable and make them equal.
3. Solve the equation.
Later in the chapter we will meet problems in which there are two quantities
we don’t know (so we’ll need two variables). In that case the information
we’re given will lead to two equations, each involving both variables. We call
these simultaneous equations, because we need to find values for the
variables that satisfy both equations simultaneously.
Real life problems may require a great many variables and as many
equations. There are some techniques for solving such equations, but
sometimes a (computer based) numerical solution is needed.

Exercises 20
1. Solve these equations. In each case express the answer as an integer or

a fraction.
(a) 4x + 1 = 13 (b) 2y − 5 = 3 (c) 7 + x = 16
(d) 3 + 2a = 11a (e) 7y − 1 = 3(y + 5) (f ) 4(1 + 3x) = 2x + 9
(g) 8b = 4 − 2b (h) x = 1 − x (i) 18a − 2(5a − 3) = 0
1 1 2y
(j) 3y + 4 = 8y (k) 7t − = 3t + (l) =1
2 2 5
6 3 x 4
(m) =1 (n) 2x+ = 4x (o) + = 1
2x 2 3 3
y 4 7a 2
(p) 4+ = 2y (q) x− = 1 + 2x (r) 3 = a+
5 3 3
2y+1 y 2x 1 4 3
(s) − =0 (t) + =x (u) y = y + 1
4 8 5 3 7 14
7x 1 11 5
(v) + = 3x (w) y = − y
8 12 10 6
2. Solve these equations. If expressing the answer as a decimal, give it

correct to 3 S.F.
(a) 0.2x = 3 (b) 1.6y = 1 (c) 8 = 0.417x
3 1 1
(d) =8 (e) = (f ) 0.707L = 3.5
x y+1 4
1 y 2 25.9
(g) 0.03x = 8 (h) = (i) = 0.88
0.7 3 x
1
(j) 0.31 =
9y
3. Solve these equations. Give each answer as a rational number.

2 1 2 y
(a) 4x− = (x + 1) (b) (6y − 1) = = 2
3 3 5 5
1 3y−2 1
(c) 3z− (1 − z) = 0 (d) = y
2 8 10
2 1
(e) x(4x − 1) = (2x + 1)(2x − 3) (f ) =
x 3x
+1
2y+1 5 3 1
(g) =6 (h) + − =2
y 4x 2x x
4 1 11 2
(i) = (j) − =0
1+2x 2x x x+1
0.4 3 1 3
(k) = (l) =
y y−2 1−0.1x 0.1x
3a−1 1 x 2 +5
(m) = (n) x =
3a+4 2 x+1
4x 2 +5 3+x
(o) = 2x + 1 (p) 4x− =1
2x 2
2 1−3y 3 2x+1
(q) y = y− (r) − = 4−x
5 10 x 2

4. In the table below the sum of the numbers in each row is given in the
right hand column. Find the values of the variables a to e.
2 a 1 7
b −1 a 8
3a −b 3 c
c −2a −b d
1 ad e −4
5. Find the ages of Teddy and Harry by formulating and solving equations.
(a) In 11 years Teddy will be four times as old as he was last year.
(b) In twenty years time Harry’s age will be twice what it’s going
to be in 3 years.
6. The sum of three consecutive numbers is 78. Find the smallest of the
numbers.
7. When I divide a certain number into 10 the answer is one more than the
answer I get when I divide the same number into 6. Find the number.
8. Electricians Douggie and Elsa charge differently for the work they do.
Douggie charges a £50 call-out fee and then £35 per hour; Elsa’s
call-out charge is only £30 but her hourly rate is £45.
(a) Write down formulae for how much Douggie and Elsa would
charge (in pounds) for a job that takes h hours.
(b) They both estimated the same time for a job and quoted the same
charge. How much did they quote for the job?
9. Amira is taken to school by car and walks home, along the same route.
Her average walking speed is 4 km h−1 and the car travels at an average
speed of 12 km h−1 . The total time she spends travelling to and from
school each day is 40 mins.
(a) How far is the journey to school?
(b) On a particular day they leave for school at the usual time but
Amira wants to get there 5 mins earlier than usual. What would the
average speed of the car need to be in order to achieve this?
10. What is the earliest time after 3 pm that the hour hand and the minute
hand of a (12 hour analogue) clock are aligned?

Solutions 20
1. (a) x = 3 (b) y = 4 (c) x = 9 (d) a = 13

1 2 1
(e) y = 4 (f) x = (g) b = (h) x =
2 5 2
3 4 1 5
(i) a = − (j) y = (k) t = (l) y =
4 5 4 2
3 20
(m) x = 3 (n) x = (o) x = −1 (p) y =
4 9
7 1 1 5
(q) x = − (r) a = (s) x = (t) x =
3 2 3 9
14 2 2 3
(u) y = (v) x = (w) y = − (x) x =
5 51 3 5
1
In (k) we could get rid of fractions by adding to both sides.
2
3
2. (a) x = 15 (b) y = 0.625 (c) x = 19.2 (d) x =
8
14
(e) y = 3 (f ) L = 4.95 (g) x = 4.17 (h) y = or 0.467
3
(i) x = 29.4 (j) y = 0.358
In (e) the denominators must be the same so y + 1 = 4.
3 12 1 10
3. (a) x = (b) y = (c) z = (d) y =
7 13 7 11
5 1 7
(e) x = 1 (f) x = (g) y = (h) x =
3 4 8
7 1 11 4
(i) x = (j) x = (k) x = − (l) y = −
8 6 9 13
5
(m) x = 7.5 (n) a = 2 (o) x = 5 (p) x =
2
5 1 2
(q) x = (r) y = (s) x =
7 9 3
4 30
In (k) multiply (both numerators) by 10, to give =
y y−2
4. a =4 since a + 3 = 7,
b=5 since b − 1 + 4 = 8,
c = 10 since 12 − 5 + 3 = c,
d = −3 since 10 − 8 − 5 = d ,
e =7 since 1 − 12 + e = −4.
5. (a) Toby is 15. T + 11 = 4(T − 1)
(b) Harry is 14. H + 20 = 2(H + 3)

6. The smallest number is 25.

If the number is x, the three numbers are x, x + 1 and x + 2,
and their sum is 3x + 3. So 3x + 3 = 78.
More directly, the middle number is 78 ÷ 3 = 26.
7. The number is 4.
10 6
If x is the number then = +1 ⇒ 10 = 6 + x.
x x
8. (a) Douggie: 50 + 35h; Elsa: 30 + 45h.
(b) £120. 50 + 35h = 30 + 45h ⇒ h = 2.
9. (a) 2 km.
40 mins = 32 hour. Let distance between home and school be d km.
d d 2
+ = ⇒ d =2
4 12 3
(b) 24 km h−1 .
2 1
Normal journey time to school is = hour = 5 mins.
12 6
So need to halve the journey time, i.e. double the speed.
2 1
Or set up an equation. = , where v = new car speed.
v 12
180
10. minutes after 3.
11
We work out how many degrees each hand has moves in t minutes:
minute hand hour hand
360◦ in 60 mins 30◦ in 60 mins
1◦
6◦ in 1 min in 1 min
2
t◦
6t ◦ in t mins in t mins
2
(At each step, we either multiply or divide by our chosen number.)
If the hands are aligned at t mins past 3, then
t
6t = 90 +
2
which we solve in the usual way (12t = 180 + t , etc).

6.3 Changing the subject of a formula

If you are very good at rearranging formulae you may be able to skip this
section, but if you’re considering doing this check the examples in the
Exercises at the end, to make sure you can do the harder ones.
So why is this topic here? Because the process involved is an exact parallel of
the method we use for solving linear equations.
To change the subject of a formula, treat it like an equation,

in which everything except the subject variable is a number.
What do we mean by this? Take the formula10

v = u + at
Suppose we want to make t the subject. If we knew the values of u, v and a
(say 7, 25 and 3), finding t would be very straight forward:
25 = 7 + 3t
(−7) 18 = 3t
(÷3) 6 = t
The steps are the same ones we use for solving equations (see page 6-12):
• Fractions
• Brackets
• Collect Terms
• Divide
In this simple example we only needed the last two steps.
It’s a small step from here to the completely algebraic version. We treat all
the variables as if they were numbers, expect t , the variable we are trying to
find.
Example 13 Make t the subject of the formula v = u + at .
v = u + at
(−u) v − u = at
v −u
(÷a) = t
a
Next an example that uses all four steps.
10
It describes the velocity of an object, initially traveling with velocity u, after
accelerating steadily at a rate a for a length of time t .
You may have met it if you are taking M1, or in Physics.

2a
Example 14 Make x the subject of the formula y = .
a+x
Solution
Now a and y we regard as numbers and we treat the formula as an equation
in x.
× (a + x) y(a + x) = 2a [Fractions]
y a + y x = 2a [Brackets]
(−a y) y x = 2a − y a [Collect Terms]
a(2−y)
(÷y) x = y [Divide]
Notice that we factorised the numerator in the last line; this is not essential
but it is good practice.
QQ
Rearrange these formula. The required subject is in the square bracket.
π
(a) y = k(1 + x), [x] (b) v 2 = u 2 + 2as, [a] (c) L = , [x]
2x
1 3
(d) ax + b y = c, [y] (e) +y = , [t ]
2t t
The next example looks quite similar to these, but there is a crucial difference:
the subject variable appears more than once. We’ll see that this means we
have to introduce a new step.
11
y−k v 2 −u 2 π c−ax
(a) x = (b) a = (c) x = (d) y =
k 2s 2L b
5
(e) Multiply by 2t to give 1 + 2t y = 6 ⇒ t =
2y

k+3x
Example 15 Rearrange the formula y = to make x the subject.
2k+x
Solution
We start in the usual way.
y(2k + x) = k + 3x [Fractions]
2yk + y x = k + 3x [Brackets]
y x − 3x = k − 2yk [Collect Terms]
This is the stage where we would normally divide, leaving x on the left hand
side. Because there are now two x terms we can’t do this.
The problem is solved by factorising the left hand side. Because we have
collected terms, it will always factorise.
x(y − 3) = k − 2yk [Factorise]

k(1 − 2y)
x = [Divide]
y −3
The factorising step we can think of as combining the two terms on the left
hand side into a single term, in this case an x term with coefficient (y − 3).
We’ll add this factorise step to the algorithm.

At the same time we’re going to “top and tail" the list with two more steps,
whose purpose will soon be clear. That completes the list; any rearragement
will yield to this algorithm.
• Roots
• Fractions
• Brackets A
!
A
• Collect Terms A
• Factorise
A
• Divide
• Roots

1 1 1
Example 16 Make v the subject of the formula + = .
u v f
Solution
This is a formula from optics. The subject variable only seems to appear
once, so it looks as if we won’t need to do any factorising. What we’ll see is
that after we’ve got rid of fractions v does appear more than once.
We start by multiplying every term by the LCM of u, v and f , which is uv f .
(We don’t know the values of these variables, so they have no known
common factor. The smallest number we can be sure they all divide into is
uv f .) Remember: treat u and f as if they are numbers12 .
v f +uf = uv [Fractions]
uf = uv − v f [Collect Terms]
uf = v(u − f ) [Factorise]
uf
= v [Divide]
u−f
Notice that we have chosen to collect the v terms on to the right hand side,
to avoid having a negative term , −u f , on the right hand side13 .
QQ
2 p px
(a) y = , [x] (b) y = , [x] (c) y = , [x]
1−x p+q x p+q x
1 1 a+x b
(d) + = a, [y] (e) = , [x]
x y b−x a
It’s time to see where the Roots steps at the start and finish of the algorithm
come in.
12
If you can’t immediately see what to do, try putting in some values for u and f ,
1 1 1
e.g. + = . Multiplying by the LCD 12v we have 3v + 12 = 4v.
4 v 3
13
This is not a must; it just makes life a bit simpler. See the Example on page 6-5
for a fuller discussion of this.
14
y−2 p(1−y) py x
(a) x = (b) x = (c) x = (d) y =
y qy p−q y ax−1
b 2 −a 2 (b+a)(b−a)
(e) x = b − a. Because a(a − x) = b(b − x) ⇒ x = = .
b+a b+a

q
l
Example 17 Make l the subject of the formula T = 2π g
.
Solution
The formula gives the period of oscillation T of a simple pendulum of length
l , where g is gravitational acceleration 15 .
Don’t be put off by all the symbols. Think of everything, except l , as a

number. You may even qwant to put in some numbers to make this crystal
l
clear, for example 5 = 6 10
.
We start by squaring (everything on) both sides to get rid of the square root.
³ ´
T 2 = 4π2 gl [Roots]
T 2g = 4π2 l [Fractions]
2
T
= l [Divide]
4π2
³ ´
A very common mistake is to write the first line: T 2 = 2π2 gl . It’s one of
those mistakes that, in isolation few student would make, but with so much
else to think about, easily slips under the radar.
So the Roots step at the beginning generally means get rid of any roots
(usually square roots). Now let’s look at an example where we need to use the
Roots step at the end.
2+t 2
Example 18 Make t the subject of the formula y= .
1+t 2
Solution
We want the last line of our rearrangement to start t = . . ., but the formula is
in t 2 . That’s not a problem: we’ll rearrange it making t 2 the subject and then
take the square root. (Or we could substitute t 2 = T , make T the subject and
then replace T with t 2 . It amounts to the same thing.) Notice in the first line
we need to put the 1 + t 2 in a bracket.
y(1 + t 2 ) = 2 + t 2 [Fractions]
y + yt2 = 2+ t2 [Brackets]
yt2 − t2 = 2− y [Collect Terms]
t 2 (y − 1) = 2 − y [Factorise]
2− y
t2 = [Divide]
y −1
s
2− y
t = ± [Roots]
y −1
15
A constant that is approximately 9.8 m s −2 . “Simple" means a mathematical
pendulum in which the “string" is weightless and the thing hanging on the end
has no size.

QQ

q
p A k
(a) L = x+y [y] (b) r = [x] (c) y = [x]
πh 1+x 2
y2
(d) x 2 + y 2 = a 2 [x] (e) x = [y]
1−y 2
That’s all you need at the moment. But if you want something to get your teeth
into before doing some exercises, read on.
q
16
k−y
(a) y = L 2 − x (b) A = πr 2 h (c) x = ±
y
q
p x
(d) x = ± a 2 − y 2 [Note that this is not a − y] (e) y = ±
1+x

p
Example 19 Make x the subject of the formula y = a x 2 + b2.
Solution
p p p
To start with, don’t be tempted to write x 2 + b 2 as x 2 + b 2 , i.e. x + b. If
you are not sure why this is wrong, see page 2-45.
The presence of the square root means that as it stands we can’t separate out
the x term. Remembering that we can do what we like to an equation as long
as we do the same to both sides, we can squaring both sides. This gets rid of
the square root.
We’re going to take this slowly, because there are some pitfalls to look out for.
First it’s important to remember that the whole of the left hand side and the
whole of the right hand side must be squared.
p
y 2 = [a x 2 + b 2 ]2 [Roots]
p
= a 2 ( x 2 + b 2 )2
y2 = a 2 (x 2 + b 2 )
y2 = a2 x 2 + a2b2 [Brackets]
y 2 − a2b2 = a2x2 [Collect Terms]
y 2 −a 2 b 2
= x2 [Divide]
a2 q
y 2 −a 2 b 2
x = ± a2
[Roots]
p
y 2 −a 2 b 2
x = ± a

kx
Example 20 Make x the subject of the formula y =p .
m(k 2 −x 2 )
Solution
k2x2
y2 = m(k 2 −x 2 )
[Roots]
y 2 m(k 2 − x 2 ) = k x 2 2
[Fractions]
m y 2k 2 − m y 2 x 2 = k 2 x 2 [Brackets]
2 2 2 2 2 2
my k = k x +my x [Collect Terms]
m y 2 k 2 = x 2 (k 2 + m y 2 ) [Factorise]
2 2
my k
= x2
k 2 +m y 2
[Divide]
r
m y 2k2
x = ± k 2 +m y 2 [Roots]
q
m
or x = ±k y k 2 +m y2
Example 21
(a) Find the surface area S of a solid cylinder of radius r and height h.
(b) Find a formula for h in terms of S and r .
(c) Find a formula for the volume of the cylinder in terms of S and r only.
Solution
Try not to be put off by all the symbols; remember they are just numbers
(whose values we don’t yet know). One other thing to mention: when doing
work involving π, we tend to leave it as the symbol π rather than give it a
numerical value.
(a) S = 2πr 2 + 2πr h.

S−2πr 2
(b) Rearranging this to make h the subject: h = .
2πr
(c) The volume is given by: V = πr 2 h, but we are asked to find a formula that
doesn’t contain h, so we get rid of it by substitution, using the formula we
πr 2
derived in part (b). Notice we write πr 2 as as a prelude to cancelling.
1
S − πr 2¶
µ
V = πr 2
2πr
Ã !µ
πr 2 S − πr 2
¶

=
1 π
2 r
r (S − πr 2 ) 1
= or r (S − πr 2 )
2 2

Exercises 21
1. In each case derive the second formula from the first.

iR i −I
(a) I = R+r , r = I
x2 y2 b
p
(b) a 2 + b 2 = 1, y = a (a 2 − x 2 )
x−2 2
(c) y = x , x = 1−y
bx+a yb−a
(d) y = b−x x = b+y
Er IR
(e) I = R+r r = E −I
1 1 1 Rx
(f) R = x + y y = x−R
2. Rearrange each formula to make the variable given in a square bracket

the subject.
(a) y = 41 x + 21 [x] (b) y = x+2
3
[x]
(c) E = mc 2
(c > 0) [c] (d) V = 34 πr 3 [r ]
k
(e) V = 13 πr 2 h (r > 0) [r ] (f) y = x 2 [x]
1 p
(g) I = 2
m(v 2 − u 2 ) (v > 0) [v] (h) y = a x + b [x]
q
l +x
(i) T = 2π g
[x] (j) A = π(R 2 − r 2 ) (R > 0) [R]
y 2 +1 y a
(k) W = y 2 −1 [y] (l) x = x−1 [x]
B B
(m) Y = A+ X [X ] (n) Y = A+ X 3 [X ]
1−t 2
q
(o) x = 1+t 2 (t > 0) [t ] (p) y = k a+x
a−x
[x]
1
3. Rearrange the formula y = 1+ 1 to make x the subject.
1+ 1+x
Check your answer by putting the value x = 0 into the above formula,
and substituting the resulting value of y into the rearranged formula.
4. Given that v = u + at and s = ut + 12 at 2 , find a formula for s in terms of

v, a and t .
1−t 1
5. If y = 1+t , where t = x , express x in terms of y.
6. The surface area and volume of a sphere are given by the formulae
4
S = 4πr 2 and V = πr 3 respectively.
3
Find a formula for V in terms of S.

Solutions 21
3−2y
q
E
2. (a) x = 2(2y − 1) (b) x = y (c) c = ± m
q q q
(d) r = ± 3 3V
4π
(e) r = ± 3Vπh
(f) x = ± ky
p
2I −mu 2 y−b 2 g T 2 −4π2 l
³ ´
(g) v = ± m
(h) x = a (i) x = 4π2
q
y
q
2
(j) R = ± A−πr
π (k) y = ± W +1
W −1 (l) x = y−a
q q
B
(m) X = Y −A (n) X = 3 Y B−A (o) t = ± 1−x
1+x
y 2 −k 2
(p) x = a y 2 +k 2
3−2y
3. x = y−2
1 2+x 2+x 3+2x
Numerator 1+ 1+x = 1+x , so y = 1+ 1+x = 2+x .
3
When x = 0, y = 2 .
3−2y 3
x = y−2 . When y = 2 , x = 0.
4. s = v t − 12 at 2 .
u = v − at , s = (v − at )t + 21 at 2 = v t − at 2 + 21 at 2 .
1−y
5. x = 1+y .
1− x1 x−1
y = = x+1 [Multiplying top and bottom by x.]
1+ x1
3 q
S2 1 S3
6. V = 1 or 3 4π . [No ± needed as V is a physical quantity.]
3(4π) 2
q
S S
¢ 12 S
¢ 23
r3 =
¡ ¡
r= 4π = 4π so 4π
3 3
S
¢ 32 4π S 2 S2
V = 43 π
¡
4π = 3 3 = 1.
(4π) 2 3(4π) 2

6.4 Linear Inequalities

Inequalities17 are like equations but instead of an = sign there will be one of
these 18 : <, ≤, > or ≥.
They might be described using phrases like
< “less than"

≤ “less than or equal to"; “no greater than"
> “greater than"
If you have skipped over the previous section on Linear Equations—because

you are comfortable with them—it may still be a good idea to remind
yourself of the algorithm we use to solve them. This is because with one
exception exactly the same algorithm solves inequalities. You will find the
method summarised on page 6-12. We’ll cone to the exception shortly; first
let’s see some examples.
An equation will have a specific number of solutions: a linear equation will
have one, a quadratic equation will have up to two, a cubic equation up to
three etc. An inequality can, and generally will, have an infinite number of
solutions.
Example 22 Solve the inequality 7x + 1 < 22.
Solution
We don’t need a formal method in this case: by inspection 7x must be less
than 21, so x must be less than 3. The solution is x < 3.
We’ll solve it using the formal method nevertheless, to prepare us for
examples where the answer is far from obvious.
7x + 1 < 22
(−1) 7x < 21
(÷7) x < 3
Or in set notation {x : x < 3}.

Represented own the number line

H
H ◦
−2 −1 0 1 2 3 4
Notice that we use an empty circle to indicate that the value x = 3 is not
included.
17
They are occasionally referred to as inequations.
18
If you don’t automatically know the difference between < and >, just remember
that the bigger number is at the “fat" end of the sign, and the smaller number
at the “thin" end.

Example 23 Solve the inequality 2x ≥ 3(10 − x).
Solution
We can’t see just by looking at it which values of x satisfy this inequality.
Multiplying out the bracket we have:
2x ≥ 30 − 3x
(+3x) 5x ≥ 30
(÷5) x ≥ 6
Let’s pause for a moment. We are saying that for every number bigger than 6,
it is also true to say that 2x ≥ 3(10 − x). Take x = 8 for example: 2x = 8 and
3(10 − 8) = 6; the l.h.s. is indeed bigger than the r.h.s. We are also saying that
if x = 6 the inequality also works: the l.h.s. is 12 and so is the r.h.s. It all
makes sense.
Now an example involving fractions. And after that, what is this exception we
referred to at the start of the section?

4x
Example 24 Solve the inequality ≤ 2x − 5.
3
Solution
(×3) 4x ≤ 3(2x − 5)
4x ≤ 6x − 15
(−4x) 0 ≤ 2x − 15
(+15) 15 ≤ 2x
15
(÷2) ≤ x
2
Compare this with Example 3. But for the sign in the middle every line is
identical.
So far everything we have done with linear inequalities carries over directly
from our work with linear equations.

So when does it not? What works for equations that doesn’t work for
inequalities? To see what the problem is, we’ll take an extremely simple
equation:
−x = 6
There are two ways to solve it. (Please bear with me, there is a point to it.)
(a) By adding and subtracting:
−x = 6
(+x) 0 = 6+x
(−6) −6 = x
(b) By multiplying
−x = 6
−
(× 1) x = −6
The two answers are the same, which is exactly what we’d expect to find.
Now let’s see what happens if we do the same with the inequality
−x ≥ 6
As before we’ll do it using both methods.

(a) By adding and subtracting:
−x ≥ 6
(+x) 0 ≥ 6+x
(−6) −6 ≥ x
or x ≤ −6
(b) By multiplying:
−x ≥ 6
−
(× 1) x ≥ −6 ???
The answers are different! What it tells us is that
if we multiply or divide an inequality by a negative number,

we must also reverse the sign.
Try this out for your self. Write a down an inequality you know to be correct,
like
−3 < 7 or − 2 < −1.
Multiply each side by −1 (or by any negative number). It only remains
correct if we also reverse the sign.

Let’s go back to Example 24 to see this rule in action.
4x
≤ 2x − 5
3
(×3) 4x ≤ 3(2x − 5)
4x ≤ 6x − 15
(−6x) − 2x ≤ −15
(×− 1) 2x ≥ 15 (reversing the sign)
15
(÷2) x ≥
2
The answer agrees with the one we got before, on page 6-34.
It’s important to stress that the sign doesn’t need to change if all we are doing
is adding or subtracting negative numbers, nor if we multiplying or dividing
by positive numbers—only if we are multiplying or dividing by negative
numbers.
We appear to have a method for solving any linear inequality. Well not quite,
as you will see in a later module. But we have everything we need for now.
QQ
x
(a) 8x < 5(x + 3) (b) 2x > 4 + 5
x 2x
(c) 17 − 3y ≥ 2(y + 1) (d) 4 − 3 ≤ 1
19 20
(a) x < 5 (b) 8x > x + 20 ⇒ x > (c) y < 3
7
12
(d) −5x ≤ 12 ⇒ 5x ≥ −12 ⇒ x ≥ −
5

Exercises 22
1. Solve these inequalities.

(a) 3x − 1 < 5 (b) 2 + 6y > 9y (c) 4x − 1 ≥ 2(x + 3)
3y
(d) 2(1 − 3a) < a + 1 (e) 3x + 2(4x + 1) ≤ 0 (f ) 2 > 1
2. Solve these inequalities.

4 1 2y
(a) x+ 5 ≤ 3x (b) 3 − 3 < 1
3 4x+1 x
(c) y+ 4 ≤ 2 + 3y (d) 3
−2> 0
2 1
(e) y < 9 − 3 y (f) 2.5x < −0.1
3 1
(g) 0.6t > 4 (h) 3z− 2 (1 − z) ≤ 0
3+y
(i) 3y(2y − 3) < (6y + 1)(y − 1) (j) 4y− 2 > 1
3 2x−1
(k) x − 2 ≤ 4 − x
3. In 14 years time Myra will be more than 3 times as old as she is now.
Write this statement in the form of an inequality and simplify it. Start
by assuming Myra’s current age to be x years.
4. (a) Show that if a and b are two numbers such that

a(b + 4) < b(a + 2)
then b > 2a.
(b) Comment on this reasoning:
a(b + c) < b(a + c)
⇒ ab + ac < ba + bc
⇒ ac < bc
⇒ a < b
5. Consider the inequality

x
<1
x +1
and whether it always holds true, whatever the value of x.
The following argument appears to prove that it does:
“The fraction must be less than 1 because the denominator is larger
than the numerator."
Find a value of x for which the fraction is not less than 1. (What we call
a counterexample.)
6. x and y are real numbers such that x > y > 0.
(a) Show that x + y > 2x.

(b) Hence, or otherwise, prove that
1 2 1
< <
x x+y y
justifying every step in your reasoning.
(c) Use the result in part (b) to show that
1 2
p <p < 1.
2 2+1

Solutions 22
2 7 1
1. (a) x < 2 (b) y < 3 (c) x ≥ 2 (d) a > 7
2 7
(e) x ≤ − 11 (f) y > 3
2 5 2
2. (a) x ≥ 5 (b) y > 1 (c) y ≥ − 8 (d) x > − 5
1 5 1
(e) y < 6 (f) x < −0.04 (g) t > 4 (h) z ≤ 7
1 5 6
(i) y > 14 (j) y > 7 (k) x ≥ 7
3. x + 14 > 3x. x > 7.
4. (a) ab + 4a < ba + 2b ⇒ 4a < 2b ⇒ b > 2a.

(b) Correct up to penultimate line, but cannot divide both sides by c,
because c might be negative.
5. If we multiply both sides by x + 1 we get the inequality x < x + 1, or 0 < 1,

which suggests the statement is true for all values of x.
However the direction of the inequality is reversed if we multiply by a negative
number, i.e. if x + 1 < 0. So any x < −1 provides a counterexample.
−2
For example, if x = −2 the fraction is −1
= 2.
6. (a) Add x to both sides of the inequality y < x, to give x + y < 2x

(b) Divide the inequality by x + y and by x.
(We can do this because x and y are both positive.)
1 2
Hence x < x+y . The other inequality is proved similarly.
p
(c) Substitute x = 2 and y = 1, which satisfy the condition x > y > 0.

6.5 Linear Simultaneous Equations

We shall be returning to inequalities in the next section, where we will be
combining more than one inequality. Before we do that we are going to look
at the simpler case of combining two linear equations.
For this section you’ll need to be comfortable with solving linear equations
(see Section 6.1).
All the equations we’ve studied so far in the book have involved a single
variable. For example the equation 3x − 8 = x is a statement about that
variable x; it specifies a relationship that the value of x must obey. In this
case x must be 4, and it’s the only value of x that satisfies the equation.
What about an equation with more than one variable? For example x + y = 9.
There isn’t just one solution to this. We could have x = 2 and y = 7, x = 15 and
y = −6, and so on. There are infinitely many solutions, each one consisting
of an x,y pair. The pair could be represented by a point on a graph.
If we plot all the solutions of x + y = 9 what do we get? The line x + y = 9, of
course. The equation is linear so we get a straight line.
Now suppose x and y need to satisfy two equations at the same time. For
example:
x+y = 9
and y = 2x.
Each equation taken on its own has an infinite number of possible solutions,
but typically a solution to one will not be a solution to the other. For example
x = 4, y = 5 fits the first but not the second; x = 12, y = 24 fits the second but
not the first. Solving them simultaneously means finding a value of x and a
value of y that fits both equations.
In our example we are looking for two numbers, one of which is twice as big
as the other, and which add up to 9. The answer is x = 3, y = 6. If we can see
what answer is without doing any work, we call this solving by inspection.
If we can’t see the solution, we could try a few numbers, in other words use
trial and error. In an exam we are unlikely to be able to find the solution this
way. We need a systematic method.
You are going to see three methods. Which one you use will depend on the
problem. 20 .
20
Real life is usually a complicated affair, so when we use mathematical models to
solve problems we typically find ourselves with equations involving a great many
variables—sometimes hundreds. If this interests you and you want to see some
practical applications, turn to page A-19 in Appendix A (or click on the page number).

The three methods for solving simultaneous equations are:
(i) Equating expressions for y.

(ii) Substitution.
(iii) Combining the equations.
Actually the first method is just a particular case of the second, as you will
see, but we tend to think of them as different. The objective of each method
is the same: to eliminate one of the variables. That will take us from two
equations in two variables to one equation in one variable 21 . In this chapter
that equation will be linear (which we know we can solve), so it will give us a
strategy for solving any pair of linear simultaneous equations. (In Chapter 7
we’ll meet simultaneous equations that are non-linear.)
Sometimes it makes no difference which method we choose; at other times

one will be decidedly simpler than the others two. We’ll work through a
couple of the examples using more than one method, to bring out their
differences. Choosing the best method can sometimes make life quite a bit
simpler.
In the first example we will use all three.
21
If we have more than two variables the approach is the same. For example if we
have three equations in three unknowns we eliminate one of them to give two
equations in two unknowns; we then eliminate another variable to give one equation
in one unknown. Once we are down to one equation in one variable we know we can
solve it. This strategy works for any number of variables.

Example 25
Solve these simultaneous equations:
y = 5−x
y = 2x − 1
Solution
(i) We start with the first method: equating expressions for y. To satisfy the
first equation y must be 5 minus the value of x; to satisfy the second y must
be one less than twice the value of x. The only way y can satisfy both
equations, therefore, is if
5 − x = 2x − 1
or 6 = 3x
x = 2
Now we know that x has to be 2, we can substituting this value into either
equation (written in any form) to tells us that y must be. The answer is y = 3.
We say the solution is x = 2, y = 3.
(ii) The substitution method means finding an expression for one of the
variables—using one of the equations—and substituting this into the other
equation. In fact this is what we just did, substituting y. But we could have
chosen x.
To substitute x we need to re-write the first equation:
y +x = 5
x = 5− y
and where we see x in the second equation, replace it with this expression:
y = 2(5 − y) − 1
y = 10 − 2y − 1
3y = 9
y = 3
Once again putting this value back into either equation, we get x = 2.
(iii) We can also use this example to describe the third method, which is
combining equations. We start by writing each equation with the constant
term on the right and the variable terms on the left—in the same order22 .
One way to do this would be:
x+y = 5
2x − y = 1
Now we are going to add these two equations. Why can we do this? We will
explain shortly.
x + y + 2x − y = 6
3x = 6
x = 2
Because we have we have a +y in one equation and a −y in the other, adding
the equations makes the y’s disappear; it leaves an equation in only one
variable. We say we have eliminated y.
It remains to find x, which is easily done by substituting in any of the
equations.
22
So that the x’s and he y’s line up vertically.

That’s all three methods.

Now we’ll explain why adding equations makes sense. If fact we’ll see that
the equations can be combined in many different ways.
All becomes clear if we see what happens with these two (extremely simple)
equations:
A = 4
B = 1
Adding the two equations gives A + B = 5, which is evidently correct. Or we

could subtract them, to give A − B = 3, also evidently correct.
Taking this a step further we could take a multiple of each of them and then
add or subtract. For example 3A − 2B = 10. This is also correct.
In Example 1 we were “lucky": we had a y in one of the equations and a −y

in the other. . One way of re-writing the equations would be:
y +x = 5
y − 2x = −1
Now adding them isn’t going to help: we’d get 2y − x = 4. (Neither variable
disappears and we are no further forward.)
Because we have +y in both equations,subtracting them does the trick:
y + x − (y − 2x) = 5 − (−1)
y + x − y + 2x = 6
3y = 6
x = 2
and the solution is the same as before. Notice we have to be particularly

careful when subtracting negative numbers.
Before moving on to some slightly tricker examples, a brief look at the

geometric significance of solving simultaneous equations.

Geometrical interpretation of linear simultaneous

equations
Any linear equation can be represented by a straight line.
We’ll use the equations in Example 1, which can be written
y +x = 5
y = 2x − 1.
For every point on y + x = 5, its x-coordinate plus its y-coordinate equals 5;

The coordinates of the point on y = 2x + 1 satisfy a different relationship.
At the point where the two lines cross, the x and y coordinates satisfy both
relationships, so the solution of the simultaneous equations tells us where
the two lines cross.
This is illustrated in the diagram below: the lines cross (or intersect) where
x = 2 and y = 3.
y

6

@ 6
+1

@
2x
@
5@
y=

@
@
4 @

@
@ • (2, 3)
3
@
@
y+
@
2
x
@
=
@
5

1 @
@

@
x
@ -

−1 0 1 2 3 4 5@ 6
@
−1 @
@

−2

−3
We shall return to the graphical interpretation of linear relationships later in

the chapter when we consider linear inequalities.

Example 26
2x + 3y = 5
x − 2y = 6
Solution
First by substitution (Method (ii)). We can use either variable but it’s much
easier to substitute x in this case, because using y will immediately
introduce fractions. (For this reason Method (i) would not be the most
sensible way to go.) From the second equation
x = 6 + 2y
Substituting x into the first (i.e. replacing x with 6 + 2y) we have
2(6 + 2y) + 2y = 5
12 + 4y + 3y = 5
7y = −7
y = −1
Substituting y = −1 back into x = 6 + 2y
x = 6 + 2(−1) = 4.
We could have substituted in any of the equations, but since we have already
taken the trouble to find an expression for x in terms of y, we may as well use
that version.
The solution is x = 4, y = −1.
We could also use Method (iii), combining equations, although in this case
simply adding or subtracting the equations isn’t going to work. But we can
make them match if we do some multiplying first (as we described on page
6-41).
Doubling the second equation would mean both equations having a 2x
term:
2x + 3y = 5
2x − 4y = 12
Subtracting these we have:

3y − (−4y) = 5 − 12
7y = −7
y = −1.
Substituting y into either of the original equations gives x = 4, as expected.

By the way we could have chosen to eliminate y rather than x, but that
would have meant multiplying both equations. In the next example we have
no choice but to do this.

Example 27
7q = 6p + 5
4(p + q) = 1
Solution
With no easy way of finding an expression for p in terms of q (or vice versa)
we’ll use Method (iii) (combining the equations). We start by getting rid of
the brackets:
7q − 6p = 5
4q + 4p = 1
We can choose to eliminate q by multiplying the first equation by 4 and the

second by 7:
28q − 24p = 20
28q + 28p = 7
Adding 52p = −13
1
p = − .
4
Substituting back we have:
4q − 1 = 1
4q = 2
1
q = .
2
The solution is 1 1
p =− , q= .
4 2
We could of course have started by eliminating q rather than p.

QQ
Solve these simultaneous equations using any method. The answers are in
the footnote23 .
(a) y = 2x − 1 (b) y =4 + x (c) a + 4b = 2

y = 8−x x+y = 7 3a − b = 19
(d) 7x − 2y = 1
3x + 4y = 15
23
(a) x = 3, y = 5 (b) x = 1.5, y = 5.5 (c) a = 6, b = −1 (d) x = 1, y = 3

Example 28 3y
x+ +1 = 0
2
17 − 2y
= 5
x
Solution
The issue here is that both equations include fractions, however as we saw in
Section 6.1, we can always get rid of fractions from an equation by
multiplying by a suitable number, in this case 2 and x respectively. This gives
us:
2x + 3y + 2 = 0
17 − 2y = 5x
rearranging into the standard format
2x + 3y = −2
5x + 2y = 17.
Complete this example for yourself. The answer is in the footnote24 .
Example 29
'$
A drive belt passes over two cylindrical drums,
r each of radius r m. The axes of the drums are
6 parallel and a distance L m apart.
&%
L The total length of the belt is 3 m, and for
'$optimum efficiency the distance between the
centres of the two drums must be four times
r
? their radius.
&%
What is the ideal radius of the drums?
Solution
The belt can be thought of comprising the complete circumference of a
circle radius r m plus two straight sections each of length L m, so
2L + 2πr = 3
The other relationship is L = 4r .
By substituting L into the first equation (and remembering that π is just a
number)
8r + 2πr = 3
2r (4 + π) = 3
3
r =
2(4 + π)
= 0.210 m (3SF).
Finally, simultaneous equations can arise in the context of function notation.
24
x = 5, y = −4. With no variable having a coefficient 1 anywhere use method (iii).

Example 30
A function f is defined by the formula
f (x) = ax + b.
Given that f (1) = 3 and f (3) = −1. Find the value of f (4).
Solution
f (1) = a + b and f (3) = 3a + b, so we have simultaneous equations:
a +b = 3
3a + b = −1
Subtracting then gives

2a = −4
a = −2
Substituting this value of a back into the first equation
−2 + b = 3
b = 5.
The function is
f (x) = −2x + 5,
so f (4) = −3.

Exercises 23
1. In each case determine which of the points represent possible
solutions to the equation.
(a) y = 4x
(0, 0), (4, 1), (1, 4), ( 12 , 4 21 ), ( 12 , 2)
(b) x + y = 6
(0, 0), (5, 1), (−3, 9), (2a, 4a), (4b, 6 − 4b)
(c) 5x + 3y = 1
(3, −5), (2, −3), (0, 13 ), ( 15 , 0), ( 21 , − 12 )
y x
(d) − =1
5 2
(0, 5), (−2, 0), (2, 10), (2, 5), (−1, 2.5)
(e) y = 2x + k, where k is a constant
(1, 2 + k), (k, 3k), (− k2 , 0), ( k2 , 2k), (−k, k)
2. Solve these simultaneous equations using the substitution method.

(a) y = 3x (b) x + y +1 = 0
x + 2y = 14 2x + 3y = 1
(c) 5y − x = 38 (d) y −x +8 = 0
x = 1 − 2y 5x + 4y + 5 = 0
(e) 2(p − q) = 1 (f ) 3x − 2y = 11
4p − 5 = q 2x + 3y = 0
3. Solve these simultaneous equations by equating expressions for one of

the variables.
(a) y = 3x + 4 (b) y = x −1
y = 12 − x x+y = 4
(c) a = 5+b (d) y = 6x

b = 2a − 8 3x − y = 1
1 x−1
(e) y = x +1 (f ) y =
2 3
y = 2x − 5 y − 4x = 7
4. Solve these simultaneously by the method of combining equations.

(a) 3x + y = 7 (b) 2x + 5y = −6
5x − y = 9 2x − 3y = 11
(c) y − 2x = 3 (d) 4s − 3t = 5
3y + x = 16 8t − 3s = 7
(e) 3(x + y) = 2 (f ) 2x = 5y
5x + 7y = 2 8y − 7x = 38

5. Solve these simultaneous equations using any method.

(a) 4x + 5y − 6 = 0 (b) y = x +1
x +3 = y 7x + 3y = 0
(c) x+y = 3 (d) y = 2x + 5

3x + 5y − 12 = 0 y = −3x
2
(e) 0.4x − 1.3y = 9 (f) y = x −1
3
3
2x − 0.5y = 3 y = x +4
2
6.
The diagram shows a rectangular pane of
glass measuring x m wide by y m high.
y
The perimeter of the pane is 7.0 m. and the
ratio of its height to its width is 3 : 2.
Find the dimensions of the pane of glass.

x
7. A small company makes a specialised component which it delivers

directly to customers. The component comes in two sizes, small and
large, weighing 3 kg and 4 kg respectively (including packaging). On a
particular day the delivery driver loaded her van with 12 of the
components. Their total weight was 46 kg.
How many component of each size were delivered on that day?
8. A flag pole is partly supported by a wire of length 13 m attached to two

points A and B on horizontal ground and making angles of 30◦ and 60◦
with it, as shown in the diagram. The points A and B are 9.5 m apart.
Find the height of the pole (to 2 SF).
C
""J
" J
"
" J
" J
"
" J
"
" J
" J
"
"
" 30◦ 60◦ J B
J
A "
9. In this question, p(x) and q(x) are polynomial function, a and b are
constants.
(a) p(x) = 6x 3 + ax + b, p(0) = 3 and p(1) = 11. Find a and b.

Hence show that p(−1) = −5.
(b) q(x) = ax 2 + bx + 1. Find the values of a and b,
given that q(2) = 5, and q(−1) = −4.
10. A 135 mile journey takes in motorways, where an average speed of

60 mph is maintained, and other types of road where the average
speed is 45 mph. The whole journey takes 2 h 20 mins.
How much of the journey is on motorways?

11. A healthy breakfast cereal only has two ingredients, rolled oats and
freeze dried berries. The oats cost the manufacturer 0.15 p per gram
and the berries 5.0 p per gram. A typical pack contains 250 g of cereal,
costing the manufacturer 86 p.
Find the breakdown by weight of oats and berries in a typical pack, and
hence find the percentage (by weight) of berries in the cereal.
12. (a) Solve the simultaneous equations
2x + y = 5
5x − 2y = 1
(b) Solve, in terms of a, the simultaneous equations
2x + y = a
ax − 2y = 1
(c) Explain why the solution you found in (b) is only valid if a 6= −4.
13. Solve the simultaneous equations

1
x
− 1y = 1
2
y = 6x
by making the substitutions X = x1 , Y = 1y .
14. (a) Find, in terms of p, q and r , the solution to the equations
2px + q y = 7r
px − 2q y = r.
(b) Show that, for these values of x and y
px = 3q y.
15. The following is a set of linear equations in three variables.
2x + y − z = 6
4x + 2y + z = 9
x − y + 3z = −2
(a) Find an expression for z in terms of x and y.

(b) By substituting this expression for z into both of the other
equations, show that the variables x and y satisfy the two
equations
6x + 3y = 15
7x + 2y = 16
(c) Solve these two equations simultaneously and hence find the
solution to the three simultaneous equations.
(d) Verify that this is the correct solution by substituting the values
into all three equations.

16. Find the ages of all the children by formulating and solving equations.
(a) In seven years time Sekai will be three times as old as she was five
years ago.
(b) Ethan is 7 years older than Fred and the sum of their ages is 12
years.
(c) Ralphie is three times as old as Sam and their ages differ by 12
years.
(d) In two years Toby will be twice as old as Zoe is now, and their ages
will add up to 35 years.

Solutions 23
1. (a) (0, 0), (1, 4), ( 21 , 2)

(b) (5, 1), (−3, 9), (4b, 6 − 4b)
(c) (2, −3), (0, 13 ), ( 15 , 0), ( 21 , − 12 )
(d) (0, 5), (−2, 0), (2, 10), (−1, 2.5)
(e) (1, 2 + k), (k, 3k), (− k2 , 0), ( k2 , 2k), (−k, k)
2. (a) x = 2, y = 6 (b) x = −4, y = 3 (c) x = −3, y = 7

3
(d) x = 3, y = −5 (e) x =2, q = 1 (f ) x = 3, y = −2
5 3
3. (a) x = 2, y = 10 (b) x =2, y =2 (c) a = 3, b = −2
1
(d) x = − 3 , y = −2 (e) x = 4, y = 3 (f) x = −2, y = −1
4. (a) x = 2, y = 1 (b) x = 1, y = −3 (c) x = 1, y = 5

1 4 2
(d) s = −1, t = 2 (e) x =3, y = −3 (f) x = −10, y = −4
3 7 3 3
5. (a) x = −1, y = 2 (b) x = − 10 , y = 10 (c) x =2, y =2
(d) x = −1, y = 3 (e) x = 3, y = 6 (f) x = −6, y = −5
6. x = 1.4 m, y = 2.1 m.
y 3
= , so 2y = 3x and the other equation is 2x + 2y = 7.
x 2
Substituting 2y = 3x into the second equation, 5x = 7.
7. 2 small, 10 large.
Assume delivery of x small components, y large components.
Two equations are: x + y = 12 and 3x + 4y = 46.
8. Height 4.1 m.
Let AC = x and BC = y. Then AB = x cos 30◦ + y cos 60◦ .
Two equations are: x + y = 13 and 0.866x + 0.5y = 9.5.
Doubling the second equation and subtracting: 0.732x = 6.
Height of pole = x sin 30◦ = 4.098 m.
9. (a) a = 2, b = 3.
p(0) = 3 ⇒ b = 3. p(x) = 6x 3 + ax + 3, so p(1) = 11.
(b) a = −1, b = 4
q(2) = 5 ⇒ 4a + 2b + 1 = 5, which simplifies to 2a + b = 2.
Similarly a − b = −5.

10. 120 miles on motorways.

7
2 h 20 mins = hours.
3
Let x = distance on motorways, y = distance on other roads.
x y 7
Equations are: x + y = 135 and + = .
60 45 3
Second equation reduces to 3x + 4y = 420.
11. 4% berries.
Assume x g of berries and y g of oats in a 250 g pack.
Cost (in pence) of ingredients in a pack = 5x + 0.15y.
5x + 0.15y = 86
x + y = 250
Solving gives 4.85y = 1164, y = 240.
10 1
Proportion of berries = = .
250 25
11 23
12. (a) x = ,y=
9 9
2a+1 a 2 −2
(b) x = ,y=
a+4 a+4
2x + y = a
ax − 2y = 1
Hence 4x + ax = 2a + 1 ⇒ x(4 + a) = 2a + 1
(remember a is a constant)
Hence a y + 4y = a 2 − 2 ⇒ y(a + 4) = a 2 − 2
(c) If a = −4 the values of x and y are undefined (the

denominators are zero), and the equations become
2x + y = −4 and 2x + y = 21
which are inconsistent (they can’t both be true).
13. x = 53 , y = 10.
The equations become X − Y = 12 and X = 6Y ,
which solve to give X = 53 , Y = 10
1
.

3r r
14. (a) x = ,y=
p q
2px + q y = 7r
px − 2q y = r
Hence 5px = 15r ⇒ px = 3r

and: 5q y = 5r ⇒ q y = r
(b) We have two equation connecting the three variables

p, q and r :
px = 3r and q y = r.
The relationship we are asked to prove does not contain r ,
so we eliminate it by substitution
px = 3(q y)
15. (a) z = 2x + y − 6.
(c) Hence 9x = 18,

so x = 2, y = 1, z = −1.
16. We use the first letter of the name to denote the age.
(a) S = 11. S + 7 = 3(s − 5)
1 1
(b) E = 2 2 , F = 9 2 . E = F + 7, E + F = 12
(c) R = 6, S = 18. R = 3S, R − S = 12
(d) T = 20, Z = 11. T + 2 = 2Z , (T + 2) + (Z + 2) = 35

6.6 Combining inequalities

6.6.1 Examples involving one variable
(a) Suppose for example that x is such that x > 2 and x ≤ 5.
Rather like with simultaneous equations, the variable x has to obey both
inequalities at the same time.
In set language x belongs to the intersection (or overlap) of the two sets
X = {x : x > 2} and Y = {x : x ≤ 5}, which we write: x ∈ X ∩ Y . Because x is
bounded below by 2 and above by 5, we cam write these two conditions in a
single satement:
x ∈ X ∩ Y = {x : 2 < x ≤ 5}
We don’t need to use set language, we could just say 2 < x ≤ 5.
On the number line diagram we indicate that 2 is not included but 5 is using
an empty circle and a filled circle respectively:
X ◦ H
H

Y

H
H •
0 1 2 3 4 5 6
X ∩Y ◦ •
0 1 2 3 4 5 6
When we discussed pairs of sets in Chapter 1 we saw that there are three
possible configurations: one set could be entirely inside the other (one is a
subset of the other), they could partially overlap or they could not overlap at
all. The example above is the middle case; here are example of the other two
cases.
(b) One is a subset of the other. For example x ≥ 5 and x > 3
X = {x : x > 3} ◦ H
H

Y = {x : x ≥ 5} • H
H

1 2 3 4 5 6 7
The intersection is the set Y :
X ∩Y = Y • H
H

0 1 2 3 4 5 6
(c) The third case is where there is no overlap. For example x < 3 and x ≥ 5.
X = {x : x < 3}

H ◦
H
Y = {x : x ≥ 5} • H
H

1 2 3 4 5 6 7
In set language we would say the intersection is the empty set: X ∩ Y = φ.

Alternatively there may be more more than one inequality but it is sufficient
for x to obey only one of them.
(a) For example x < 3 or x ≥ 5.
In set language x belongs to the union of the two sets: X = {x : x < 3} and
Y = {x : x ≥ 5}. On the number line it looks like this:
X = {x : x < 3}

H ◦
H
Y = {x : x ≥ 5} • H
H

1 2 3 4 5 6 7
x ∈ X ∪ Y = {x : x < 3 or x ≥ 5}.
Unlike example (a) in the previous section it is not possible to combine these
two conditions into a single statement (in that case 2 < x ≤ 5).
As with the intersection of sets there are two other cases to consider.
(b) If one set is a subset of the other. For example x ≥ 5 or x > 3
X = {x : x > 3} ◦ H
H

Y = {x : x ≥ 5} • H
H

1 2 3 4 5 6 7
Now the set {x : x ≥ 5} is entirely subsumed within {x : x > 3}:
X ∪Y = X ◦ H
H

0 1 2 3 4 5 6
(c) The third case is where there is a partial overlap, for example x > 2 or
x ≤ 5.
X = {x : x > 2} ◦ H
H

Y = {x : x ≤ 5}

H
H •
0 1 2 3 4 5 6

H
H
X ∪Y = R H
H

0 1 2 3 4 5 6
The union of these two sets is the entire number line, i.e. every real number.
Any number we choose will obey one of these inequalities; some will obey
both. For example x = 3.5, because 3.5 ∈ X ∩ Y .
We say that x ∈ X ∪ Y if x belongs to X , Y or both X and Y .

QQ
Find the set of values of x for which:
(a) x > 10 and x > 8 (b) 4x < 3 and x ≥ −1 (c) x < 1 or x ≤ 6

x
(d) 2x + 1 < 9 and 2(1 + x) ≤ 3x + 1 (e) 2 > 3 or 1 − x > 4
3x
(f) x < 5 and (x − 1)(x + 3) < 1 + x 2 (g) 2 ≥ 1 or x < 2.
25 3
(a) x > 10 (b) −1 ≤ x < (c) x ≤ 6 (d) 1 ≤ x < 4 (e) x < −3 or x > 6 (f ) x < 2 (g) R
4

6.6.2 Examples involving two variables

When we only had one variable it was a case of pinning down where x could
lie on the number line. If there are two variables we need to move into two
dimensions.
Take the example y < 2x + 1. This involves two variables, x and y and the
possible values that one of the variables can take depends on the value the
other one takes.
For example x = 3 , y = 1 satisfied the inequality; x = 1, y = 4 doesn’t.
The point (3, 1) lies below the line; (1.4) lies above it.
y
At every point
6
+1

6 above the line: y > 2x + 1
2x

5 on the line: y = 2x + 1
y=

4 •(1, 4) below the line: y < 2x + 1

3
•
2

It is not hard too see why
this simple rule of “above,
1 •(3, 1) on and below" works,

- x because y tells us the
−1 0 1 2 3 4 5 6 height of the line above the

−1 x-axis.

−2

So if the inequality is expressed in the for y < ......, or y > ..... we can easily
determine where all the solutions are: above the line > applies; below it <
applies.
If the inequality signs are ≤ or ≥ we simply include the points on line as well.
So for example y ≥ 2x + 1 corresponds to all the point on or above the line
y = 2x + 1.

So far so good, but if the inquality is not in this form, the simple “above" or
“below" rule won’t necessarily apply. Look at this example.
Check 26 first that the equation of the line 2x − 3y = 12 is right:
y
In this case, at every point
12
6
y = above the line: 2x −3y < 12
12 −3
3 y< 2x on the line: 2x − 3y = 12
x−

2 below the line: 2x −3y > 12

• •
- x
(6, 0)

12

(0, −4)•
>
3y

−
2x

So how can we tell which inequality applies on which side of the line?
We pick a point that we know for sure to be one side of the line and check to
see which inequality it satisfies. Since every point on that side of the line will
satisfy the same inequality, we have our answer.
We call this a test point and the best one to use is undoubtedly (0, 0). (Unless
the line passes through the origin.)
In our example (0, 0) satisfies27 the inequality 2x − 3y < 12, so every point on
that side of the line satisfies that inequality.
y
Another example. Consider the line y = 3 − 2x. AA 6
A
A
(0, 0) satisfies the inequality y < 3 − 2x A
A
So the origin lies in the region y < 3 − 2x A
y>
and the other side of the line is y > 3−2x. A

y<
A
3−
A
But we only chose the origin to keep the •
3−
2x
A
calculations simple. We could equally A
2x
A
well have chosen any point we know for
• A - x
certain to be one side of the line or the 0 A
AA
other.
y=
For example (−1, 1). Not surprisingly

3−
this point satisfies y < 3 − 2x.

2x
So to decide which side of a line corresponds to which inequality, pick any

test point.
If the line doesn’t pass through the origin the simplest one to use is the
origin.
26
(6, 0) and (−4, 0) both satisfy the equation.
27
Because 0 − 0 < 12.

Now suppose more than one inequality applies (where we have two
variables).
Example 31
Use shading to show on a graph the region for which

x + y < 5, y ≤ 2x and y > 1.
Solution
The three boundary lines are drawn on the graph below: x + y = 5, y = 2x

and y = 1.
We can use (0, 0) as the test point for the two lines that don’t pass through the
origin, and say (1, 0) for y = 2x.
This will tell us that all the inequalities are satisfied by the points in the
shaded region.
On the boundary lines themselves:
a solid line means points on the line are included;
a dotted line means the points are not included.
4 y = 2x
3
x+y =5
2
y =1
1
0 x
1 2 3 4 5 6
−1

Example 32
Show on a graph the region which satisfies all these inequalities:
x +1
y ≥ −2, y ≤ x, 2x + y ≤ 8 and y≤
2
Hence find the point in this region for which y takes its largest possible
value.
Solution
There is no easy way of answering the question algebraically, but a graph
makes it clear. First we draw the boundary lines. A simple way to do this is to
identify a pair of points on each of them.
For y = x we could use, say, (0, 0) and (4, 4), for 2x + y = 8, the points (4, 0) and
x+1
(0, 8) and for y = 2 the points (−1, 0) and (0, 12 ). It doesn’t matter which
points we choose but if we pick values of x or y that are simple to work with,
arithmetic errors are less likely.
5 2x + y = 8
4 y =x
3 x+1
y= 2
0 x
−2 −1 1 2 3 4 5
−1
−2
Using a test point for each inequality, they are all satisfied in the shaded area
(including its entire boundary). The largest value y can take within this region
x+1
is where y = 2 meets 2x + y = 8. Reading from the graph the point is (3, 2).
We can check this by substituting (3, 2) into both equations28
28
If we were to find they don’t satisfy both equations we can find the correct values by
solving the two equations simultaneously.

Not all inequalities are linear of course. In the next chapter we will be
exploring quadratic functions, including quadratic inequalities such as
y < x 2 − 3x + 2 or y + x 2 ≥ 4.
As a taster we will look at a simple example here.
Example 33
The graph shows the curve y = (x − 1)2 and a straight line that passes
through (0, 3) and (3, 0).
(a) Find the equation of the straight line.
(b) Write down the inequalities satisfied by the points in the shaded region.
(c) Write down the inequality in y satisfied by all the points in the shaded
region for which x = 1.
y
5
y = (x − 1)2
4
0 x
−1 1 2 3
−1
Solution
(a) The equation of the line is x + y = 3.
(b) Using (0, 0) as the test point, in the shaded region x + y < 3.
In the same way as with linear boundaries, points on one side of the curve
will satisfy y < (x − 1)2 , points on the other side y > (x − 1)2 and point on the
curve y = (x − 1)2 .
The test point (0.0) satisfies y < (x − 1)2 , so in the shaded region y ≥ (x − 1)2 .
It is ≥ not > because the solid curve tells us that points on the line are to be
included.
The answer is x + y < 3 and y ≥ (x − 1)2 .
(c) Within the shaded region, if x = 1, y must lie between 0 and 2, where 0 is
included but 2 is not. The inequality is therefore 0 ≤ y < 2.

Combining inear inequalities: practical applications

If the notion of combining linear inequalities seems rather too abstract to be
of much practical use, you would be mistaken. There is a mathematical
technique called Linear Programming which can be applied to a whole range
of problems that can only be specified using inequalities. The objective will
be to find an optimum value of some quantity (for example the maximum
profit or minimum cost) subject to a number of constraints, which is where
the inequalities come in.
Let’s take a simple example. Each day a bakery has a delivery of raw
materials (flour, sugar, etc) which it uses to make a range of products: bread,
cakes etc. The quantities used differ from product to product and they sell at
different prices. Various pieces of equipment are needed, of which the
bakery only has a limited number, and the amount of time needed on each
varies according to the product. Some products require more time on
manual tasks like packing. The problem is to find what combination of
products the bakery should make each day to maximise profits 29 .
What are the constraints? The total amount of each ingredient cannot
exceed the amount delivered. The total time spend on each piece of
equipment cannot exceed the length of the working day. A fixed number of
employees means a limit on the total time spent on manual tasks.
Each of these constraint can be expressed as an inequality, but there won’t
just be two variables x and y. For a large bakery there could be hundreds of
variables. As long as there are more than two we wouldn’t be able to visualise
where the solutions lie using a graph; we have moved into n dimensions.
Such problems are solved using an algorithm that systematically narrows
down the possibilities until it arrives at the optimum solution; a common
one is known as the simplex method. It’s possible to apply this method by
hand, but this is very tedious and invariably a computer package will be
used.
Airlines use Linear Programming for their flight scheduling: having aircraft
of sufficient capacity in the right place at the right time, with suitably trained
crew available. It’s an enormous task, and because the situation constantly
changes (aircraft develop faults, flights are delayed etc) scheduling decisions
need to be continuously updated. The constraints will run into the
thousands, the objective will be to minimise total costs.
If you are interested in a career that uses mathematical techniques like this
to solve practical problems, the field of work is called Operational Research
30
.
29
We may need to make some simplifying assumptions, for example that every item sells,
or that the machines don’t break down, but even these can often be expressed a further
constraints.
30
Or Operations Research. These days it may be grouped together with the related field
of Data Analytics, which is about finding patterns and relationships in large datasets.
Both involve the use of cutting edge mathematical and statistical techniques.

Exercises 24
1. Find the values of x for which

(i) x > 0 and 2x − 5 < 1 − x (ii) 2x − 13 ≤ 7 + x and 3(x + 5) < 5x + 3
(iii) x + 5 > 0 and 2(x − 1) < 0 (iv) x ≥ 3 and x ≥ 1
x 5x
(v) x ≤ 2 and 4 > 1 (vi) 4 + x 2 > (2 − x)2 and 2 < 1
(vii) 3x < 2 and 4x + 3 > 6(1 − 2x) + 1 (viii) x 2 ≥ (x − 2)(x + 4) and x > 1
2. The sets A, B and C are defined by

A = {x : x > 4}, B = {x : −1 ≤ x < 7}, and C = {x : x ≤ 2}.
Using set notation, describe each of these sets:
(i) A ∩ B (ii) A ∪ B (iii) B ∪C (iv) B ∩C (v) A ∩C (vi) A ∪C
3. Which of the regions defined by these inequalities contains the origin?

(a) y < 4x + 3 (b) 2y > 6 − x (c) x + y ≥ 1 (d) 4y − 3x < 2
x 2y+x
(e) 3y ≥ 5x (f ) 2 > 1 − y (g) 11y ≤ 2(x + 3) (h) 3
>1
4. Repeat question 3 for the point (0, 2).

5. The shaded region in the y
diagram is bounded by
6
the lines y = x, x = 5 and
x + 2y = 6. 5
x + 2y = 6 4 y =x
No points on the boundary
lie within the region 3 x =5
FInd the three inequalities 2
that define the region. 1
0 x
1 2 3 4 5 6 7
6. The vertices of the shaded triangle y

in the diagram are (1, −2), (7, −2) 6
and (3, 2).
(a) Find the equation of each of 4
the boundary lines.
(b) Hence determine the three 2 (3, 2)
inequalities that describe this
region, given that it includes all
0 x
the points on its boundary. 1 2 3 4 5 6 7 8
−2
(1, −2) (7, −2)
−4
7. By drawing the appropriate straight lines, shade the region on a graph

defined by the inequalities
1
y ≥ x − 2, x + y ≤ 8, and x > 2.
3

8. The diagram shows a square with

y
vertices A(4, 0), B (0, 4), C (−4, 0)
and D(−4, −4). The midpoints of B
its sides are at P , Q, R and S.
(a) Write down the equation of Q P
each of the sides.
(b) Using inequalities describe C A
O x
each of these regions, where
a solid line indicates that the
boundary points are included R S
and a dotted line indicates that
they are not.
D
(i) ABC (ii) O AS (iii) OQC R
9. The diagram shows a quadrilateral O AP B formed by joining A(4, 6) to
C (−2, 0) and B (−4, 6) to D(2, 0). P is where AC and B D intercept the
y-axis. O is the origin coordinates.
y
B 6 A
P
2
C D
x
−4 −2 0 2 4
(a) By finding the gradient of AC show that the coordinates of P are
(0, 2).
(b) Find the area of the quadrilateral O AP B .
(c) Find the inequalities that describe the region inside or on the
perimeter of the triangle O AP .
(d) Hence describe the region inside or on the edges of the
quadrilateral O AP B .

Solutions 24
1. (i) 0 < x < 2 (ii) 6 < x ≤ 20 (iii) −5 < x < 1 (iv) x ≥ 3

2 1 2
(v) No solutions (vi) 0 < x ≤ 5 (vii) 4 < x ≤ 3 (viii) 1 < x ≤ 4
2. (i) {x : 4 < x < 7} (ii) {x : x ≥ −1} (iii) {x : x < 7}

(iv) {x : −1 ≤ x ≤ 2} (v) φ (empty set) (vi) {x : x ≤ 2 or x > 4}
3. (a), (d), (e), (g)
4. (a), (c), (e), (f ), (h)
5. y < x, x + 2y > 6, x < 5.
6. (a) y = −2, y = 2x − 4, x + y = 5
[Gradients are 2 and −1, so (y −− 2) = 2(x − 1) and (y −− 2) = −(x − 7)]
(b) y ≥ −2, y ≤ 2x − 4, x + y ≤ 5 .
7.
y
[The lines intersect at
8
(2, 6), (7.5, 0.5) and (2, − 43 )]
0 x
2 4 6 8 10
−2
8. (a) x + y = 4, y − x = 4, x + y = −4, x − y = 4 (or the equivalent).

(b) (i) y ≥ 0, x + y ≤ 4 and y − x ≤ 4 (ii) y ≤ 0, x − y ≤ 4 and y > −x
(iii) y < −x, y > x, y − x ≥ −4 and x + y ≥ −4.
9. (a) The gradient of AC is 1, so OP = 2.

(b) Area = 8 [For triangle O AP , base = 2, height = 4]
(c) x ≥ 0, y < x + 2 and y > 23 x.
(d) x ≥ 0, y ≤ x + 2 and y ≥ 23 x, or x ≤ 0, y ≤ −x + 2 and y ≥ − 32 x

6.7 Linear modelling

There are numerous situations in which the relationship between two
variables is linear. The simplest is when the variables are in direct
proportion, which we write as
y α x.
If we double x, it doubles y, if we halve x, it halves y and so on. This means
there will a constant number k such that
y = kx.
To find what k is all we need is a value of x and the corresponding value of y.
We know that inn general a linear relationship will be of the form
y = mx + c.
An example of direct proportion would be the cost of mobile phone calls
under a so-called Pay As You Go (PAYG) fee structure. There is no fixed
monthly fee, simply a charge per minute on calls made. Suppose that cost is
12p per minute, then the cots of 100 minutes would be exactly £12. In
general m minutes of calls cost C 1 = 12m pence. This kind of fee structure is
attractive for someone who makes few calls.
For a person who makes a lot of calls it may be better to have price plan with
a fixed monthly fee. Suppose that is £5, with calls costing say 2p per minute.
The monthly cost for them would be C 2 = 500 + 2n pence.
The breakeven point would be when C 1 = C 2 , i.e. 12m = 500 + 2n, or n = 50,
at which point the total cost under either tariff would be £6. We can illustrate
the difference using a graph.
C pence
1500
1000
500
m mins
0 50 100 150
The first tariff would be better for someone who makes fewer than 50
minutes of calls each month; above this it is better to be on the second
tariff31
With the first tariff we say the marginal cost of calls is 12p per minute (the
additional cost for each extra minute); with the second tariff the marginal
cost is 2p. The marginal cost is represented by the gradient of the
corresponding straight line.
A fixed marginal cost means a constant gradient which means the
relationship is a linear one.
31
In practice the fixed monthly fee is likely to be higher and include text messages and data,
but the principle is the same.

With gas and electricity charges the situation is similar: there is normally a
standing charge (the fixed cost) and a marginal cost i.e. the cost per unit of
energy consumed (usually expressed in Kilowatt hours32 ).
Directly proportional tariffs are available but most have a fixed cost element.
What this is, and the corresponding marginal cost, for each tariff is readily
available on suppliers’ websites. These can be used to to find the best tariff.
Alternatively a price comparison website like USwitch will do the calculation
for you based on your anticipated annual consumption.
In these examples there is an exact relationship between the variables. In

practical situations we often don’t know what the relationship is. We may
suspect is a linear one, we may not. To find out we can look at some data,
and if we get a straight line when we plot them on a graph deduce that it is
linear. In real life situations the points will not lie exactly on a straight line
but they may be sufficiently close that we can reasonably suppose that the
underlying relationship is linear, and we can express not as a formula.
We would call this formula a model. It’s not a physical model, in the way that
we might make a scale model of a building or a car, but a mathematical
model. It is a model in the sense that it represents a physical reality. If we
have such a model it helps us to understand the exact nature of the
relationship and from this we can infer things beyond anything we have
observed.
Models are used in such different contexts as economics, engineering and
business planning. The only way to make any kind of economic forecast is to
build a model. This will be a set of equations that describe relationships
between key factors such as tax and wages, or inflation and borrowing. In
the days before mathematical modelling engineers would sometime build a
physical scale model, or a bridge say. Nowadays the model will be
programmed in a computer, allowing the engineer to subject the bridge to
extreme forces before construction even starts. In a business environment
management decisions rely on calculations about future events; that means
modelling.
Not all these models will be linear ones, and in many cases, where
forecasting is involved, there will be a lot of uncertainty around how reliable
the model will prove to be. Techniques have been developed to help deal
with this. One of these is computer simulation, where a set up or a decision
is tested against a range of future scenarios.
For the present we will only be concerned with simple linear models.
32
1 KWh is the energy consumed by a 1 Kilowatt appliance left running for 1 hour.

Example 34
The table gives the speed of a car as it accelerates away from its starting
point over a period of 5 seconds.
Time t secs 0 1 2 3 4 5
Speed v m/s 0 3.2 6.4 9.6 12.8 16.0
(a) Draw a graph to show that during this time the speed increases at a linear
rate.
(b) Find the formula for the speed of the car v after t seconds.
(c) Write down the acceleration of the car over this period.
(d) If the car continues to accelerate at the same rate, find how long it takes
to reach a speed of 24 m/s.
Solution
(a) y
20
15
10
x
0 1 2 3 4 5
(b) Equation: v = 3.2t .

(c) Acceleration is 3.2 m−2 . It is the amount by which the speed increases
every second. It is a constant amount in this case. We call this uniform
acceleration.
Notice the units we use: every second the speed increases by 3.2 metres per
sec (or ms −1 ), so the acceleration is 3.2 metres per second per second (or
ms −2 ).
(d) 24 = 3.2t , i.e. t = 7.5 seconds.
This example is very artificial because there is a straight line that passes
exactly through every data point. The graph was not strictly necessary to
answer all the questions. The next example is more realistic because the data
points only lie approximately in a straight line. We call this a regression line.
There are techniques for determining statistically the line that fits the
points“best". For now we shall simply plot the point on a graph—called a
scatter graph— and fit a line by eye. That means when you do the examples
below your line may not have exactly the same equations as mine. That is
fine provided the coefficients are not too different.
We can use a regression line to estimate values within the range of values we
have already observed. This is called linear interpolation. Estimating values
outside this range is called linear extrapolation.

Example 35
The table shows the average fuel consumption (measured in miles per
gallon) for a number of saloon cars with different engine capacities
(measured in litres).
Car model A B C D E F
Engine capacity E (l ) 1.5 1.2 2.0 2.4 1.5 1.8
Ave. fuel consumption F (mpg) 34.3 36.0 27.1 23.3 35.1 31.7
(a) By drawing a scatter graph show that the relationship between average
fuel consumption and engine size is approximately linear.
(b) If the formula for F in terms of E is F = aE + b, use the graph to estimate
the values of a and b.
(c) Describe in words what the gradient represents physically.
(d) Estimate the fuel consumption for a car with an engine capacity of 2.2 l .
(e) Give one reason why it would be unwise to use the formula for values of E
too far outside the range 1.2 l litres to 2.4 l .
Solution
(a) F mpg
40 (1, 39.5)
30
(2.5, 21.0)
20
E litres
0 0.5 1 1.5 2 2.5 3
The points lie approximately on a straight line, so for the relationship for
these data points is approximately linear.
(b) If we could read off where the line crosses the F -axis, the intercept would
give us the value of b. In this case we can’t see that point, but it passes
through (1, 39) and (2.5, 21), so we can find a and b by solving the
simultaneous equations:
9.5 = a + b
21.0 = 2.5a + b
⇒ a = −12.33 and b = 51.83 (2 DP)
We could have chosen any two points on the line, but to keep the calculation
as simple as possible, we have picked easy values for E .
(c) The estimated change in average fuel consumption for a 1 l increase in
engine capacity.
(d) F = −12.33(2.2) + 51.83 = 24.7 mpg (to 1DP).
(e) The relationship cannot be linear for all vales of E otherwise it would be
possible to estimate the fuel consumption for a car with zero engine capacity.

Exercises 25
1. A tour guide charges a flat fee of $20 plus $4 per person. A second guide
charges $8 euros per person and no flat fee.
(a) Write down a formula for how much each would charge for a group
of N people.
(b) For what size group would each guide group charge the same?
(c) In a particular week each guide operates 12 tours and they take the
same total number of people on a tour. However the first guide earns
twice as much. How many people did they take on a tour during that
week?
2. A lady buys a new car costing £20, 000 on 1st Jan 2018. While she has the
car its value will depreciate at the rate of £2, 500 per year. She decides
she will sell it when it has depreciated by 60%.
(a) With down a formula for its value after t years.
(b) When does she plan to sell the car?
3. A website designer offers two fee structures :

(A) £500 plus £40 per page, or
(B) £1, 200 plus £16 per page,
(i) Write down the formula for producing a p page website under each
fee structure.
(ii) For what size website is the costs the same under both structures?
To attract new clients the designer decides to offer a discount on the
fixed part of fee (A), so that a website with 25 pages would be 20%
cheaper. What should the new fixed part of that fee be?
4. A company estimates that in addition to their wages the cost of

employing each member of staff is £200 per week in fixed costs
(accommodation etc). If every member of staff works a 40 hour week,
write down the formula for the cost of employing someone whose
hourly rate is £p.
In a particular office there are x staff members earning £6 per hour and
y earning £10 per hour. There are 7 members of staff altogether and
the total staffing cost for that office is £3, 400 per week. Find the values
of x and y.
5. On a particular type of savings account Tansander Bank charges

customers a monthly fee of £5 and interest is payable at 1.5% per year.
(a) Draw a graph to show the net annual income I a customer will
receive on a fixed balance £B . (Take values of B from zero to £10, 000).
(b) What is the minimum balance a customer should have in her
account to make it worthwhile holding it?
(c) On the same graph draw a line to show the net income for an
account with ABC Bank that charges no fee and pays 1.1% interest on
balances above £2, 000.
(d) It is more profitable to save with ABC Bank for balances up to £P .
Find the value of P .

6. The table below shows annual sales figures for a product and what the
company that makes it spent advertising that product during the
corresponding years. Plot the figures for sales S against advertising
spend A on a scatter graph and draw a straight line that fits the points
well.
Year 2010 2011 2012 2013 2014 2015 2016
Sales (£ m) 36.4 38.0 41.8 42.4 55.0 48.1 59.1
Advertising (£ m) 1.52 1.25 1.75 1.60 2.35 2.10 3.06
(a) Write down the coordinates of the point where the line crosses the
S-axis.
(b) Find the gradient of the line by reading off the value of S when
A = 3.5. Hence, or otherwise, find the equation of the line.
(c) Say what the gradient of the line represents in terms of sales and
advertising expenditure.
(d) The company is relocating as as a result output will be limited to
sales of £50m. What would the recommended advertising spend be?
(e) Give one reason why it would it be unwise to assume the
relationship between advertising expenditure and sales is
approximately linear for all values of A?

Solutions 25
1. (a) 20 + 4N and 8N (b) N = 5 (c) 60 people. [12 × 20 + 4N = 8N ]
2. (a) 20, 000 − 2, 500t (b) May 2024 [20 − 2.5t = 8 ⇒ t = 6 yrs 4.8 months]
3. (i) 500 + 40p, 1200 + 16p (ii) 50 pages

New fixed fee = £200 [0.8(500 + 1000) = x + 1000]
4. 200 + 40p. x = 5, y = 2 [x + y = 7, 440x + 600y = 3400]
5. (a) and (c)

I (£)
100
50
0 B (×£1, 000)
1 2 3 4 5 6 7 8 9 10
−50
−100
(b) £4, 000 [x × 0.015 = 12 × 5] (d) £15, 000 [x × 0.015 − 60 = x × 0.011]
6. S (£m)
70
(3.5, 68)
60
50
40
30
20
(0, 18)
10
0 A (£m)
0.5 1 1.5 2 2.5 3 3.5
68−18
(a) (0, 18) (b) 14.3 (to 3 SF) S = 14.3A + 18 [gradient = 3.5
]
(c) The extra sales generated by every additional £m spent on advertising.
(d) £2.24m [50 = 14.3A + 18] This is an example of linear interpolation.
(e) If it were linear there would be in principle no limit to the sales that could
be generated. In practice there is a so-called saturation point when all the
consumers in the market bought as much as they could use. (Extrapolating
beyond the range of existing data in invariably subject to a lot of uncertainty.)


• Linear equations. For example
7x 1 25.9 3y−2 1
7y − 1 = 3(y + 5), 8
+ 12 = 3x, x
= 0.88, 8
= 10 y.
• Formulating linear equations. Solving problems by converting textual

information into mathematical equations.
• Changing the subject of a formula. The technique follows on from the

method for solving linear equations: treat the new subject as a variable,
everything else as a constant. Formulae such as
1−t 2
q
1 1 x−2 4
y = 4 x+ 2 , y= 2 , V = 3 πr 3 , T = k a+x
a−x
, x = 2+t 2
• Linear inequalities. For example 3x − 1 < 5, 2(1 − 3a) ≤ a + 1.

Rearrange just like an equation except never multiplying or divide by
a negative number.
• Simultaneous linear equations. Solving pairs of equations in two

variables. For example
x−1
x + y +1 =0 2(p − q) =1 3
=y
and 2x + 3y =1 and 4p − 5 =q and y − 4x =7
• Combining inequalities Linear and non-linear. One variable case

x
e.g. x > 2 and x ≤ 5, x ≤ 3 or x ≤ −1, 2 > 3 or 1 − x > 4
Two variable case. Graphical solution e.g. x + y < 5, y ≤ 2x and y > 1,
y ≥ (x − 1)2 and x + y < 3.

Chapter 7
QUADRATIC RELATIONSHIPS
Overview
7.1 Completing the square. For example
¢2
x 2 + 6x − 5 ≡ (x + 3)2 − 4, 3x 2 + 4x − 7 ≡ 3 x + 23 − 25
¡
3
7.2 Maximum or minimum values of a quadratic expression.
7.3 Solving quadratic equations. Three methods:
p
−b± b 2 −4ac
(a) Factorising, (b) Completing the square, (c) the Formula x = 2a
7.4 The discriminant. Determining the number of roots.
Applications to problems.
7.5 Disguised quadratic equations. For example
p
x 4 − 2x 2 − 8 = 0, x −2 x −3 = 0
7.6 Sketching quadratic functions. Which way up is it? Intersection with
the axes. No roots, distinct roots, multiple roots.
7.7 Quadratic inequalities. Reminder about linear inequalities. Solving by
curve sketching. Modulus notation, for example |x| < 3 meaning −3 < x < 3.
7.8 Inequalities with an algebraic denominator. If an inequality includes a
term with an algebraic denominator, multiplying through by that
denominator is not possible because it may be negative.
1 x+1
For example x−4 < 3 or x ≥ 2.
7.9 Non-linear simultaneous equations. For example
y = x 2 − 2x + 1 and x(y − 1) = 18
y = x +3 2x = 3y
7.10 Non-linear simultaneous inequalities in two variables.
For example using a sketch to show where the values of x and y are for which
y > x(x − 2) and y < x + 4.
7.11 Circles. Theorems and properties.
7.12 Equation of a circle with centre (a, b) and radius r :
(x − a)2 + (y − b)2 = r 2
7-1
CHAPTER 7. QUADRATIC RELATIONSHIPS
7.1 Completing the square.

A quadratic relationship is inherently more complex—and therefore more
interesting—than a linear one. Its graph is a curve rather than a straight line
and there is a fascinating interplay between the the features of the graph and
its algebraic structure. We will see this emerge as the chapter unfolds.
For any quadratic function there are three ways we can write it. Each one
tells us something different about it.
Here’s an example:
x 2 + 6x + 5
(x + 1)(x + 5)
(x + 3)2 − 4.
If we expand the two that involve brackets we can see they are equivalent1 :
(x + 1)(x + 5) = x 2 + x + 5x + 5 = x 2 + 6x + 5
(x + 3)2 − 4 = x 2 + 6x + 9 − 4 = x 2 + 6x + 5
Of the three versions the form (x + 3)2 − 4 is probably the least familiar, but
it tells us something vital about the function that is completely opaque in the
other two.
In this section we’ll be concentrating on this version, and what it tells us.
We’ll be returning to the other two versions later in the chapter.
First we’ll be loooking at how any quadratic expression can be written in this
form, using a process called completing the square.
1
For more on multiplying out brackets see Section 3.2 on page 3-4.

7.1.1 x 2 coefficient = 1
We start with the simplest case, where the coefficient of x 2 is 1—like in the
example above. The process involves three steps.
Step 1.
Starting with the expression x 2 + 6x + 5 we construct the perfect square2
which, when multiplied out, generates the x 2 term and the x term.
In this case it is:
(x + 3)2 = x 2 + 6x + 9.
Don’t worry that the constant terms are different—we’ll come to that.
The perfect square will always be of the form (x + ?)2 . Can you see3 how to
work out the missing number (in this case +3)?
It will simply be half the coefficient of x, (including its sign).
Step 2.
We replace the two terms involving x, using the perfect square but adding or
subtracting a number. In this case
x 2 + 6x = (x + 3)2 − 9
Step 3.
We finish by adding back in the correct constant term:
x 2 + 6x + 5 = (x + 3)2 − 9 + 5
= (x + 3)2 − 4.
2
For a reminder about perfect squares see page 3-58.
3
(x + 3)2 = x 2 + 6 x + 9.
(x − 7)2 = x 2 − 14 x + 49, etc.

Two more examples.
Example 1 Write the quadratic x 2 − 10x − 3 in the form (x + a)2 + b.
Notice the way the question is posed: it just means complete the square.
Step 1. Construct a perfect square: (x − 5)2 = x 2 − 10x + 25.
Step 2. Replace terms involving x: x 2 − 10x = (x − 5)2 − 25.
Step 3. Add the constant term:

x 2 − 10x + 3 = (x − 5)2 − 25 − 3
= (x − 5)2 − 28.
Answer: x 2 − 10x + 3 ≡ (x − 5)2 − 28.
N.B. Every line above is an identity, i.e. it holds for all values of x, so strictly
speaking the sign in each case should be ≡ not =. It’s conventional to use the
= sign unless we want to emphasis the fact that the relationship is an
identity.
Example 2 Write the quadratic x 2 − 6x in completed square form.
Step 1. Construct a perfect square: (x − 3)2 = x 2 − 6x + 9.
Step 2. Replace terms involving x: x 2 − 6x = (x − 3)2 − 9.
That’s it—no Step 3 needed.
QQ
For each of these quadratic expressions, complete the square.
(a) x 2 + 8x + 11 (b) x 2 + 2x − 7 (c) x 2 − 6x + 10

(d) x 2 + 12x − 8 (e) x 2 + 4x
4
(a) (x + 4)2 − 5 (b) (x + 1)2 − 8 (c) (x − 3)2 + 1 (d) (x + 6)2 − 44 (e) (x + 2)2 − 4

So far so good. The fun starts when we find fractions creeping in.
Example 3 Write the quadratic x 2 − 3x + 4 in completed square form.
In this example the x-coefficient is an odd number, so halving it inevitably

introduces fractions5 .
Step 1. Construct a perfect square: (x − 23 )2 = x 2 − 3x + 94 .

3 9
Step 2. Replace the terms involving x: x 2 − 3x = (x − 2 )2 − 4
Step 3. Add the constant term:

3 2 9 16
µ ¶
2
x − 3x + 4 = x − − +
2 4 4
µ ¶2
3 7
= x− +
2 4
Answer:
3 2 7
µ ¶
x 2 − 3x + 4 = x − +
2 4
Get into the habit of a doing a quick check at the end, by expanding he result,
preferably in your head:
µ ¶2
3 7 9 7 9 16
x− + = x 2 − 3x + + = x 2 − 3x + +
2 4 4 4 4 4
Forming the perfect square bracket has been quite straightforward. But that is
because the coefficient of x 2 was 1.
What if it is not 1? That’s what we are going to look at next.
5
To remind yourself about manipulating fractions see pages 4-11 and 4-26.

7.1.2 x 2 coefficient 6= 1
This puts the cat among the pigeons when it comes to forming the perfect
square.
In the example below you’ll see how we cunningly sidestep the problem. It
needs two extra steps—one at the beginning and one at the end.
Example 4. Complete the square for the quadratic 3x 2 + 4x − 7.
Step 0. Form a (large) bracket

· ¸
2 4 72
3x + 4x − 7 = 3 x + x −
3 3
Step 1. Construct a perfect square
2 2
µ ¶
4 4
x+ = x2 + x +
3 3 9
Step 2. Replace the terms involving x
2 2 4
µ ¶
2 4
x + x= x+ −
3 3 9
Step 3. Add the constant term
2 2 4 21
µ ¶
24 7
x + x− = x+ − −
3 3 3 9 9
µ ¶2
5 9
= x− −
8 64
Step 4. Remove the square bracket
2 2 25
·µ¶ ¸
2
3x + 4x − 7 = 3 x + −
3 9
µ ¶2
2 25
=3 x+ −
3 3
µ ¶2
2 2 25
Answer: 3x + 4x − 7 ≡ 3 x + −
3 3

Dealing with fractions

It’s worth stepping back for a moment to think about the work we needed to
do with fractions. Our objective is for this to become second nature—so that
we can do it without it interrupting out train of thought on other issues.
4
In Step 1, we have to divide the fraction 3 by 2.
To divide a fraction by 2, we can either halve the numerator or double the

denominator.
2
First we’ll verify that the answer is 3 by applying the hammer and boot rule
(see page 4-20):
4
3 4 2
= = .
2 6 3
Now see how this can be achieved by either halving the numerator:
4 4÷2 2
→ = ,
3 3 3
or doubling the denominator.
4 2 4 2
→ = = .
3 3×2 6 3
This makes sense, if you think about it6 .
In Step 4 there’s a pitfall to watch out for: we might be tempted to write

µ ¶2
2
3 x+ as (3x + 2)2 .
3
Why is it wrong to multiply everything inside the bracket by 3?

The reason is in the footnote7 .
Exam questions will sometimes be written in a way that obscures the nature of
the problem. It makes it look harder than it is. Next we are going to see an
example of this.
6
In the first case we have 2 thirds instead of 4; in the second we are dividing 4
into 6 equal parts rather than 3
7
What’s inside the bracket is squared but the 3 isn’t squared. To be correct the
¤2
original expression would need to have been 3(x + 32 ) .
£

Example 5 Write the expression 9x − (2x + 1)2 in the form a(x + b)2 + c.
Although it looks a bit like a completed square expression, it isn’t: in a

completed square there is no x term outside the bracket.
So before we can start we need to expand it:
9x − (2x + 1)2 = 9x − (4x 2 + 4x + 1)

= −4x 2 + 5x − 1
Now we can apply the standard algorithm
Step 0. Form a (large) bracket

· ¸
2 5 1 2
−4x + 5x − 1 = −4 x − x +
4 4
Step 1. Construct a perfect square
5 2
µ ¶
5 25
x− = x2 − x +
8 4 64
Step 2. Replace terms involving x
5 2 25
µ ¶
2 5
x − x= x− −
4 8 64
Step 3. Add the constant term
5 2 25 16
µ ¶
25 1
x − x+ = x− − +
4 4 8 64 64
5 2 9
µ ¶
= x− −
8 64
Step 4. Remove the square bracket
5 2 9
·µ¶ ¸
2
−4x + 5x − 1 = −4 x − −
8 64
µ ¶2
5 9
= −4 x − +
8 16
5 9
Answer: a = −4, b = − 8 , c = 16 .
Now you are familiar with the technique, you may want a slicker way to set it
out. Next we are going to illustrate one such layout.

The red box highlights how we capture both x terms using a perfect square.
Of course you don’t need to put that in yourself.
Example 1 [Short version.] x 2 − 10x − 3.
x 2 − 10x −3
= (x − 5)2 − 25 −3
= (x − 5)2 − 28.
Example 2 [Short version.] x 2 − 3x + 4.
x 2 − 3x +4
= (x − 32 )2 − 94 + 16
4
= (x − 32 )2 + 74 .
Example 3 [Short version.] 3x 2 + 4x − 7.
3x 2 + 4x − 7
h
= 3 x 2 + 34 x − 73
¤
h
= 3 (x + 23 )2 − 49 − 21
¤
9
= 3 (x + 23 )2 − 25
£ ¤
9
= 3 (x + 32 )2 − 25
3
.
You might like to practice this layout on Example 4.
QQ
For each of these quadratic expressions, complete the square (the answers
are in the footnote8 ).
(a) x 2 + 5x − 2 (b) x 2 + x + 7 (c) 2x 2 + 8x − 5 (d ) 2 − 5x − x 2
(e) 3x 2 − 2x + 1
To see how completing the square is used to solve quadratic equations, go to

page 7-30.
¢2 ¢2 ¢2
(a) x + 52 − 33 (b) x + 12 + 27 (c) 2(x + 2)2 − 13 33
− x + 52
¡ ¡ ¡
8 4 4 (d) 4
¢2
(e) 3 x − 31 + 23
¡

7.2 Maximum and minimum values

So why have we gone to all this trouble learning how to express a quadratic in
this completed square form? One important reason is this:
every quadratic expression has either a minimum or maximum value,

and we can see what that value is merely by looking at the completed
square form.
What exactly do we mean by this statement? We’ll explain using the function
y = x 2 − 4x + 5.
We will see that, regardless of the value of x, the valaue of x 2 − 4x + 5 (which

we are labelling y) can never be less than 1.
What’s going on here? We’ll start by putting in a few values. For example if we
let x take the value 1, the value of y is 2; if we let x take the value 20, the value
of y is 475, and so on.
Obviously we can’t check out every possible value of x, but we can look at a
range of values, like in the table below. It shows how the value of y (i.e.
x 2 − 4x + 5) changes as x takes different values.
x −1 0 1 2 3 4 5
2
x 1 0 1 4 9 16 25
−4x 4 0 −4 −8 −12 −16 −20
+5 5 5 5 5 5 5 5
y 10 5 2 1 2 5 10
Look at the top line and the bottom line, and think about what’s happening
to the value of y as x takes different values. As we go from left to right, x gets
bigger. At the same time the value of x 2 − 4x + 5 (i.e. y) falls—as far as 1—and
then it starts to go up again.
It looks as if the function never takes a value less than 1 (which it hits when
x = 2).
But how can we be sure it doesn’t go lower at some point, perhaps just to the
left or right of x = 2, or for some value of x outside the range of the table
altogether.
Completing the squares removes all doubt. That’s what we are going to do
next.

Finding maximum and minimum points

The completed square form9 of x 2 − 4x + 5 is
(x − 2)2 + 1.
As a (number)2 , the (x − 2)2 term can never be negative—regardless of the

value of x.
Make sure you are absolutely clear about this, because it’s crucial10 .
So we express this more concisely by saying that for every value of x,
(x − 2)2 ≥ 0, and (x − 2)2 = 0 only when x = 2.
Therefore the lowest value (x − 2)2 + 1 can possibly have is 1, which it does
when x = 2.
By completing the square we have found that, whatever the value of x
x 2 − 4x + 5 ≥ 1
and that the function takes its minimum value of 1 when x = 2.
All this is leading very naturally into the notion of using a graph to show the
relationship visually.
y
C 6
C
C
C
5
C
y = x 2 − 4x + 5
= (x − 2)2 + 1
1 •
- x
2
It shows graphically how the function y = x 2 − 4x + 5 takes its lowest value, of

1, when x = 2.
We say it has a minimum point at (2, 1).
9
Multiply the bracket out in your head to satisfy yourself this is correct.
10
Because the square of any number, positive or negative, is positive or zero.

In this next example we’ll get a maximum point, because the curve is the
other way up. This corresponds to a negative x 2 coefficient.
Example 6 Find the maximum point of the curve
y = 3 + 10x − x 2 .
We showed (in Example 1 of the previous section) that

x 2 − 10x − 3 = (x − 5)2 − 28.
Reversing all the signs
3 + 10x − x 2 = 28 − (x − 5)2 .
This time the perfect square term (x − 5)2 (which is always positive or zero
regardless of the value of x) is subtracted from 28. So the function 3+10x − x 2
can’t be greater than 28. And, as before, it’s value will equal 28 when (x −5)2 =
0, i.e. when x = 5.
The power of the completed square is that it makes evident something that is
far from obvious when the function is written in another way, namely that
3 + 10x − x 2 ≤ 28 for all values of x

2
and 3 + 10x − x = 28 when x = 5.
This is confirmed by the graph of the function: there is a maximum point at

(5, 28).
y
6
28 •
y = 3 − 10x − x 2
= 28 − (x + 5)2
3
E
E - x
EE
5
The next example contains nothing new. But it will look more complicated,
because we have to introduce fractions. It is fairly typical of the sort of
problem you could get in an exam.

Example 7 Find the value of x for which the function x 2 + 4x − 7 takes its
least value, and find what that least value is.
This is the function we used in Example 3.

2 2 25
µ ¶
2
y = 3x + 4x − 7 ≡ 3 x + − .
3 3
It makes it slightly clearer if we write the terms the other way round:
2 2
µ ¶
25
y = − +3 x + .
3 3
Now 2 2 2 2
µ ¶ µ ¶
x+ ≥ 0 so 3 x + ≥ 0,
3 3
25
and we can see that y ≥ − 3 .
25 2
The smallest value y can take is − 3 , which it does when x = − 3 .
It is worth taking a close look at the graph in this case.
We know that the curve has a minimum point at (− 32 , − 25

3
) (because the x 2
coefficient is positive).
Also Putting x = 0 in 3x 2 + 4x − 7 tells us that the y-intercept is −7.
The graph must therefore look like this:
y
6
CC
C - x
−7
•
(− 32 , −8 13 )
Notice that in sketching the curve we let the position of the key points
dictate how we draw the axes.
In this case we are most interested in the bit of the curve where x and y are
both negative. And we used a different scale on each axis, to bring out the
fact that the minimum point is close to the y-axis but not on it.
Taking a step back for a moment, why might we be interested in a maximum
or minimum value? The answer is simple: in real life the function describes
something. It might be a stress force on an aircraft wing or the flow of gas
through a pipeline or the cost of a major building project. Whatever it is,
understanding its maximum value or minimum value can be critically
important.
To that end we complete this section with a practical illustration. It’s a little
bit artificial, but it does show how quadratic functions can pop up in
unexpected places.

Example 8
A company making widgets needs to decide how much to sell them for. The
higher the selling price the greater the profit on each one, but the fewer it
sells. Based on past sales data, the company finds that the following
approximate relationship applies:
N = 400(10 − x),
where x is the unit price and N is the number it sells each month. Each
widget costs £2 to make.
What is the optimum selling price?
Solution.
If £p is the profit on each one (called the marginal profit) then p = x − 2.
It’s important to get a feel for what’s going on, so we start by drawing the
graph corresponding to each of the formulae. They are both straight lines.
£p N
6 6
4, 000
J
J
J
J
J
J
J
J
-x J -x
10
−2
The graphs reflect the fact that the break-even 11 price of a widget is £2, and
that the company would never 12 sell more than 4, 000 per month. Nobody
buys the widgets unless they are priced under £10.
Suppose we let the profit per month be £Y . Since Y = N p,
Y = 400(10 − x)(x − 2) = −400[x 2 − 12x + 20]
= −400[(x − 6)2 − 16] = 6400 − 400(x − 6)2
The optimum13 price is £6, delivering a profit of £6, 400 per month.
The profit function Y looks like this:
£Y
6
6, 400
-x
2 6 10
11
p < 0 if x < 2. We make a loss if we sell them for less than £2.
12
It is what the demand would be even if the price were zero.
13
In reality Y will depend on many variables and relationships may not be linear.

Exercises 26
In these exercises, feel free to use either the short or the longer layout when
completing the square. Eventually you will find the shorter preferable.
1. Rewrite each of these quadratic functions, complete the square and

write down the coordinates of the maximum or minimum point.
(a) x 2 + 4x + 7 (b) x 2 + 2x + 5 (c) x 2 + 6x + 4
(d) x 2 + 6x − 3 (e) x 2 − 14x + 50 (f) x 2 − 2x + 1
(g) 3x 2 − 24x + 3 (h) x 2 + 3x + 2 (i) 4x 2 − 20x − 20
(j) x 2 − x + 1 (k) x 2 + 9x − 4 (l) x 2 + 13x + 40.
2. For each of the function in the left hand column of Question 1, sketch
its curve, marking the y-intercept and the position of its maximum or
minimum point.
3. Write each function in completed square form, and hence write down
the coordinates of its maximum or minimum point.
(a) 37 + x(x − 12) (b) 1 − 6x + x 2 (c) t 2 + 10t + 3
(d) w(w − 16) (e) 12 − 7x + x 2 (f) (x − 1)2 − x
(g) 4x + (x + 3)2 (h) 8 + y(y + 1) (i) (x + 32 )2 + (x + 32 ).
4. Write each of these expressions in completed square form.

(a) 2x 2 + 8x + 3 (b) 1 + 6x − x 2 (c) 3x 2 − 6x + 2
(d) 2x 2 + 3x + 5 (e) (x + 1)2 + (x + 2)2 (f) 4x 2 − x − 3
(g) 1 + 2x − 3x 2 (h) −5x 2 + 8x + 3 (i) 2x(1 − x) − 1
2 2
(j) (3x + 1)(x + 1) + 2x (k) (y − 1) + 3y (l) 2t + t (4 − t )
5. The graph shows a sketch of the quadratic

function y
y = (x + a)2 + 3. 6
(0, k)
The minimum point is (1, b).
(1, b)
(a) Write down the values of a and b.
(b) Show that k = 4. - x
6. y The sketch shows the function

(−2, 6) 6 y = p − (x + q)2 .
r
Its maximum point is (−4, 6) and its
y-intercept is at (0, r ).
-x
Find the values of p, q and r .

7. The graph shows a sketch of the function

f (x) = 2(x + 1)2 + β. y
The minimum value of f (x) is −8 and x =α - 6
the axis of symmetry of the curve is x =
- x
α. P Q
(a) Write down the values of α and β. HH
Y
R
(b) Find the positions of the points P ,
Q and R, where the curve crosses the
axes.
8. A farmer plans to make a rectangular enclosure to hold some sheep.

He only needs to construct three sides, the fourth side being provided
by an existing straight wall.
L
x x
WALL
He has 20 m of fencing and all the necessary posts. (Assume that the
entrance to the enclosure will be through an opening in the wall.)
(a) Assuming the width of the enclosure to be x m, find its length L in
terms of x.
(b) Show that its area is 20x − 2x 2 square metres.
(c) What dimensions provide for the largest possible enclosed area?
9. The height of an object t seconds after being projected directly up into

the air with velocity v m s−1 is given approximately14 by the formula
h = v t − 5t 2 .
If v = 30, find how high it reaches before falling back to earth.
10. A closed rectangular cuboid box of height 2 cm is constructed using a

framework of rigid wire, as shown in the diagram. The frame is made
using 40 cm of wire and every surface of the box is covered in fabric.
Let the length of the box be x cm.

(a) Prove that d + x = 8. d cm
(b) Show that the total surface area
of the box is given by the formula 2 cm 6

A = 32 + 16x − 2x 2
-
x cm
(c) Find the largest possible outside surface area of the box.
11. Repeat Question 10 but with the right hand end open.
14
The calculation ignores air resistance, and the formula assumes gravitational
acceleration is 10 m s−2 . A more accurate value would be 9.81 m s−2 .

12. What is the smallest value that each of these functions can take:
(a) f (x) = (x + 1)2 + (x 2 + 1)
(b) f (x) = (x − 1)2 + (x 2 − 1)
(c) f (x) = (x + 1)2 − (x 2 − 1)
(d) f (x) = (x 2 + 1)2 + (x 2 − 1)2
13. Show that the minimum point on the curve

y = x(x − 2k)
where k is a constant, is at (k, −k 2 ).
Find two values of k for which the minimum point of the curve lies on
the line y = x.
14. A small direct marketing company finds that the profits made in
previous years followed a pattern, described quite closely by the
formula:
P = 10s(8 − s) − 20,
where P is the annual profit figure (in thousands of pounds), and s is
the number of sales staff it employed that year.
Based on this model, what is the maximum profit the company can
expect to make, and how many sales staff should it employ to achieve
it?
15. Show that for the function

f (x) = x 2 (x + 6) − x(x + 2)2 + 5,
f (x) ≥ 3, whatever the value of x.
Hence write down the solution of the equation
x 2 (x + 6) − x(x + 2)2 + 2 = 0.

Solutions 26
1. (a) (x + 2)2 + 3 (b) (x + 1)2 + 4 (c) (x + 3)2 − 5

(d) (x + 3)2 − 12 (e) (x − 7)2 + 1 (f ) (x − 1)2
¢2 ¢2
(g) 3[(x − 4)2 − 15] (h) x + 23 − 14 (i) 4 x − 52 − 45
¡ ¡
¢2 ¢2 ¢2 9
(j) x − 21 + 43 (k) x + 92 − 97 (l) x + 13
¡ ¡ ¡
4 2
−4
Notice that in parts (g) and (i) there is a common factor, which we
can take out before completing the square.
2. (a) y (b) y (c) y (d) y

6 6 6 6
7 1
-x -x
−3
1
(−2, 3) (4, −15) ( 12 , 34 )

-x
(−3, −12) -x
3. (a) (x − 6)2 + 1 (b) (x − 3)2 − 8 (c) (t + 5)2 − 22

¢2 ¢2
(d) (w − 8)2 − 64 (e) x − 27 − 41 (f) x − 32 − 45
¡ ¡
¢2
(g) (x + 5)2 − 16 (h) y + 12 + 31 (i) (x + 2)2 − 14
¡
4
4. (a) 2(x + 2)2 − 5 (b) 10 − (x − 3)2 (c) 3(x − 1)2 − 1

¢2 ¢2 ¢2
(d) 31 + 2 x − 34 (e) x − 72 + 15 (f ) 4 x − 81 − 49
¡ ¡ ¡
8 8 16
¢2 4 2 1 2
(g) 34 − 3 x − 13 1 1
¡ ¡ ¢ ¡ ¢
(h) 5 − 5 x + 5 (i) − 2 − 2 x + 2
¢2
(j) 3(x + 1)2 − 2 (k) 4 y − 14 + 34 (l) 9 − (t − 3)2
¡
5. (a) a = −1, b = 3. (b) y = (x − 1)2 + 3 = x 2 − 2x + 4.
6. p = 6, q = 2, r = 2. [y = 6 − (x + 2)2 = 6 − (x 2 + 4x + 4)].
7. (a) α = −1, β = −8. [The minimum point is (−1, −8).]

(b) R(0, −6). [y = 2(x + 1)2 − 8 = 2x 2 + 4x − 6.]
P (−3, 0) and Q(1, 0).
y = 0 ⇒ 2(x + 1)2 = 8 ⇒ (x + 1)2 = ±4 ⇒ x + 1 = −2 or +2
8. (a) L = 20 − 2x. (b) Area A = Lx = x(20 − 2x).

(c) x = 5 m, L = 10 m.
A = −2[x 2 − 10x] = −2[(x − 5)2 − 25] = 50 − 2(x − 5)2 ,
so maximum area is 50 m2 , with x = 5 and L = 20 − 2x = 10.
9. Maximum height reached = 45 m (after 3 seconds).

h = 30t − 5t 2 = −5[t 2 − 6t ]
= −5[(t − 3)2 − 9] = 45 − 5(t − 3)2 .
10. (a) Total length of wire = 4d + 4x + 8 = 40

(b) A = 2xd + 4d + 4x = 2x(8 − x) + 4(8 − x) + 4x
(c) Maximum surface area = 64 m2 .
A = −2[x 2 − 8x − 16] = −2[(x − 4)2 − 32] = 64 − 2(x − 4)2 .

11. (a) As Question 10.

(b) A = 2x(8 − x) + 2(8 − x) + 4x = 16 + 18x − 2x 2 .
(c) Maximum surface area = 56 12 m2 .
A = −2[x 2 − 9x − 8] = −2[(x − 92 )2 − 113 113 9 2
4 ] = 2 − 2(x − 2 )
Notice that having a side missing leads to a different shape.
(a) 23 2[x 2 + x + 1] = 2 (x + 21 )2 + 34
£ ¤
12.
(b) − 12 2[x 2 − x] = 2 (x − 21 )2 − 14
£ ¤
(c) No minimum 2x + 2
(d) 2 2x 4 + 2 (x 4 ≥ 0)
13. y = x 2 − 2kx = (x − k)2 − k 2 , giving a minimum point of (k, −k 2 ).

Because k doesn’t have a specific value, the equation is really a lot
of different curves, one for each value of k. The vertex of one of
these curves lies on the line y = x if k = −k 2 . This is an equation
in k which factorises to: k(1 + k) = 0, so k = 0 or k = −1.
The corresponding minimum points are (0, 0) and (−1, −1).
To solve the problem we treated k (a constant) as a variable.
14. Profit of £140, 00, with 4 sales staff.

P = −10s 2 + 80s − 20 = −10[s 2 − 8s + 2] = −10[(s − 4)2 − 14]
= 140 − 10(s − 4)2
15. f (x) = x 3 + 6x 2 − x(x 2 + 4x + 4) + 5

= x 3 + 6x 2 − x 3 − 4x 2 − 4x + 5
= 2x 2 − 4x + 5
= 2 x 2 − 2x − 52
£ ¤
= 2 (x − 1)2 + 32
£ ¤
= 3 + 2(x − 1)2
x = 1.
We can write the equation as x 2 (x + 6) − x(x + 2)2 + 5 = 3
i.e. f (x)=3, its minimum value, which it takes when x = 1.

7.3 Solving quadratic equations

In Chapter 5 we solved linear equations in one variable by rearranging the
terms. The procedure is straightforward: we get an x on its own on one side
of the equation; what we are left with on the other side is the solution.
This doesn’t work for quadratic equations, like for example

x 2 = 8x + 20.
This is because however we re-write it we can never separate out one single x
term. Sometimes students try by dividing both sides by x:
8x + 20
x= .
x
It certainly gives us a single x on the left had side, but it gets us nowhere
because we still have an x on the right hand side 15 . If we think about it,
solving a quadratic equation by rearranging it is doomed to failure, because
it normally has two solutions: if we were able to solve it this way we would
only get one solution.
15
In an act of desperation, some students “cancel" the x’s on the right hand side and
conclude that x = 28. We can’t do that because cancelling a fraction means dividing
the (whole of the) top and the (whole of the) bottom by the same number.
(See section 4.3 on page 4-5).

7.3.1 The zero product principle

This is a really important principle. It crops up again and again. Imagine
there are two boxes. In each one there is a card with a number written on it.
We don’t know either number, but we are told their product is zero.
? × ? =0
What could the numbers be? Think about it.

It can’t work unless one of the numbers is zero. Or both could be zero.
A very simple application is solving an equation like
5x = 0.
One of the two numbers must be zero. The left hand one clearly isn’t, so x
must be zero.
In an equation like
x(x − 7) = 0
we can think of his as the two numbers x and x − 7 multiplied together16 . So
either
x =0 or x − 7 = 0
i.e. x =0 or x = 7.
If the product is in the form of two brackets multiplied together, we have

something like this
(x − 10)(x + 2) = 0 ⇒ x − 10 = 0 or x + 2 = 0
⇒ x = 10 or x = −2.
Actually we have just solved the quadratic equation we began the section
with, because
x 2 = 8x + 20
⇒ x 2 − 8x − 20 = 0
⇒ (x − 10)(x + 2) = 0
16
Being adjacent to each other means they are multiplied.

It is not difficult to see that the zero product principle applies however many
numbers are multiplied together: the product of the numbers can’t be zero
unless at least one of the numbers is zero.
For example
x(x − 2)(x − 4) = 0.
This is a cubic equation: when we multiply it out we get
x 3 − 6x 2 + 8x = 0
but we can easily solve it. The zero product principle tells us that
x = 0, x − 2 = 0, or x − 4 = 0
so it has three solutions: x = 0, x = 2 and x = 4.
We will return to equations like this later. For now we’ll go more deeply into
solving quadratic equations.

7.3.2 Writing quadratic equations in the form f (x) = 0

Look at this equation.
x 2 = 8x.
It is a quadratic equation (all the terms are either x 2 , x or constant terms).
Since the constant terms happens to be zero we might be inclined to do divide
each side by x:
x A2 = 8Z
x
and deduce that the solution is x = 8.
It is certainly true that x = 8 satisfies the original equation. But the notion of
solving means finding all possible answers. We can see that in this instance
another possible solution is x = 0; that satisfies the equation as well.
If this were an exam question there might be one mark for each solution, so
if we didn’t “spot" the zero solution we’d lose half the marks for that part of
the question.
The answer is always to write the quadratic with a zero on one side of the
equation:
x2 = 8x
x 2 − 8x = 0
x(x − 8) = 0
⇒ x = 8 or x −8 = 0
⇒ x = 8 or x =8
What was wrong with the cancelling method? The answer is that
we can’t divide by zero—or a quantity that may be zero.
If a quantity is algebraic the possibility exists that it may be zero.

This means we can never divide by an algebraic quantity—unless we know
for sure it’s value isn’t zero.
For our purposes it’s important that
we always write a quadratic equation with a zero on one side.

“Proof" that 2 = 1
There is wonderful, quite simple, piece of algebraic reasoning that appears to
conclude that 2 = 1.
Of course it’s not a valid proof: it exploits the cancelling pitfall we just looked
at. But it’s an interesting bit of fun that contains a valuable message. To see
the “proof", turn to page A-14 of Appendix A (or click on the page number).
You may want to challenge your friends to spot the error in the logic.

7.3.3 Quadratic equations. Method 1: factorising

This brings us to the first method for solving quadratic equations.
If we go back to the quadratic equation we started with:

x2 = 8x + 20
which can be written x 2 − 8x − 20 = 0

or (x − 10)(x + 2) = 0
and the solution follows by the zero product principle:

(x − 10)(x + 2) = 0 ⇒ x − 10 = 0 or x + 2 = 0
⇒ x = 10 or x = −2.
Summarising, to solve a quadratic equation by the factorising method:
Step 1. Rearrange the equation so one side equals zero.
Step 2. Factorise the other side.
Step 3. Put each factor = 0.
Step 4. Solve these two linear equations.
If you need to remind yourself about factorising quadratics, see Chapter 3

Section 3.6 on page 3-43. To remind yourself about solving linear equations
look at the first section of Chapter 5.

A word of warning
The zero product principle is exactly what it says on the tin: it applies if the
product is zero. We could (perfectly correctly) write the equation above as
x 2 − 8x = 20
or x(x − 8) = 20
But there is nothing we can deduce from this. Can you see why?
There are lots
p of pairs
p of numbers that multiply together to give 20 (4 and 5,
8 and 2.5, 20 and 20 to name just a few.), so it’s fruitless to try and infer
something about the value of x from this form of the equation.
Only if there is a zero on one side can we infer anything useful.

Example 9 Solve the equation 2x 2 − x − 3 = 0.
Solution
2x 2 − x − 3 = 0
(2x − 3)(x + 1) = 0
⇒ 2x − 3 = 0 or x + 1 = 0
3
⇒ x = or x = −1
2
Not all equations will be quite as straightforward. The following examples

illustrate some of the twists you will meet, and the things to look out for.
For example the equation may not even appear to be a quadratic.
Another twist is that the factorising could be one-bracket factorising.
And remember, if the equation contains fractions, we can always get rid of
them by multiplication, in the same way that we did with linear equations
(see Section 6.1).
Example 10 Solve the equation 2x(x + 1) = x 2 + 15.

Solution
2x(x + 1) = x 2 + 15
2x 2 + 2x = x 2 + 15
x 2 + 2x − 15 = 0
(x + 5)(x − 3) = 0
⇒ x +5 = 0 or x − 3 = 0
⇒ x = −5 or x = 3
Next we are going to look at three examples At first sight they seem very similar,
but the subtle differences affect how we tackle them.

Example 11 Solve the equation x 2 − 9 = 0
Solution
x2 − 9 = 0
(x + 3)(x − 3) = 0 (the difference of two squares)
⇒ x + 3 = 0 or x − 3 = 0
⇒ x = −3 or x = +3
Look out for the difference of two squares pattern (see Section 3.6.4).
We might meet this equation as: x2 = 9
which we could solve by by taking the quare root of each side:
x = ± 3.
This is fine, of course. It is just as good as the factorising method—provided
we don’t forget the negative solution. It’s another example of the value of
always writing a quadratic equation with a zero on one side.
Example 12 Solve the equation x 2 − 9x = 0
Solution
x 2 − 9x = 0
x(x − 9) = 0
⇒ x = 0 or x − 9 = 0
⇒ x = 0 or x = 9
We are still using factorisation in this example, but in this case one-bracket
factorising not two-bracket factorising.
Notice how similar this example looks to Example 10.
Example 13 Solve the equation x 3 − 9x = 0
Solution x 3 − 9x = 0
x(x 2 − 9) = 0
x(x + 3)(x − 3) = 0
⇒ x = 0 or x + 3 = 0 or x − 3 = 0
⇒ x = 0 or x = −3 or x = 3
Here we applied the zero product principle with more than two factors17
17
Alternatively we could view x(x 2 − 9) = 0 as two factors ⇒ x = 0 or x 2 − 9 = 0
(the equation in Example 14).

12
Example 14 Solve the equation x− = 1
x
Solution 12
x− = 1
x
(× x) x 2 − 12 = x
(−x) x 2 − x − 12 = 0
(x − 4)(x + 3) = 0
⇒ x − 4 = 0 or x + 3 = 0
⇒ x = 4 or x = −3
It’s reassuring to check the solutions occasionally.

If x = 4, l.h.s. = 4 − 3 = 1.
12
If x = −3, l.h.s. = −3 − −3 = −3 + 4 = 1.
4x
Example 15 Solve the equation = 3−x
1+x
Solution 4x
= 3−x
1+x
(× (1 + x)) 4x = (1 + x)(3 − x)
4x = 3 + 2x − x 2
x 2 + 2x − 3 = 0
(x + 3)(x − 1) = 0
⇒ x +3 = 0 or x − 1 = 0
⇒ x = −3 or x = 1
QQ
Solve these equations by factorising (the answers are in the footnote18 ).
(a) 6x = 16 − x 2 (b) x 2 + 6x = 0 (c) x 2 = 16

6
(d ) x + 3 = 4−x (e) x(x + 1) = 2
18
(a) x = 2 or x = −8 (b) x = 0 or x = −6 (c) x = ±4
(d) x = 3 or x = −2 (e) x = −1 or x = 2

7.3.4 Quadratic equations. Method 2: completing the square

Using factorising to solve a quadratic equation is all very well (I hear you say)
provided the expression factorises. What if it doesn’t? That’s where
completing the square comes in (see page 7-2), and when we have to use this
this technique square roots are inevitable.
But first we will solve the quadratic equation we met at the beginning of the
last section (which won’t need square roots):
x 2 = 8x + 20
First we rearrange it so that one side is zero (as always, then we complete the
square):
x 2 − 8x − 20 = 0
(x − 4)2 − 36 = 0
(x − 4)2 = 36
Taking the square root of each side
x −4 = ±6
⇒ x = +6 + 4 or x = −6 + 4
⇒ x = 10 or x = −2
This (of course) is the same as the answer we found by factorising.

When we discussed Example 10, we said the equation
x2 = 9
can be solved simply by taking the square root of each side:
x = 3 or x = −3
But there is a risk we will forget the possible negative root. The factorising
route is safer.
However here we seem to have introduced the same risk when solving:
(x − 4)2 = 36.
We might forget that x − 4 could equal −6.
Usually we just try to remember! But if you’d like a “safer" way we could
substitute X = x − 4.
Then the equation (x − 4)2 − 36 = 0 becomes
X 2 − 36 = 0
(X + 6)(X − 6) = 0 (the difference of two squares)
⇒ X +6 = 0 or X −6 = 0
⇒ X = −6 or X = +6
⇒ x − 4 = −6 or x − 4 = +6
⇒ x = −2 or x = +10
The meat of this section, of course, is quadratic expressions that don’t

factorise. That’s what we will look at next.

Example 16 Solve the equation x 2 + 6x + 2 = 0 by completing the square.
Solution
x 2 + 6x + 2 = 0
(x + 3)2 − 7 = 0
(x + 3)2 = 7
p
x +3 = ± 7
p p
⇒ x = + 7 − 3 or x = − 7 − 3
We would normally leave an answer in square root form rather than

decimal form, unless the question suggested otherwise (e.g. by specifying
the number of significant figures).
By the way, it’s conventional to write the solutions with the square root last:
p p
x = −3 + 7 or x = −3 − 7.
This is to avoid the risk of the square root sign inadvertently “creeping" over
the −3 like this p
7 −3
which you may then misread. (Mathematicians are not known for their great
handwriting.)
Example 17 Solve the equation 3x 2 + 6x + 2 = 0 by completing the square.
Solution
Completing the square is more involved when the coefficient of x 2 is not 1.
The good news, with quadratic equations, is that we can always avoid
this—by dividing every term by that coefficient (in this case 3).
3x 2 + 6x + 2 = 0
2
(÷3) x 2 + 2x + = 0 [0 ÷ 3 = 0]
3
2
(x + 1)2 − 1 + = 0
3
1
(x + 1)2 =
3
rp
1 1 1
x +1 = ± = ±p = ±p
3 3 3
1
x = −1 ± p
3

Example 18 Solve the equation 9x = (2x + 1)2 by completing the square.
Solution
At first sight this looks as if it might be in completed square form already. It
isn’t because the term on the left hand side is not a constant.
9x = 4x 2 + 4x + 1
0 = 4x 2 − 5x + 1
5 1
(÷4) 0 = x2 − x +
4 4
µ ¶2
5 25 16
0 = x− − +
8 64 64
5 2
µ ¶
9
= x−
64 8
5 3
x− = ±
8 8
3 5 3 5
x = + or − +
8 8 8 8
1
x = 1 or .
4
By the way, the absence of any square roots in the answer tells us we could
have solved it by factorising:
4x 2 − 5x + 1 = (x − 1)(4x − 1) = 0.
Sometime we don’t spot this until after we have solved it. Later we will learn
a neat little calculation we can do to tell if a quadratic can be factorised.

QQ
1. Solve these equations by completing the square.
(a) x 2 + 6x − 16 = 0 (b) x 2 + 4x + 1 = 0 (c) x 2 − 5x + 6 = 0

7
(d) 2x 2 − x + 5 = 0 (e) x = x−4
1
2. Verify, by sustitution, that x = −1 + p is a solution of the equaition we met
3
in Example 16:
3x 2 + 6x + 2 = 0.
The answers (and a hint for Question 2, if you get bogged down with it) are in
the footnote19 .
Just occasionally quadratic equations arise in the context of every day

problems.
p
1. (a) (x + 3)2 = 25 ⇒ x = 2 or x = −8 (b) (x + 2)2 = 3 ⇒ x = −2 ± 3
¢2 ¢2 p
39
(c) x − 52 = 14 ⇒ x = 2 or x = 3 (d) x − 41 = 39 1
¡ ¡
16 ⇒ x = 4 ± 4
p
(e) (x − 2)2 = 11 ⇒ x = 2 ± 11
p p p
p3 3 3
2. Hint: = p = 3
3 3
p p
19 x2 = 1 − p2 + 1 = 4 − p2 . Now 3x 2 . Since p3 = 3, 3x 2 = 4 − 2 3.
3 3 3 3 3
p p p
Similarly 6x = −6 + 2 3 and we have 4 − 2 3 + −6 + 2 3 + 2 = 0
A real live problem
While I have been writing this chapter, my wife and I have been planning an
orangery (a posh name for a building or extension with lots of glass).
Imagine my surprise when I found that in order to solve a particular design
problem I needed to solve a quadratic equation. (I promise I am not making
this up!)
The problem was this. We were told that if at least 75% of the roof is glass,
the building is deemed to be a conservatory for Building Regulation
purposes20 , making life simpler in relation to planning permission. In
practice this means having a large opening in the ceiling above which there
is some kind of glass structure (known as a roof lantern).
The orangery is to be 6 m long by 3.5 m wide, and for aesthetic reasons it
seemed sensible to have the non-glass part of the roof the same width all the
way round; call this distance x.
x
6 6
?
x - GLASS x - 3.5 m
x
6
? ?
-
6m
The area of glass in the roof is
(3.5 − 2x)(6 − 2x) m 2 .
For the smallest lantern possible consistent with the 75% rule
3
(3.5 − 2x)(6 − 2x) = (3.5 × 6)
4
21
which simplifies to 4x 2 − 19x + = 0.
4
Solving this:
19 21
x2 − x+ = 0
4 16
19 2 361 84
µ ¶
x− − + = 0
8 64 64
19 2
µ ¶
277
x− =
8 64
p
19 277
x = ± = 4.46 or 0.295.
8 8
The solution x = 4.46 m makes no sense practically (it is greater than the
width of the room), leaving the solution x = 0.295 m.
The width of the unglazed part of the roof can be no wider than about 30 cm.
Whereas we would normally prefer the exact answer (involving a square root),
in practical problems such as this we want an actual measurement.
20
After I had done the calculation we learned that this rule doesn’t apply where we
live.

7.3.5 Quadratic equations. Method 3: the Formula.
The quadratic formula is:

p
−b ± b 2 − 4ac
A
x=
!
A
2a A
A
where the a, b and c in the formula refer to the coefficients of the quadratic
equation when it is written in the form
ax 2 + bx + c = 0
i.e. with zero on one side of the equation. It doesn’t matter which side 21 .
This should probably be called Method 2 12 because the formula is derived by

completing the square, so it’s not really a different method. The completing
square process is captured within the formula.
Example 19. Solve the equation x 2 − 4x − 3 = 0 using the formula.
Solution
a = 1, b = −4, c = −3, so substituting into the formula:
p
+4 ± (−4)2 − 4(1)(−3)
x = [take care with the signs]
2
p p
4 ± 16 + 12 4 ± 28
= =
2 2
This is the correct answer, but you would be expected to exploit any
opportunity for simplifying.
p
In this case there is factor of 2 embedded within 28, so we can simplify 22 it
in this way: p p p p
28 = 4 × 7 = 2 7.
So the best way to express the answer is
p
4±2 7 p
x= = 2 ± 7.
2
21
If we swap everything to the other side, all three signs change but the answer remains
the same. If this seems puzzling replace a with −a, b with −b and c with −c
in the formula; you’ll see that all the changes cancel each other out.
22
Using the rules of indices in Chapter 2 (see page 2-48). In particular the rule that
says (ab)n = a n b n , with n = 21 .

Example 20 Solve the equation
3x 2 + 7x − 1 = 0
giving your answers correct to 3 S.F.

Solution
Because the question tells us the level of precision needed (3 S.F.) we know it
can’t be solved by factorising. That means we’ll need to complete the square
or use the formula. In this case we’ll use the formula.
Before we start the calculation, an important point about precision. To be
sure the answer is correct to 3 S.F. the working needs to be to a greater
precision. Where possible do the whole calculation from start to finish in
your calculator, using the memory or memories necessary. That way you
won’t lose any precision. If you do it in stages, make sure you write down any
intermediate values to a precision of 5 or 6 S.F.
With a = 3, b = 7, c = −5
p
−7 ± 49 − 4(3)(−5)
x =
6
p
−7 ± 109 −7 ± 10.4403
= =
14 6
= −5.81344 or 1.14677
= −5.81 or 1.15 (to 3 S.F.)
QQ
Solve these equations using the quadratic formula (the answers are in the
footnote23 ).
(a) x 2 + 6x − 16 = 0 (b) x 2 + 4x + 1 = 0 (c) x(1 − 2x) = 5

7
(d ) x 2 + 6x = 0 (e) x = x−4
23 p p p
39
(a) x = 2 or −8 (b) x = −2 ± 3 (c) x = 14 ± 4 (d) x = 0 or −6 (e) x = 2 ± 11

Do you like a challenge? If so you might like to have a crack at solving this
equation using the formula:
x 2 + k(k + 1)x + k 2 = 0.
The algebraic coefficients raise the game somewhat. But it doesn’t require
anything we haven’t covered: it’s all about multiplying out brackets and
factorising, with a bit of fraction work thrown in.
You will almost certainly know if you have been successful because, although
the algebra is fairly lengthy, the solutions boil down to beautifully simple
expressions.
If you get stuck, the solution is in Appendix A, on page A-22. (Click on the
page number.)

Derivation of the quadratic formula.
We derive the formula by solving the equation ax 2 + bx + c = 0 by the

method of completing the square.
You don’t need to know how this is done; you won’t be examined on the proof.
But if you are interested, and like a challenge, try deriving it for your self. It
starts like this:
We begin by making the coefficient of x 2 equal to 1:
ax 2 + bx + c = 0
b c
x2 + x + = 0.
a a
See if you can carry on from here.

If you get stuck the complete derivation is in Appendix A, on page A-21.

7.4 The Discriminant

Does this mean we can now solve any quadratic equation? Not exactly. To see
why, we only have to consider an equation such as
x 2 − 6x + 11 = 0.
When we apply the formula we get

p p
6 ± 36 − 44 6 ± −8
x= = .
2 2
It has no solution because it’s impossible to find the square root of a negative
number, at least in the normal sense of what we mean by a number24 .
Unless you take Further Maths, you will only ever meet Real Numbers at A
Level, so for our purposes, it is impossible to find the square root of a
negative number.
Whether the equation has roots or not will therefore depend on the sign of
what’s under the square root, i.e. b 2 − 4ac.
We call this expression the discriminant of the equation.
This simple looking expression is surprisingly powerful. You will come to
treasure it because it is extremely useful. So much so that it has it’s own
symbol
∆ = b 2 − 4ac.
To be precise, the value of the discriminant will either be > 0, < 0 or = 0.
If ∆ > 0 it has two (distinct) roots, because the ± in the formula gives two
possible answers;
If ∆ = 0 it only has one root, because the square root term in the formula
disappears altogether.
If ∆ < 0 it has no roots (no square root of a negative number);
The “no root" and “two root" cases speak for themselves. In the one root case
there is something interesting going on, which, as you will se later, ties up
beautifully with what the graph of the function looks like.
But here we just need to be clear about the language you may meet in the one
root case.
24
p on page A-6 that in the world of Complex Numbers they do exist. In that world,
We saw
replacing −1 with the letter i we could write the solutions of this equation as
p p p p
6 ± −1 8 6 ± i 8 6 ± i 2 2 p
x= = = = 3 ± i 2.
2 2 2
We say they are complex roots, each comprising
p p the same real part (the number 3) and a
different imaginary part (the number i 2 or −i 2).

The one root case

If ∆ = 0, the one root case, we sometimes say that the equation has one
repeated root or two equal roots.
Why might we might use this language, rather than just say “one root"? To see
why let’s solve an equation in which ∆ = 0, for example
x 2 − 6x + 9 = 0.
We can confirm that

b 2 − 4ac = (−6)2 − 4(9) = 0
and the equation is easily solved by factorising:
(x − 3)(x − 3) = (x − 3)2 = 0 ⇒ x = 3.
As usual we have two brackets, each yielding a root. It just so happens that in
this instance the two roots are the same. Hence “one repeated root" or “two
equal roots".
This explains why we have said two (distinct) roots in the case ∆ > 0. We are
underlining the fact that the two roots are different.

Using the discriminant to solve problems

We said that the discriminant is a powerful tool. How come? Well, the
statements we made connecting the value of the discriminant and the
number of roots works both ways. If we know an equation has no real roots,
for example, its discriminant must be negative. If we know it only has one
root, the discriminant must be zero, and so on.
Mathematically we can use the phrase “if and only if" to signify this. So for
example, we could say
∆ < 0 if and only if the equations has no roots, and so on.
Many exam questions make use of this fact, as do the following examples.
Example 21 Show that the equation
x 2 + 10x + k = 0,
where k is a constant, has real roots if and only if k ≤ 25.
Solution
This is a direct application of what we have just learnt about the
discriminant, and vice versa (this is the “if and only if" bit).
For the equation to have real roots the discriminant must be ≥ 0. (They don’t
have to be distinct roots, so we must include the case ∆ = 0.)
Since a = 1, b = 10 and c = k, that means
b 2 − 4ac ≥ 0
100 − 4k ≥ 0
100 ≥ 4k
25 ≥ k
We have solved the problem, by relying on the concept of a discriminant.

What would happen if we tried to answer the question without using the
discriminant? We can’t find the solutions because we don’t know the value k,
but we can start the process. Completing the square:
x 2 + 10x + k = 0
(x + 5)2 − 25 + k = 0
(x + 5)2 = 25 − k.
Now we can see why k can’t be more than 25. If it were, (x + 5)2 would be a
negative number!

Example 22 Find the value of λ for which the equation

x 2 + 2λx + λ + 2 = 0
has only one root.
Solution
In this case the discriminant must be zero. Since a = 1, b = 2λ and c = λ + 2
b 2 − 4ac = (2λ)2 − 4(λ + 2)

so 4λ2 − 4(λ + 2) = 0
What’s interesting here is that, in the process of exploring the original

equation, we find ourselves with a new quadratic equation, in the variable λ.
The solution to our problem lies in finding the values of λ that satisfy this
new equation.
(÷4) λ2 − (λ + 2) = 0
λ2 − λ − 2 = 0
(λ + 1)(λ − 2) = 0
The original equation has only one root if λ = −1 or λ = 2.

We can check that this is right. If λ = −1 the equation is
2 2
x − 2x + 1 = (x − 1) = 0, which has a single root: x = 1.
Check the case λ = 2 for yourself.
In the next example the algebra is a little more interesting.

Example 23 For what values of p does the equation
p(x + 1)2 = x 2 + 3
have real roots?
Solution
This is an outline solution; the detail of the algebra has been left out. Make
sure you are happy with each step.
First we have to write in the usual ax 2 +bx+c = 0 form. Expanding the bracket
and writing it in the form f (x) = 0:
(p − 1)x 2 + 2px + (p − 3) = 0.
Now for reals roots the discriminant must be ≥ 0, i.e.
4A p 2 − 4A (p − 1)(p − 3) ≥ 0
which simplifies to
4p − 3 ≥ 0.
3
The equation has real roots if p ≥ 4 .
With this kind of problem we are “flying by our instruments", to borrow an

aeronautical phrase. If we had made a mistake, how would we know?
3
It is far from obvious that values of p less than 4 fail to yield any (real)
solutions!
In these circumstance we can at least try to get comfortable with the result.
One way to this would be to pick a value of p, one that makes the equation
nice and simple. An obvious one that springs to mind is p = 0. If our
conclusion is correct, the equation will have no solutions. Sure enough
0 = x2 + 3
has no solutions.
Another obvious value to choose would be p = 1, because then there is an x 2
term on each side, so the equation reduces to a linear one:
2x + 1 = 3.
We don’t have to solve it; we just have to recognise that it does have a
solution, as we expected.
This approach won’t prove we haven’t made any mistakes, but it may reveal
it if we have.

The language used in exam questions

When an expression we meet in a problem is obviously a quadratic, we
automatically look to techniques like completing the square and using the
discriminant.
In exams you will be expected to see through the way a question is described
and understand what the nature of the problem is. The next example is like
this.
Students sometimes complain about this: is the examiner being deliberately

awkward? The answer is an emphatic no. If you work as a mathematician
you won’t be asked to apply a particular technique: you will just be presented
with a compicated problem—expressed in whatever terminology the client
wants to use. The job—and of course the fascination—is in unravelling the
problem. Only then can you decide on a technique for solving it.


x 2 = α(x − α), α 6= 0
has no real roots.
Solution
There will be no real roots (solutions) if the discriminant is negative,
whatever the value of α.
Writing the equation in a form suitable for extracting the discriminant:
x 2 − αx + α2 = 0.
a = 1, b = −α, c = α2 , so
b 2 − 4ac = (−α)2 − 4α2

= −3α2
< 0 (since α2 > 0)
There is often more than one way to solve a problem. in this case, if we
complete the square the quadratic equation becomes25
1 2 3 2
µ ¶
x− α + α =0
2 4
In this form we have two terms. Both are ≥ 0 , and yet their sum is zero. This
is impossible so the equation has no solutions.
You may (or may not) have wondered what the α 6= 0 was doing in the
question. This might seem pedantic, but to a mathematician it is just being
precise. If α = 0 the original equation becomes x 2 = 0, which does have real
roots.
You don’t need to be overly concerned with such details at this stage.
However it does underline an important idea: mathematics is the only area
of knowledge where we are able to state absolutely immutable truths. But for
this to be the case we need scrupulous rigour in our reasoning.
25
You may want to check that this multiples out to give the correct equation.

Summarising, the discriminant discriminates between the three cases:
b 2 − 4ac > 0 ⇐⇒ two (distinct) roots

A
b 2 − 4ac = 0 ⇐⇒ one root
!
A
A
b 2 − 4ac < 0 ⇐⇒ no roots. A
We have used the symbol ⇐⇒ (which we met in Chapter 1) to capture the

fact that the relationships work both ways.
For example ∆ < 0 ⇒ “no roots", and “no roots" ⇒ ∆ < 0.
QQ
1. For each equation find the value of the discriminant and say whether the
equation has two (distinct) roots, one root or no roots.
(a) x 2 + 3x + 2 = 0 (b) 3x 2 + 2x + 1 = 0 (c) x 2 − 6x + 9 = 0

2
(d ) x(4x − 3) = 2 (e) x = 1−x
2. Find the value of k for which the equation x 2 + 6x + k = 0 has

(a) two (distinct) roots (b) one root and (c) no roots.
3. For which values of α does the equation x 2 + αx + α = 0 have equal roots?
3
4. Show that if k > 2 , the equation (k − 1)x 2 + 2kx + (k + 3) = 0 has no real
roots.
26
1. (a) ∆ = 1. 2 distinct roots. (b) ∆ = −8. No roots. (c) ∆ = 0. 1 root.
(d) ∆ = 41. 2 distinct roots (e) ∆ = −7. No roots.
2. (a) k < 9 (b) k = 9 (c) k > 9.
2
3. α − 4α = α(α − 4) = 0 ⇒ α = 0, or α = 4.
4. No roots ⇒ 4k 2 − 4(k − 1)(k + 3) < 0, which simplifies to −2k + 3 < 0.

7.5 Disguised quadratic equations

Sometimes an equation doesn’t appear to be a quadratic but an underlying
quadratic structure emerges when we do one (or more) of the following:
• rearrange it;
• take out a common factor;
• make a substitution.
We start with the common factor idea—checking for (one-bracket) factors.
Example 25. Solve the equation x 3 − 7x 2 − 10x = 0.
Solution
Apparently a cubic equation (for which we don’t know a formula 27 ), taking
out the common factor x, we have
x(x 2 − 7x + 10) = x(x − 2)(x − 5)
and the equation is
x(x − 2)(x − 5) = 0.
Using the zero product principle (with three numbers multiplied together)
x = 0, x − 2 = 0 or x − 5 = 0.
⇒ x = 0, x = 2 or x = 5.
As we saw on page 7-23, a common mistake is to divide through by x at the

beginning, leaving
x 2 − 7x − 10 = 0.
This only gives the solutions x = 2 and x = 5. The solution x = 0 has
disappeared, and since the equation is cubic we should expect to see three
solutions.
Two out of three is not bad. Unfortunately examiners don’t see it that way.
It’s a question of being rigorous: the omission would always lose you marks
in an exam.
The way to avoid falling into this trap is to remember that we can
A
!
A
never divide by a variable, or cancel a variable A
A
unless we can be certain its value isn’t zero.
27
There is a formula for solving a cubic equation. It is quite complicated and rarely
used today.

The following two examples bring out the power of substitution in solving
equations. (Here it is applied to quadratic equations but it’s also useful with
other types of equation).
Example 26. Solve the equation y 4 − 2y 2 − 8 = 0.

Solution
There is no common factor, but the fact that we only have even powers of y
gives us a clue. It means we can replace y 2 with, say Y , giving:
Y 2 − 2Y − 8 = 0
because y 4 = (y 2 )2 = Y 2 .
Factorising this: (Y − 4)(Y + 2) = 0
whose solutions are Y = 4 and Y = −2. But Y = y 2 , so our original equation

becomes two equations:
y2 = 4 and y 2 = −2.
The second has no (real) solutions, so the answer is y = ±2.

p
Example 27. Solve the equation x − 2 x − 3 = 0
Solution
Looking like anything but a quadratic, this equation is another that yields to
p
substitution. If we put y = x we can write x = y 2 , and we have
p
x − 2 x − 3 = y 2 − 2y − 3
= (y − 3)(y + 1)
p p
= ( x − 3)( x + 1)
Now we solve 28 the equation
p p
( x − 3)( x + 1) = 0
p p
⇒ x −3 = 0 or x +1 =0
p p
⇒ x = 3 or x = −1
⇒ x = 9 or x =1
Look back at the original equation. With the benefit of hindsight you may be
p
able to see it in your mind’s eye as a quadratic in x:
p p
( x)2 − 2( x) − 8 = 0
If so, you could skip the substitution stage and go straight to

p p
( x − 3)( x + 1) = 0,
and solve the equation from there.
28
This time we substituted back to x before solving; it doesn’t matter which way
round we do it.
We could have said: (y − 3)(y + 1) = 0 ⇒ y = 3 or y = 1, so x = 9 or x = 1.

In the next example we combine several techniques together, including the

difference of two squares (see page 3-53).
Example 28 Solve the equation x 5 = 16x.
Solution
Remember we can’t divide by x (see Example 24 above). Instead we move
everything on to one side and factorise:
x 5 − 16x = 0
x(x 4 − 16) = 0
⇒ x = 0 or x 4 − 16 = 0
To solve the equation x 4 − 16 = 0, we substitute x 2 = X , giving
X 2 − 16 = 0
(X + 4)(X − 4) = 0
⇒ X = x 2 = −4, (no solutions) or X = x 2 = 4
The only remaining solution arises from x 2 − 4 = (x + 2)(x − 2) = 0.

This quadratic has two roots and we found a solution x = 0 at the start.
The answer is: x = 0, −2 or 2.

In the previous example we solved the equation x 2 − 4 = 0 by factorising. We

could have solved it by saying
x2 = 4
⇒ x = 2 or − 2.
This is perfectly fine. Where the factorising approach has the slight edge is
that it makes it impossible to forget the negative root.
But this raises another issue. If we want to make a habit of factorising rather
than taking the square root, where does that leave us with an equation like
x 2 = 5.
Can we factorise x 2 − 5 = 0?
The answer is yes: p p
(x + 5)(x − 5) = 0
p p
which lead to the solutions we expect: x = − 5 and x = + 5.
QQ
Solve these equations (the answers are in the footnote29 ).
(a) x 3 − 7x 2 + 10x = 0 (b) x 6 + 7x 3 − 8 = 0 (c) x 4 − x 2 = 0

p 5 14
(d) x − x − 6 = 0 (e) x 3 = 9x (f) 1 + x − x 2 = 0
(a) x(x − 2)(x − 5) = 0. x = 0, 2 or 5 (b) y = x 3 . y = 1 or −8. x = 1 or −2

p
29 (c) x 2 (x + 1)(x − 1) = 0. x = 0, 1 or −1 (d) y = x. y = −2 or 3. x = 4 or 9.
(e) x(x + 3)(x − 3) = 0. x = 0, −3 or +3. (f) Times by x 2 or y = x1 . x = 2 or −7.

7.6 Sketching quadratic functions

Being able to sketch graphs is an invaluable tool. Why? Because the brain is
far better at dealing with pictures than it is with abstract concepts.
I vividly remember getting completely lost as one of my lecturers tried to
explain a theorem that referred to n-dimensions. A much braver student
than I ventured to ask how he managed to get his head round it. The reply
was instantaneous: “I think of it in 3 dimension, write down what that would
mean in n-dimensions and set about proving it algebraically".
Even the greatest30 mathematical brains rely on concrete analogies and
visual images, the most fundamental of which are graphs. They show us how
a function behaves as we vary the value of x: does it go up, or go down, or a
bit of both? How fast does it change? By plotting the value of the function
(which we will call y) against the value of x we get to understand instantly
the key features of the function.
What do we mean by sketching a curve, as opposed to plotting it? The word
sketch is used to indicate that it doesn’t need to be drawn accurately, or even
to scale. It’s a drawing that captures the key features of a curve, for example
where it crosses the coordinate axes. It doesn’t need to be particulary neat,
but if it’s too small or untidy it will be difficult to use. Always draw a sketch in
pencil, so you can correct or improve it if you need to.
At the start of the chapter we said there are three ways to write a quadratic
function, each one telling us something different about it. Here we shall use
this example:
y = x 2 − 4x − 5
y = (x + 1)(x − 5)
y = (x − 2)2 − 9.
We will find that all the information these three different perspectives tell us
about the function are captured succinctly in a simple graph.
30
This young down-to-earth lecturer would go on to develop ground-breaking ideas
connecting geometry and algebra which “had a profound impact on developments in
Quantum Physics". (Wikipedia: Sir Michael Atiyah F.R.S.)

Using this example let’s summarise what each of the three versions of the
expression tells us about its graph:
y = x 2 − 4x − 5 y-intercept y = −5
y = (x + 1)(x − 5) roots x = −1 and x = 5
y = (x − 2)2 − 9 minimum value y = −9, when x = 2
The maximum or minimum point of the curve we call its vertex.
From the three versions of the equation, we can read off the coordinates of
the key points (we have marked them with a red dot) and hence sketch the
curve.
y
6 x =2
BB

9
B
B
B• • - x
−1 5
−5 •
y = −9
Notice that its axis of symmetry, in this case at x = 2, is also exactly midway
between the two roots:
1
x = (−1 + 5) = 2.
2
This will always be the case for a quadratic curve.
In fact before we have begun to find out any of the details, the first version,
y = x 2 − 4x − 5, tells us something else important:
the sign of the x 2 term tells us which way up the curve is:
positive x 2 coefficient ⇒ curve up this way up
negative x 2 coefficient ⇒ curve up this way up

@
@
R
@

Example 29. Sketch the graph of y = x 2 − 6x + 11.
Solution.
The y-intercept is at (0, 11).
Completing the square with a view to finding the roots we have
y = (x − 3)2 + 2.
This tells us that y has a minimum point at (3, 2) so the curve must looks like
this:
y
AA 6
A
11A•
2 •
- x
3
Let’s take stock for a moment, and see how this ties up with the other ideas
we have met.
(1) We knew the curve had to be this way up because the x 2 coefficient is
positive.
(2) The curve clearly never reaches the x-axis, so the equation
x 2 − 6x + 11 = 0
has no roots. We can also see this from the completed square form:
(x − 3)2 + 2 can never be = 0.
(3) This is the function we met on page 7-40, whose discriminant is
(−6)2 − 44 < 0, confirming again that that the equation has no roots.

Example 30 Sketch the curve y = 4 + 2x − x 2 .
Solution
We’ll tackle it in stages: first we’ll just see what the curve looks like, then
gradually fill in the detail.
The y-intercept is at (0, 4) and the x 2 -coefficient is negative so the curve

must look something like this:
•
• or
We will know which it is when we we either find the roots or we find the x-
coordinate of the maximum point (is it positive or negative?).
The quadratic doesn’t factorise, so finding the roots is not quick and easy.
We’ll complete the square.
y = −[x 2 − 2x − 4]
= −[(x − 1)2 − 5]
= 5 − (x − 1)2
⇒ y ≤ 5
confirming that the vertex is indeed a maximum point, and that is has
coordinates (1, 5).
The graph looks like the one on the left.
We know the y-intercept is at y = 4 and the vertex at (1, 5). It remains to find
the roots. Then we’ll have a complete picture of what the curve looks like.

From the completed square form of the function (putting y = 0):
5 − (x − 1)2 = 0
(x − 1)2 = 5
p
x −1 = ± 5
p
x = 1± 5
x = −1.24 or 3.24 (to 3 SF)
The reason we have gone into decimals in the last line is to show us roughly
where the roots are on the x-axis. This could be particularly important if
there are other lines or curves on the same graph.
We end up with this beautiful picture.
y
6
y =5

5 •
4•
x =1

p p
1− 5 1+ 5
A
A
U
A
• •
-
x
p - p -
5 5
We have drawn in the axis of symmetry of the curve: the line x = 1.

The reason we’ve done this is to bring out the fact that the roots are situated
symmetrically about this line:
p p
the smaller root is 5 to the left of the line of symmetry, i.e. at 1− 5;
p p
the larger root is 5 to the right of the line of symmetry, i.e. at 1+ 5.

7.6.1 Repeated roots

There is an important special case that the last example points to: when the
vertex is on the x-axis. When this happens we say the curve touches the
x-axis, and it corresponds to the one root case.
In Example 28 we met this function:
y = (x − 3)2 + 2.
It has a minimum point at (3, 2) and the curve can only look like the graph
below.
y
AA 6
A
11 A•
2 •
- x
3
Without the “+2", the function would be
y = (x − 3)2 ,
the minimum point is now at (3, 0) and x = 3 is a root.
y
6
AA
A
9 A•
2
• - x
3
The effect on the curve is to move it down31 2 units, so that it is touching the
x-axis.
And where before there were no roots, now there is one, x = 3.
31
What we call a translation.

So why do we sometimes use the term repeated root when this happens? Or
two equal roots?
All will become clear.
Imagine the curve starting slightly lower (the top graph), and step by step
moving up until it corresponds to the function y = (x − 3)2 .
As the curve gets closer to y = (x − 3)2 the roots get closer together.
y
6
A
A
A
• • -x
PP
i
P
Two distinct roots
P
y
6
A
A
A
• • -x
P
i
PP
PP
Two distinct roots - but
P
getting closer to each other

y
6
A
A
A
y = (x − 3)2 -
• -x
P
i
PP
PP
P
Two equal roots
P
We can think of the equation (x − 3)2 = 0 as having one root, x = 3. But, as the
graphs show, it is rather more meaningful to think if it has having two roots,
both equal to 3.

To summarise, we have seen that there are three versions of the quadratic
function available to us:
y = x 2 − 6x + 5
y = (x − 1)(x − 5)
y = (x − 3)2 − 4.
To sketch a quadratic curve, we can find
which way up it is, from the sign of the x 2 coefficient

the y-intercept by putting x = 0 into the first version;
the x-intercepts (roots) by putting y = 0 into the second version;
the maximum or minimum point (vertex) from the third version.
QQ
Sketch these quadratic functions. In each case mark the position of the
vertex and where the curve crosses the axes.
(a) y = x 2 − 6x + 5 (b) y = x 2 + 4x (c) y = x 2 + 4
(d) y = x 2 − 4 (e) y = 12 − 4x − x 2 (f) y = x 2 − 6x − 3
(g) y = x 2 − 10x + 25 (h) y = 3x − x 2
The working is shown in the footnote32 and the sketches on the next page.
(a) y = (x − 1)(x − 5) = (x − 3)2 − 4 (b) y = x(x + 4) = (x + 2)2 − 4

(c) No roots. (d) y = (x + 2)(x − 2)
2
32 (e) y = −(x + 6)(x − 2) = 16 − (x + 2) .
p p
(f) y = (x − 3)2 − 12. Roots x = 3 ± 12 = 3 ± 2 3 (g) y = (x − 5)2
9 3
(h) y = x(3 − x) = −(x− )2
4 2

Notice that the scale need not be the same on the two axes. What is important
is the overall configuration (which way up the curve is and where it sits in
relation to the axes) and the coordinates of the key points.
y y
(a) 6 (b) 6
5
- x
- x −4 0
1 5 •
• (−2, −4)
(3, −4)
(c) y (d) y
6 6
4• - x
−2 2
- x −4 •
(e) y (f) y
6 6
(−2, 16)
• 12
- x
−6 2
- x
@
I
p • @ p
3−2 3 (3, −12) 3+2 3
(g) y (h) y
6 6 ( 32 , 94 )

•
25 - x
0 3
• - x
5

A method of finding maximum and minimum values for all types of function
eluded mathematicians for many hundreds of year, until Calculus was
developed.
The foundations of Calculus were laid by Isaac Newton in England and

Gottfried Leibniz in Germany in the latter half of the 17th century. There is
some debate about who should take the lion’s share of the credit.
It was without a doubt one of the greatest breakthroughs in the history of
Mathematics. Without Calculus there would be no space travel. Calculus is
deeply integrated in every branch of the physical sciences. It is also found in
computer science, statistics, economics, business, and medicine.
In Chapters 9 and 11 these key principles of curve sketching come

wonderfully into their own as we study a whole range of functions, and we
get our first taste of the wonderful world of calculus.
It takes a bit of practice to become fluent in sketching curves, but stick with
it; it very often provides a beautiful—and often powerful—window into an
abstract world.
Not all problems involving quadratic curves involve sketching them. The
following examples are primarily algebraic.

Example 31 Find the equation of the quadratic curve that passes through
P (−4, 0), Q(2, 0) and R(1, −15).
Solution
The roots are x = −4 and x = 2, so the quadratic function must have factors
(x + 4) and (x − 2).
That means we must be able to write the function as:
y = k(x + 4)(x − 2) for some constant k.
We also know that the curve passes through (1, −15), i.e x = 1, y = −15 must
satisfy the equation, so substituting these values into the equation we have
−15 = k(1 + 4)(1 − 2) ⇒ −15 = −5k.
i.e. k = 3 and the equation is
y = 3(x + 4)(x − 2)
or y = 3x 2 + 6x − 24.
Is it always as easy to construct the equation of a quadratic curve from three

points that lie on it? No it isn’t. In the example we just looked at two of the
points happened to be roots.
In the next example only one of the points is a root, which makes it a little
tricker.
(If none of the points are roots it is trickier still. Solving problems like this
points the way into the fascinating field of matrices. But that’s an adventure
for much later.)

Example 32 Find the equation of the quadratic quadratic curve that passes
through (1, 0), (3, 18) and (−1, −2).
Solution
Because x = 1 is a root, one of the factors must be (x − 1).
We don’t know where the other root is so we can’t say anything about the
other linear factor. We will have to assume it is the most general linear factor
possible, i.e. (ax + b).
We can safely say that its equation is
y = (x − 1)(ax + b).
All that remains is to find a and b, which we can do because we know two
other points that the curve passes through. Substituting the coordinates in
we have the simultaneous equations:
18 = (2)(3a + b)
−2 = (−2)(−a + b)
or more simply
3a + b = 9
−a + b = 1
Solving33 these, the coefficients are a = 2 and b = 3.

The equation is
y = (x − 1)(2x + 3).
or y = 2x 2 + x − 3.
QQ
Find the equation of the quadratic curve that touches the x-axis at (−2, 0)
and passes through (1, 27).
33
For example subtracting them gives 4a = 8 ⇒ a = 2, so b must be 3.
If you need to remind yourself about solving linear simultaneous equations
see page 6-39.
34
y = 3(x + 2)2 . The equation must be of the form y = k(x + 2)2 .
Substituting gives 18 = 9k ⇒ k = 3.

Exercises 27
1. Solve these equations.
(a) (x − 2)(x + 5) = 0 (b) (3x − 2)(x + 1) = 0 (c) x(x − 6) = 0

(d ) (w − 7)(3 − w) = 0 (e) (x + 1)(x + 2)(x + 3) = 0
( f ) x 2 (1 − 5x) = 0 (g ) 4y(y − 3) = 0 (h) x2 (x − 4) = 0
2. Solve these quadratic equations by factorising.
(a) x 2 − 6x + 5 = 0 (b) x 2 + 2x − 8 = 0 (c) x 2 − 8x + 15 = 0

(d ) x 2 − 8x = 0 (e) 3x 2 − 6x − 9 = 0 ( f ) 10x = x 2
(g ) x 2 − 16 = 0 (h) x 3 − x = 0 (i ) 4x 2 − 13x + 10 = 0
(If you need to brush up on your factorising see page 3-43.)
3. Repeat Question 2, but solve the equations by completing the square.
(a) x 2 + 2x = 3 (b) x 2 = 9x − 14 (c) x(x − 1) = 6

2
(d ) x = 10 + 3x (e) x(3x − 1) = 4 ( f ) (x − 3)(x + 1) = 21
(g ) 1 2
2
x − 3x + 4 = 0 (h) x + 1 = 6
x
(i ) x4 (1 − 3x) = 1
(a) 8 − 2x − x 2 = 0 (b) 28 + 24x − 4x 2 = 0 (c) 4 − x 2 = 0

(d ) 7x + 2x 2 = 0 (e) 9x 2 − 1 = 0 ( f ) 9x 3 − x = 0
(g ) 9x 3 − x 2 = 0 (h) x 4 − 4x 2 = 0 (i ) x 4 + 4x 2 = 0
6. Solve these quadratic equations by completing the square. Leave your

answer in surd (square root) form. Simplify where possible.
(a) x 2 − 2x − 1 = 0 (b) x 2 + 4x + 1 = 0 (c) x 2 − 6x + 4 = 0

(d ) x 2 + 3x − 1 = 0 (e) x 2 − x − 11 = 0 ( f ) x2 + x − 2 = 0
(g ) x 2 − 3x + 2 = 0 (h) 2x 2 + 4x − 3 = 0 (i ) 3x 2 − x − 2 = 0
7. Solve these equations using the quadratic formula. Leave any square
roots in your answer. Simplify where possible.
(a) 2x 2 + x − 5 = 0 (b) x 2 + 2x − 7 = 0 (c) x 2 − 3x − 9 = 0

(d ) x 2 − 8x + 1 = 0 (e) 5x 2 + x − 4 = 0 ( f ) x 2 − 7x + 1 = 0
(g ) 20x 2 + x − 1 = 0 (h) 7x 2 + 2x + 3 = 0 (i ) x 2 − 6x + 9 = 0
8. Solve these equations. Make a substitution if you need to.
(a) x 3 − x 2 − 6x = 0 (b) x 4 − 9x 2 = 0 (c) x 4 + 9x 2 = 0

(d ) x 4 + 9x 2 − 10 = 0 (e) x 2 (x 2 − 1) = 2 ( f ) x5 = x3
p 1 1
(g ) x 3 − x73 = 8 (h) x − 5 x − 14 = 0 (i ) 2x 2 = 13 + 6x − 2
9. Tamsin is eight years older that Sara. The product of their ages is three
times the sum of their ages. How old are they?

10. The mean of x and x 2 is 1. One possible value for x is x = 1.

What other value could x have?
11. The mean of 1, x and 4x is 2x 2 . Find the possible values for x.
12. The sides of a right angles triangle are x, x + 1 and x + 2.

(a) Which one is the hypoteneuse?
(b) Find the value of x.
13. The sides of a right angles triangle are x, x + 1 and x + 3.

(a) Which one is the hypoteneuse?
(b) Find the value of x.
2
14. Find the values of x for which 2x = 4(2x ).
2
15. For what values of x is 3x = 27(32x ).
16. The function f (x) is defined by the formula

1 − x1
f (x) = , x 6= −1
1 + x1
(a) Find the values of f (1) and f ( 12 ).
(b) Solve the equations
(i) f (x) = 2, and (ii) f (x) = 6x.
(c) Show that the equation f (x) = 2x has no solution.
(d) Why can we x not take the value −1?
3−x
17. The function f (x) = .
x+1
(a) Find the values of f (2) and f ( 13 ).
(b) Solve the equations
(i) f (x) = 5, (ii) f (x) = x, (iii) f (x) = x1 .
18. A house has a rectangular garden measuring 12 m by 8 m. The owner

decides to set it out as a rectangular lawn completely surrounded by a
flower bed of width w m. The lawn area is to be 45 m2 . Find w.
19. Show that the equation

1 1
1+ + 2 = k, x 6= 0
x x
can be written as
(1 − k)x 2 + x + 1 = 0.
Hence prove that the equation only has a solution if k ≥ 43 .
20. Using the identity

(x + 2)3 ≡ x 3 + 6x 2 + 12x + 8
show that
(x + 2)3 − 9(x + 2) ≡ x 3 + 6x 2 + 3x − 10.
By making the substitution u = x + 2, find the three roots of the
equation
x 3 + 6x 2 + 3x − 10 = 0.

21. Solve these equations. Substitute each solutions back into its equation
to check that it is correct. (There is a reason you have been asked to do
this.)
p
(a) x = x + 12.
¢2
(b) x + x1 = 4
¡
22. (i) Show that (p + q)2 − 4pq = (p − q)2 .

(ii) The function f (x) is defined by
f (x) = (x − 2p)(x − 2q)
where p and q are constants.
(a) Write down the roots of f (x) = 0 and deduce the equation of the
line of symmetry of the curve y = f (x).
(b) Show that for all values of x
f (x) ≥ −(p − q)2 .

Solutions 27
1. (a) x − 2 = 0 or x + 5 = 0 ⇒ x = 2 or −5
(b) 3x − 2 = 0 or x + 1 = 0 ⇒ x = 23 or −1
(c) x = 0 or x − 6 = 0 ⇒ x = 0 or −6
(d) w − 7 = 0 or 3 − w = 0 ⇒ w = 7 or 3
(e) x + 1 = 0, x + 2 = 0 or x + 3 = 0 ⇒ x = −1, −2 or −3
2 1
(f) x = 0 or 1 − 5x = 0 ⇒ x = 0 or 5
(g) 4y = 0 or y − 3 = 0 ⇒ y = 7 or 3
2
(h) x
= 0 (not possible) or x − 4 = 0 ⇒ x =4
2. (a) 1, 5 (x − 1)(x − 5) = 0 (b) 2, −4 (x + 2)(x − 4) = 0

(c) 3, 5 (x − 3)(x − 5) = 0 (d) 0, 8 x(x − 8) = 0
2
(e) −1, 3 [÷3] x − 2x − 3 = (x + 1)(x − 3) = 0
(f) 0, 10 x(10 − x) = 0 (g) −4, +4 (x + 4)(x − 4) = 0
2
(h) −1, 0, 1 x(x − 1) = x(x + 1)(x − 1) = 0
5
(i) 4,2 (4x − 5)(x − 2) = 0
3. (a) 1, 5 (x − 3)2 − 4 = 0 (b) 2, −4 (x + 1)2 − 9 = 0

(c) 3, 5 (x − 4)2 − 1 = 0 (d) 0, 8 (x − 4)2 − 16 = 0
(e) −1, 3 [÷3] (x − 1)2 − 4 = 0 (f) 0, 10 (x − 5)2 − 25 = 0
(g) −4, +4 Already in completed square form.
(h) −1, 0, 1 x 2 − 1 already in completed square form.
(i) 54 , 2 (x − 13 2 9
8 ) − 64 = 0
4. (a) 3, −1 x 2 + 2x − 3 = (x − 3)(x + 1) = 0
(b) 2, 7 x 2 − 9x + 14 = (x − 2)(x − 7) = 0
(c) 3, −2 x 2 − x − 6 = (x − 3)(x + 2) = 0
(d) 5, −2 x 2 − 3x − 10 = (x − 5)(x + 2) = 0
(e) −1, 34 3x 2 − x − 4 = (x + 1)(3x − 4) = 0
(f) −4, 6 x 2 − 2x − 24 = (x + 4)(x − 6) = 0
(g) 2, 4 x 2 − 6x − 8 = (x − 2)(x − 4) = 0
(h) −3, 2 x 2 − x − 6 = (x + 3)(x − 2) = 0
(i) 43 , −1 3x 2 − x + 4 = (3x − 4)(x + 1) = 0
5. (a) −4, 2 [×− 1] x 2 − 2x − 8 = (x + 4)(x − 2) = 0

(b) −1, 7 [÷− 4] x 2 − 6x − 7 = (x + 1)(x − 7) = 0
(c) 2, −2 (2 + x)(2 − x) = 0 or [×− 1] (x + 2)(x − 2) = 0
(d) 0, − 72 x(7 + 2x) = 0
(e) 13 , − 13 (3x − 1)(3x + 1) = 0
(f) 0, 13 , − 31 x(9x 2 − 1) = 0 ⇒ x = 0 plus solutions of part (e)
(g) 0, 91 x 2 (9x − 1) = 0
(h) 0, −2, 2 x 2 (x 2 − 4) = x 2 (x + 2)(x − 2) = 0
(i) 0 x 2 (x 2 + 4) = 0

p p
6. (a) 1 ± 2 (x − 1)2 − 2 = 0 (b) −2 ± 3 (x + 2)2 − 3 = 0
p p
13
(c) 3 ± 5 (x − 3)2 − 5 = 0 (d) − 32 ± 2
(x + 3 2
2
) − 13
4
=0
p p
(e) 21 ± 3 2 5 (x − 12 )2 − 45
4
=0 (f ) − 12 ± 2
3
(x + 1 2
2
) − 49 =0
p
(g) − 23 ± 213 (x − 32 )2 − 54 = 0
p
(h) −1 ± 25 [÷2] (x + 1)2 − 54 = 0
(i) 1, − 23 [÷3] (x − 16 )2 − 25
36
= 0 ⇒ x = 16 ± 65
p p p
7. (a) 14 (−1 ± 41) (b) 21 (−2 ± 32) = −1 ± 2 2
p p p p
(c) 21 (3 ± 45) = 32 (1 ± 5) (d) 12 (8 ± 60) = 4 ± 15
1
p p p
(e) 10 (−1 ± 81) = 1 or − 45 (f) 12 (7 ± 45) = 12 (7 ± 3 5)
1
p
(g) 40 (−1 ± 81) = − 14 or 15 (h) No solution (discriminant < 0)
p
(i) 12 (6 ± 0) = 3 (one solution)
8. (a) −2, 0, 3 x(x 2 − x − 6) = x(x + 2)(x − 3) = 0

(b) −3, 0, 3 x 2 (x 2 − 9) = x 2 (x + 3)(x − 3) = 0
(c) 0 x 2 (x 2 + 9) = 0
(d) −1, 1 [y = x 2 ] y 2 + 9y − 10 = (y − 1)(y + 10) = 0
⇒ x 2 − 1 = 0 or x 2 + 10 = 0
p p
(e) − 2, 2 [y = x 2 ] y 2 − y − 2 = (y + 1)(y − 2) = 0
⇒ x 2 + 1 = 0 or x 2 − 2 = 0
(f) −1, 0, 1 x 3 (x 2 − 1) = x 3 (x + 1)(x − 1) = 0
(g) −1, 2 [y = x 3 ] y 2 − 7y − 8 = (y + 1)(y − 8) = 0
⇒ x 3 = −1 or x 3 = 8
p
(h) 4, 49 [y = x] y 2 − 5y − 14 = (y + 2)(y − 7) = 0
⇒ y = −2 or y = 7
1 p 1
(i) 14 , 36 y = x 2 = x and x − 2 = 11 = p1x
x2
⇒ 2y 2 − 13y − 6 = 0 ⇒ y = 12 or y = 6
9. Sara is 4, Tamsin is 12.

Let Sara’s age be x. Then x(x + 8) = 3(2x + 8) ⇒ x 2 + 2x − 24 = 0
(x − 4)(x + 6) = 0 ⇒ x = 4.
10. x = −2.
x+x 2
= 1 ⇒ x 2 + x − 2 = (x + 2)(x − 1) = 0.
2
11. x = 1 or x = − 16 .
1+x+4x
= 2x 2 ⇒ 6x 2 − 5x − 1 = (6x + 1)(x − 1) = 0.
3
12. (a) Hypoteneuse is x + 2.
(b) x 2 + (x + 1)2 = (x + 2)2 ⇒ x 2 − 2x − 3 = (x + 1)(x − 3) = 0
Length must be positive, so x = 3.
This is the familiar 3, 4, 5 triangle, and we have proved that this is
the only right angled triangle whose sides are consecutive integers.
13. (a) Hypoteneuse is x + 3.

(b) x 2 + (x + 1)2 = (x + 3)2 ⇒ x 2 − 4x − 8 = (x − 2)2 − 12 = 0
p p p
x = 2 ± 12 = 2 ± 2 3. Length must be positive, so x = 2 + 2 3.

14. x = −1 or 2.
2
2x = 22 .2x = 22+x ⇒ x 2 = x + 2 ⇒ x 2 − x − 2 = (x + 1)(x − 2) = 0
15. x = −1 or 3.
2
3x = 27(32x ) = 33 .32x = 33+2x ⇒ x 2 = 3 + 2x
x 2 − 2x − 3 = (x + 1)(x − 3) = 0
x−1
16. Multiplying top and bottom by x: f (x) = .
x+1
1 1
1 2 −1 − 2
(a) f (1) = 0, f (2) = 1 = 3 = − 31 ,
2 +1 2
1− x1
or more simply x = 12 ⇒ 1
= 2, so f (x) = = 1−2 = − 13
1+ x1
x 1+2
(b) (i) x−1

x+1
= 2 ⇒ 2(x + 1) = x − 1 ⇒ x = −3
x−1
(ii) x+1
= 6x ⇒ 6x(x + 1) = x − 1
⇒ 6x 2 + 5x + 1 = (3x + 1)(2x + 1) = 0 ⇒ x = − 13 or − 12 .
(c) 2x(x + 1) = x − 1 ⇒ 2x 2 + x + 1 = 0: discriminant is < 0.
(d) x = −1 makes the denominator zero. (We can’t divide by 0.)
1 1
3− 31 83
17. (a) f (2) = 3
, f (3) = 1 = 4 = 2.
3
+1 3
3−x
(b) (i) x+1 = 5 ⇒ 5(x + 1) = 3 − x ⇒ x = − 13
(ii) 3−x
x+1 = x ⇒ x(x + 1) = 3 − x
2
⇒ x + 2x − 3 = (x + 3)(x − 1) = 0 ⇒ x = −3 or 1.
3−x 1
(iii) x+1
= x
⇒ 3 − x = x(x + 1)
⇒ x − 2x + 1 = (x − 1)2 = 0 ⇒ x = 1.
2
18. 12 m
(12 − 2w)(8 − 2w) = 45

8m w
- 4w 2 − 40w + 51 = 0
(2w − 3)(2w − 17) = 0
w
6
⇒ w = 1.5 m. (w can’t be 8.5 m.)
?
19. Multiplying through by x 2 gives x 2 + x + 1 = kx 2 .

⇒ (1 − k)x 2 + x + 1 = 0, which has (real) roots provided
the discriminant = 12 − 4(1 − k) ≥ 0 or k ≥ 34 .
20. (x + 2)3 − 9(x + 2) = x 3 + 6x 2 + 12x + 8 − 9x − 18
= x 3 + 6x 2 + 3x − 10. Substituting y = x + 2 gives y 3 − 9y = 0
or y(y 2 − 9) = y(y + 3)(y − y) = 0 ⇒ y = 0, −3 or +3.
i.e. x + 2 = 0, x + 2 = −3 or x + 2 = 3 ⇒ x = −2, −5 or 1.

21. (a) Squaring both sides of the equation: x 2 = x + 12.

(x − 4)(x + 3) = 0 ⇒ x = 4 or −3.
p
Substituting x = 4: r.h.s = 16 = 4 = l.h.s.
p
Substituting x = −3: r.h.s = 9 = 3 6= l.h.s.
p
x = −3 is not a solution. It is is the solution of x = − x + 12,
which also leads to the equation x 2 = x + 12 when squared.
(b) Method 1. Square root both sides: x + x1 = ±2
⇒ x 2 ± 2x + 1 = (x ± 1)2 = 0. Solutions are x = ±1.
³ 2 ´2 2 2
Method 2. x x+1 = (x x+1) 2 = 4 ⇒ (x 2 + 1)2 = 4x 2
Substituting y = x 2 gives (y + 1)2 = 4y which simplifies to
(y − 1)2 = 0 ⇒ y = 1, i.e. x 2 = 1 ⇒ x = ±1.
Method 3. The same as Method 2 up to (x 2 + 1)2 = 4x 2
Then take square root of each side: x 2 + 1 = ±2x
giving x 2 ± 2x + 1 = 0 as in Method 1.
Checking the solutions. If x = 1, l.h.s = (1 + 1)2 = r.h.s.
If x = −1, l.h.s. = (−2)2 = r.h.s. Both solutions are correct.
22. (i) (p + q)2 − 4pq = p 2 − 2pq + p 2 = (p − q)2 .
(ii) (a) The roots are x = 2p and x = 2q.
The axis of symmetry bisects the line segment between the roots,
so it is the line x = p + q.
(b) The simplest way is to say that the coefficient of x 2 is positive,
so the function has a minimum value where x = p + q ⇒ for all x
f (x) ≥ f (p + q) = (p + q − 2p)(p + q − 2q) = −(p − q)(p − q).
You could also do it by completing the square:
f (x) = x 2 − 2(p + q)x + 4pq = [x − (p + q)]2 − (p + q)2 + 4pq
= [x − (p + q)]2 − (p − q)2 .

7.7 Solving quadratic inequalities

First a reminder about the different inequality signs:
< “less than"
≤ “less than or equal to"; “no greater than"
> “greater than"
Remember that the smaller quantity is beside the “sharp" end of the sign
(where “smaller" means further to the left on the number line). For example
2 < 7 and − 50 < 1.
7.7.1 Linear inequalities

Before we turn to quadratic inequalities, a reminder about linear ones. You
may have come across the graphical representation of linear inequalities
involving two variables. For example one like
2y + x > 4.
What happens is that the line divides the plane into three distinct regions:
the points on the line, the point on one side of the line and the points on the
other side of the line.
In the case of the line 2y + x = 4, these regions are shown on the graph
below:
at every point above the line 2y + x > 4

at every point on the line 2y + x = 4
at every point below the line 2y + x < 4.
HH 6
HH
H
HH
H 2y
HH +x
2 H >4
HH
H
2y HH
+x H
<4 HH
H
HH
x
-
0 4 HH
HH
H
H
2y
H
+x
=4
Exam questions specifically on this are rare in C1 or C2, but to understand it

is very useful.

In Chapter 5 we learned how to solve linear inequalities (in one variable),

such as
2x + 3 > 0.
It is a straight forward process—very similar to solving a linear equation. If
you want a reminder about the method see page 6-32 in Chapter 5.
We are about to see that to solve quadratic inequalities in one variable, such
as
4x > x 2 + 3
requires an entirely different approach: one that ivolves sketching quadratic
curves.

7.7.2 Quadratic inequalities

With quadratic inequalities the same principle applies as it does with linear
inequalities, except that now we have a curve not a straight line.
At every point
on the curve the = sign will apply;

on one side of the curve the > sign will apply;
on the other side of the curve the < sign will apply.
For example, take the relationship shown on the graph below:
y = x 2 − 4x + 3 = (x − 1)(x − 3).
At every point on the curve y = x 2 − 4x + 3;

at every point above the curve y > x 2 − 4x + 3
at every point below the curve y < x 2 − 4x + 3;
on the x + 3
y
curve
4
C 6
y = x2−
C
C
C
C
C
3 above the curve
y > x 2 − 4x + 3
- x
1 3
below the curve
y < x 2 − 4x + 3

We said that solving quadratic inequalities is very unlike solving linear ones.
Let’s see why.
Suppose we are trying to find the values of x for which
4x > x 2 + 3.
If we imagine x taking lots of different values, from −∞ to +∞, 4x goes from
large negative values to large positive ones. The expression x 2 + 3 starts off
large and positive, gets smaller and then gets bigger again. It’s not obvious for
which values of x the l.h.s. is greater than the r.h.s. We can’t solve the problem
by rearranging, in the way that we did with linear inequalities, because of the
square term, so how do we do it?

The key is to be found in curve sketching. We start by writing the inequality

with a zero on one side 35 , just like we do with quadratic equations:
x 2 − 4x + 3 < 0
and sketch the curve y = x 2 − 4x + 3. This is the curve on the previous page.
Now we can think of the inequality

x 2 − 4x + 3 < 0
as the inequality
y <0
and the problems becomes:
for what values of x is y < 0?
In other words: where is the curve below the x-axis. We’ve marked it in red.
y
C 6
y = x 2 − 4x + 3
C
C
C
C @
@
3C
C
@
R
@
y >0
y >0 H
HH
j
- x
1 3
H
Y
H
HH
H
y <0
So the sketch tells us that x 2 − 4x + 3 < 0 when x lies between 1 and 3.
Or mathematically: x > 1 and x < 3.

We can combine these two statements into one and say: the solution of the
inequality is
1 < x < 3.
Or in set language {x : 1 < x < 3}.
35
Make sure you are comfortable with this. We have moved everything over on to
the right: 0 > x 2 − 4x + 3, then swapped sides and reversed the inequality.

We can check this by putting some some values into the inequality. For
example when x = 2 we have 4 × 2 > 22 + 3: the inequality holds; when x = 0
we have 0 > 3: it doesn’t hold.
To solve the same inequality but with the sign the other way round, i.e.
4x < x 2 + 3
the reasoning is exactly the same as before, except now we are looking for the
values of x for which
x 2 − 4x + 3 > 0,
that is to say where the curve y = x 2 − 4x + 3 is above the x-axis.
y = x 2 − 4x + 3
C 6
C
C
C A
3C
A
A
AU
- x
1 3
These are the values to the left of 1 and to the right of 3, that is:
x < 1 or x > 3.
When the inequality was the other way round, we were able to combine the
two inequalities x > 1 and x < 3 into the single statement 1 < x < 3. That isn’t
possible when the solution is two separate regions of the number line.
Notice that in neither case are the values x = 1 and x = 3 included. When
these “end points" are not included we say it is a strict inequality.
Had the inequality been

x 2 − 4x + 3 ≥ 0
(with ≥ rather than >) then the solution would have been
x ≤ 1 or x ≥ 3.
Notice that whereas the solution of a linear equation is a single number, the
solution of an inequality is (in most cases) a range of values 36 .
36
Depending on the type of equation, it may have a number of solutions; inequalities
typically have an infinite number.

Another approach to the problem would be to compare the graphs of y = 4x

and the graph of y = x 2 + 3, and see where one curve is above the other.
y
6
2
y = x +3

- x
1 3
y = 4x
@
@
R
@
The graphs cross when 4x = x 2 + 3, that is when x 2 − 4x + 3 = 0.

We know this is when x = 1 and x = 3, which helps us draw the sketch.
The solution of the inequality
4x > x 2 + 3
now corresponds to the values of x where the line y = 4x is above the curve
y = x 2 + 3, i.e. when
1 < x < 3.
Human beings find pictures easier to understand than abstract concepts,

which is why we use a graphical method to solving inequalities wherever we
can.
Although you won’t meet them at this level, there are occasions when a purely
algebraic approach has advantages, so next we are going to solve the same
problem without a graph.

To solve the inequality 4x > x 2 + 3 without using a graph.
We start by rearranging the inequality and writing it in factorised form:
x 2 − 4x + 3 < 0
(x − 1)(x − 3) < 0
Since their product is negative, one bracket must be positive and the other
negative.
We consider the two cases in turn:
Either (a) x − 1 > 0 and x −3 < 0 ⇒ x > 1 and x < 3,

Or (b) x − 1 < 0 and x −3 > 0 ⇒ x < 1 and x >3
Case (b) is impossible, so the solution is (a) which we write as
1 < x < 3.
If the inequality were the other way round, i.e. 4x < x 2 + 3,
(x − 1)(x − 3) > 0
so now the two factors must have the same sign.
As before we have two cases:
Either (a) x − 1 > 0 and x −3 > 0 ⇒ x > 1 and x >3

Or (b) x − 1 < 0 and x −3 < 0 ⇒ x < 1 and x <3
Case (a) boils down to saying x > 3 and case (b) to x < 1. Both are perfectly
valid and therefore the the solution is
x < 1 or x > 3.

Example 33 Solve the inequality x(4 − x) ≥ 1, leaving your answer in surd

form.
Solution.
The inequality can be written: x 2 − 4x + 1 ≤ 0.
Notice by choosing which way round to have the inequality sign we can
always write the inequality with a positive x 2 coefficient.
We want to sketch the curve y = x 2 − 4x + 1, knowing that inequality holds

where y ≤ 0.
For the sketch we need the roots, and in this case the quadratic doesn’t
factorise. We’ll find the roots by completing the square:
(x − 2)2 − 3 = 0
p
x −2 = ± 3
p
x = 2± 3
(We could have used the formula, of course.)
The curve looks like this:
y
6
y = x 2 − 4x + 1
@
@
@
R
@
- x
BMB
B
B
p
pB
2− 3 2+ 3
The solution is the values of x for which the curve is either below or on the
x-axis, i.e. p p
2 − 3 ≤ x ≤ 2 + 3.
Notice that the sketch doesn’t need to be at all accurate. Even if wephadn’t
looked closely at the values of the roots and thought that 2 − 3 was
negative, the solution would still be correct. Of course it’s better to make the
sketch reasonably accurate.

There is a very important special case. Look at this inequality:
x 2 > 9.
You may be tempted to deduce that x must be greater than 3. Yes?

Sketching the graph of

y = x2 − 9
we can see that this is not necessarily the case.
-
−3 0 3
We want to know which values of x make x 2 − 9 > 0. This is where y > 0.
Certainly x > 3 is part of the solution, but so also is x < −3.
Expressing in mathematical language, we can say it’s certainly true that
x >3 ⇒ x2 > 9
but
x2 > 9 6⇒ x >3

Modulus notation
In situations like the one we just described there is a very useful new
notation. It’s called the modulus function, written |x|.
|x| is the magnitude of a number: how big is the number, regardless of its sign.
So for example
| − 7| = 7, |3| = 3, | − x 2 | = x 2 .
Put another way
on the number line |x| is the distance of the number x from zero.
Notice that
|1 − 5| = | − 4| = 4 = |5 − 1|
which we can generalise to
|x − y| = |y − x|
It gives us a way of describing the distance between x and y, regardless of

which is the larger.
Using this notation we can replace
−3 < x < 3 with |x| < 3
i.e. x is less than three units away from the origin.
−6 −5 −4 −3 −2 −1 0 1 2 3 4 5 6 7
And we can replace
x < −3 or x > 3 with |x| > 3
i.e. x is more than three units from the origin.
−6 −5 −4 −3 −2 −1 0 1 2 3 4 5 6 7
Needless to say we can use the modulus notation with ≤ and ≥, for example
|x| ≤ 1 means −1 ≤ x ≤ 1
|x| ≥ 1 means x ≤ −1 or x ≥ 1.
The next example is in the form of a practical problem.

Example 34 A solid cylindrical block of height 10 cm needs to have a

volume no greater than 90π cm3 and a total surface area no less than
48π cm2 . Find the range of possible values for its radius.
Solution
r -
6
10 cm
Suppose the radius is r . The restriction on the volume implies that
10πr 2 ≤ 90π i.e. r 2 ≤ 9.
We know this means |r | ≤ 3. Or rather, since r is a physical dimension,

0 < r ≤ 3.
The restriction on the surface area (it has two ends) implies that
2πr 2 + 20πr ≥ 48π

r 2 + 10r − 24 ≥ 0
(r + 12)(r − 2) ≥ 0
A sketch of the curve

y = r 2 + 10r − 24
(and the fact that the radius can’t be a negative number) tells us that r ≥ 2.
Since both restrictions apply, we must have
2 ≤ r ≤ 3.
The radius must lie between 2 cm and 3 cm.

QQ
Solve these inequalities by sketching a suitable quadratic curve (the answers

are in the footnote37 ).
(a) x 2 − 6x + 8 < 0 (b) x 2 − 6x + 8 ≥ 0 (c) x(x + 1) < 2
(d ) 7x > 2x 2 + 5 (e) x 2 ≤ 4 ( f ) 9x 2 > 25
In Section 7.4 we used inequalities involving the discriminant to solve

problems concerning the roots of equations. All those inequalities were linear
ones. Now we are in a position to solve the problem if it gives rise to a
quadratic inequality. That’s what we are going to look at next.
5
37 (a) 2 < x < 4 (b) x ≤ 2 or x ≥ 4 (c) −2 < x < 1 (d) 1 < x < 2
5
(e) |x| ≤ 2 (f) |x| > 3

7.7.3 Harder quadratic inequalities

Before we embark on this new application of quadratic inequalities, here is
what on the face of it looks like a trivial observation.
We established that
x 2 − 4x + 3 < 0 ⇒ 1 < x < 3.
It follows, for example, that
k 2 − 4k + 3 < 0 ⇒ 1 < k < 3,
or
p 2 − 4p + 3 < 0 ⇒ 1 < p < 3,
and so on. We could substitute any variable we like. Now look at this next
problem.
Trivial and obvious though this seems, you’ll see how this simple insight can
unlock some tricky looking problems.

Example 35 For what values of k does the equation

x 2 + 2kx + 4k − 3 = 0
have no roots?
Solution
It has no roots if the discriminant is negative, i.e. if
(2k)2 − 4(4k − 3) < 0

4A k 2 − 4A (3k + 4) < 0
k 2 − 4k + 3 < 0
We know that the solution to this quadratic inequality is
1<k <3
so when 1 < k < 3 the quadratic equation x 2 + 2kx + 4k − 3 = 0 has no roots.
What’s going on here?
We started not with a single quadratic equation but a family of quadratic

equations: one for each value of k (so there are infinitely many of them). The
problem is, in effect, to identify which of these quadratic equations have no
roots.
In the process of exploring this question (using the discriminant) we found
ourselves with a new quadratic expression, and it was this that gave rise to
the quadratic inequality—not the original quadratic.
We can check some values of k.
The equation corresponding to k = 2 is x 2 +4x +5 = 0. It should have no roots.

It’s discriminant is 16 − 20 < 0, confirming that this is indeed the case38 .
If k = 0, the equation is x 2 − 3 = 0 which does have roots, as expected.
38
Or we could say the equation is (x + 2)2 + 1 = 0 which doesn’t cross the x-axis.


x 2 + 4px + 4(p + 6) = 0
only has roots if p ≤ −2 or p ≥ 3.
Solution
For a quadratic equation to have (real) roots the discriminant must be ≥ 0, i.e.
(4p)2 − 16(p + 6) ≥ 0
16p 2 − 16(p + 6) ≥ 0
p2 − p − 6 ≥ 0
or (p + 2)(p − 3) ≥ 0
As with any quadratic inequality we can solve by sketching the function. We

will call the function Y . We are looking for the values of p for which:
Y = (p + 2)(p − 3) ≥ 0.
The sketch (of Y against p) looks like this
Y
6
y = p2 − p − 6
- p
−2 3
The values of p for which Y ≥ 0 are shown on the p-axis with red lines, i.e.
where p ≤ −2 or p ≥ 3.
A word about what happens when p = −2 or p = 3, i.e. at the boundaries.

This is where the discriminant is equal to zero, so corresponds to the equal
root scenario. Let’s check this out.
If p = −2, the equation we have is
x 2 − 8x + 16 = 0 i.e. (x − 4)2 = 0.
Check for yourself that p = 3 also leads to an equal roots quadratic equation.

Example 37 For which values of k does the equation
x 2 − 6x + k = 0
have (real) roots?
Solution
We can look at this purely algebraically (without appealing to a graph at all).
The discriminant needs to be ≥ 0, i.e.
36 − 4k ≥ 0 ⇒ k ≤ 9.
This is a very simply solution. But it doesn’t give us any idea why! A sketch
shows us what’s going on.
We are going to draw y = x 2 − 6x + k. Don’t be put off by the fact that we don’t
know k. After all it’s a sketch—we are not plotting points.
It helps to write the quadratic as
y = (x − 3)2 + (k − 9).
The minimum value the function takes is at its vertex, i.e. at (3, k − 9).
The number of roots the equation x 2 − 6x + k = 0 has (if any), will depend on
whether the vertex is above, on or below the axis. That in turn depends on
whether k is greater than, equal to or less than 9.
y k >9 y k =9 y k <9
C 6 6 6
C
C C
C
C C
C
C
-x -x -x
no roots one root two roots

(equal roots)

Example 38 For which values of k, where k is a constant, does the equation
x 2 + kx + 9 = 0
have (a) no roots, (b) one root, and (c) two roots?
Solution
a = 1, b = k and c = 9 so the discriminant is k 2 − 36.
Case (a). No roots if k 2 − 36 < 0 ⇐⇒ |k| < 6.

Case (b). One root if k 2 − 36 = 0 ⇐⇒ |k| = ±6.
Case (c). Two roots if k 2 − 36 > 0 ⇐⇒ |k| > 6.
The problem is solved. To understand fully what’s going on here we are going
to draw some graphs—one for each case.
In each case the y-intercept is at (0, 9).

Case (a). No roots. Example k = 4.
y = x 2 + 4x + 9 = (x + 2)2 + 5. Minimum point (−2, 5).
Case (b). One root. Example k = 6.

y = x 2 + 6x + 9 = (k + 3)2 . Minimum point (−3, 0).
Case (c). Two roots. example k = 8.

y = x 2 + 8x + 9 = (k + 4)2 − 7. Minimum point (−4, −7).
y
6

AA J
A k =8 J k =6 @ k =4
@
9
• -x
−3 0
and it makes a simple, graceful design too...

QQ
1. In each case find the values of k for which
(a) x 2 + kx + 2k = 0 has no (real) roots.

(b) kx 2 + 6x + k + 8 = 0 has real roots.
(c) x 2 + 2kx + (5 − 4k) = 0 has two distinct real roots.
(d) x 2 + kx + 1 = 0 has real roots.
2. Show that the equation x 2 + px + p − 1 = 0 has at least one real root

whatever the value of p.
1. (a) 0 < k < 8 (b) −9 ≤ k ≤ 1 (c) k < −5 or k > 1 (d) |k| > 2
39
2. Discriminant is p 2 − 4(p − 1) = p 2 − 4p + 4 = (p − 2)2 ≥ 0 for all values of p.

Exercises 28
1. Solve the following inequalities.
(a) x 2 − 6x + 8 < 0 (b) x 2 + 2x − 3 < 0 (c) x 2 − 3x + 2 > 0

(d ) x 2 − 7x ≥ 0 (e) t 2 − 4t − 12 < 0 ( f ) 2x 2 − 11x + 8 ≤ 0
2. Solve the following inequalities. Where appropriate express your

answer in modulus form.
(a) x 2 < 4x (b) x 2 ≤ 4 (c) k 2 > 9

(d ) a 2 ≥ a (e) 4x 2 > 1 ( f ) 3y 2 < y
(a) 4 − 3x − x 2 > 0 (b) x 2 < x + 2 (c) 1 − x 2 ≤ 0

(d ) 2x 2 − 5x > 0 (e) x − 4x 2 ≥ 0 ( f ) 3x 2 < 1
(a) x(2x − 1) ≤ 10 (b) x 2 > 2(3x − 4) (c) x 2 < 11x−6

3
(d ) x 2 ≥ 12 (1 + x) (e) (2x + 3)2 ≤ (x − 1)2 ( f ) 4x 2 < (3x − 1)2
5. In each case find the values of the constant k for which the equation
has (i) no (real) roots, (ii) one root, (iii) two roots.
(a) x 2 + 6x + k = 0 (b) x 2 − 4x + 2k = 0 (c) k − 5x − 3x 2 = 0

(d ) x = 1 + 2kx 2 (e) kx 2 = (1 + x)2 ( f ) k 2 x 2 + 8x + k1 = 0
6. In each case find the values of the constant λ for which the equation
has at least one real root.
(a) x 2 + λx + 1 = 0 (b) λx 2 + 2x + λ = 0
(c) λx 2 + 2λx + 3 = 0 (d ) x 2 + λx + λ − 1 = 0
(e) (2λ + 3)x 2 + 2λx + 1 = 0 ( f ) 3x 2 + λx − λ = 0
7. Find the values of x for which

(a) x 2 − 6x + 7 > 0 (b) x(1 + x) ≤ 12 (c) x(1 + x) < 21 (x − 1)2 .
leaving your answers in surd form.
8. Find the values of the constant a for which

x
= 3.
x + a2
2
has real roots.
9. For what values of m does the equation
m 1−x
=
x m
have real roots, where m is a constant?

10. (i) Show that the equation

1 1
+ =1
x +a x
has real roots whatever the value of the constant a.
(ii) For what values of a does the equation
1 a
+ =1
x +1 x
have real roots?
11. Using the a substitution y = x 2 , find the values of x for which
x 4 − 14x 2 − 32 < 0.
12. (i) Find the values of the constant k for which the equation
1
x2 + =k
x2
has real solutions.
(ii) Show that the equation
1
x2 −
=k
x2
has real solutions whatever the value of k.
13. A small rectangular box measures x cm by (x + 1) cm by 2 cm. What are

the possible values of x if the surface area of the box is to be greater than
16 cm 2 , but it’s volume less than 12 cm 3 ?
14. A storage tank is in the shape of a solid right circular cylinder 2 m high.
Find the range of values that its radius can take if its volume is to be less
than than 98π m 3 and its surface area is to be at least 70π m 2 .
15. A coin has been tampered with so that when tossed it is more likely to
show a head than a tail. Suppose that probability is p (i.e. p > 21 ).
The coin is tossed twice. Write down, in terms of p, the probability that
the result is (a) two heads, (b) a head followed by a tail, (c) a head and a
tail (in either order).
As part of a game the coin is tossed twice many times. At the end of the
game two heads have come up more often than a head and a tail. Show
that if this is generally the case, then p must be greater than 23 .
16. (i) Show that it is impossible to find three consecutive integers for which
the square of the middle one is less than the product of the other two.
(Let the middle number be n.)
(ii) Is it possible for the square of the largest of the three integers to be
less than the product of the other two? If not, explain why not. If so,
give an example of three such numbers.
17. The area between two concentric circles is more than three times the
area of the inner circle. The radius of the outer circle is 2 units greater
than the radius of the inner circle. What can we say about the radius of
the inner circle?
18. A triangle ABC is such that AB = R, AC = 2R − r , BC = R + r and A is an

acute angle. Show that r < 32 R. (There is a hint in the footnote40 .)
40 If the angle A were a right angle, we could say BC 2 = AB 2 + AC 2 .

Write down an in equality that holds if A is acute, i.e. < 90◦ .

Solutions 28
The quadratic curve from which the solution is derived is given in brackets.
We use the Greek letter ∆ (capital “D") to stand for the discriminant.
1. (a) 2 < x < 4 [y = (x − 2)(x − 4) < 0]
(b) −3 < x < 1 [y = (x + 3)(x − 1) < 0]
(c) x < 1 or x > 2 [y = (x − 1)(x − 2) > 0]
(d) x ≤ 0 or x ≥ 7 [y = x(x − 7) ≥ 0]
(e) −2 < x < 6 [y = (x + 2)(x − 6) < 0]
3
(f) 2 ≤x ≤4 [y = (2x − 3)(x − 4) ≤ 0]
2. (a) 0 < x < 4 [y = x(x − 4) < 0]

(b) |x| ≤ 2 i.e. −2 ≤ x ≤ 2 [y = (x − 2)(x + 2) < 0]
(c) |k| > 3 i.e. k < −3 or k > 3 [y = (k − 3)(k + 3) > 0]
(d) a ≤ 0 or a ≥ 1 [y = a(a − 1) ≥ 0]
1
(e) |x| > 2
i.e. x < − 12 or x > 1
2
[y = (2x − 1)(2x + 1)]
1
(f) 0 < y < 3 [Y = y(3y − 1)]
3. (a) −4 < x < 1 [y = (x + 4)(x − 1) < 0]

(b) −1 < x < 2 [y = (x + 1)(x − 2) < 0]
(c) |x| > 1 [y = x 2 − 1 = (x + 1)(x − 1) ≥ 0]
(d) x < 0 or x > 52 [y = x(2x − 5) > 0]
1
(e) 0 ≤ x ≤ 4
[y = 4x 2 − x = x(4x − 1) ≤ 0]
p p
(f) |x| < p1 [y = 3x 2 − 1 = ( 3x − 1)( 3x + 1) < 0]
3
4. (a) 2 ≤ x ≤ 52 [y = 2x 2 − x − 10 = (2x − 5)(x + 2) ≤ 0]

(b) x < 0 or x > 4 [y = x 2 − 6x + 8 = (x − 2)(x − 4) > 0]
2
(c) 3
<x <3 [y = 3x 2 − 11x + 6 = (3x − 2)(x − 3) < 0]
(d) x ≤ − 12 or x ≥ −1 [y = 2x 2 − x − 1 = (2x + 1)(x − 1) ≥ 0]
(e) −4 ≤ x ≤ − 23 [y = 3x 2 + 14x + 8 = (3x + 2)(x + 4) ≤ 0]
(f) x < 51 or x > 1 [y = 5x 2 − 6x + 1 = (x − 1)(5x − 1) > 0]
5. (a) (i) k > 9 (ii) k = 9 (iii) k < 9 [∆ = 36 − 4k]

(b) (i) k > 2 (ii) k = 2 (iii) k < 2 [∆ = 16 − 8k]
(c) (i) k < − 25
12
(ii) k = − 25
12
(iii) k > − 25
12
[∆ = 25 + 12k]
(d) (i) k > 81 (ii) k = 81 (iii) k < 18 [∆ = 1 − 8k]
(e) (i) k < 0 (ii) k = 0 (iii) k > 0 [∆ = 4k]
(f) (i) k < 16 (ii) k = 16 (iii) k > 16 [∆ = 64 − 4k]
6. (a) |λ| ≥ 2 [∆ = λ2 − 4 = (λ + 2)(λ − 2) ≥ 0)]

(b) |λ| ≤ 1 [∆ = 4 − 4λ2 ≥ 0 ⇒ 4(λ2 − 1) ≤ 0)]
(c) λ ≤ 0 or λ ≥ 3 [∆ = 4λ2 − 12λ = 4λ(λ − 3) ≥ 0)]
(d) All values of λ [∆ = λ2 − 4(λ − 1) = (λ − 2)2 ≥ 0)]
(e) λ ≤ −1 or λ ≥ 3 [∆ = 4λ2 − 4(2λ + 3) = 4(λ + 1)(λ − 3) ≥ 0)]
(f) λ ≤ −12 or λ ≥ 0 [∆ = λ2 + 12λ = λ(λ + 12) ≥ 0)]

p p p
7. (a) x < 3 − 2 or x > 3 + 2 [Roots: 3 ± 2]
p p p
(b) − 12 − 23 ≤ x ≤ − 12 + 23 [2x 2 + 2x − 1 ≤ 0. Roots: − 12 ± 23 ]
p p p
(c) − 2 − 5 < x < −2 + 5 [x 2 + 4x − 1 ≤ 0. Roots: −2 ± 5]
8. |x| ≤ 61 . [3x 2 − x + 3a 2 = 0. ∆ = 1 − 36a 2 ≥ 0]
9. |m| ≤ 12 . [x 2 − x + m 2 = 0. ∆ = 1 − 4m 2 ≥ 0]
10. (i) Equation is x 2 + (a − 2)x − a = 0.

∆ = (a − 2)2 + 4a = a 2 − 4a + 4 + 4a = a 2 + 4 ≥ 0 for all a.
(ii) a ≤ −4 or a ≥ 0. Equation is x 2 − ax − a = 0. ∆ = a 2 + 4a.
11. |x| < 4. [(y + 2)(y − 16) < 0 ⇒ −2 < x 2 < 16. But x 2 ≥ 0.]
12. (i) k ≥ 2. [X 2 − k X + 1 = 0 where X = x 2 . ∆ = k 2 − 4 ≥ 0

⇒ |k| ≥ 2. But x 2 + x12 > 0, so k ≤ −2 doesn’t work.]
(ii) Equation: X 2 − k X − 1 = 0. ∆ = k 2 + 4 ≥ 0 for all k.
13. 1 < x < 2. [Area = 2x 2 + 10x + 4 > 16 ⇒ 2(x + 6)(x − 1) > 0.

⇒ x > 1, since x can’t be < −6.
Volume = 2x(x + 1) < 12 ⇒ (x + 3)(x − 2) < 0
⇒ 0 ≤ x < 2, since x ≥ 0.]
14. 5 < r < 7. [S = 2πr 2 + 4πr ≥ 70π ⇒ (r + 7)(r − 5) > 0.

⇒ r > 5, since r can’t be < −7.
V = 2πr 2 < 98π ⇒ r 2 < 49 ⇒ 0 ≤ r < 7, since r ≥ 0.]
15. (a) p 2 (b) p(1 − p) (c) 2p(1 − p).

p 2 > 2p(1 − p) ⇒ 3p 2 − 2p = p(3p − 2) > 0 ⇒ p > 32 (p is positive).
16. (i) Would need n 2 > (n − 1)(n + 1) i.e. n 2 > n 2 − 1 ⇒ 0 > −1.
(ii) Now we need (n + 1)2 < n(n − 1). Common sense may suggest
this is impossible, but it simplifies to 3n + 1 < 0 ⇒ n < − 31 .
It works for any integer n ≤ −1. For example −4, −3, −2.
17. Radius < 2. [π(r + 2)2 − πr 2 > 3πr 2 ⇒ 3r 2 − 4r − 4 < 0.

⇒ (3r + 2)(r − 2) < 0 ⇒ r < 2.]
18. Since A is acute BC 2 < AB 2 + BC 2 ⇒ (R + r )2 < R 2 + (2R − r )2

⇒ 4R 2 − 6Rr = 2R(2R − 3r ) > 0.

7.8 Inequalities with an algebraic denominator

We must be careful not to multiply both sides a inequality by a negative
number (see Chapter 6).
An example. Suppose −x < 4. Multiplying by −1 would give x < −4. But this
is wrong, which we can see by adding x and then −4 to both sides. This gives
the (correct) answer: −4 < x or x > −4.
Now let’s look at this example

7
<3
x
It may be tempting to multiply both sides by x, to give 7 < 3x or x > 73 .
But this assumes that x is a positive number. We don’t know this.
There is an easy way round this: we multiply x 2 , which we know must be
positive.
7x < 3x 2
0 < x(3x − 7)
7
x < 0 or x >
3
You remember we solve quadratic inequalities by sketching, in the case the
curve y = x(3x − 7).
y
6
C
C
C
-x
0 7
3
The solution x > 37 that we came up with initially is valid. But any negative
number is also a possible. Only by using a correct method do we get all the
solutions.

4 3
Example 39 Find the values of x for which > 1+ 2
2x x
Solution
Multiplying both sides by x 2
2x > x2 + 3
x 2 − 2x + 3 < 0
(x − 3)(x + 1) < 0
y = (x − 3)(x + 1)
6
C
C
C
C
- x
−1 3
We are looking the values of x for which y < 0, where y = (x − 3)(x + 1).
The sketch tells us that −1 < x < 3.
4
Example 40 Solve the inequality ≥1
x−1
Solution
Multiplying both sides by (x − 1)2
4(x − 1) ≥ (x − 1)2
4x − 4 ≥ x 2 − 2x + 1
0 ≥ x 2 − 6x + 5
0 ≥ (x − 5)(x − 1)
y = (x − 5)(x − 1)
6
C
C
C
C
- x
1 5
The solution: x ≤ 1 or x ≥ 5.

Example 41
(a) Find the roots of the equation f (x) = 0, where f (x) = x 4 − 5x 2 + 6.
(b)Sketch the curve y = f (x).
1
(c) Find the values of x for which > 1.
x 2 −2
solution
(a) Substituting X = x 2 we have
X 2 − 5X + 6 = 0
(X − 2)(X − 3) = 0
x 2 = 2 or x 2 = 3
p p
x = ± 2 or x = ± 3
(b) y
6
p p p p -
x
− 3 − 2 2 3
Notice that as a sketch the positions of the intercepts don’t need to be

accurate. It’s the overall configuration that matters.
(c)The denominator x 2 − 2 could be positive or negative so we need to
multiply both sides by (x 2 − 2)2
x 2 − 2 > x 4 − 4x 2 + 4
0 > x 4 − 5x 2 + 6
p p p p
From the sketch the values of x are − 3 < x − 2 or 2 < x 3.
Check this yourself by picking a few values of x within these ranges and
outside them.

QQ
1. Solve these inequalities
9
(a) x
> 5.
x+4
(b) x ≤ 2.
1 3
(c) 5x > x .
1
(d) x
+ x22 ≤ 1.
4 3
(e) x2
< x
− 12 .
1
(f) 3−x
> 2.
3x−1
(g) x−2
< 4.
41 9 5
(a) 0 < x < 4 (b) x ≤ 0 or x ≥ 4 (c) x < 0 (d) x ≤ −1 or x ≥ 2 (e) 2 < x < 4 (f) 2 <x <3
(g) x < 2 or x > 9.

7.9 Non-linear simultaneous equations

This is an extension of the work we did on linear simultaneous equations in
Chapter 5 (Section 6.5 on page 6-39). Geometrically we can think of it as
finding where two curves intersect, or a curve and a line, rather than two
lines. To make this step up we are going to have to call on many of the ideas
from this chapter as well ones from Chapters 5 and 6. It is very much in the
nature of Maths that it’s in combining techniques, each of which is quite
unremarkable, that we can often achieve beautiful results, or surprising
ones—and occasionally startling ones.
Example 42 Solve these equations simultaneously:
y = x 2 − 2x − 1
y = x + 3.
Solution
We are looking for a pair of numbers (a value for x and a value for y) that
satisfy both these equations. In fact we will find there are two possible pairs.
Equating 42 the two expressions for y, we have
x 2 − 2x − 1 = x + 3
x 2 − 3x − 4 = 0
(x + 1)(x − 4) = 0
⇒ x = −1 or x = 4
Each of these values for x gives a value for y, which can find using either
equation. Obviously it easiest to use y = x + 3, i.e. y = 2 or y = 7.
Solution: x = −1 and y = 2, or x = 4 and y = 7.
We can see the geometrical interpretation on a sketch 43 . The two solutions

correspond to the two points where the line crosses the curve.
y
6

•(4, 7)
BB
B
B
B

3
x+
=
y

y = x 2 − 2x − 1

(−1, 2) •

- x
42
The values of x and y are the same in both equations, so where we see y in one
equation we can replace it with the y from the other equation.
43
Satisfy yourself that the sketch looks like this.

xy = 2
x + y − 3 = 0.
Solution
Since neither equation is in the form y = . . ., we can’t simply equate them like
we did in the last example. But the substitution method works. In fact this is
the the method we frequently adopt when one of the equations is not linear.
In case you have forgotten, substitution works like this: we use one of the
equations to find an expression for y, and then substitute that into the other
equation.
Rearranging the second equation we have y = 3 − x. Now wherever we see y

in the first equation, we replace it with 3 − x (in this case y only appears
once):
x(3 − x) = 2
0 = x 2 − 3x + 2
0 = (x − 1)(x − 2)
⇒ x = 1 or x = 2
Substituting back into any of the equations:
when x = 1, y = 2, and when x = 2, y = 1.
The solutions can also be expressed as coordinates, i.e. (1, 2) and (2, 1).
A quick check tells us that both these solutions satisfy both equations.
In this case we could equally well have started by finding an expression for x
(i.e. x = 3 − y ) and substituted that.
We don’t have this luxury in the next example.

x(y − 1) = 18
2x = 3y.
Solution
The linear equation doesn’t allow a nice simple expression, either for x or y.
We could launch into fractions, e.g. substitute y = 2x
3
into the first equation.
(A solution using this approach is shown in the footnote44 .)
Alternatively we could use the fist equation to find an expression for x or y; in
this case it’s obviously easier to find x:
18
x =
y −1
36
= 3y
y −1
12 = y(y − 1)
0 = y 2 − y − 12
0 = (y − 4)(y + 3)
whose roots are y = 4 and y = −3.

18
Substituting these values of y into x = , the solutions are
y−1
9
x = 6, y = 4 and x = − , y = −3.
2
In exam questions the solutions to equations are very often whole numbers or
simple fractions.
But not always...
44
µ ¶
2x
x −1 = 18
3
x 2x − 3
µ ¶
= 18
1 3
x(2x − 3) = 54
2
2x − 3x − 54 = 0
(x − 6)(2x + 9) = 0
9
x = 6 or x = −
2

3x 2 − y 2 = 3
2x − y = 1.
Solution
We are going to substitute y = 2x − 1, from the second equation.
1+y
This is obviously a lot easier than substituting x = 2 .
3x 2 − (2x − 1)2 = 3
3x 2 − (4x 2 − 4x + 1) = 0
3x 2 − 4x 2 + 4x − 1 = 0
x 2 − 4x + 1 = 0
(x − 2)2 − 5 = 0
p p
⇒ x = 2 + 5 or x = 2 − 5
Substituting back into the second equation, using the form y = 2x − 1

p p
y = 2(2 + 5) − 1 = 3 + 2 5
or p p
y = 2(2 − 5) − 1 = 3 − 2 5.
p p p p
Solution: x = 2 + 5, y = 3 + 2 5 or x = 2 − 5, y = 3 − 2 5.
Notice that when finding the values of y at the end, we used y = 2x − 1,

rather than its original form 2x − y = 1. It is generally easiest to do this.
It would definitely not be a good idea p to use2the 2first equation to find the
values of y, i.e. to substitute x = 2 ± 5 into 3x − y = 3!
We haven’t attempted to do sketches for the last two examples, because

neither of the curves are quadratic. In Chapter 9 we will be extending the
repertoire of functions we know how to sketch.
It is sometimes possible to simplify a pair of non-linear simultaneous equation

by convert them into a linear pair.

4y 2 − x 2 = 16
2y − x = 2.
Solution
Using the principle of the difference of two squares:
4y 2 − x 2 = (2y − x)(2y + x) = 16
so since 2y − x = 2, we have the simultaneous equations
2y + x = 8
2y − x = 2.
This is easily solved, e.g. by adding and then subtracting, to give
5
x = , y = 3.
2
Simultaneous equations may both be non-linear, i.e. represented by two

intersecting curves.

Example 47 Find the coordinates of the points A and B where the curves
1
y = x2 and y = (x − 3)2 intersect.
4
If the two curves meet the x-axis at C and D respectively, show that the line
joining A and B bisects the line C D.
Solution
First let’s find out where the points A and B are. Equating expressions for y,
the curves intersect where
x2
x 2 − 6x + 9 =
4
4x 2 − 24x + 36 = x 2
3x 2 − 24x + 36 = 0
3(x − 2)(x − 6) = 0
⇒ x = 2 or x = 6
⇒ y = 1 or y = 9,
by substitution into either of the original equations.
The points where the curves cross are A(2, 1) and B (6, 9).
Both curves touch the x-axis,
1
y = x 2 at C (0, 0) and y = (x − 3)2 at D(3, 0).
4
We are now ready to address the question about AB bisecting C D. Let’s sketch
the curves.
y
6
##
#
#
y = x 2 − 4x + 1 #
#
•#B (6, 9)
#
y = 14 x 2 #
#
#
#
#
#
#
#
#

#
#
A(2, 1) •#
• # • - x
C (0, 0) #
# D(3, 0)
#
#
9−1
The gradient of the joining A and B is 6−2 = 2 , so its equation (using the
formula) is
y − 1 = 2(x − 2) or y = 2x − 3
which crosses the x-axis at x = 23 , midway between C and D.

Example 48 Find the values of k for which these two curves meet:
y = 9 − x2 and y = x 2 − 4x + k + 4.
Solution
Writing the equations as
y = (3 + x)(3 − x) and y = (x − 2)2 + k
we can easily sketch the curves. We’ll choose to draw them not intersecting:
y
6
9
•
(2, k)
- x
−3 0 3
The vertex of the curve y = (x − 2)2 + k is at (2, k). Its line of symmetry is x = 2.
Because this curve sits to the right of y = 9 − x 2 , the value of k can be less
than 9 with the curves still not meeting—as shown in the diagram.
If we progressively lower the right hand curve (by reducing the value of k)
the two will eventually touch. Can you see that they won’t touch at (2, k)?
Lower it further and they will cross.
Algebraically we are looking for the values of k for which
9 − x 2 = (x − 2)2 + k
has real roots. This equation simplifies to
2x 2 − 4x + k − 5 = 0
which has real roots provided

16 − 8(k − 5) ≥ 0
2−k +5 ≥ 0
k ≤ 7.
The two curves meet when k ≤ 7.

QQ
1. In Example 45 state the value of k for which the two curves
y = 9 − x2 and y = x 2 − 4x + k + 4
touch. Find the coordinates of the point where they do so.
2. Solve these simultaneous equations. Express your answers in coordinate

form.
(a) xy = 6 (b) y = 2x (c) y = x2 − 4
y = x +1 y = x(x − 5) x+y = 8
(d) x2 = y2 + 3 (e) y = 2x 2 (f ) 4x + 3y = 5
y = x −1 y = x(6 − x) 2x y = 1
45
1. k = 7. (1, 8, ).
2. (a) (2, 3), (−3, −2) (b) (0, 0), (7, 14) (c) (3, 5), (−4, 12) (d) (2, 1)
(e) (0, 0), (2, 8) (f ) ( 21 , 1), ( 34 , 23 )

7.10 Non-linear simultaneous inequalities in two

variables
In Chapter 6 we discussed simultaneous inequalities in linear variables, the
solution which graphically is a region bounded by straight lines. For example
x + 2y < 6, y ≥x and x ≥ 1
2 R
0 x
−1 1 2 3 4 5 6
The only issue here is to decide which side of the line x + 2y = 6 corresponds
to the inequality x + 2y < 6. To determine this we can use any test point. The
origin is the easiest one to pick in this case. Substituting we have 0 + 2(0) < 6
which tells us that the origin lies the right side of the line, so every point on
that side of the line lies in the region.
Where the inequality includes equality (i.e. ≤ or ≥) points on the line are
included.
In this case we have marked the region with R.
Notice that we use a solid line to show that points on the line are to be
included and a dotted line when they are not.
When one or more of the inequalities is non-linear exactly the same

principles apply. We can use a test point or if the inequality is of the form
y < . . . it is points below the curve, if the inequality is of the form y > . . . it is
points above the curve. An example:
y ≥ x(x − 2) and y < x + 4.
y
8
0 x
−1 1 2 3 4

QQ
In the following examples show by shading or labelling the region for which
both the inequalities apply.
(a) y > x 2 − 1 and x + y ≤ 3 (b) y < 3 + 2x − x 2 and y > x + 1
x2
(c) y ≤ 1 − x 2 and y > x(x − 1) (d) y < 4 − x 2 and y ≥ 1−
4
The answers are in the next page.

(a)
y
6
0 x
−3 −2 −1 1 2 3 4
(b)
y
4
3
2
1
0 x
−2 −1 1 2 3
(c)
y
0 x
−1 1
(d)
4
3
y
0 x
−1 1

Exercises 29
1. Solve each of these inequalities.

4 x+3 4 1 1 x−7
(a) x
> 1 (b) x
≥ 2 (c) x−1
< 1 (d) x+1
+ (x+1)2
> 2 (e) 2−x
≥ 4.
2. Solve these simultaneous equations. Express your answers in

coordinate form.
(a) y = x2 − x − 1 (b) xy = 2 (c) y 2 + 2y + x = 7
y = 1 − 2x 6y = x −1 y + 2x = 0
(d) xy = x + y +1 (e) y = 2(x + 1) (f ) x − 3y = 4

1 1
y = 4x − 5 y = x+x 2+y
= x
3. The diagram shows the line y = x + 2 intersecting the curve

y = −x 2 + 6x − 2.
(i) Find the coordinates of the points A, B and C .
(ii) Hence or otherwise show that B is the mid-point of AC .
y
6

C

B

A - x

4. Find the dimensions of a rectangle whose perimeter is 11 units and

whose area is 6 square units.
5. Find the pair of integers that differ by 6 and whose squares differ by 24.
6. Find all the possible pairs of integers that differ by 6 and whose squares
add up to 68.
7. The diagonal of a rectangle is 13 cm long and its perimeter 34 cm. Find

the lengths of the sides.
8. The hypoteneuse of a triangle is 6 mm long and its perimeter 14 mm.

Find the lengths of the other sides.
9. (a) Show that a 2 − b 2 = (a + b)(a − b).

(b) Hence show that if a 2 − b 2 = 12
and a +b = 2
then a − b = 6.
Hence or otherwise find the values of a and b.

10. In each case, show on a separate coordinate grid the region that satisfies
the inequalities
(a) y < (x + 2)(3 − x) and y ≥ 2x
(b) y < 9 − x 2 and y ≥ x 2 − 4
(c) y ≥ (x − 2)2 and x + y ≤ 4
(d) y ≤ (x + 6)(2 − x) and y + 3x > 10
11. Find the coordinates of the points where the curves y = x(4 − x) and
y = x 2 (x − 1) intersect.
Sketch the curves and show the region where
x ≥ 0, y ≤ x(4 − x) and y ≥ x 2 (x − 1).
12. Use the factor theorem to find all the roots of the equation f (x) = 0,
where
f (x) = x 3 − 2x 2 − x + 2
Sketch the curve y = x 2 (2−x) and the line x+y = 2. Find the coordinates
of the points where they intersect.
On a coordinate grid shade the region where y > x 2 (2 − x) and x + y ≤ 2.
State with your reason whether the point (1, 1) lies within the region.
13. Show that the perimeter pof a triangle whose hypoteneuse is 5 units
long, cannot exceed 5(1 + 2) units. [There is a hint in the footnote46 .]
14. (a) Show that (3y − 2x)2 = 9y 2 + 4x 2 − 12x y.

(b) Two variables x and y are such that
3y − 2x = 8
and 9y 2 + 4y 2 = 40.
Show that x y = −2 and hence solve the simultaneous equations.
15. Show that the line x + y = p meets the curve 3x 2 + 2y 2 = 30 provided

|p| ≤ 5.
State the value of p for which the line touches the curve and find the
coordinates of the point where it does so.
16. A storage tank required by a

chemical factory is to be 8 m tall 6
and be in the shape of a circular
cylinder with a hemisphere on
top, as shown in the diagram. 8m
For cooling purposes, the tank 6
H
needs to have a total exposed (i.e.
curved) surface area of 80π sq m. -? ?
R
If the radius of the of the cylinder is R and its height H , write down two
simultaneous equations connecting R and H . Hence find the diameter
of the tank.
[The formula for the surface area of a sphere is 4πr 2 .]
46
Let the sides containing the right angle be x and y and start by assuming x + y = k.

3
17. The diagram shows the curve of the function y = x 2 and where it
meets curve y = 4 − x 2 .
(a) Find the coordinates of the points A, B , C and D.
p
(b) Show that the gradient of the line C D is −(1 + 3).
y
6
D
B
C
- x
A
[If you need it there is a hint in the footnote47 .]
18. Solve the simultaneous equations
8x = 4 y−1 and x y = y + 4.
47
Substitute X = x 2 .

Solutions 29
1. (a) 0 < x < 4 (b) 0 ≤ x ≤ 3 (c) x < 1 or x > 5 (d) x < − 32 or x > 0 (e) 2 ≤ x ≤ 3
2. (a) (0, 1), (−1, 3) (b) (4, 12 ), (−3, − 32 ) (c) (−1, 2), ( 47 , − 72 )
p p p p
(d) (2, 3), ( 21 , −3) (e) (−1 + 2, 1 + 2 2), (−1 − 2, 1 − 2 2)
(f) (1, −1), (−3, − 37 )
3. (i) A(−2, 0), B (1, 3), C (4, 6)
(ii) B is 3 units to the right of A, and C is 3 units to the right of B .
Or we could compare vertical distances. Or use Pythagoras.
3 11
4. 4 units by 2 units. x + y = 2 , x y = 6.
5. −1 and 5. x − y = 6, x 2 − y 2 = 24.
6. 2 and 8, or −8 and −2. x − y = 6, x 2 + y 2 = 68.
7. 5 cm and 12 cm. x + y = 17, x 2 + y 2 = 132 .
p p
8. (4 − 2) mm and (4 + 2) mm. x + y = 8, x 2 + y 2 = 62 .
9. (a) Multiply out the brackets. (This is the difference of two squares relationship.)
(b) (a + b)(a − b) = 2(a − b) = 12.
Now we have linear simultaneous equation. a = 4, b = −2
10.
y
10. (a) y (b)
6 8
3
4
0 x
−2 3
−3 0 x
−3 −2 −1 1 2 3
−6
−4
(c)
y
6
0 x
2 4

(d)
y
20
(−2, 16)
10
(1, 7)
0 x
−6 −4 −2 2 4
In this example we need to know where the line crosses the curve to be able
to draw the sketch.
This is where
12 − 4x − x 2 = 10 − 3x
0 = x2 + x − 2
0 = (x + 2)(x − 1)
i.e. x = −2 or x = 1
11. (−2, −12), (0, 0) and (2, 4) [x 3 − x 2 = 4x − x 2 ⇒ x(x 2 − 4) = 0]
0 x
−2 −1 1 2 3 4
−4
−8
−12

12. x = −1, x = 1 and x = 2 [ f (1) = 13 − 2(1)2 − 1 + 2 = 0 ⇒ (x − 1)(x 2 − x − 2)]

Curves intersect where 2x 2 − x 3 = 2 − x, i.e. f (x) = 0.
Line and curve intersect at (−1, 3), (1, 1) and (2, 0).
0 x
−1 1 2
The point (1, 1) does not lie within the region, because it’s coordinates do not
satisfy the inequality y > x 2 (2 − x).
13. Let the sides be of length x and y, and let x + y = k. By Pythagoras x 2 + y 2 = 25.
Substituting: 2x 2 − 2kx + k 2 − 25 = 0. Seen as an equation in x, this must have
real solutions, so the discriminant must be ≥ 0,
p
i.e. 4k 2 − 8(k 2 − 25) ≥ 0 ⇒ k 2 ≤ 50 ⇒ 0 < k ≤ 5 2. (k can’t be negative.)
p
The hypoteneuse is 5 so the maximum possible perimeter = 5 + 5 2.
14. (−1, 2) and (−3, 32 ).
2
Substitute y = − x into 3y − 2x = 8.
15. Substituting y = k − x into 3x 2 + 2y 2 = 30 gives 5x 2 − 4kx + 2k 2 − 30 = 0 .
The discriminant = 16k 2 − 20(2k 2 − 30) ≥ 0 ⇒ k 2 ≤ 25.
k = 5. Touch at (2, 3). [Equal roots. k = 5 ⇒ 5x 2 − 20x + 20 = 0.]
16. R + H = 8, 2πR 2 + 2πR H = 80π, i.e. R 2 + R H = 40 ⇒ R = 5, H = 3,
so the diameter is 10 m.
p
17. (a) A(2, 0), B (0, 4), C ( 3, 1), D(1, 3)
3
Substituting y = x 2 gives x 4 − 4x 2 + 3 = 0.
With X = x 2 : X 2 − 4X + 3 = (X − 1)(X − 3) = 0 ⇒ x 2 = 1 or x 2 = 3.
p
1−3 2( 3+1)
(b) Gradient of C D = p =− 2
(rationalising the surd).
3−1
5 3
18. (2, 4) and (− 3 , − 2 )
8x = (23 )x = 23x , and 4 y−1 = (22 ) y−1 = 22y−2 , so 3x = 2y − 2
y+4 3(y+4)
Substitute x = y into 3x = 2y − 2, giving y = 2y − 2,
which simplifies to 2y 2 − 5y − 12 = (2y + 3)(y − 4) = 0

7.11 Circle Theorems

Before we begin a reminder of some geometrical theorems that you will need.
(These are repeated from Chapter 6.)
Perpendicular bisector of a chord

@
@ Each of the diagrams opposite shows the chord
t
@
@• C of a circle and its perpendicular bisector (the
@ line through its midpoint, at right angles to it).
@ The theorem is that
@
@
the perpendicular bisector of a
chord passes through the centre
of the circle.
In the bottom circle we can see that it’s

• C obviously true, by symmetry. If we rotate this
t circle we get the diagram at the top—where
the result is less self-evident.
Circumcircles and circumcentres

The diagram opposite shows
the circle passing through all
the vertices of a triangle (there
will only be one circle that does
this). Such a circle we call the
circumcircle of the triangle, and HH
its centre the circumcentre. The HH
CH•
circumcentre of a triangle (the H H

centre of the circle that passes HH
H
H
through its vertices) may be H

outside the triangle, like in the one
opposite.
Visualise the third perpendicular
bisectors, also passing through C .

The angle in a semicircle

An important special case is when the
circumcentre of the triangle lies neither
inside nor outside but actually on one of
its sides. A

A
If this is the case then that side is a A
diameter of the circle, and we can say C A
t
• A
two things:

(a) the centre of the circle will be at its

midpoint;

(b) its opposite angle is a right-angle.
(You may recall the geometrical
theorem that says: the angle in a
semicircle is a right-angle.)
Tangent properties
B
t
@
@
@
@• C
@

D

A
The diagram shows a tangent from the point A

touching the circle at B .
C is the centre of the circle.
The radius through B is perpendicular to the tangent at B .
We can therefore use Pythagoras to find the length of the tangent AB if we
know the coordinates of A and C , and the radius of the circle.
The tangents from A to the circle are the same length: AB = AD.

7.12 Equation of a circle

7.12.1 From geometry to equation
As with any curve (or straight line) the equation of a circle is the relationship
that holds between the x-coordinate and the y-coordinate of any point on
the curve.
The circle below has equation
x 2 + y 2 = 25
and we can see that the coordinates of all points marked fit this equation.
y
6
(3, 4)
•
p
(− 24, 1) •
- x
O
• (4, −3)
•
(0, −5)
This is hardly surprising: all the points are 5 units from the origin O. The
circle has a radius of 5 units.
It must be therefore that
the equation of a circle, centre the origin and radius r is
x2 + y2 = r 2

Example 49
P and Q are the points where the line x = 2 intersects a circle of radius 7
whose centre is the origin. R is the point (7, 0).
(a) Write down the equation of the circle.
(b) Find the coordinates of P and Q.
p
(c) Show that the equation of the line joining P and R is 3x + 5y = 21.
(d) Find where the line RP crosses the y-axis.
Solution
y
6
(a) The equation of the circle
P is
x 2 + y 2 = 49.
(b) This meets the line x = 2
where y 2 = 45, i.e. at
p p
- x y = ± 45 = ±3 5.
O R
The points of intersection are
p p
P (2, 3 5) and Q(2, −3 5).
Q
(c) R is the point (7, 0), so using the formula

y − y1 y2 − y1
=
x − x1 x2 − x1
the equation of P R is p p
y 3 5 3 5
= =
x −7 2−7 −5
p
−5y = 3 5(x − 7)
−5y
p = 3(x − 7)
5
p
− 5y = 3x − 21
p
21 = 3x + 5y
(d) RP crosses the y-axis where x = 0, i.e. where
21
y=p .
5
This is correct, however it is usual to rationalise surds (i.e. write them so there
is no square root in the denominator). The best way to express the answer
would therfore be Ã p !
21 5
0, .
5

7.12.2 The equation of a circle with any centre and radius

What we’ll do is construct the equation of a particular circle whose centre is
not at the origin, and then generalise it to any circle.
Suppose the centre of our circle is C (8, 5) and its radius is 3.
Most points in the plane are either inside or outside the circle. A point P will
only lie on the circle if P is exactly 3 units from C .
y
6
P (p, q)
•

q − 5

3 9

•
C (8, 5) @ I
@
@
@
@
@ p −8
- x
O
P is p units to the right of the Y -axis and C is 8 units to the right of it.
So P is p − 8 units to the right of C .
Similarly q − 5 is units above C .
Since PC = 3, Pythagoras’ Theorem tells us that:
(p − 8)2 + (q − 5)2 = 32 .
Wherever P is on the circle, p and q must be connected by this relationship.

But this is how we define the equation of a curve. And we generally describe
it using (x, y), so
the circle with centre (8, 5) and radius 3 is (x − 8)2 + (y − 5)2 = 32 .
Rather than leave the equation in this form we will generally expand the
brackets and simplify:
(x − 8)2 + (y − 5)2 = 32
x 2 − 16x + 64 + y 2 − 10y + 25 = 9
x 2 + y 2 − 16x − 10y + 80 = 0

It’s a small step to generalise this to any circle:
P (x, y)
•

y − b

r

9

•
@
I
C (a, b) @
@
@
@
@
@
x −a
- x
O
A
2 2 2
the circle centre (a, b) and radius r is (x − a) + (y − b) = r
!
A
A
A
If we expand this out we get:
x 2 − 2ax + a 2 + y 2 − 2b y + b 2 = r 2
x 2 + y 2 − 2ax − 2b y + a 2 + b 2 − r 2 = 0
Every circle must have an equation like this, and the values of a, b and r will
determine its position and radius.

Example 50 Find the equations of the circles

(a) with centre (4, −3) and radius 2.
3 2
(b) with centre (−1, 2 ) and radius 3 .
Solution.
(a) The equation is:
(x − 4)2 + (y + 3)2 = 22
x 2 − 8x + 16 + y 2 + 6y + 9 = 4
x 2 + y 2 − 8x + 6y + 21 = 0
(b) This is no harder but for the need to introduce fractions:

µ ¶2 µ ¶2
2 3 2
(x + 1) + y − =
2 3
9 4
x 2 + 2x + 1 + y 2 − 3y + =
4 9
13 4
x 2 + y 2 + 2x − 3y + − = 0
4 9
101
x 2 + y 2 + 2x − 3y + = 0
36
36y 2 + 36y 2 + 72y − 108y + 101 = 0
If the equation is that of a circle,
• the coefficients of x 2 and y 2 will always be the equal;
• there will be no x y term.

QQ
Say if the equation represents a circle? (Be careful. Some are designed to try
and trip you up.)
(a) x 2 + y 2 = 5(x + 2y)

(b) x 2 + y 2 + 2x y + x − 2y + 5 = 0
(c) x 2 + y 2 + 25 = 0
(d) 2x 2 + 2y 2 = 1
(e) (x + y)2 = 2(x y + 4)
y x
(f) + = 4.
x y
(g) y = 1 − x 2
(h) y 2 = 1 − x 2
The solutions are in the footnote48 .
(a) Yes. (b) No. (c) No. (r 2 = −25). (d) Yes. (r 2 = 21 )

48
(e) Yes. Expanding the brackets, 2x y cancels.
(f ) No. (getting rid of fractions, there is a 4x y on rhs) (g) No. (h) Yes.
Example 51
Write down the equation of the circle centre (−3, 2) with radius 5.
(i) Find where the circle crosses the y-axis.
(ii) The circle cuts the x-axis at P and Q. Write down the coordinates of the
mid-point of PQ.
Solution
The equation of the circle is
(x + 3)2 + (y − 2)2 = 25
x 2 + 6x + 9 + y 2 − 4y + 4 = 25
x 2 + y 2 + 6x − 4y − 12 = 0.
(a) This circle crosses the y-axis where

y 2 − 4y − 12 = 0
(y − 6)(y + 2) = 0
⇒ y = 6 or y = −2.
Although the question didn’t require a sketch, drawing one can help to weed
out errors. Does my answer look about right?
y
6
• (0, 6)
•
(−3, 2)
- x
P Q
• (0, −2)
(b) The perpendicular bisector of a chord, in this case PQ passes through the
centre of the circle. The mid-point of PQ must therefore be immediately
below the centre of the circle.
The mid-point of PQ is (−3, 0).

From equation to centre and radius
The natural next step is to work backwards: from the equation of a circle to to
its centre and radius.
Suppose we had been given the equation as
x 2 + y 2 + 6x − 4y − 12 = 0.
(This is the equation in Example 46.)

It’s a matter of completing the square, in both variables, x and y.
x 2 + 6x we can write as (x + 3)2 − 9

and
y 2 − 4y we can write as (y − 2)2 − 4,
so the equation becomes
(x + 3)2 − 9 + (y − 2)2 − 4 − 12 = 0
(x + 3)2 + (y − 2)2 = 52 .
The centre of the circle is (−3, 2) and its radius 5.
When we met completing the square before (see page 7-2) we saw that the
working is slightly more tricky if the x-coefficient is an odd number, or if the
x 2 -coefficient is not 1. So next we’ll look at two example that remind us what
we have to do in this case.

Example 52
Find the centre and radius of the circles
(a) x 2 + y 2 − 3x + 7y + 2 = 0
(b) 3x 2 + 3y 2 − 12x + 9y − 2 = 0
Solution
(a) x 2 − 3x + y 2 + 7y − 2 = 0
3 2 9 7 2 49 8
µ ¶ µ ¶
x− − + y+ − + = 0
2 4 2 4 4
3 2 7 2 50
µ ¶ µ ¶
x− + y+ − = 0
2 2 4
¶2 µ Ã p !2
7 2
µ ¶
3 50
x− + y+ =
2 2 2
p p
2×25
3 7
The centre is ( 2 , − 2 ), and the radius 2
= 522.
(b) 3x 2 + 3y 2 − 6x + 9y + 7 = 0
7
x 2 − 2x + y 2 + 3y + = 0
3
µ ¶2
3 9 7
(x − 1)2 − 1 + y + − + = 0
2 4 3
µ ¶2
3 13 7
(x − 1)2 + y + − + = 0
2 4 3
3 2
µ ¶
2 11
(x − 2) + y + =
2 12
q
3 11
The centre is (1, − 2 ), and the radius 12
.

Example 53
(a) Find the centre and radius of theses circle:
x 2 + y 2 − 2x + 8y + 7 = 0.
(b) Without solving any equations, state whether the circle intersect the x-
axis.
(c) For which values of k does the equation
x 2 + y 2 − 2x + 8y + k = 0
intersect the x-axis?
Solution
(a) The equation can be written
x 2 − 2x + y 2 + 8y + 7 = 0
(x − 1)2 − 1 + (y + 4)2 − 16 + 7 = 0
(x − 1)2 + (y + 4)2 = 10.
p
The centre of the circle is at (1, −4) and its radius is 10.
(b) The centre of the circle is 4 units below the x-axis.
p
Its radius = 10 < 4, so it does not cross the x-axis.
(c) For the circle to cross the x-axis we must get real roots when we put y = 0.
The equation is
x 2 − 2x + k = 0
and the discriminant 4 − 4k, so the circle crosses the x-axis when
4 − 4k ≥ 0 i.e. k ≤ 1.
This is consistent with what we concluded in part (b), namely that there are
no roots when k = 7.

Example 54
In each case find where the two circles meet.
(i) x 2 + y 2 + 2x − y − 10 = 0 and x 2 + y 2 + 3x − 3y − 14 = 0
(ii) x 2 + y 2 + 4x − 5y + 3 = 0 and 2x 2 + 2y 2 + 12x − 12y + 7 = 0
Solution
Part (i) is easily solved—because the numbers have been rigged to make sure
it is. Part (ii) is more difficult than you would be expected to tackle at this
level, so you can skip it if you wish. Needless to say, though, it is more
interesting and brings out a couple of neat little tricks.
(i) It isn’t possible to find an expression, either for x or for y that we can use
to substitute, and so eliminate one of the variables. We need to be a little
cunning: we do it in two stages. The first stage is to find an expression for
x 2 + y 2 , and substitute that.
From the first equation,
x 2 + y 2 = −2x + y + 10.
Substituting this into the second equation, we have
−2x + y + 10 + 3x − 3y − 14 = 0
which simplifies to
x = 2y + 4.
Now we do have an expression for x, and we can substitute it into either
circle equation to eliminate the variable x.
(2y + 4)2 + y 2 + 2(2y + 4) − y − 10 = 0
4y 2 + 16y + 16 + y 2 + 4y + 8 − y − 10 = 0
5y 2 + 19y + 14 = 0
14
(y + 1)(5y + 14) = 0 ⇒ y = −1 or y = −
5
Substituting into x = 2y + 4 tells us the points are (2, −1) and (− 85 , − 14

5 ).

(ii) The two circles are
x 2 + y 2 + 4x − 5y + 3 = 0 and 2x 2 + 2y 2 + 12x − 12y + 7 = 0.
Apart from the fact that the coefficients of x 2 and y 2 in the second equation
are 2 rather than 1, this appears to very similar to part (i).
So let’s see why it isn’t.
The coefficient problem is easily overcome: we can substitute
x 2 + y 2 = −4x + 5y − 3
from the first equation into the second, giving
2(−4x + 5y − 3) + 12x − 12y + 7 = 0.

This simplifies to
2y = 4x + 1.
Unfortunately in this case, whether we choose to substitute x or y, it will
inevitably introduce fractions:
2y − 1 4x + 1
x= or y=
4 2
If you want to pursue this, feel free. It will be excellent practice at algebraic
fractions.
There is an easier way.
We don’t have an expression for y but we do have one for 2y, so in the second
equation we can say
−12y = −6(4x + 1).
So far so good.
But what about the y 2 term? We are tantalisingly close but, 2y 2 6= (2y)2 = 4y 2 .
This is the clue.
We double every term in the second equation so we can think of it as
4x 2 + (2y)2 + 24x − 12(2y) + 14 = 0.
Now we can substitute 2y = 4x + 1:

4x 2 + (4x + 1)2 + 24x − 12(4x + 1) + 14 = 0.
This simplifies to 20x 2 − 16x + 3 = 0, i.e. (2x − 1)(10x − 3) = 0. The solutions
3
are x = 21 and x = 10 .
A final bit of substitution tells us that the points of intersection are
µ ¶ µ ¶
1 3 3 11
, and , .
2 2 10 10

QQ
1. Find the equation of each of these circles. Express your answer in the form
ax 2 + a y 2 + bx + c y + d = 0, where a, b, c and d are integers.
p
(a) Centre (4, 2), radius 3 (b) Centre (2, −1), radius 5.
(c) Centre (0, 21 ), radius 32 . (d) Centre (5, −1) passing through (1, 1).
(e) Centre (1, 23 ), radius 31 . (f ) The points (2, 1) and (6, 1) lie at the
ends of a diameter.
2. Find the centre and radius of each of these circles.

(a) x 2 + y 2 − 4x + 2y − 4 = 0
(b) x 2 + y 2 − 2x − 3 = 0
(c) 4x 2 + 4y 2 − 4x + 4y + 1 = 0
(d) x 2 + y 2 − 2ax − 2b y = 0
1. (a) x 2 + y 2 − 8x − 4y + 11 = 0 (b) x 2 + y 2 − 4x + 2y = 0 (c) x 2 + y 2 − y − 2 = 0

(d) x 2 + y 2 − 10x + 2y + 6 = 0. Substitute x = 1, y = 1 into (x − 5)2 + (y + 1)2 = r 2
(e) 3x 2 + 3y 2 − 6x − 4y + 4 = 0 (f) x 2 + y 2 − 8x − 2y + 13 = 0. Centre is (4, 1), radius is 2.
49
p
2. (a) (2, −1), radius 3. (b) (1, 0), radius 2. (c) ( 12 , − 12 ), radius 21 . (d) (a, b). Radius a2 + b2.

Something amazing and something wonderful

First something amazing. It’s an astonishingly quick and easy way to find the
equation of a tangent to a circle, if we know its equation and the coordinates
of the point of contact.
We’ll first show you how it works for an equation whose centre is the origin:
the tangent to the circle x 2 + y 2 = 20 at the point P (4, 2).
By substituting x = 4 and y = 2 into the equation we see that P lies on the
circle.
To find the equation of the tangent we use what’s called “semi-substitution".
What this means is best explained by showing you. Where we have
x 2 = x × x, we replace only one of the x’s with x = 4, to give 4x. Similarly we
replace only one of the y’s in y 2 , to give 2y.
The circle equation has been changed to 4x + 2y = 20. That’s it. This is the
equation of the tangent at (4, 2).
It is very easy to show that this is correct.
y
A
6 A g r ad i ent = −2
A
A
A
A
A
1
g r ad i ent =
A
2 A P (4, 2)
@
@
R A
A
A - x
O A
A
A
A
A
A
First we can see that P lies on this line (its coordinates satisfy the equation
4x + 2y = 20).
Then the gradient of the line is −2 and the gradient of OP is 21 , so the radius
to P is perpendicular to this line which passes through P . The line must be
the tangent at P .
Here is another example.
The equation of the tangent to the circle 4x 2 + 4y 2 = 5 at (1, − 21 ) is 4x − 2y = 5.

Isn’t that amazing?
If you think so, turn to the next page for something even more extraordinary.

What is even more amazing is that by learning what semi-substitution means

for linear terms, we can find the tangent for any circle.
Take the tangent at (2, −1) to the circle
x 2 + y 2 + 8x + 10y − 11 = 0.
We know how to semi-substitute x = 2 and y = −1 into x 2 and y 2 .

To semi-substitute x = 2 into 8x we simply regard it as 4x + 4x and replace
only one of the x’s, i.e. we write 4x + 8.
Similarly, since y = −1, we replace 10y with 5y + 5(−1) = 5y − 5.
Completing the semi-substituting gives us the equation of the tangent:
2x − y + 4x + 8 + 5y − 5 − 11 = 0 which simplifies to 3x + 2y = 4.
You may like to check for yourself that this is indeed the tangent to the circle
at (2, −1), in the same way that we did in the previous example.

Conic sections
Now something wonderful. We have looked in some detail at the parabola
(the curve associated with a quadratic function) and also at the circle. In fact
they both belong to a family of curves called conic sections, or “conics" for
short.
The complete family of conics consists of circles, ellipses, parabolas and
hyperbolae.
They are referred to as a family because they are connected in an exquisitely
elegant way50 that we are about to discover.
Circles appear in all manner of places. Parabolas are not so ubiquitous. A
parabola is the shape used to make a satellite dish or the lens of a reflecting
telescope, since it focusses electromagnetic waves onto a single point.
Although we occasionally see hyperbolic shapes too, it is the function (its
formula if you like) that is so important. For example the curve produced by
1
y=
x
is a type of hyperbola. We’ll be studying this function in a later chapter.
To see how all these conic sections are related we start by imagining a
vertical straight line (the dotted line in the diagram) and a second straight
line moving round it and maintaining a constant angle with it. This second
line, called a generator, traces out a mathematical cone, which looks like
two physical cones one on top of the other. Since mathematical “lines" are
infinitely long these are very large cones!
J

P
P

J

J

J

J

J

J

J

J

J
Now we are going to imagine slicing through this infinitely large object with
different planes. Where it cuts through the cone we get a curve. Any curve
produced in this way is, by definition, a conic section. Depending on the
angle of the plane, we get a different member of the family.
If the angle is shallow, i.e. the plane is not far off being horizontal,we get a
closed curve. Progressively steepen the angle and it starts producing curves
that are not closed. Eventually, if the angle is steep enough, the plane cuts the
cone twice.
These “cuts" or “sections" are illustrated on the next page.
50
Identified by Menaechmus, born around 380 BC in what is now Turkey.

(a) Circle (b) Ellipse

Angle: zero (horizontal). Angle: less than generator.
!!
!!
! !!
!
! !!
!
!
!!
!!
JJ !

J !!
!
J
! J!
J
! ! J
!
J
!
! J

J
J

J
J

J
J

J
J

J
J

J
J

J
J

J
J

JJ

JJ
(c) Parabola (d) Hyperbola

Angle: same as generator. Angle: greater than generator.
(The curve has two branches.)

J

J @

@
J @

J @

@

JJ

J

J

J J

J
J
J
J

J
J

J
J

J
J

J J

J J

J J

J J

J J

J J

J J
@ @
@
@
@

Exercises 30
1. Find the equation of the circle centre (1, −2) that passes through (4, 1).
2. Find the equation of the circle whose centre is (6, 2) and which touches
the y-axis.
3. A circle passes through (0, 1) and (0, 7) and its centre lies on the line
y = 2x.
Find the coordinates of its centre and its radius. Hence determine its
equation.
4. Show that these two circles touch:
x 2 + y 2 + 6x + 2y + 1 = 0 and x 2 + y 2 − 6y + 5 = 0.
There is a hint in the footnote51 .
5. A circle touches the positive x-axis and the positive y-axis. Its centre
lies on the line 3x + y = 8.
Find the centre of the circle.
6. A circle passes through the points P (4, 0), Q(9, 1) and R(3, 5). Show that
the triangle is right angled and hence, or otherwise, find the equation
of circle.
7. The line x = 3 is a tangent to the circle x 2 + y 2 + 4x − 2y − 12 = 0. Find

the point of contact.
8. Show that the line x + 3y = 8 is a tangent to the circle

x 2 + y 2 − 8x + 4y + 10 = 0.
9. Find the length of the tangent from A(6, 4) to the circle

x 2 + y 2 − 2x − 6y + 1 = 0.
10. Show that the line y = mx + c is a tangent to the circle x 2 + y 2 = a 2

provided c 2 = a 2 (1 + m 2 ).
11. The radius of the circle x 2 + y 2 +10x −8y +c = 0 is 8. Find the value of c.
12. Find the centre of the circle that touches the line 2y + x = 7 at A(1, 3)
and passes through the point B (−1, 5).
13. Show that the line y = 2x + 5 touches the circle x 2 + y 2 = 5 at (−2, 1).
If the line intersects the y-axis at P ,write down the equation of the
other tangent to the circle that passes through P . [There is a hint in the
footnote52 .]
14. The point T (p, q) lies on the circle x 2 + y 2 = a 2 . Show that the tangent
to the circle at T is px + q y = a 2 .
15. Sketch the circle x 2 + y 2 − 2px − 4p y + 4p 2 = 0.

51
Find the distance between the centres.
52
Draw a sketch.

16. Show that the point P (5, 1) lies on the circle x 2 + y 2 − 2x + 4y − 20 = 0.

Find the centre of the circle C and the gradient of PC . Hence find the
equation of the tangent to the circle at P .
17. Find the centre and radius of the circle 2x 2 + 2y 2 + 2x − 10y + 5 = 0.
18. Find the two points wherep the line 3x + y = 4 crosses the circle with
centre (1, −4) and radius 5.
19. The line 3y = x + 5 meets the circle x 2 + y 2 = 5 at A and B .

Find the equation of the perpendicular bisector of AB .
20. (a) Show that the centre of the circle 9x 2 + 9y 2 − 30x + 6y + 22 = 0 is

( 35 , − 13 ).
(b) Find the radius of the circle.
(c) Explain why the circle crosses the x-axis but not the y-axis.
21. Where do the circles x 2 + y 2 = 20 and x 2 + y 2 = 10x intersect?
22. The line y = x is a tangent to a circle whose centre is (4, 0). Find its
radius.
Hence determine the equation of the circle.
23. A circle passes through the points A(1, 0), B (5, 0) and C (3, 4).
(a) Using a diagram to explain why C lies on the perpendicular bisector
of AB .
(b) Find the equation of the perpendicular bisector of AC .
(c) Hence, or otherwise, find the position of the centre of the circle.
(d) Write down the radius of the circle.
(e) Find the equation of the circle.
24. A parabolic bowl sits on a horizontal table. Referred to axes along the
table and vertically upwards the equation of the bowl’s cross section is
2y = x 2 .
A spherical ball of radius 2 units is dropped inside the bowl. When the
ball comes to rest, how far is its centre above the level of the table?
[The diagram is not drawn to scale.]
y
6
B
B
- x

25. By sketching the curve, show that the y-coordinate of every point on
the circle
x 2 + y 2 − 2x − 4y + 1 = 0
satisfies the condition 0 ≤ y ≤ 4.
Write down a similar condition for the x-coordinate of every point on
the circle.
26. (a) Determine the two values of k for which the line y = 2x + k is a
tangent to the circle
x 2 + y 2 − 12x − 2y + 17 = 0.
(b) Find the equation of the line through the centre of the circle
perpendicular to the two tangents.
(c) Hence, or otherwise, find the points of contact of each tangent.
27. The diagram shows two circles with centres (−2, −1) and (2, 1), each
touching the line y = 0. The other common tangent is the line y = mx.
Find the value of m.
y
6
•
(2, 1)
- x
•
(−2, −1)
y
28. The curve x y = 3 crosses the
circle x 2 + y 2 = 10 at the points 6
A, B , C and D, taken in order
anticlockwise.
Show that ABC D is a rectangle
-x
and find its area. O
29. (a) Show that the points P (−1, 0), Q(1, 4) and R(5, 2) form a right angled
triangle, stating which side is the hypoteneuse.
(b) Write down the coordinates of the centre of the circle that passes
through P , Q and R.
(c) Find the coordinates of the other point on the x-axis through which
the circle passes.
[Hint: it is not necessary to find the equation of the circle.]

Solutions 30
1. x 2 + y 2 − 2x + 4y − 13 = 0.
Substitute x = 4, y = 1 into (x − 1)2 + (y + 2)2 = r 2 .
2. x 2 + y 2 − 12x − 4y + 4 = 0. Touches y-axis so radius = 6.

p
3. (2, 4). Radius 10. Equation x 2 + y 2 − 4x − 8y + 10 = 0.
Centre is on perpendicular bisector of (0, 1) and (0, 7), i.e. y = 4 and also on y = 2x.
For the radius find the distance between (say) (0, 1) and (2, 4).
4. The circles are (x + 3)2 + (y + 1)2 = 9 and x 2 + (y − 3)2 = 4. The radii are 3 and 2.
p
The centres are at (−3, −1) and (0, 3) so the distance between them is 32 + 42 = 5.
Since this distance is the sum of their radii the circles must touch.
5. (2, 2).
The centre lies on 3x + y = 8, and it also lies on y = x (by symmetry), so 3x + x = 8.
6. The gradient of PQ is 15 , the gradient of P R is −5 so PQ ⊥ P R and QR must be

a diameter. The centre is therefore C (6, 3). For the radius PC 2 = 22 + 33 = 13.
Equation: (x − 6)2 + (y − 3)2 = 13, or x 2 + y 2 − 12x − 6y + 32 = 0.
7. (3, 1).
The equation of the circle is (x + 2)2 + (y − 1)2 = 25. Now there are a couple of methods.
One way is to solve this equation simultaneously with x = 3, which gives
y 2 − 2y + 1 = (y + 1)2 = 0. Repeated root so the line touches the circle.
A neater way is simply to notice that, since the tangent is vertical, the centre must be
on the same "horizontal" level as the centre.
8. We solve the two equations simultaneously, by substituting x = 8 = 3y.

(8 − 3y)2 + y 2 − 8(8 − 3y) + 4y + 10 = 0 simplifies to y 2 − 2y + 1 = (y − 1)2 = 0
Though not asked to find where it touches, it does so at y = 1 and therefore at x = 5.
p
9. 17.
Suppose the point of contact is B . The circle is (x − 1)2 + (y − 3)2 = 9,
so the centre is C (1, 3), AC 2 = 26 and BC 2 = 9. Hence AB 2 = 26 − 9.
10. For the line to be a tangent, x 2 + (mx + c)2 = a 2 must yield equal roots,
i.e. discriminant = 0. Writing as (1 + m 2 )x 2 + 2mc x + (c 2 − a 2 ) = 0, discriminant is
4m 2 c 2 − 4(1 + m 2 )(c 2 − a 2 ) = 0, which simplifies to the required relationship.
11. c = −23. The equation of the circle must be (x + 5)2 + (y − 4)2 = 64.
12. (3, 7).

The centre lies on the line through A perpendicular to the tangent, and also on
the perpendicular bisector of AB .
The gradient of the tangent is − 12 , so the line perpendicular to it is
y − 3 = 2(x − 1), i.e. y = 2x + 1.
The gradient of AB is −1, and its midpoint (0, 4), so its perpendicular bisector is
y = x + 4. The two lines meet at (3, 7).

13. y = −2x + 5
The line y = 2x + 5 has gradient 2.
From the sketch, the other tangent must have the same y-intercept,
but gradient −2 (by symmetry).
y
6
5A
A
g r ad i ent = 2 A g r ad i ent = −2
@ A
@ A
R
@ A
A
A
A
A
A - x
A
A
AA
q
14. The centre of the circle C is the origin, so the gradient of C T is p .
p
The gradient of the tangent at T is therefore − q .
Since p 2 + q 2 = a 2 , T lies on this line px + q y = a 2 .
p
But this line also has gradient − q so it must be the tangent at T .
Or we could simply construct the equation of the tangent.
p
It is y − q = − q (x − p), which simplifies to px + q y = p 2 + q 2 .
But p 2 + q 2 = a 2 because (p, q) lies on the circle.
15. The equation of the circle is

(x − p)2 + (y − 2p)2 = p 2 .
Its centre at (p, 2p) and radius p
means the circle touches the y-axis. p
•
2p
16. The coordinates of P satisfy the equation.

The equation is (x − 1)2 + (y + 2)2 = 25, so C is (1, −2) and
the gradient of C P is 34 .The tangent is therefore
y − 1 = − 43 (x − 5), which simplifies to 3y + 4x = 23.
17. (− 12 , 52 ). Radius 2.
The equation is x 2 + y 2 + x − 5y + 25 = 0, which can be written
(x + 12 )2 + (y − 52 )2 = 4.

18. (2, −2) and (3, −5).

The equation is (x − 1)2 + (y + 4)2 = 5.
It is easiest to substitute y = 4 − 3x without multiplying this out, giving
(x − 1)2 + (8 − 3x)2 = 5, i.e. 10x 2 − 50x + 60 = 0,
or 0 = x 2 − 5x + 6 = (x − 2)(x − 3).
19. (1, 2) and (−2, 1). y + 3x = 0.

The substitution of x = 3y − 5 into the equation of the circle simplifies to
y 2 − y − 2 = 0.
Gradient of AB is 13 , so gradient of the perpendicular bisector is −3.
This is all we need to write down the equation of the bisector, because it must go
through the centre of the circle, i.e. the origin.
20. (a) The equation is x 2 + y 2 − 10 2 22

3 x + 3 y + 9 = 0 which can be written
(x − 53 )2 + (y + 13 )2 = 94 .
q
(b) The radius is 49 = 23 .
1
(c) The centre is 3 from the x-axis, whereas the radius is 32 .
5
The centre is 3 from the y-axis. The sketch makes the configuration very clear.
y
6
- x
•
5
3 HH
Y
H r ad i us = 32
21. (2, 4) and (2, −4).

We can’t substitute either x or y from one equation into the other.
But we can substitute x 2 + y 2 , giving 10x = 20 ⇒ x = 2.
Now using either equation (we’ll choose the first) 4 + y 2 = 20 ⇒ y = ±4.
p
22. Radius is 8. Equation x 2 + y 2 − 8x + 8 = 0.
A geometric approach is probably the easiest way to go.
Draw a sketch. O P̂C = 90◦ , P ÔC = 45◦ , so P ĈO = 45◦ .
Triangle OPC is isosceles, so P is (2, 2).
p
By Pythagoras PC 2 = 22 + 22 , i.e. PC = 8. Equation is (x − 4)2 + y 2 = 8.
y
6
@P
@
@
@
@
@ -x
O C (4, 0)

Alternatively we could use coordinate geometry to find the radius.

The line y = x is a tangent to the circle (x − 4)2 + y 2 = r 2 , for some value of r .
Solving simultaneously must result in a quadratic equation with equal roots:
(x − 4)2 + x 2 = r 2 or 2x 2 − 8x + 16 − r 2 = 0, where 64 − 8(16 − r 2 ) = 0
p
⇒ r 2 = 8 (and r 6= − 8).
23. (a) The perpendicular bisector of AB is the line x = 3, so it passes through C .
y
6 C (3, 4)
M (2, 2)
- x
A(1, 0) B (5, 0)
(b) 2y + x = 6
The mid-point of AC is M (2, 2) and its gradient 2, so its perpendicular bisector is
y − 2 = − 21 (x − 2).
(c) (3, 32 ).
The centre is where x = 3 on 2y + x = 6, i.e. where y = 32 .
(d) Radius is 52 .
By symmetry C is the highest point on the circle, so radius is 4 − 32 = 25 .
(e) x 2 + y 2 − 6x − 3y + 5 = 0.
Equation is (x − 3)2 + (y − 23 )2 = 25
4
.
24. 5 units.
Suppose the centre of the ball is at (0, h). The equation of its cross-section is
x 2 + (y − h)2 = 9.
We need to solve this simultaneously with the equation of the cross-section of the bowl.
x2
The easiest way to do this is to substitute x 2 = 2y, rather than y = 2 .
The curves will touch if the resulting equation has equal roots.
We have 2y + (y − h)2 = 9, or y 2 + 2y(1 − h) + h 2 − 9 = 0.
The discriminant must be zero, so 4(1 − h)2 − 4(h 2 − 9) = 0,
which simplifies to 1 − 2h = −9.
2
Had you substituted y rather than x 2 , it would have led to x 2 + ( x2 − h)2 = 9.
Although technically a quartic equation we can think of it as a quadratic in x 2 ,
which suggests we substitute X = x 2 and first solve the equation X + ( X2 − h)2 = 9.
This simplifies to
X2
4
+ X (1 − h) + h 2 − 9 = 0 and the zero discriminant condition
⇒ (1 − h)2 − (h 2 − 9) = 0.
Notice we didn’t need to revert to back to the variable x.

25. The equation can be written

(x − 1)2 + (y − 2)2 = 22 , so the centre is
4
at (1, 2) and the radius 2.
The sketch looks like this. The circle touches the x-axis.
From the sketch we can see that the
y-coordinate of the highest point of the circle is 4; 2 • (1, 2)
the lowest is on the x-axis.
The x-coordinates lie in the range −1 ≤ x ≤ 3.
−1 0 1 2 3
26. (a) k = −1 and k = −21.

x 2 + (2x + k)2 − 12x − 2(2x + k) + 17 = 0

simplifies to y = 2x −
1
5x 2 + 4(k − 4)x + k 2 − 2k + 17 = 0. @
R

The discriminant being zero leads

(eventually) to k 2 + 22k + 21 = 0.

(b) 2y + x = 8.

•
The equation of the circle can be written

(x − 6)2 + (y − 1)2 = 20, so its centre is (6, 1).
The gradient of each tangent is 2, so

the gradient of any line ⊥ to it is − 21 , and the

equation of the line we are after is y = 2x − 21
y − 1 = − 12 (x − 6). @
R
@

(c) (2, 3) and (10, −1).
The radius to the point of contact of a tangent will be perpendicular to it,
so the two points of contact will be where 2y + x = 8 crosses each tangent.
For example, solving 2y + x = 8 and y = 2x − 1 simultaneously gives x = 2.
Another (longer) method would be to put each value of k into the equation
5x 2 + 4(k − 4)x + k 2 − 2k + 17 = 0. For example, with k = −1, 5x 2 − 20x + 20 = 0
i.e. x 2 − 4x + 4 = (x − 2)2 = 0 ⇒ x = 2. Notice the equal roots that we anticipated.
27. By symmetry if the line touches one circle it touches the other.
We will use the right hand circle, which is (x − 2)2 + (y + 1)2 = 22 .
Solving the equations simultaneously we have (x − 2)2 + (mx + 1)2 = 4,
which simplifies to (1 + m 2 )x 2 − 2(2 + m)x + 4 = 0. The line touches the circle
where the discriminant is 0, i.e. 4(2 + m)2 = 16(1 + m 2 ) ⇒ m(3m − 4) = 0.
The solution m = 0 is the x-axis tangent that we already know about;
4
the answer is m = 3 .
There is a much quicker method, in fact it only takes two lines of working.
However it uses a trigonometry formula you haven’t met yet. A treat you can
look forward to.

28. The points A, B , C and D are (3, 1), (1, 3), (−3, −1), (−1, −3).
Area is 16 square units.
3
Substituting y = x into x 2 + y 2 = 10 gives x 2 + x92 = 10.
Putting x 2 = X , we have X + X9 = 10.
Multiplying every term by X gives X 2 − 10X + 9 = (X − 1)(x − 9) = 0.
3
Now reverting to x we have x 2 = 1 or x 2 = 9. Then, for example, if x = −1, y = −1 = −3.
Now we need to show that quadrilateral ABC D is a rectangle.
We’ll label the points A(3, 1), B (1, 3), C (−3, −1), D(−1, −3).
You may have labelled them starting in a different place. It won’t matter.
First we show that opposite sides are parallel (it is a parallelogram).
3−1 −3+1
Gradient of AB is 1−3 = −1 and gradient of C D is −1+3 = −1.
Similarly the gradient of BC and D A is 1.
We can see that BC and D A are perpendicular to AB and C D, so ABC D is a rectangle.
For the area of the rectangle.
p p p
AB 2 = (3 − 1)2 + (1 − 3)2 = 8, so AB = 8 = 2 2. Similarly C D = 4 2.
p p
Area of the rectangle = 2 2 × 4 2 = 16.
p p p p p p
Or, from AB = 8 and C D = 32, area = 8 × 32 = 8 × 32 = 256 = 16.
We’ll find out how to sketch the curve x y = 3 in a later chapter.
29. (a) Gradient of PQ is 2, and of QR is − 12 . The product of these gradients is −1,

so PQ ⊥ QR. P R is the hypoteneuse.
P R is a diameter of the circle (the angle in a semicircle = 90◦ ) so the centre is at
( −1+5
2
, 0+2
2
) = (2, 1).
If S is the other point (apart from P ) where the circle crosses the x-axis,
its centre lies on the perpendicular bisector of AP . Since the x-coordinate of the
centre is 4, S must be at (5, 0). The diagram looks like this:
y
6 Q(1, 4)
R(5, 2)
•
- x
P (−1, 0) S


• Completing the square. For example
¢2
x 2 + 6x − 5 ≡ (x + 3)2 − 4, 3x 2 + 4x − 7 ≡ 3 x + 23 − 25
¡
3
• Maximum or minimum values of a quadratic expression
• Solving quadratic equations. Three methods: (a) Factorising,

p
−b± b 2 −4ac
(b) Completing the square, (c) the formula x = 2a
• The discriminant. Determining the number of roots.

∆ > 0 ⇒ 2 root ∆ = 0 ⇒ 1 (repeated) root, ∆ < 0 ⇒ no roots,
where ∆ = b 2 − 4ac
• Disguised quadratic equations. For example

p p
x 4 − 2x 2 − 8 = 0 (substitute X = x 2 ), x − 2 x − 3 = 0 (substitute X = x)
• Sketching quadratic functions. Which way up is it?

Positive x 2 coefft like ∪, negative x 2 coefft like ∩.
Intersection with the axes. No roots, distinct roots, multiple roots.
• Quadratic inequalities. Solve by curve sketching.

e.g. x(x − 3) > 0 when y > 0, where y = x(x − 3), i.e. curve above x-axis
Modulus notation, e.g. |x| < 3 meaning −3 < x < 3
• Inequalities with an algebraic denominator. If an inequality includes

a term with an algebraic denominator, multiplying through by that
denominator is not possible because it may be negative. Multiply both
sides by (denominator)2
x+1 1
For example x ≥ 2 [multiply by x 2 ] or x−4 < 3 [multiply by (x − 4)2 ]
• Non-linear simultaneous equations. For example

y = x 2 − 2x + 1 and x(y − 1) = 18
y = x +3 2x = 3y
Solve by substituting a variable isolated from one equation into the
other equation.
• Non-linear simultaneous inequalities in two variables.

For example using a sketch to show where the values of x and y are for
which y > x(x − 2) and y < x + 4
• Circles. A reminder of the important Theorems and properties
• Equation of a circle with centre (a, b) and radius r :

(x − a)2 + (y − b)2 = r 2
Constructing the equation. Deducing the centre and radius from the
equation.

Chapter 8
SKETCHING AND TRANSFORMING

CURVES
Overview
8.1 Sketching polynomials. What does “sketching" mean? Roots and the
zero product principle.
Multiple roots. Higher order polynomials. Which way up is the curve?
1 1
8.2 The reciprocal functions. y = x and y = x 2 . Asymptotes.
8.3 Transforming functions.
Translations y = f (x) + a and y = f (x + a)
Stretches y = a f (x) and y = f (ax)
Reflections (in the coordinate axes) y = − f (x) and y = f (−x)
8.4 Periodic functions. Definition. Amplitude and period. Transforming the
sine, cosine and tan curves.
8.5 An elegant way of thinking about transformations. (Optional.)
8-1
CHAPTER 8. SKETCHING AND TRANSFORMING CURVES
8.1 Sketching Polynomials

Roots and the zero product principle
The notion of roots, or solutions, lies at the head of curve sketching, and
fundamental to the idea of finding roots is the zero product principle. We
met this in the last chapter; here is a quick reminder. If you don’t need it, skip
to the next page.
zero.
? × ? =0
What could the numbers be?

It can’t work unless one of them numbers is zero. Or both could be zero. If
the product is in the form of two brackets multiplied together, we have
something like this
(x − 10)(x + 2) = 0 ⇒ x − 10 = 0 or x + 2 = 0
⇒ x = 10 or x = −2.
This will apply however many elements there are in the product, for example
x(x + 1)(x − 2)(x − 7) = 0.
Either x = 0, x + 1 = 0, x − 2 = 0 or x − 7 = 0.
The roots are x = −1, 0, 2 and 7.

What is sketching?
We talked about the idea of sketching graphs, and why it is such a vital skill,
in the Chapter 7. It is because the brain is far better at dealing with pictures
than it is with abstract concepts.
By sketching a curve, as opposed to plotting it, we mean making a drawing
of it that captures the key features, for example where it crosses the
coordinate axes. It doesn’t need to be particulary neat, but if it’s too small or
untidy it will be difficult to use. Always draw a sketch in pencil, so you can
correct or improve it if you need to.
We defined what a polynomial is in Chapter 3. If you want a reminder see

page 3-21.
But first a summary of what we learnt in Chapter 7 about sketching the

simplest1 polynomials, i.e. quadratics. If this is very familiar to you, skip the
next page.
1
Strictly speaking a straight line is a polynomial of order 1.

Below are three different versions of the same quadratic function:

y = x 2 − 4x − 5
y = (x + 1)(x − 5)
y = (x − 2)2 − 9.
We can identify
• which way up the curve is, from the sign of the x 2 coefficient;
• what the y-intercept is by putting x = 0 into the first version;
• what the roots are by putting y = 0 into the second version;
• the maximum or minimum point from the third version.
Using a combination of these will always be enough to sketch a quadratic

curve.
y
6 x =2
BB

9
B
B
B• • - x
−1 5
−5 •
y = −9
If you want to refresh your memory on how we arrived at all these

conclusions, see page 7-53.
We’ll pick up the story from here. The maximum/minimum issue only
applies to quadratics; we are going to need calculus to find maximum and
minimum points for polynomials (see Chapter 10). We’ll be seeing how the
others apply—to polynomials in general.

8.1.1 Cubic functions and beyond

Take this function
y = (x + 1)(x − 2)(x − 4).
It’s a cubic and its roots, i.e. where y = 0, are at x = −1, 2 and 4.
To find the y-intercept we don’t need to multiply out the brackets—just put
x = 0 into the formula: y = (1)(−2)(−4) = 8.
The function looks like this (notice the axes use different scales):
y
6
- x
−1 2 4
It’s a beautiful curve. More interesting than a quadratic curve. For example
the maximum and points do not lie midway between the roots. You can see
why this is: the curve at x = −1, for example, is noticeably steeper than it is at
x = 2, so the shape of the curve between these points is skewed to the left.
Similarly the curve between 2 and 4 is skewed to the right.
The cubic curve below has exactly the same roots, but it’s the other way up.
y
6
- x
−1 2 4
−8
This is the function

y = −(x + 1)(x − 2)(x − 4).
The roots are the same, but every y-value has the opposite sign. The curve
has been reflected in the x-axis. Not surprisingly the new y-intercept comes
out as −(1)(−2)(−4) = −8.
Checking the y-intercept is a good way to tell which way up the curve is.
That’s fine as long as the curve doesn’t go through the origin.

Which way up?

There is a quick and easy way to tell which way up the curve, whether or not
it goes through the origin.
All we need to do is find the sign of the highest power of x. We think of it like
this: if we make x big enough the highest power will always dominate. Take
this function, for example
y = 50x + 10x 2 − x 3 .
For small values of x the function is decidedly positive. e.g. when x = 1,

y = 59. When x = 10 it’s even more positive: y = 500. By the time we reach
x = 20, the negative x 3 term is “winning" hands down:
y = 1000 + 4000 − 8000 = −3000. The further we go the more negligible the
first two terms are compared with the x 3 term.
We say “as x tends to infinity, y tend to minus infinity", writing it like this:
as x → ∞, y → −∞.
In our example the diagram below on the right applies.
Armed with the sign of the highest power of a function we can see which way
up the curve is. Check this for yourself with the curves on the previous page.
Coeff. highest power of x positive Coeff. highest power of x negative
y y
6 6
-x -x
The sign of the coefficient of the highest power of x will tell us

which way up the curve is.

You may have spotted that this is a generalisation of what we found for
quadratics, where the highest power of x is 2.

@
I
@
@
positive x 2 coefficient ⇒ curve up this way up
negative x 2 coefficient ⇒ curve up this way up
B
B

Example 1
Sketch the curves of these functions

(i) y = x 3 − 4x (ii) y = (x − 3)(5 − x)(2x + 1).
Solution
(i) y = x(x 2 − 4) = x(x + 2)(x − 2) so its roots are at x = 0, −2 and 2.

The coefficient of x 3 is +1, so the curve looks like this:
y
6
- x
−2 0 2
(ii) The roots are where x − 3 = 0, 5 − x = 0 and 2x + 1 = 0, so the roots are at

3, 5 and − 12 .
The x 3 coefficient comes from multiplying together the three x terms—one
from each bracket
y = (x − 3)(5−x)(2x + 1).
The coefficient is −2.
We can also pick out the y-intercept (without expanding the brackets fully):
y = (x −3)(5 − x)(2x+1).
It is −15, so the curve looks like this
y
6
- x
− 21 3 5
−15
We have made no attempt to keep to the scale on each axis. Imagine how
odd the graph would look if we did! It means we can’t use the sketch to infer
anything intuitive about distances or gradients. Rather like the map of the
London Underground, it’s purpose is to show key points and the relationships
between them, not the actual physical layout.

Example 2
Sketch the function y = x(x − 2)(9 − x 2 ).
Solution
This a polynomial of order2 4, i.e. a quartic.

Factorising the last bracket, we can see that its roots are at x = 0, 2, 3 and −3.
The coefficient of x 4 is (1)(1)(−1) = −1, i.e. it is negative.
The y-intercept is the origin. The function looks like this:
y
6
-x
−3 0 2 3
2
The highest power of x when the polynomial is multiplied out.

The shape of a polynomial curve

We have already met the term “order" in the context of polynomials.
The order of a polynomial is the highest power of x that it contains (in its
multiplied out form). You have probably noticed that every time we increase
from one order to the next we get an extra “wiggle" in the curve.
Order 2 Order 3 Order 4
and so on. . .

8.1.2 Multiple roots

At the heart of understanding how a polynomial behaves is knowing where
the roots are, and what kind of roots they are. In Chapter 7 we talked about
repeated roots. Here is a quick reminder.
In the top graph the function has two roots, say x = 2 and x = 4. We say they
are distinct.
If we imagine moving the curve up slightly (the middle graph), the roots
move closer together, to say 2.8 and 3.2. But the roots are still distinct.
Continuing to move the curve up, at a certain point the two roots come
together. Now the curve touches the x-axis and the two roots are both equal
to 3. We say there is a repeated root at x = 3.
y
6
A
A
A
• • -x
P
i
PP
P
Two distinct roots
y
6
A
A
A
• • -x
PP
i
PP
P
Two distinct roots
- but getting closer to each other
y
6
A
A
A
• -x
PP
i
P PP
P
Two equal roots
We can think of the equation (x − 3)2 = 0 as having one root, x = 3. But, as the
graphs show, it is rather more meaningful to think if it has having two roots,
both equal to 3.

We shall see that a polynomial can have any number of roots: a cubic can
have three, a quartic four etc. That means the scope for repeated, or multiple
roots is much greater.
In the top graph the function has three distinct roots.
In the middle graph we imagine the curve “straightening out" a bit, forcing
the three roots closer together.
In the bottom graph they concur. The curve approaches from the left with
negative values, levels out, and then switches to positive values.
y
6
• • • - x
PP
i
PP
Three distinct roots
y
6
• • • - x
PP
i
PP
Three distinct roots
- but getting closer together
y
6
• - x
PP
i
PP
Three equal roots
A triple root like this looks like a single root, except that the curve flattens out
at the root, i.e. it has zero gradient at that point.

Of course we could go on forever with this, but we’ll end with a quick look
quadruple roots.
• • • •
P
i
PP
PP
P
Four distinct roots
•
H
Y
HH
H
HH
Four equal roots
Yes, you guessed it, it ends up looking very much like a double root. The
curve is flatter than at a double root, but it’s not possible to show this clearly
on a sketch.

QQ
Sketch these functions, in each case marking where the curve crosses the
axes.
y = x 2, y = x 3, y = x 4, y = x 5.
The answers are on the next page.

y = x2 y = x3
6 6

-x -x
y = x4 y = x5
6 6

-x -x
The y = x 4 curve is, for sketching purposes, the same shape as y = x 2 . It is

“steeper" of course, because the y values increase more rapidly.
Clearly the same is true of y = x 5 , which is the same shape but steeper than
y = x 3 . And so on...

Example 3
Sketch these functions.

(a) y = x(2x − 3)2 .
(b) y = x 2 (16 − x 2 ).
(c) y = 3(x + 2)(x − 1)3 .
Solution
(a) This polynomial is a cubic,
y
which when multiplied out
would start 6
y = 4x 3 . . .
which tells us which way up it is.
It has a single root at x = 0 and a
double one at 32 . - x
0 3
2
So the function must looks like
this:
y
(b) The coefficient of x 4 is −1
and the function factorises to 6
y = x 2 (4 + x)(4 − x)
so it has single roots at ±4 and a
double root at the origin.
-x
The function must look like this: −4 0 4
y
4
(c) The coefficient of x is 3. 6
It has single roots at x = −2 and

a triple root at the x = 1.
Putting x = 0 into the
formula, the y-intercept is
3(2)(−1)3 = −6.
-x
The function must look like this: −2 1
−6

So far in this chapter our quadratic function had two roots; our cubic
functions have all had three roots and our quartic functions four roots. We
know from our study of quadratics that this is not necessarily the case. The
order of the polynomial corresponds to the maximum number of roots.
Take this cubic function
y = (x − 2)(x 2 + 1).
Since the second bracket doesn’t have any solutions, the curve only has one
root. The coefficient of x 3 is positive (and the y-intercept is negative) so the
curve must look something like this
y
6
- x
0 2
−2
We can’t tell at this stage exactly where the “humps" are: they may both be to
the right of the y-axis, or both to the left. Calculus will allow us to pin this
down. (So we’ll return to this function in Chapter 10 to find out.)
Another example:
y = (x + 3)2 (2x 2 + 1).
Because 2x 2 + 1 = 0 has no roots this quartic has only two roots, both at
x = −3. You can probably pick out the coefficient of x 4 in your head if we see
the function as
y = (x + 3)(x + 3)(2x 2 + 1).
It is +2. In the same way we can pick out the y-intercept:
y = (x + 3)(x + 3)(2x 2 + 1).
It is +9. So the curve look like this
y
6
- x
−3

A cubic curve will always have to cross the x-axis somewhere: it has at least
one root.
Not so a quartic curve—it can escape altogether, just as quadratic curve can.
Two examples:
y
6
-
x
y
6
- x
Increasing the order of the polynomial further, the number of possible

configurations increases. For example look at this polynomial of order 5; it
has 3 roots.
y
6
-x

Example 4
The graphs below show the curves of two cubic functions. Find their
equations.
(a) y
6
- x
−2 1
−6
(b) y
6
- x
− 32 0 2
(1, −10)
Solution
(a) With a single root at −2 and a double root at 1 the equation of the curve
must be of the form
y = k(x + 2)(x − 1)2 .

The point (0, −6) lies on the curve, so substituting these coordinates into the
equation we have
−6 = k(2)(−1)2 or − 6 = 2k,
so k = −3 and the equation is
y = −3(x + 2)(x − 1)2 .
(b) With roots at − 32 , 0 and 2, the equation must be of the form

µ ¶
3
y = kx x + (x − 2).
2
Since the curve passes through (1, −5)
µ ¶
5 −5k
−10 = k (−1) = ⇒ k = 4.
2 2
The equation of the curve is
µ ¶ µ ¶
3 3
y = 4x x + (x − 2) = 2x × 2 x + (x − 2) = 2x(2x + 3)(x − 2).
2 2

Exercises 31
Sketch the curves in questions 1 to 8.

1. (a) y = x 2 − 2x − 8 (b) y = 8 + 2x − x 2
2. (a) y = (x − 1)(x + 2)(x + 3) (b) y = (1 − x)(x + 2)(x + 3)
3. (a) y = x(1 − x 2 ) (b) y = x 2 (1 − x)
4. (a) y = 2(3 + x)(x − 1)2 (b) y = 2(3 − x)(1 − x)2
5. (a) y = (x − 2)3 (b) y = 14 (x − 2)3
6. (a) y = x 2 (6 − x)2 (b) y = x(6 − x)3
7. (a) y = (x 2 − 1)(x + 3)2 (b) y = (x − 1)2 (x 2 + 3)
8. (a) y = 2x 3 (x − 1)2 (b) y = 2x 3 (x − 1)3
9. Sketch the curve of the function y = x 4 − x 3 − 6x 2 .
10. Are the curves of these two functions identical?

y = (2x − 7)(x − 3)3 and y = (7 − 2x)(3 − x)3 .
11. Find the equation of each of these cubic polynomials. (Make sure the
y-intercepts are correct.)
(a) y (b) y
6 6
6
-x - x
−2 1 −1 1 2
−4
(c) y
6 (d )
y
(−1, 2) 6
•
-x
-x −2 0 3
−1
−3
12. This question is about counting the number of possible roots a

polynomial can have. Do sketches to investigate. When counting them, the
roots need not be distinct, i.e. they can be multiple roots.
(a) For a polynomial of order 7, what are the possible number of roots it can
have?
(b) For a polynomial of order 8, what are the possible number of roots it can
have?

(c) For a polynomial of order 2n, what are the possible number of roots it can
have? For this polynomial, how many possibilities does this represent?

Solutions 31
1.(a) y (b) y
6 6
8
-x
−2 4 -x
−2 4
−8
2.(a) (b)
y y
6 6
6
-x -x
−3 −2 1 −3 −2 1
−6
3.(a) y (b) y
6 6
- x
−1 0 1 - x
0 1
4.(a) y (b) y
6 6
- x - x
−3 0 1 −3 0 1
5.(a) y (b) y
6 6
-x -x
2 2
−2
−8

6.(a) (b) y
y
6
6
-x
-x 0 6
0 6
7.(a) (b)
y
y
6
6
-x 3
−3 −1 0 1
-x
1
−9
8.(a) y (b) y
6 6
-x -x
0 1 0 1
9. Factorising the function: y

6
4 3 2 2 2
x − x − 6x = x (x − x − 6)
= x 2 (x + 2)(x − 3)
-x
−2 3
10. Comparing brackets, (7 − 2x) = −(2x − 7) and (3 − x)3 = −(x − 3)3 , so the
functions are identical.
11. (a) y = 3(x + 2)(x − 1)2

(b) y = 2(x + 1)(x − 1)(x − 2) or y = 2(x 2 − 1)(x + 2)
(c) y = −3(x + 1)3 (d) y = 21 x(x + 2)(x − 3)
12. (a) 1, 3, 5 or 7. (b) 0, 2, 4, 6 or 8

(c) 0, 2, 4, 6 . . . 2n, i.e. n + 1 possibilities.

8.2 The functions y = x1 and y = x12

1
We can also think of the function y = as the beautifully simple function
x
x y = 1, or its more general form x y = c 2 . The shape of the curve is a
hyperbola, one of the family of so-called conic sections we met at the end of
the last chapter.
The curve of this function has three properties that set it apart from every
function we have met so far. Firstly it has two distinct parts or branches to
it; secondly it has two asymptotes (we’ll explain what these are later); thirdly
there is a value of x for which there is no corresponding value of y.
1
Here’s what the graph of y = looks like.
x
1
y=
x
6
• ( 41 , 4)
• (1, 1)
1
• (2, 2 )
-x
•
(−5, −0.2)
(−1, −1) •
In the formula for y, the bigger the value of x, the smaller the value of y—
because we are dividing 1 by a bigger and bigger number.
In the right hand branch the shape of the curve is consistent with this: points
further to the right have smaller y-coordinates. (In the left hand branch the
values of x with larger magnitude are further to the left.)

4
Another example. This is the graph of y =
x
y
6
• (1, 4)
• (2, 2)
• (4, 1)
•
-x
(−8, − 21 )
(−2, −2) •
Can you see what the symmetries are? (See the footnote3 .)
3
The lines of symmetry are y = x and y = −x. The function also has rotational
symmetry of order 2 about the origin. (When rotating the curve about the origin
there are 2 two positions it could in that look identical.)

1 1
The curve of y = has many the features of y = .
x2 x
It differs in that the function is positive for all values of x and the curve
approach the x-axis more rapidly as x gets larger (although it is hard to show
this clearly on a sketch).
1
y=
x
6
• ( 21 , 4)
(−1, 1) • • (1, 1)
1
• (2, 4 )
• -x
(−5, 0.04)

Example 5
k
The curve y = passes through the point (4, 32 ).
x
Find the value of k and determine where the curve crosses the line y = x − 1.
Solution
By substitution 3 k
= ⇒ k = 6.
2 4
6
The curve is y = , and this crosses the line y = x − 1 where
x
6
x −1 =
x
x2 − x − 6 = 0
(x − 3)(x + 2) = 0.
The points of intersection are (3, 2) and (−2, −3).

Example 6
Show that the line y + 2x = 8 is a tangent to the curve x y = 8 and find the
point where it touches.
Use a symmetry argument to determine the equation of the other tangent to
the curve which has the same gradient.
Solution
To find where they meet we can substitute y = 8 − 2x into the equation of the
curve:
x(8 − 2x) = 8
0 = 2x 2 − 8x + 8
0 = x 2 − 4x + 4
(x − 2)2 = 0.
The equation has equal roots so the line is a tangent, touching where x = 2,
i.e. at. (2, 4).
y
6
(2, 4)
-x
(−2, −4)
By symmetry the other tangent will touch at (−2, −4), and since its gradient
is −2, its equation must be
y + 4 = −2(x + 2)
y + 4 = −2x − 4
y = −2x − 8
or y + 2x + 8 = 0.

Example 7 [Ignore this example if you haven’t yet met the equation of a circle.]
By sketching the curves x y = 1 and x 2 + y 2 = a 2 show that the equation
x4 − a2x2 + 1 = 0
has either 0, 2 or 4 distinct roots.

For what values of a does the equation have no roots?
Solution
The equation x 2 + y 2 = a 2 is a circle centred on the origin with radius a. The

sketch shows that there are three possible configurations: the circle will
1
intersect the curve y = four times, twice or not at all. The symmetries of
x
the functions prevent there from being an odd number of intersections.
But finding where the curves meet corresponds to solving the equations
simultaneously.
1
Substituting y = into the equation of the circle we have
x
1
x2 + = a2
x2
which can be written
x 4 − a 2 x 2 + 1 = 0.
So this equation must have 0, 2 or 4
roots.
The easiest way to find when there are y
no roots is to look at a borderline case. 6
The middle circle
touches both branches of the curve
1
y= . (1, 1)
x
By symmetry we can see that it does so
-x
at (1, 1) and (−1, −1). Any circle smaller
won’t cut the curve so won’t yield any (−1, −1)
solutions.
By
p Pythagoras the points of contact are
2 units from the origin, so the equation
p
has no roots when the radius ap< 2, or
to be more precise when |a| < 2.
There is a language issue to clarify. Strictly speaking when the circle touches
the curve there are still 4 roots—split into two equal pairs. This is why the
question says “distinct roots": depending on the value of r the equation has
0, 2 or 4 distinct, i.e. numerically different, roots.

Asymptotes
An asymptote in the graph of a function is a line4 that the curve gets closer
and closer to, but never reaches.
1
The curve y = x has two asymptotes: the coordinate axes.
1
y=
x
6
-x
If we imagine extending the x-axis, say as far as x = 100, the the point on the
curve with this x-coordinate has y-coordinate equal to 0.01.
(100, 0.01)
• -x
No matter how far we go the curve will never cross the x-axis. Because the
same thing happens whether we make x a very large positive number or a
very large negative number, we simply say: as x tends to infinity, y tends to
zero. Or more concisely
as x → ∞, y → 0.
The other asymptote is the y-axis. Let’s make the value of x close to zero. For
example x = 0.001, the value of y would be 1000, and so on. In this case
as x → 0, y → ∞.
As before, we don’t distinguish between +∞ and −∞ in this context.

For the curve to cross the y-axis we would need x = 0. But that’s impossible
because we can’t divide 1 (or any other number) by zero. This explains why
there is no value of y corresponding to x = 0. We say the function is
undefined for x = 0.
At this second asymptote something very strange happens. If we make x
approach zero from the right y → +∞; if we make it approach from the left
y → −∞. We say the curve is discontinuous (or there is a discontinuity) at
x = 0. It means there is a sudden jump in the y-value at that point.
4
For our purpose here it will be a straight line; an asymptote can be a curve.

Many other functions have asymptotes. And asymptotes aren’t necessarily

the coordinate axes. When they are not we generally indicate them using a
dotted line. Here are some lovely examples. We are not going to do anything
with them; this is just to get you familiar with the concept.
y y
x
x
y y
Can you see that all except one of the curves has two asymptotes?

QQ
By drawing two sketches on the same graph, say how many solutions (roots)
there are to each of these equations.
1 1 1 1 1
(a) x − 1 = x 2 (b) x = x 2 + 3 (c) x 2 = 10 − x (d) 100 x +1 = x
The answers are on the next page. If you are not sure where to start, there is a
hint in the footnote5 .
5 1 1
The solutions of the equation x − 1 = are where y = x − 1 and y = meet.
x x

(a) 1 solution. (b) 1 solution.
y y
x
x
−1
(c) 3 solutions
y
10
(d) 2 solutions.
y 1
The line y = 100 x +1
Will eventually meet the
curve again (at a point just
below the x-axis).
1
x

8.3 Transforming functions and curves

We’ll be looking at six6 different types of transformations. In each case we
will be making a change to the function and seeing what effect that has on
the curve of the function.
The best way to describe the transformations is to see what effect they have
on a simple function like
y f (x) = x 2 − 4.
6 This is a quadratic function; its
curve is a parabola, which has an
axis of symmetry—in this case the
line x = 0.
The y-intercept is −4.
-x The roots are where x 2 − 4 = 0,

−2 2 i.e. where x = 2 and x = −2.
−4
When we have done that we’ll look in detail at how it applies to a whole
range of different functions, including ones that have asymptotes.
The six types of transformation can be be neatly summarised using the

function notation. If you want a reminder about function notation see page
1-31.
Two translations
(i) y = f (x) + a Translation a distance +a parallel to the y-axis
(ii) y = f (x + a) Translation a distance −a parallel to the x-axis
Two stretches
(iii) y = a f (x) Stretch scale factor a parallel to the y-axis
1
(iv) y = f (ax) Stretch scale factor a parallel to the x-axis
Two reflections
(v) y = − f (x) Reflection in the x-axis
(vi) y = f (−x) Reflection in the y-axis
We shall see that transforming a curve is best seen as moving all the points
on the curve, rather than moving the curve as a whole. The “destination" for
each point we call its image.
We are going to look at each type of transformation in turn.
6
Actually two of them are just special cases of two of the others. But they have a
different name and it’s simplest to treat them separately.

8.3.1 Translation y = f (x) + a

This moves the curve up a distance a (parallel to the y-axis).
Starting with a function y = f (x) we add some constant value a to it. The
effect is to increase every y value by a.
We’ll look at two examples of this applied to our function f (x) = x 2 − 4.
y = f (x) → y = f (x) + 3.
y = f (x) y = f (x) + 3
6= x 2 − 4 6= x 2 − 1
4
-x -x
−2 2 −1 1
−1
3
−4
Notice how the y-intercept has move up 3. At the same time the roots have
moved closer to the origin, to exactly where we expect them to be: at the
solutions of x 2 − 1 = 0.

y = f (x) → y = f (x) − 5.
y = f (x) y = f (x) − 5
6= x 2 − 4 6= x 2 − 9
5
-x -x
−2 2 −3 3
5
−4
−9
We have marked the new y-intercepts in each case, and the new roots. Satisfy
yourself that they are correct.

8.3.2 Translation y = f (x + a)
This moves the curve a units to the left (parallel to the x-axis).
In the formula for f (x), we replace every occurrence of x with x + a.
How come the translation is in the opposite direction to the one we’d expect?
It seems to be inconsistent with translation in the y direction. When we’ve
finished describing how all the transformations work, we’ll exlain why there
is actually no inconsistency at all.
We’ll look at two examples of this transformation applied to our function

f (x) = x 2 − 4.
y = f (x) → y = f (x + 2).
y = f (x)
6= x 2 − 4
-x
−2 2
−4
C 2
y = f (x + 2)
6 = (x + 2)2 − 4
= x 2 + 4x
-x
−4 0
Moving the function y = f (x) to the left 2 units will move the two roots to −4
and 0.
This is consistent with the fact that the roots of y = f (x +2) = x(x +4) are x = 0
and x = −4.

y = f (x) → y = f (x − 1).
y = f (x)
6= x 2 − 4
-x
−2 2
−4
1 B
y = f (x − 1)
6= (x − 1)2 − 4
= x 2 − 2x − 3
-x
−1 3
Because the curve of the function y = f (x − 1) is obtained by moving the the

curve of y = f (x) 1 unit to the right, we know what its roots are, i.e. x = −1
and x = 3.
Satisfy yourself that these are the roots of the equation x 2 − 2x − 3 = 0.

8.3.3 Stretch y = a f (x)

Starting with a function y = f (x) we multiply every y value by a. The effect is
to stretch the curve by a scale factor a in the y direction—rather like pulling on
a rubber sheet, in the positive and negative directions, with the x-axis fixed.
We’ll look at two examples of stretches applied to our function
f (x) = x 2 − 4.
y = f (x) → y = 2 f (x).
y = f (x) y = 2 f (x)
2
6= x − 4 2
6= 2(x − 4)
• (3, 10)
• (3, 5)
4
-x ×2 -x
−2 2 −2 2
5
−4
−8
Think of the stretch as applying to each point individually. For a stretch in

the y direction, it means the x-coordinate of a point is unchanged. Its y-
coordinate is multiplied by a, so in this case it is doubled. To take a couple of
examples, the image of (3, 5) is (3, 10), and the y-intercept moves to (0, −8).
Any points on the x-axis are unchanged by the stretch (because 2 × 0 = 0), so
the roots are unchanged.

y = f (x) → y = 15 f (x).
As before think of transforming the curve point by point. The x-coordinate

of each case remains the same and its y-coordinate is multiplied by 51 .
y = f (x) y = 15 f (x)
2
6= x − 4 1
6= 5 (x 2 − 4)
• (3, 5)
4
• (3, 1)
-x × 15 -x
−2 2 −2 2
5 − 45
−4
The roots remain at 2 and −2. for example the image of (3, 5) is (3, 1) and the
y-intercept moves to (0, − 45 ).
We still call it a stretch even though it is more like a compression. We could

imagine drawing the graph on a block of rubber, and squashing the block in
the y direction, from both directions, in a way that leaves the x-axis fixed.

8.3.4 Stretch y = f (ax)

Starting with a function y = f (x) we replace x in the formula for f (x) with
1
ax. The effect is to stretch the curve in the x direction by a scale factor a .
We are using the term stretch in its technical sense: if a > 1 it means the
points actually get closer to the y-axis. The points get further away if a < 1.
y = f (x) → y = f (3x).
y = f (x)
2
6= x − 4
• (3, 5)
-x
−2 2
−4
C B
× 31
y = f (3 x)
2
6= (3x) − 4
= 9x 2 − 4
• (1, 5)
-x
− 23 2
3
−4
As before we think of the stretch as applying to each point individually: now

the y-coordinate of the point is unchanged and its x-coordinate is (in this
case) divided by 3. So, for example the image of (3, 5) is (1, 5).
The roots move to (− 32 , 0) and ( 23 , 0). The y-intercept is unchanged.
These conclusions are consistent with what the new function looks like:
y = 9x 2 − 4.

¡1 ¢
y = f (x) → y=f 2x .
y = f (x)
2
6= x − 4
• (3, 5)
-x
−2 2
−4
C B
×2
y = f 12 x
¡ ¢
6= x 2 − 4
¡ ¢
2
x2
= 4 −4
• (6, 5)
-x
−4 4
−4
As before we think of the stretch as applying to each point individually: the

y-coordinate of the point is unchanged and its x-coordinate is (in this case)
multiplied by 2. So, for example the image of (3, 5) is (6, 5).
The roots move to (−4, 0) and (4, 0). The y-intercept is unchanged.
These conclusions are consistent with what the new function is:
x2
y= − 4.
4

8.3.5 Reflection in the axes: y = − f (x) and y = f (−x)

Strictly speaking these transformations are just stretches with a = −1. Their
effect is very simple.
Let’s look at y = − f (x) first. Starting with a function y = f (x) we change the
sign of every y value. The effect is to reflect the curve in the x-axis. Points on
the x-axis, i.e. any roots, remain unchanged.
To show reflections graphically we’e going to use the function
y = f (x) = x(x − 4)
which has roots at x = 0 and x = 4.
y = f (x) → y = − f (x).
y = f (x) y = − f (x)
6= x(x − 4) 6= −x(x − 4)
(2, 4)
•
4
-x -x
0 4 0 4
5
•
(2, −4)
The roots are unchanged but the vertex moves from (2, −4) to its reflection in
the x-axis, i.e. (2, 4).

y = f (x) → y = f (−x).
y = f (x)
6= x(x − 4) = x 2 − 4x
-x
0 4
•
(2, −4)
4 5
y = f (− x)
6 = (−x)2 − 4(−x)
= x 2 + 4x
-x
−4 0
•
(−2, −4)
The vertex (2, −4) moves to (−2, −4).

The root at zero remains unchanged; the root at x = 4 moves to x = −4. We
can confirm this by looking7 at the function f (−x).
7
y = 0 ⇒ x(x + 4) = 0 ⇒ x = 0 or x = −4.

To sum up, we’ve met these transformations:
Translations
(i) y = f (x) + a Translation +a parallel to the y-axis
(ii) y = f (x + a) Translation −a parallel to the x-axis
Stretches
(iii) y = a f (x) Stretch scale factor a parallel to the y-axis A
1
(iv) y = f (ax) Stretch scale factor a parallel to the x-axis
!
A
A
Reflections A
(v) y = − f (x) Reflection in the x-axis

(vi) y = f (−x) Reflection in the y-axis
At GCSE level you will have met the enlargement transformation. This is
equivalent to two stretches with the same scale factor—one in the x
direction and one in the y direction.
Remember: in the case of translations and stretches, when we are changing

the entire function (e.g. multiplying it by something, or adding something to
it), the effect on the graph is in the direction we expect; when we are
replacing x with something else, it’s in the opposite direction .
QQ
In each case describe the effect of the transformation on the graph of the
function.
(a) y = x 2 − 5 → y = x2 − 3
(b) y = x 2 → y = (x + 6)2
(c) y = x 2 + 1 → y = 4x 2 + 4
x2
(d) y = x 2 → y = 25
(e) y = 3x 2 − 1 → y = 1 − 3x 2
(f) y = x 2 (1 + x) → y = x 2 (1 − x)
(a) x 2 − 3 = (x 2 − 5) + 2. Translation +2 parallel to the y-axis.

(b) x → x + 6. Translation −6 parallel to the x-axis.
(c) 4x 2 + 4 = 4(x 2 + 1). Stretch scale factor 4 parallel to the y-axis.
8
x2 x 2
(d) = ( ) i.e. x → x5 . Stretch scale factor 5 parallel to the x-axis.
25 5
(e) 1 − 3x = −(3x 2 − 1). Reflection in the x-axis.
2
(f ) (−x)2 (1 + (−x)) = x 2 (1 − x). Reflection in the y-axis.

Example 8
The diagram shows the curve of a function y = f (x).

y = f (x)
6
P (2, 3)
•
2
-x
−2 0 6
In separate diagrams sketch the curve

(a) y = f (x − 2)
(b) y = 3 f (x)
(c) y = f (−x)
(d) y = f (2x)
marking where the transformed curve crosses the coordinate axes, and the
image of P .
Solution
In each case decide what transformation is involved, and work out what that
transformation does to the points marked on the graph. (For the sake of
clarity we have shown the original curve y = f (x) together with the
transformed curve. You won’t need to do this.)
(a) Translation +2 units parallel to the x-axis.
y = f (x − 2)
6
(4, 3)
•
2
-x
−2 0 6 8

(b) Stretch scale factor 3 parallel to the y-axis.
y = 3 f (x)
6 (2, 9)
•
-x
−2 0 6
(c) Reflection in the y-axis.

y = f (−x)
6
(−2, 3)
•
2
-x
−6 −2 0 2 6
1
(d) Stretch scale factor 2
parallel to the x-axis.
y = f (2x)
6
(1, 3)
•
2
-x
−2 −1 0 3 6

Example 9
(a) Sketch the curve y = f (x), where
f (x) = x(x 2 − 4).
(b) By applying a suitable transformation, sketch the curve y = f (x − 2).

(c) Use the sketch in part (b) to solve the equation f (x − 2) = 0.
Solution
(a) The function is a cubic polynomial, whose roots are where
x(x + 2)(x − 2) = 0,
i.e. x = 0, x = −2 and x = 2. The x 3 coefficient is positive (because the

function is y = x 3 − 4x) so the curve looks like this:
y
6
- x
−2 0 2
(b) The transformation x → x − 2 changes y = f (x) into y = f (x − 2).

This corresponds to a translation of +2 units parallel to the x-axis, so the
new curve is:
y
6
- x
0 2 4
(c) The roots have moved to x = 0, x = 2 and x = 4; these must therefore be

the solutions of the equation f (x − 2) = 0.
Although the question hasn’t asked us to, we can verify the positions of these
roots algebraically:
f (x − 2) = (x − 2)[(x − 2)2 − 4] = (x − 2)[x 2 − 4x] = (x − 2)x(x − 4)
whose roots are x = 0, x = 2 and x = 4.

Example 10
(a) By applying a suitable transformations of the function y = x 2 , find the

equation of the quadratic curve whose vertex is at (2, 1) and which passes
through the point (0, 5).
(b) Find the equation of the quadratic function that crosses the y-axis at the
same point but whose vertex is at (1, 1).
Solution
First let’s translate the vertex from the origin to (2, 1).
y = f (x) = x 2 y
6 6
(2, 1)
•
- x - x
The curve has been translated +2 unit parallel to the x-axis and +1units
parallel to the y-axis, so the transformed function is
y = f (x − 2) + 1
= (x − 2)2 + 1
= x 2 − 4x + 5
We can see from this that the curve passes through (0, 5) as required.
(b) The curve looks like this:
y
6
• (1, 1)
- x
1
the resultof a stretch scale factor 2 parallel to the x-axis.
This corresponds to the transformation x → 2x, and the new function is
y = (2x)2 − 4(2x) + 5 = 4x 2 − 8x + 5.
A quick check on our algebra: f (1) = 4 − 8 + 5 = 1, confirming that the curves

passes through (1, 1).

Example 11
On the same graph, sketch and label the curves

1 1
(i) y = − 1 and (ii) y = −1
x x+3
Solution
1
(i) We can start with the curve y = , which has the axes as asymptotes.
x
1
If we translate this curve “down" 1 unit, we have y = −1 .
x
The “horizontal" asymptote, formerly y = 0, is now y = −1. We can think of
the asymptote as moving with the curve.
Putting y = 0, the curve crosses the x-axis, at x = 1.
y y
6 6
- x - x
1
−1
1 1
(ii) The only difference between y = −1 and y = −1 is that the
x x+3
x has been replaced by x + 3.
1 1
Using function notation, if f (x) = −1, f (x + 3) = −1.
x x+3
So the next step is to translate the curve we found in (i) 3 units to the left,
which moves the “vertical" asymptote from x = 0 to x = −3. .
You may want to see if you can complete this for yourself before turning the
page. That includes pinning down where the curve crosses the axes.

5 C 3
x
-
−3 −2
− 23
−1
1 2
Putting x = 0 into y = −1 gives y = − .
x+3 3
Putting y = 0 we have
1
0 = −1
x +3
0 = 1 − (x + 3)
x +3 = 1
x = −2
You will have noticed that positions of the key points in our diagram are not
particularly accurate. It’s a sketch: all we need is for the overall configuration
to be right (the relationship between the points), and for the coordinates of
key points to be marked. That includes the position of any asymptotes.
As an exercise repeat this question for the functions

1 1
(i) y = 2 − 1 and (ii) y = −1
x (x+3)2

Example 12
By applying suitable transformations to the function y = x 2 , sketch the

function y = x 2 − 4x + 1
Find where the curve crosses the coordinate axes.
Solution
Let’s call f (x) = x 2 and g (x) = x 2 − 4x + 1. To see what transformations are

needed we complete the square:
g (x) = (x − 2)2 − 3 = f (x − 2) − 3.
So we need to translate y = x 2 a distance 2 units parallel to the x-axis and −3
units parallel to the y-axis. The vertex moves to (2, −3).
y
6
C
C
C
PC
-x
0 Q R
•
(2, −3)
[You may recall that the vertex of the curve y = (x − a)2 + b is at (a, b)
confirming that the vertex is at (2, −3).]
It remains to find the positions of P , Q and R.
For the y-intercept we need g (0), where g (x) = x 2 − 4x + 1. P is (0, 1).
We can find Q and R from g (x) = 0 using the completed square:
(x − 2)2 = 3
p
x −2 = ± 3
p
x = 2± 3
p p
P is 2 − 3, Q is 2 + 3.
p p
Do these look about right? Well, 3 ≈ 1.732, so 2 − 3 is a small positive
number— which is consistent with the sketch.
We have line symmetry too: the axis of the parabola should be exactly
midway between the roots, which it is.

Example 13
p
The graph shows the function y = x.
y
6
- x
p
(i) Sketch the curve y = − x.
(ii) On a separate graph, sketch the two curves
p p
y = x + 1 and y = 2 x + 1.
Solution
Notice that the curve only exists to the right of the y-axis. This makes sense
because we can’t take the square root of a negative number.
The curve is also unusual in another way. Look at the way it behaves at the
origin: the gradient there is infinite. None of this affects how we treat the
function when transforming it.
p
(i) This is the reflection of y = x in the x-axis.
y
6
- x
p p
(ii) If f (x) = x, y = x + 1 = f (x + 1), i.e. f (x) translated −1 unit parallel to
the x-axis. It crosses axes at (−1, 0) and (0, 1).
p
y = 2 x + 1 is this curve strteched by a scale factor 2 parallel to the y-axis.
y
6
- x
−1 0

8.4 Periodic functions

For some functions their values repeat regulary, so the shape of the curve
does the same. The commonest example is to be found in wave functions.
These desribe any kind of wave, such as sound waves, light waves, waves in
water etc.
In most of these cases the underlying function can be described using the
function y = sin x.
The distance before the pattern starts repeating itself we call the period.
The maximum height above or below the middle position is the amplitude9 .
y = sin x
6
per i od -
1
6
ampl i t ud e
BB ? - x
−180◦ 0 ◦
180 ◦
360 ◦
540 ◦ 6 720◦
ampl i t ud e
−1 ?
There is a wide variety of periodic functions.

What is the period function shown in the graph below?
(The answer is in the footnote10 .)
y
6
-x
−4 −2 2 4 6 8
−2
First we will look at the trig functions, and the effect of transforming them.
After that we will return to consider other periodic functions.
9
Not all periodic functions have an amplitude. An example would be the tan function.
10
The period is 6.

Example 14
Sketch the curves

(i) y = 1 + sin x (ii) y = 3 sin x (iii) y = sin 2x (iv) y = sin(x − 30◦ )
Solution
The sin x function looks like this:

y = f (x) = sin x
6
J
J
- x
−180◦ −90 ◦
0 ◦
90 ◦
180 ◦
270 ◦
360◦
−1
It takes values between −1 and +1.

It has a root at every multiple of 180◦ .
Mathematically we could write this as x = 180n ◦ , where n is an integer.
(i) 1 + sin x = f (x) + 1, so the curve looks like this:

y = 1 + sin x
6
J
J 1
- x
−180◦ −90◦ 0 90◦ 180◦ 270◦ 360◦
This function takes values between 0 and 2.

It has roots 360◦ apart and they are all double roots (because the curve
touches the x-axis at these points).
Can you write down a mathematical formula for the position of all these
roots? One possible answer is in the footnote11 .
The period is still 360◦
The amplitude is still 1 unit.
11
x = 270◦ + 360n ◦ , where n is an integer (which can be positive, negative or zero).

(ii) 3 sin x = 3 f (x) so the curve looks like this:
y = 3 sin x = 3 f (x)
6
- x
−180◦ −90◦ 0◦ 90◦ 180◦ 270◦ 360◦
−1
−2
−3
The y-coordinate of each point on the curve has been trebled, so the function
takes values between −3 and +3, leaving the roots unchanged.
We can imagine the graph drawn on sheet of rubber, which is stretched in
both directions (parallel to the y-axis), while leaving the x-axis in the same
position.
The period remains at 360◦ .
The amplitude is 3 units.

1
(iii) sin 2x = f (2x), so we need to stretch f (x) by a scale factor 2
parallel to
the x-axis.
This is much less intuitive than parts (i) and (ii). Rather than thinking of
transforming the curve as a whole, it is safer to
transform individual points.
To do this you need to start with a careful sketch of y = sin x, with some of
the key points labelled, like the ones marked below. (If the sketch is too
rough you won’t be able to see what’s going on.)
y = f (x) = sin x
6
1 •
J
J• • •

•
- x
−180◦ −90 ◦
0◦ 90 ◦
180◦ 270 ◦
360◦
−1 •
Under the transformation, the distance of each dot from the y-axis is halved.
y = f (x) = sin x
6
1 •
J
J • • • •
- x
−180◦ −90◦ 0◦ 90◦ 180◦ 270 ◦
360◦
•−1 •
So the curve looks like this
y = f (2x) = sin 2x
6
1 •
B
B• • • • - x
◦
−180 −90◦ 0◦ 90◦ 180◦ 270◦ 360◦
−1 •
Copy this sketch and complete the curve, so it goes from −180◦ to +360◦ .
The formula for the position of all the roots is very simple: 90n ◦ where n is
an integer.
The effect of the transformation has been to halve the period from 360◦ to
180◦ . The amplitude remains unchanged at 1.

(iv) y = sin(x − 30◦ )
This is the function y = sin x translated +30◦ parallel to the x-axis.
y = f (x − 30◦ ) = sin(x − 30◦ )

6
J
J
- x
◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦
−180 −90 0 30 90 180 270 360
−1
It takes values between −1 and +1.

It has a root 30◦ to the right of every multiple of 180◦ .
Mathematically we could write this as x = (180n + 30)◦ , where n is an integer.
The period is still 360◦ and the amplitude is still 1.

Example 15
Sketch the curve y = cos( 21 x + 45◦ ).
Solution
In principle this is no different from the examples we have met; there is

clearly some enlarging going on and some translating. The important thing
is to get the order right, because—unlike in previous examples—it makes a
difference in this case.
If you rely on your intuition, you could come unstuck; the only safe method
is to use transforming functions. There are two ways we can go, The first
method, as we shall see, if definitely the easier of the two.
Method 1
Start with the function y = f (x) = cos x. Apply the transformation
x → x + 45◦ , giving the function g (x) = f (x + 45◦ ) = cos(x + 45◦ ).
1
Now apply the transformation x → 2x to g (x), giving the final function
h(x) = g ( 21 x) = cos( 12 x + 45◦ ).
In other words a translation of −45◦ parallel to the x-axis, followed by a
stretch scale factor 2 in the x-direction.
Under the translation (90◦ , 0) moves to (45◦ , 0), and (0, 1) moves to (−45◦ , 1).
Under the stretch (45◦ , 0) goes back to (90◦ , 0), (225◦ , 0) moves to (450◦ , 0) and
(−45◦ , 1) moves to (−90◦ , 1).
y
6
y = cos(x + 45◦ ) y = cos x
• 1• @
R
@
• • - x
◦ ◦ ◦ ◦
−270 −180 −90 0 90◦ 180 ◦
270 ◦
360 ◦
450 ◦
−1
y
y = cos( 21 x + 45◦ ) 6
y = cos(x + 45◦ )
@ • • 1
R
@
• • • •
- x

◦
−270 −180◦ −90◦ 0◦ 90◦ 180◦ 270◦ 360◦ 450◦

−1

Method 2
Suppose we choose to do the stretch first, i.e.x → 21 x, and we have
g (x) = f ( 12 x) = cos( 21 x).
So far so good—or so it seems.
Now we apply the transformation x → x + 45◦ , giving
h(x) = g (x + 45◦ ) = cos( 12 (x + 45◦ )) = cos( 12 x + 22.5◦ ).
To correct this the translation needs to be x → x + 90◦ —hardly the most
obvious transformation. Then
h(x) = g (x + 90◦ ) = cos( 12 (x + 90◦ )) = cos( 12 x + 45◦ ).
With the sin and cos functions stretch before translating.
Construct the curve for yourself, using Method 2. The solution is on the next
page.

y
6
y = cos( 12 x) y = cos x
1
• • • • • •
- x
−270◦
◦
−180 −90◦ 0◦ 90◦ 180◦ 270◦ 360◦ B ◦
450 540◦

B

B
−1
Under the stretch (−90◦ , 0) moves to (−180◦ , 0), (90◦ , 0) moves to (180◦ , 0) and
(270◦ , 0) moves to (540◦ , 0).
y
y = cos( 21 x + 45◦ ) 6
@ • 1 •
R
@
• • •
•
- x

◦ −180
−270
◦ −90◦ 0◦ 90◦ 180◦ 270◦ 360◦ 450◦ 540◦

−1
Under the translation (0, 1) moves to (−45◦ , 1), (180◦ , 0) moves to (90◦ , 0) and
(540◦ , 0) moves to (450◦ , 0).

8.5 Transforming functions: an elegant approach

There is an apparent inconsistency in the way we treat transformations. In
the y-direction the way curves change seems intutively sensible. In the
x-direction not so: the direction of a translation is the opposite to what we’d
expect, and the scaling factor of a stretch in that direction is the reciprocal of
what we’d expect. What’s going on?
If we describe transformation in both directions by means of a change of
variable, the problem disappears.
Let’s look at the example of a translation 2 in the positive direction. We’ll

put the two transformations side by side, so you can see the processes are
identical.
x-direction y-direction
Starting with y = f (x) Starting with y = f (x)

x → x −2 y → y −2
giving giving
y = f (x − 2) y − 2 = f (x)
or y = f (x) + 2
Now a stretch scale factor 3.
x-direction y-direction
Starting with y = f (x) Starting with y = f (x)
x → 13 x y → 31 y
giving giving
1
y = f 13 x
¡ ¢
3
y = f (x)
or y = 3 f (x)
This approach will always work, although you may prefer to stick with the one
you are used to.
In the next, and final, example it is the only possible approach .

Example 16
The circle C 0 is
2x 2 + 2y 2 + 7y − 1 = 0.
C 0 is translated a distance +3 parallel to the x-axis and −1 parallel to the
y-axis, to form a new circle C 1 .
(a) Find the equation of C 1 .
(b) Show that C 0 crosses the x-axis but C 1 does not.
Solution
(a) We could write the equation in the form (x − a)2 + (y − b)2 = r 2 , find the
position of the new centre, move that centre and hence find the new
equation. The coefficients in this case would make that a bit heavy going.
The much simpler approach is to apply the two transformations
x → x − 3 and y → y +1
to the equation of C 0 .
The equation of C 1 is
2(x − 3)2 + 2(y + 1)2 + 7(y + 1) − 1 = 0

2(x 2 − 6x + 9) + 2(y 2 + 2y + 1) + 7y + 7 = 0
2x 2 + 2y 2 − 12x + 11y + 27 = 0.
(b) If we knew where the centre of C 0 was, and its radius, we could easily find
the answer.
There is a much simpler way. Suppose we set about finding where each of
them crosses the axis, by putting y = 0 into their equations.
1
C0 : 2x 2 − 1 = 0 ⇒ x = ±p .
2
So C 0 does, and we have found where. Turning to C 1 ,
C1 : 2x 2 − 12x + 27 = 0.
For this circle to cross the x-axis this quadratic equation needs to have real
roots, in other words its discriminant must be positive or zero.
∆ = b 2 − 4ac = 122 − (4)(2)(27) = −72.
C 1 doesn’t cross the x-axis.

Exercises 32
1. In each case describe precisely the transformation that changes the curve
y = f (x) into y = g (x) (drawn in blue).
(a) y (b) y
6 6
3 3
-x -x
−5 −2 2 5
−3
(c) y (d ) y
6 6
-x -x
5 10 5 10
(e) y (f ) y
6 6
8
4 4
-x -x
−4
2. For each of the transformations in Question 1, express g (x) in terms of f (x).
3. The graph shows the curve y = f (x). Sketch the curves

y (i) y = 14 f (x),
(ii) y = f (x − 2),
6
P (4, 8) (iv) y = f (2x).

•
On each transformation mark

- x where the curve crosses the
−2 6 coordinate axes, and the
image of P .

The curve in the sketch is

4.
y = f (x).
It is transformed into the
y
curve
6
y = f (px + q).
• B (10, 3) Write down two possible
pairs of values for p and
-x
q, such that, after the
• transformation,the image of
A (4, −2) A is at (1, −2).
In each case state the
coordinates of of the image
of B .
5. The graph shows the function y = f (x).

Write down the y
coordinates of where P , 6
Q and R move to when P (−4, 3)
•
y = f (x) is transformed
into
-x
(i) y = f (x) + 2 Q
•
(ii) y = 3 f (x) R (2, −2)
(iii) y = f 12 x
¡ ¢
(iv) y = f (−x)
(v) y = f (x − 4)
6.
2
y=x The diagram shows the curve
6 2
y =
x
Copy the graph and on the
same diagram sketch the curve
2
-x y =
x +3
Write down the equations of
the asymptotes.
Find the positions of any
points where the curve crosses
the coordinate axes.
7.
y
The graph shows the function
6 y = f (x). The function has roots
at x = −2 and x = 8.
A(3, 4)
•
It has a maximum point at (3, 4).
In separate diagrams sketch the
-x curves
−2 8 (i) y = − f (x)
(ii) y = f (2x)
On each diagram mark the roots of f (x) = 0 and the image of A.

8. The graph show a function y = f (x), which takes the value 3 for all values of x
except those in the interval 2 < x < 6. The scale on each axis is the same.
y On separate graphs sketch
6 the functions
(a) y = 13 f (x)
(b) y = f (x + 4)
(c) y = f 12 x .
¡ ¢
(You need not use the

same scale for each.)
- x
7+3x
9. L is the line y = . Show that the point (1, 2) lies on L.
5
Find the equation of the line parallel to L that passes through (−4, 2).
10. Sketch these curves

1 1
(a) y = (b) (b) y = − 1.
(x+2)2 (x+2)2
1
11. Sketch these curve y = + 2.
(x−3)2
Find where the curve intersects the line y = 6.
x 1 x
12. Show that = + 1. Hence sketch the curve y = .
x−1 x−1 x−1
1 4
13. (a) Find where the curve y = crosses the curve y = 2 .
x x
1 4
(b) For what values of k do the curves y = and y = 2 + k meet.
x x
14. For the function f (x) defined in Question 7, solve the equations
(a) f (x) = 6 and (b) f (x) = 12 x.
How many solutions are there to the equation f (x) = 11 − x?
1 1
15. Sketch the curves (i) y = x+1 (ii) y = 1−x .
There is hint to part (ii) in the footnote12 .
16.
4
y = x2 The diagram shows the
4
6 curve y = x 2 .
On separate diagrams,
sketch the curves
4
(a) y = x 2 −1
4
(b) y = x 2 −4x+4 .
-x
[There is a hint to part (b) in the footnote13 .]

In each case write down the equations of the asymptotes and find
the positions of any points where the curve crosses the coordinate
axes.
12 1
Start with y = and transform this.
x−1
13
Factorise the denominator.

17. The centre of the circle 3x 2 + 3y 2 + 5x − y + 2 = 0 is at (a, b).

Find the equation of the circle with the same radius, whose centre is at
(i) (−a, b), and (ii) (a, b − 2).
p
18. The graph shows the function y = x − 2 and the line y = 3 (not drawn to scale).
Write down the x-coordinate of the point P .
y
6
Q
3
- x
0 P
The curve intersects the line y = 3 at Q, find the distance PQ.

p
Sketch the curve y = 2x − 2.
p
Write down where the curve y = 2x − 2 crosses the line y = 3.
19. Find the equation of the ellipse that passes through the points
(6, 0), (0, 3), (−6, 0) and (0, −3).
14
Stretch a circle.

Solutions 32
1. (a) translation −6 // y−axis (b) reflection in y − axi s or translation +3 // x-axis
(c) stretch s.f. 2 // x-axis (d) translation +5 // x-axis
(e) reflection in x-axis or stretch s.f. −1 // y-axis
(f) stretch s.f. 2 // y-axis
(c) g (x) = f 21 x
¡ ¢
2. (a) g (x) = f (x) − 6 (b) g (x) = f (−x) or f (x − 3)
(d) g (x) = f (x − 5) (e) g (x) = − f (x) (f) g (x) = 2 f (x)
3. (i) (ii) (iii)

y y y
6 6 6
P (4, 8) P (2, 8)
• •
P (4, 2)
•
-x -x -x
−2 6 0 8 −1 4
4. p = 1, q = 3. Translation −3 // x-axis. y = f (x + 3). B (7, 3),

1
or p = 2, q = 0. Stretch s.f. 2 // x-axis. y = f (2x). B (5, 3).
5. P Q R
(i) (−4, 5) (0, 2) (2, 0)
(ii) (−4, 9, ) (0, 0) (2, −6)
(iii) (−8, 3) (0, 0) (4, −2)
(iv) (4, 3) (0, 0) (−2, −2)
(v) (0, 3) (4, 0) (6, −2)
6.
y
6
−3 - x
2
The curve y = x+3 is shown in blue.
The curve crosses the y-axis at (0, 32 ).
Asymptotes: y = 0 and x = −3.

7. (i) y = − f (x) (ii) y = f (2x)

6 6
A( 32 , 4)
•
-x -x
−2 8 −1 4
•
A(3, −4)
8. (a) (b)
y y
6 6
-x -x
(c) y
6
-x
22+3x
9. y= 5
.
The line through (−4, 2) is L translated −5 // x-axis. i.e. f (x) → f (x + 5).
10. (a) (b) y
x
−2
x −1

11.
1
y +2 = 6
(x − 3)2
1
= 4
(x − 3)2
1
(x − 3)2 =
4
2 1
x x −3 = ±
3 2
5 7
x =
or
2 2
Notice that it is easier to rearrange the equation this way than to multiply
through by (x − 3)2 .
1 1+x−1 x x (x−1)+1 1 1
12. +1 = = . Or = = +
x−1 x−1 x−1 x−1 x−1 x−1 x−1
y
1
x
1
1
13. (a) x = 4 [multiply both sides by x 2 ] (b) |k| ≤
16
1 4
The equation = + k can be written kx 2 − x + 4 = 0.
x x2
This has roots provided the discriminant is non-negative, i.e. 1 − 16k ≥ 0
14. (a) x = 3 or 5 (b) x = 6.

3 solutions, by inspection. The line has gradient −1 and passes through
(11, 0). [The solutions are 3 12 , 5 and 8.]
1
15. Let f (x) = x .
1
(i) Curve is y = f (x + 1). (ii) Start with g (x) = x−1 = f (x − 1)
1 −1
then y = 1−x = x−1 = −g (x)
y y
6 6
- x - x
−1 1

16.
4
y = x2 −1 Asymptotes: x = 0 and y = −1.
6 x = ±2.
4
[Crosses x-axis where x 2 − 1= 0]
-x
−1
4 4
y = (x−2)2 Let f (x) = x 2 .
6 Since x 2 − 4x + 4 = (x − 2)2
the curve is
y = g (x) = f (x − 2).
Asymptotes: y = 0, x = 2.
y = g (0) = 1.
-x
2
17. There is no need to find a and b. We can simply transform the circle.
(i) reflection (in y-axis) x → −x, giving 3x 2 + 3y 2 − 5x − y + 2 = 0 .
(i) translation y → y − 2, giving 3x 2 + 3y 2 + 5x − 13y + 4 = 0 .
18. P is (2, 0). [The smallest value x can take for the square root to be possible.]
p p p
PQ = 90 = 3 10. [At Q, 3 = x − 2, or 9 = x − 2. So Q is (11, 3).]
p p
Let f (x) = x − 2, then 2x − 2 = f (2x). Stretch y = f (x) s.f. 21 // x-axis.
The image of Q is ( 11
2
, 3).
y
6
3
Q(11, 3)
- x
0 1 2
19. Start with the circle centre the origin radius 3, i.e. x 2 + y 2 = 9.
If we stretch this s.f. 2 // x-axis we get the required ellipse: x → 12 x.
x2
4
+ y 2 = 9 which simplifies to x 2 + 4y 2 = 36.
We can check this is correct by finding where it crosses the axes.


• Sketching polynomials. What does “sketching" mean? Roots and the
zero product principle.
Multiple roots. Higher order polynomials. Which way up is the curve?
1 1
• The reciprocal functions. The curves y = x and y = x 2 . Asymptotes.
• Transforming functions.
Translations y = f (x) + a and y = f (x + a)
Stretches y = a f (x) and y = f (ax)
Reflections (in the coordinate axes) y = − f (x) and y = f (−x)
The importance of focussing on the transformation of individual
points rather than broad changes.
The effect that transforming has on roots, intersections with the axes,
minimum and minimum points and asymptotes.
• Periodic functions. Definition. Amplitude and period. Transforming

the sine, cosine and tan curves.
• An elegant way of thinking about transformations which treats the x

and y directions the same.

Chapter 9
BINOMIAL EXPANSION
Overview
9.1 Binomial coefficients
Expanding expressions like (a + b)n for values of n > 2 without multiplying
out lots brackets. Pascal’s triangle.
9.2 Expanding (1 + x)n

Formula and simple problems. For example finding the coefft of x 3 in the
expansion of (1 − 2x)5 .
9.3 Binomial notation

¡n ¢
Shorthand notation: and n C r . Definition using the factorial function.
r
9.4 Expanding (y + x)n
Formula and problems. For example finding the first 4 terms in the expansion
of (4+ x)(2−3x)6 , or the x 3 coefficient in the expansion of (1− x +3x 2 )4 . Or to
1 6
find the constant term in the expansion of x 2 + 2x
¡ ¢
9.5 Approximation using the binomial expansion

For example estimating (1.04)5 or (0.996)7 correct to 4 decimal places without
using a calculator.
9-1
CHAPTER 9. BINOMIAL EXPANSION
9.1 Binomial coefficients

We know how to multiply brackets together. So if we wanted to know the
coefficient of x 2 in (2 + 3x)7 , for example, we could find it. But it would be
rather tedious. We would first have to find (2 + 3x)2 , multiply this by (2 + 3x),
multiply the answer to that by another (2 + 3x) etc until we reached (2 + 3x)7 .
The Binomial explansion provides a short-cut, enabling us to write down all
the terms straight away, with very little calculation. It has other uses too but
first we’ll look at finding the terms.
First let’s look at some expansions to see if there is a pattern.
(a + b)1 = 1
(a + b)2 = a 2 + 2ab + b 2
(a + b)3 = a 3 + 3a 2 b + 3ab 2 + b 3
(a + b)4 = a 4 + 4a 3 b + 6a 2 b 2 + 4ab 3 + b 4
(a + b)5 = a 5 + 5a 4 b + 10a 3 b 2 + 10a 2 b 3 + 5ab 4 + b 5
We’ve highlighted the coefficients in red (apart from the implied 1’s at each
end). These are binomial coefficients.
As you go from one term to the next, the power of a decreases by 1 and the
power of b increases by 1. (This applies everywhere because a 0 = b 0 = 1.)
The way the coefficients are formed is less simple. There is a way of deriving
them from the row above. For example in the last row, each of the 10’s is the
sum of the 4 and the 6 above it; similarly with the 5’s (because the coefficients
of a 4 and b 4 are 1)1 .
It follows that the expansion of (a + b)6 is
(a + b)6 = a 6 + 6a 5 b + 15a 4 b 2 + 20a 3 b 3 + 15a 2 b 4 + 6ab 5 + b 6 ]
But to get to an expansion of (a + b)n we would rather not have to write out
the previous n − 1 expansions.
1
The binomial coefficients form what is referred to as Pascal’s triangle after the
extraordinarily gifted French Mathematician, Physicist and philosopher Blaise Pascal who
lived in the middle of the 17th century and died at the age of 39.

9.2 Expansion of (1 + x)n

To see how to produce the coefficients directly we’re going to take the case
n = 5, with a = 1 and b = x. When we see what the pattern is it will be short
step to generalising it to any a, b and n. Please note that, throughout this
chapter, we are using a dot as shorthand for a multiplication symbol.
5 5.4 2 5.4.3 3 5.4.3.2 4 5.4.3.2.1 5

(1 + x)5 = 1 + x+ x + x + x + x
1 1.2 1.2.3 1.2.3.4 1.2.3.4.5
Before we simplify this look at the pattern2 . The powers of x increase from x 0
to x 5 , i.e. 6 terms altogether. The coefficients (ignoring the first term for the
moment) are fractions in which the numerator starts with 5 and is
progressively multiplied a number that is one less (5, 5.4, 5.4.3, etc).
Similarly the denominator starts with 1 and is progressivelty multiplied by a
number that is one more. The process ends when numerator and
denominator are the same. Simplifying, we have
2 2
5 5 5.4 2 5.4 .3 3 5.4 .3 .2 4 5 .4 .3 .2 .1 5
(1 + x) = 1+ x + x + x + x + x
1 1.2 1.2 .3 1.2 .3 .4 1.2 .3 .4 .5
= 1 + 5 x + 10 x 2 + 10 x 3 + 5 x 4 + x 5
You’ll see this is consistent with our expansion of (a + b)5 , with a = 1 and
b = x.
You will see there are a number of symmetries in the coefficients.

1. Starting at each end and working towards the middle they go in pairs.
2. The number of integers in the numerator and the denominator is the same
as the power of x (with the exception of the first term).
2
Remember when we have integers we can write 5.4 instead of 5 × 4.

Following this pattern will enable you be able to expand any bracket raised to
any power n 3 . You will even be able to have more than two terms inside the
bracket. But first a couple more examples.
3 3.2 2 3.2.1

(1 + x)3 = 1 + x+ x + x 3
1 1.2 1.2.3

= 1 + 3x + 3x 2 + x 3
6 6.5 2 6.5.4 3 6.5. 4.3 6.

5.4.3.2 6.5.4.3.2.1
(
(
6

4 5 (
((( 6
(1 + x) = 1+ x + x + x + x + x + (x
1 1.2 1.2.3 1.2.
3.4
1.
2.3.4.5

1.2.3.4.5.6
( ((((
= 1 + 6x + 15x 2 + 20x 3 + 15x 4 + 6x 5 + x 6
Check the rest of the cancelling in the last example to see that you agree.
Secondly, we know that there are n + 1 terms in the expansion e.g. there are
7 terms in the expansion of (1 + x)6 . If n is even there are an odd number of
terms so we get a “middle” term, whose coefficient is different from all the
others (in the case n = 6 it is 20). If n is odd there is no “middle” term; every
coefficient has a “twin” (as in the case n = 3 or n = 3).
Look at the expansions of (a + b)n on the previous page and see this is always
the case.
We point these features out not just because it is interesting, but because we
can use them as quick and simple checks after we have done an expansion.
Do I have the right number of terms (i.e. n + 1)? Do they come in matching
pairs? If not, I must have slipped up somewhere.
3
Here we will confine ourselves to the case where n is a positive integer.

Example 1
Using the expansion of (1 + x)5 , find the expand of (1 − 2x)5 .

By putting x = 14 in the expansion show that
10 40 80 80 32 1
1− 2 + 4 − 6 + 8 − 10 = 5
2 2 2 2 2 2
Solution
Putting X = −2x in the expansion of (1 + X )5
(1 + X )5 = 1 + 5 X + 10 X 2 + 10 X 3 + 5 X 4 + X 5
so (1 + (−2x))5 = 1 + 5 (−2x) + 10 (−2x)2 + 10 (−2x)3 + 5 (−2x)4 + (−2x)5
= 1 − 10x + 10(4x 2 ) + 10(−8x 3 ) + 5(16x 4 ) + (−32x 5 )
= 1 − 10x + 40x 2 − 80x 3 + 80x 4 − 32x 5
1
Substituting x =
4
10 40 80 80 32
= 1− + 2 − 3 + 4 − 5
4 4 4 4 4
1 1
Writing 4 as 22 gives the required series.
1
¡ 1 ¢5
But we know that this is the expansion of (1 − 2x)5 with x = 4 , i.e. 2 .
Example 2
¢8
In the expansion of 1 + x2
¡
(a) Show that the coefficients of x 2 and x 3 are equal.

(b) Find the coefficient of x 4 .
(c) Without expanding beyond the x 4 term, find the coefficient of x 7 .
Solution
We only need expand as far as the x 4 term.

³ x ´8 8 ³ x ´ 8.7 ³ x ´2 8.7.6 ³ x ´3 8.7.6.5 ³ x ´4
1+ = 1+ + + + +...
2 1 2 1.2 2 1.2.3 2 1.2.3.4 2
8 .7 x 2 8 .7.6 x 3 8 .7.6 .5 x 4
= 1 + 4x + . + . + . +...
1.2 4 1.2 .3 8 1.2 .3 .4
16

2
35
= 1 + 4x + 7x 2 + 7x 3 + x4 + . . .
8
(a) The coefficients of x 2 and x 3 are both equal to 7.

35
(b) The coefficient of x 4 is 8 .
(c) The coefficient of x 7 is the same as the coefficient of x, i.e. 4.
23
¡ x ¢7 1 1
So the x 7 term is 8 2 , and the coefficient is 27 = 25 = 16 .

We can generalise the pattern of coefficients for any power n.

Obviously it is not possible to write down every term, because n is not a
specific number, but we can find as many terms as we want. (We can find
the last term too, but it will be in terms of n.)
To help us see the pattern we’ll start with a specific case (we’ve used n = 5)
5 5.4 2 5.4.3 3 5.4.3.2 4 5.4.3.2.1 5

(1 + x)5 = 1 + x+ x + x + x + x
1 1.2 1.2.3 1.2.3.4 1.2.3.4.5
So for any n
n n(n − 1) 2 n(n − 1)(n − 1) 3 n(n − 1)(n − 1)(n − 3) 4

(1 + x)n = 1 + x+ x + x + x +...
1 1.2 1.2.3 1.2.3.4
It is good see that the general formula looks right. At this stage it is sufficient;
to answer exam questions you don’t need to know why this always gives the
correct coefficients.
For those of you who want to understand why, or are mystified as to why the
first term doesn’t appear to follow the same pattern, there is an explanation
on the next page. If you are content to take it all on trust you can safely skip
that page.

9.3 Binomial notation

There are two shorthand ways of writing the coefficients, illustrated below
5 5.4 2 5.4.3 3 5.4.3.2 4 5.4.3.2.1 5

(1 + x)5 = 1 + x+ x + x + x + x
1 1.2 1.2.3 1.2.3.4 1.2.3.4.5
Ã ! Ã ! Ã ! Ã ! Ã ! Ã !
5 5 5 2 5 3 5 4 5 5
(1 + x)5 = + x+ x + x + x + x
0 1 2 3 4 5
(1 + x)5 = 5
C 0 +5 C 1 x + 5C 2 x 2 +5 C 3 x 3 +5 C 4 x 4 +5 C 5 x 5
Notice that in this notation every coefficient follows the same pattern,
including the first term, which in the notation we have been using seems not
to.
We shall tend not to use to use the first two notations, although it is the third
one that a calculator will use if it has a button for calculating binomial
coefficients.
Before returning to look at problems involving the binomial expansion we

are going to take a closer look at how the coefficients are constructed. In
doing so we will see where the symmetry comes from.
First we define the factorial function, starting with an example and
generalising.
7! = 1.2.3.4.5.6.7
n! = 1.2.3.4. . . . . . . (n − 1)n
We can write the binomial coefficients terms in factorial notation. For

example Ã !
11 11.10.9.8 11.10.9.8) 7.6.5.4.3.2.1 11!
= = . =
4 1.2.3.4 1.2.3.4 7.6.5.4.3.2.1 4!7!
Similarly Ã !
5 5.4.3 5.4.3 2.1 5!
= = . =
3 1.2.3 1.2.3 2.1 3!2!
In other words Ã !
n n n!
= Cr =
r r !(n − r )!
We have cheekily glossed over the first ¡5¢ and ¡5¢last terms of the binomial
expansion in this notation. How come = = 1, for example?
0 5
¡5¢ ¡5¢ 5!
According to the formula = = .
0 5 5! 0!
But for this to equal 1 we would need 0! = 1. Strange though it may seem it is:
0! = 1. The reason is that the definition for the factorial function that we gave
you only applies to integers ≥ 1.
If you are interested to know what the general definition is, and therefore
how 0! = 1, see the next section. It is an optional (blue) section. There is
nothing in it that you need to know.

¡n ¢
The formula for has an interesting feature. Look at the formula for
¡ n ¢ r
:
n−r Ã ! Ã !
n n! n! n
= = =
n −r (n − r )![n − (n − r )]! (n − r )!r ! r
So if we want to calculate the value of something like 12

¡ ¢
9 we can use this
relationship to simplify it
Ã ! Ã !
12 12 12.11.10
= = = 220
9 3 1.2.3
But more importantly is explains the symmetry in the coefficients of the

binomial expansion.
Ã ! Ã ! Ã ! Ã ! Ã ! Ã !
5 5 5 5 5 5 5
(1 + x)5 = + x+ x2 + x3 + x4 + x
0 1 2 3 4 5
Ã ! Ã ! Ã ! Ã ! Ã ! Ã !
5 5 5 2 5 3 5 4 5 5
= + x+ x + x + x + x
0 1 2 2 1 0
In this case n is odd, so there are an even number of terms. If n is even there
are an odd number of terms and there is a “middle" term. For example
Ã ! Ã ! Ã ! Ã ! Ã ! Ã ! Ã !
6 6 6 6 6 6 6 6
(1 + x)6 = + x+ x2 + x3 + x4 + x5 + x
0 1 2 3 2 1 0
QQ
Evaluate each of these correct to 3 decimal places

¡5¢ ¡7¢ ¡ 1 ¢4 ¡ 2 ¢3 ¡20¢
(i) 2 (0.6)2 (0.4)3 (ii) 4 3 3
(iii) 17 (0.5)20
4
(i) 0.230 (ii) 0.128 (iii) 0.001

Explaining the binomial coefficients

Here we will explain where the binomial coefficients come from. In the
process of answering this question it will also become clear how the first
term (the 1 at the beginning of the expansion) follows the same pattern.
To follow what we are going to cover you will need to have come across
permutations and combinations. If you haven’t, no problem, leave this for
now and come back to it when you have.
Using (1 + x)5 as an example, we can see it as 5 brackets multiplied together:
(1 + x)(1 + x)(1 + x)(1 + x)(1 + x)
How can we find the terms without having to mulitply it all out?
This is the way to think about it: when multiplying two brackets, every term
in the first bracket must be multilplied by every term in the second. In other
words we have to pairs up the terms, taking one from each bracket. The same
principle applies if there are five brackets: we must pick one term from each
bracket and multiply them. The simplest case is the constant term, which we
get by picking the 1 from each bracket.
A more interestingly example: if we pick the terms in red, we get an x 3 term:
(1 + x)(1 + x)(1 + x)(1 + x)(1 + x)
The term is x × 1 × ×1 × x × x = x 3 . But this is not the only way we can get x 3 .
How many ways are there? It’s the number of ways we can select 3 of the x’s
out of a possible 5, where the order of selection doesn’t matter.
From your work on permutations and combinations you will know this is
¡5¢
5
C 3 , or 5C 3 or 3 (We will use the last notation.)
5.4.3
The answer is 1.2.3 = 10, which is the coefficient of x 3 we expected to see.
¡5¢
In general the coefficient of x r must be r and the whole expansion fomed
by adding all the possible powers of x:
Ã ! Ã ! Ã ! Ã ! Ã ! Ã !
5 0 5 1 5 2 5 3 5 4 5 5
x + x + x + x + x + x
0 1 2 3 4 5
We can clear up here the fact that in the expansion formula we described
earlier, the constant terms at the front doesn’t appear to follow the same
pattern.
¡n ¢
It does, if you remember that 0 = 1, for any n.
¡n ¢
Why does 0 = 1? It is the number of what of selecing no x’s. And there is
only one way to do that, namely by only selecting 1’s.
¡5¢ 5.4
¡5¢ 5
But wait a minute, I hear you say, if 2 = 1.2 and 1 = 1 what on earth does
¡5¢ ¡n ¢
the fraction 0 look like? The answer lies in knowing where the r fraction
comes from.
¡n ¢ n!
¡5¢ 5! 5.4.3.2.1
5.4.3
r
= (n−r )!r ! , so for example 3 = 2!3! =
2.1.3.2.1
= 3.2.1 .
¡5¢ ¡5¢ 5!
That means 0 = 5 = 5!0! = 1, because 0! = 1.
But wait, I hear you say again, how can 0! be 1?
(n+1)! 1!
Because n! can defined as n+1 , which allows us to define 0! = 1 = 1, and
this keeps everything consistent.

9.4 Expansion of (y + x)n

This is where we started looking at the binomial expansion, only with a and
b instead of x and y. There are two ways to go if we don’t have a 1 in the
bracket. We’ll look at the neatest method first. But it’s not a “better” method;
some people prefer the other one, which we’ll show you after.
As before, starting with the case n = 5:
5 5.4 2 5.4.3 3 5.4.3.2 4 5.4.3.2.1 5

(1 + x)5 = 1 + x+ x + x + x + x
1 1.2 1.2.3 1.2.3.4 1.2.3.4.5
5 4 5.4 3 2 5.4.3 2 3 5.4.3.2 4 5.4.3.2.1 5
(y + x)5 = y 5 + y x+ y x + y x + yx + x
1 1.2 1.2.3 1.2.3.4 1.2.3.4.5
The coefficients are the same in both cases, but in (y + x)5 there are
decreasing powers of y as well as the increasing powers of x. The result is
that the sum of the powers of x and y is always n (in this case 5):
y 5 x 0 , y 4 x 1 , y 3 x 2 etc.
Generalising for any n
n n(n − 1) 2 n(n − 1)(n − 1) 3 n(n − 1)(n − 1)(n − 3) 4

(1 + x)n = 1 + x+ x + x + x +...
1 1.2 1.2.3 1.2.3.4
n n−1 n(n − 1) n−2 2 n(n − 1)(n − 1) n−3 3
(y + x)n = y n + y x+ y x + y x +...
1 1.2 1.2.3
In exactly the same way, decending powers of y are inserted, starting with y n ,
and sum of the powers of x and y is n.

Example 3
Find the expansion of (2 + x)4 .
Solution
Method 1
4 44 3 4.3 2 2 4. 3.2

3 4.3.2.1

4
(2 + x) = 2 + 2 x+ 2 x + 2.x + x
1 1.2 1.
2.3
1.2.3.4

= 16 + 32x + 24x 2 + 8x 3 + x 4
In the case of n = 4 we could have used the binomial coefficients 1, 4, 6, 4, 1

and gone straight to
(2 + x)4 = 24 + 4.23 x + 6.22 x 2 + 4 2.x 3 + x 4
= 16 + 32x + 24x 2 + 8x 3 + x 4
Method 2
We said there was another approach. This is to start by converting the

expression into the form (1 + x)n . To illustrate how this works we’ll assume
we already know the binomial coefficients.
h ³ x ´i4
(2 + x)4 = 2 1 +
2
h ³ x ´i4
= 24 1 +
2
· ³x ´ ³ x ´2 ³ x ´3 ³ x ´4 ¸
4
= 2 1+4 +6 +4 +
2 2 2 2
6x 2 4x 3 x 4
· ¸
4
= 2 1 + 2x + 2 + 3 + 4
2 2 2
= 16 + 32x + 24x 2 + 8x 3 + x 4
In line 2 it is easy to forget to raise the 2 to the power 4 when taking it out of
the square bracket.
Notice also, in the penultimate line, we have left the denominators as powers
of 2, rather than cancelling, anticipating that each term will have to be
multiplied by 24 .
Method 1 is quicker, but to use it you have to remember how to insert the
powers of 2.

Example 4
Find the expansion of (2a + 3b)3 .
Solution
Method 1
3 3.2 3.2.1
(2a + 3b)3 = (2a)3 + (2a)2 (3b) + (2a)(3b)2 + (3b)3
1 1.2 1.2.3
= (2a)3 + 3 (2a)2 (3b) + 3 (2a)(3b)2 + (3b)3
= 8a 3 + 36a 2 b + 54ab 2 + 27b 3
Method 2
Once again going directly to using the binomial coefficients
3b 3
· µ ¶¸
3
(2a + 3b) = 2a 1 +
2a
· µ ¶¸3
3 3b
= (2a) 1 +
2a
· µ ¶ µ ¶2 µ ¶3 ¸
3 3b 3b 3b
= (2a) 1 + 3 +3 +
2a 2a 2a
= (2a)3 + 3(2a)2 (3b) + 3(2a)(3b)2 + (3b)3
= 8a 3 + 36a 2 b + 54ab 2 + 27b 3
Again, notice that it is better not to simplify or cancel too early, because we
are going to have to multiply through by a power of (2a). Keeping it in the
form we have it on the third and fourth lines also means we can still see the
pattern with all its symmetries, helping us to spot errors. (There’s a lot to
think about so it’s easy to slip up.)
Example 5
Find the coefficient of x 3 in the expansion of (3 − 2x)6 .
Solution
We will use Method 1. You may want to try it for yourself using Method 2.
6 6.5 4 6 .5.4 3
[3 + (−2x)]6 = 36 + 35 (−2x) + 3 (−2x)2 + 2 (−2x)3 + . . .
1 1.2 1. 2.3

So the coefficient of x 3 is 20(8)(−8) = −1, 280.
Strictly speaking we need not have written down the first 3 terms of the
expansion, but we normally do, even though we don’t have to simplify them.

QQ
1. Expansion as far as the x 3 term

(a) (1 + x)8 (b) (1 + x)10 (c) (1 − x)6
¢6
(d) (1 + 2x)5 (e) 1 + x2
¡
2. FInd the first four terms in the expansion of

(a) (2 + x)6 (b) (4 − x)4 (c) (2 + 3x)5
¢4
(d) (a − b)8 (e) 2y + x2
¡
1. (a) 1 + 8x + 28x 2 + 56x 3 (b) 1 + 10x + 45x 2 + 120x 3 (c) 1 − 6x + 15x 2 − 20x 3
(d) 1 + 10x + 40x 2 + 80x 3 (e) 1 + 3x + 15 2 5 3
4 x + 2x
2. (a) 64+192x+240x 2 +160x 3 (b) 256−256x+384x 2 −16x 3 (c) 32+240x+720x 2 +1080x 3
(d) a 8 − 8a 7 b + 28a 6 b 2 − 56a 5 b 3 (e) 16y 4 + 16y 3 x + 6y 2 x 2 + y x 3

Example 6
Expand (3 − x)(1 + 2x)6 in ascending powers of x as far as the x 3 term.
Solution
We’ll do this in two stages.

6 6.5
(1 + 2x)6 = 1 + (2x) + (2x)2 + 20 (2x)3 + . . .
1 1.2
= 1 + 6 (2x) + 15 (2x)2 + 20 (2x)3 + . . .
= 1 + 12x + 15 (4x 2 ) + 20 (8x 3 ) + . . .
= 1 + 12x + 60x 2 + 160x 3 . . .
so (3 − x)(1 + 2x)6 = 3(1 + 12x + 60x 2 + 160x 3 . . .) − x(1 + 12x + 60x 2 + 160x 3 . . .)
= (3 + 36x + 180x 2 + 480x 3 . . .) − (x + 12x 2 + 60x 3 . . .)
= 3 + 35x + 168x 2 + 420x 3 . . .
Example 7
Find the expansion of (1 + x + 2x 2 )4 as far as the x 4 term.
Solution
We will first treat the expression as (1 + X )4 , where X = x + 2x 2 . Notice when

4.3.2.1
we expand, the coefficient of the last term is always 1 (it is 1.2.3.4 in this case).
4 4.3 2 4.3.2 2
(1 + X )4 = 1 + X+ X + X +X4
1 1.2 1.2.3
= 1 + 4X + 6X 2 + 4X 3 + X 4
= 1 + 4(x + 2x 2 ) + 6(x + 2x 2 )2 + 4(x + 2x 2 )3 + (x + 2x 2 )4
Going through them one by one, you’ll see we’ve used several shortcuts.
1. Having multiplied out (x + 2x 2 )2 , we use the result to find (x + 2x 2 )3 and
(x + 2x 2 )4 .
2. At every stage we ignore terms involving powers of x higher than 4. The
further down we go the more of these there are. When we get to (x + 2x 2 )4 , it
gives us single term: x 4 .
(x + 2x 2 )2 = x 2 + 4x 3 + 4x 4 = x 2 + 4x 3 + 4x 4
(x + 2x 2 )3 = (x + 2x 2 )(x 2 + 4x 3 + 4x 4 ) = x(x 2 + 4x 3 . . .) + 2x 2 (x 2 + . . .)
= x 3 + 4x 4 + 2x 4 + . . . = x 3 + 6x 4 + . . .
(x + 2x 2 )4 = (x 2 + 4x 3 + 4x 4 )(x 2 + 4x 3 + 4x 4 ) = x 4 + . . .
so (1 + X )4 = 1 + 4(x + 2x 2 ) + 6(x 2 + 4x 3 + 4x 4 ) + 4(x 3 + 6x 4 ) + x 4 . . .

= 1 + 4x + 14x 2 + 28x 3 + 49x 4 . . .
The expansion of (1+x+2x 2 )4 as far as the x 4 term is 1+4x+14x 2 +28x 3 +49x 4 .

Example 8
Find the term independent of x in the expansion of

2 8
µ ¶
x+ .
x
Solution
2
We can use the expansion of (Y + X )8 with Y = x and X = , which starts
x
8 8.7 6 2 8.7.6 5 3
(Y + X )8 = Y 8 + Y 7 X + Y X + Y X +...
1 1.2 1.2.3
The constant term will be the one in which the power of X is the same as the
power of Y (the x’s will then cancel).
8.7.6.5 4 4
¡ 2 ¢4
That term will be the next one, i.e. Y X = 70 x 4 = 1, 120.
1.2.3.4 x
Example 9
Find the constant term in the expansion of

1 6
µ ¶
2
3x − .
x
Solution
1
As before we will expansd (Y + X )6 , and then substitute Y = 3x 2 and X = − .
x
6 6.5 4 2
(Y + X )6 = Y 6 + Y 5 X + Y X +...
1 1.2
This time it is the Y 2 X 4 term that is constant. It save times time if you
identifying in advance the term in which the x’s disappear. Failing that write
down an expression for each one until you get the one you are after (there ‘s
no need to simplify them all).
6.5.
4.3
¡ ¢4
2 2 1

2 4
That term is Y X = 15 (3x ) = 15 × 9 = 135.
1.2.
3.4
x
In these two examples, we have inroduced new variables X and Y . It’s not
essential to do so. It’s main purpose is very simple: there is a lot going on and
breaking it down like this reduces the likelihood of making a mistake. It’s not
just for beginners; we all slip up in our working sometimes, so don’t be afraid
to take it slowly.

Example 10
Simplify the function f (x), where

f (x) = (2 + 3x)3 − (2 − 3x)3 .
(i) Hence, or otherwise, express
p p
(2 + 3 2)3 − (2 − 3 2)3
p
in the form k 2, wher k is an integer.
(ii) Show that if x ≥ 0, then f (x) ≥ 0.
(iii) Evaluate f (2) and find its prime factorisation.
(iv) Show also that if n is any even integer f (n) is divisible by 72.
Solution
Rather than expand each one starting from scratch, it is quicker to expand
(a + b)3 and then substitute values for a and b.
3 2 3.2 2
(a + b)3 = a 3 + a b+ ab + b 3
1 1.2
= a 3 + 3 a 2 b + 3 ab 2 + b 3
Putting a = 2, and b = 3x, then −3x we have
(2 + 3x)3 = 23 + 3.22 (3x) + 3.2(3x)2 + (3x)3

= 8 + 36x + 54x 2 + 27x 3
and (2 − 3x)3 = 23 + 3.22 (−3x) + 3.2(−3x)2 + (−3x)3

= 8 − 36x + 54x 2 − 27x 3
so (2 + 3x)3 − (2 − 3x)3 = (8 + 36x + 54x 2 + 27x 3 ) − (8 − 36x + 54x 2 − 27x 3 )

= 2(36x + 27x 3 )
= 18x(4 + 3x 2 )
p p p
(i) (2 + 3 2)3 − (2 − 3 2)3 = f ( 2)
p
= 18 2(4 + 3 × 2)
p
= 180 2
(ii) 4 + 3x 2 ≥ 4, for all values of x, so if x ≥ 0, it follows that 18x(4 + 3x 2 ) ≥ 0.
(iii) f (2) = 18 × 2 × 16 = (32 × 2) × 2 × 24 = 26 32 .
(iv) f (n) = 18n(4 + 3n 2 ).

If n is an even integer then so is (4 + 3n 2 ) ⇒ n(4 + 3n 2 ) is a multiple of 4.
So 18n(4 + 3n 2 ) is a multiple of 18 × 4 = 72.
Actually in part (iv) we can go further, because if n is even 3n 2 is a mulitple of
4, and therefore 4 + 3n 2 is also a multiple of 4 ⇒ n(4 + 3n 2 ) is a multiple of 8
and f (n) has a factor of 144.

Example 11
In the expansion of (1 + kx)n , the coefficients x and x 2 are 18 and 135

respectively. Find the values of k and n.
Solution
n(n−1)
(1 + kx)n = 1 + n(kx)+ 1.2
(kx)2 + . . .
n(n−1) 2 2
= 1 + nkx+ 2
k x +...
n(n−1) 2
so nk = 18 and 2
k = 135.
From the first of these relationships we can find n in terms of k, or k in terms

of n, and substitute it into the second relationship. The manipulation is
slightly easier if we do the latter, i.e.
18 n(n−1)
¡ 18 ¢2
k = n and substitute to give 2 n
= 135
n(n−1)
i.e. 182 2 = 135 ⇒ 9 .2 .18(n − 1) = 2 .(9 .15)
2n
which simplifies to 6(n − 1) = 5n ⇒ n = 6, and so k = 3.

9.5 Approximating with the Binomial expansion

The binomial expansion can be used to obtain approximations. For example
(1.003)6 by writing it as (1 + 0.003)6 and substiting x = 0.003 in the expansion
of (1 + x)6 . Because 0.003 is such small number the terms of the expansion
rapidly become very small very quickly. This is despite the binomial
coefficients getting larger:
1 + 6 (0.003) + 15 (0.003)2 + 20 (0.003)3 + 15 (0.003)4 + 6 (0.003)5 + (0.003)6
For example the first few terms are

1 + 0.01|8 + 0.00|01|35 + 0.00|00|00|54 . . .
With so many zeros it is easy to lose track of them. It is a good idea to divide
them into 20 s or 30 s as we have done. That helps to avoid mistakes.
Suppose we want an approximation that is correct to 6 decimal places. We
obviously need to include the first 3 terms. What about the 4th and beyond?
The 4th term will contribute 0.000001. In this case we can safely ignore all
the other terms. It won’t always be so obvious, so the safest thing to do is to
include more terms in the expansion than we expect to need. Fortunately
exam questions generally give some guidance on how far you need to
expand.
At this point you may be thinking: what’s the point, I have a perfectly good
calculator! That is a good question, and where the expression is arithmetic of
course we would use a calculator. The point to learning about it is where the
expression is algebraic. We can find an approximate expression that is often
perfectly adequate, for an engineer for example. 6
6
This will typically require the use of other types of expansion than the binomial.
sin x
For example the expression p can be approximated by 1 − x + 12 x 2 − 31 x 3 if x is small.
1−x 2

Example 12
Use the binomial theorem to find the value of (1.02)8 , correct to four decimal
places.
p8
Without using a calculator use this result to show that 300 is equal to
approximately 2.04.
Solution
8.7 8.7.6 8.7.6.5
(1 + x)8 = 1 + 8x+ 1.2 x 2 + 1.2.3 x 3 + 1.2.3.4 x 3 . . .
= 1 + 8x + 28x 2 + 56x 3 + 70x 4 + . . .

Putting x = 0.02, we have
(1.02)8 = 1 + 8(0.02) + 28 (0.02)2 + 56 (0.02)3 + 70 (0.02)4 + . . .
= 1 + 0.16 + 28(0.00|04) + 56(0.00|00|08) + 70(0.00|00|00|16) . . .
= 1 + 0.16 + 0.01|12| + 0.00|04|48| + 0.00|00|11|2 . . .
= 1.17|16|59|2
Correct to four decimal places (1.02)8 = 1.1717.

We have put vertical separators in the decimals to make it eaiser to keep
track of the decimal places.
It may have been tempting to stop at the (0.02)3 term, but you will see that
adding the extra term affects the fourth decimal place.
In an exam you will be told how many terms to include.
(1.02)8 ≈ 1.1716592 (notice we use all the available figures at this stage).
(2.04)8 = 28 × (1.02)8
≈ 256 × 1.1716592
≈ 299.95 (to 2 DP)
≈ 300

Example 13
Use the binomial theorem to find the value of (0.997)10 , correct to four
decimal places.
solution In writing out the coefficients of (1 − x)10 below we are making use
of the fact that (1 − x)n is identical to (1 + x)n except that the signs alternate
between + and −. (We could equally well have expanded (1 + x)10 and
substituted x = −0.003.)
10.9 10.9.8 10.9.8.7
(1 − x)10 = 1 − 10x + 1.2 x 2 − 1.2.3 x 3 + 1.2.3.4 x 4 . . .
= 1 − 10x + 45x 2 − 120x 3 + 210x 4 + . . .

Putting x = 0.003, we have
(0.997)10 = 1 − 10(0.003) + 45 (0.003)2 − 120 (0.003)3 + 210 (0.003)4 + . . .
= 1 − 0.03 + 45(0.00|00|09) − 120(0.00|00|00|02|7) + . . .
+210(0.00|00|00|00|00|81) . . .
= 1 − 0.03 + 0.00|04|05 − 0.00|00|03|24 + 0.00|00|00|01|70|1 + . . .
= 0.970401777 + . . .
Correct to four decimal places (0.997)10 = 0.9704.

You will see that in this example, because the fourth term in the expansion is
so small, and almost certain not to affect the fourth decimal place, it would
have been safe to ignore the fifth term at that point.

Exercises 33
1. Find the first four terms in the expansion of (1 + x)8 in ascending

powers of x. Hence determine the value of (1.01)8 , correct to four
decimal places.
2. By expanding (1 − x)7 as far as the x 3 term, find the value of (0.998)7 ,

correct to four decimal places.
3. Find the expansion of each of these, ignoring powers of x higher than 3.

¢8
(a) (1 + 2x)9 (b) (1 − x)5 (c) (1 + 3x)6 (d) 1 − x2 (e) (1 + 10x)4 .
¡
4. Find the expansion of (2 + x)5 .

Use the result to find the value of (2.01)5 , correct to 4 decimal places.
5. Find the expansion, in ascending powers of x, of

(a) (3 + x)4 (b) (3 − x)4 (c) (2 + 3x)5 (d) (a + 2x)6 .
6. Use the binomial theorem to find the value of each of these, correct to
four decimal places.
(a) (1.001)7 (b) (1.002)10 (c) (0.96)8
7. (i) Find all of the terms of the expansion of (1+2x)6 in ascending powers
of x.
(ii) Write down the corresponding expansion of (1 − 2x)6 .
(iii) Find the value of (1.02)6 + (0.98)6 , correct to four decimal places.
8. Expand (1+2x)5 . Use the result to write down the expansion of (1−2x)5 .
Solve the equation (1 + 2x)5 + (1 − 2x)5 = 82.
9. Find the expansion of (1+x)(1+4x)5 in ascending powers of x up to and

including the x 3 term.
10. Expand each of these in ascending poweres of x up to and including the

x 2 term.
(a) (1 + 3x)(1 + 2x)4 (b) (1 − x)(1 + 4x)5 (c) (1 + 2x 2 )(1 − x)6
(d) (1 + x + x 2 )(1 + x)4 (e) (1 − x)8 (3 + 4x) (f) (2 − x)(2 + x)3
11. If x is so small that terms in x 2 or higher powers can be ignored, show

that (1 + x)4 (1 + 2x)6 = 1 + 16x + 114x 2 .
12. Solve the equation (1 + 2x)3 = (1 + 3x)2 .
13. Show that (1 + x)5 (1 − x)4 = (1 + x)(1 − x 2 )4 .

Hence,or otherwise, expand (1 + x)5 (1 − x)4 .
14. Factorise 1 − x 4 .
Hence, or otherwise, find the expansion of (1 + x 2 )3 (1 + x)3 (1 − x)3 .
15. In the expansion of (1 + ax + bx 2 )4 the coefficient of x is −8 and there

is no x 2 term.
Find a, b and the coefficient of x 3 .

16. Find the first three terms in the expansion of

(a) (a + b)6 (b) (x − y)8 (c) (p + 2q)4
1 6
¶ µ
17. Find the term independent of x in the expansion of x + .
x
3 4
µ ¶
2
18. In the expansion of x + 2 show that the constant term is 54.
x
2 4
µ ¶
2 2
19. Find the x term in the expansion of x − .
x
20. Show that (a + b)4 − (a − b)4 = 8ab(a 2 + b 2 ).

p p
Hence, or otherwise, find the exact value of (2 + 3)4 − (2 − 3)4 .
21. In the expansion of (1 + 2x)n , where n ≥ 2, the coefficients of x 3 is four

times the coefficient x 2 . Find the value of n.
22. In the expansion of (2+x)n , where n ≥ 2, the coefficients of x and x 2 are

equal. Find the value of n.
23. Find the first three terms in the expansion of (1 + 2x + 3x 2 )6 .
24. Show that the first three terms in the expansion of (1 + ax + bx 2 )3 are
1 + 3ax + 3(b + a 2 )x 2 . Find the coefficient of x 3 in terms of a and b.
Hence, or otherwise, find the first four terms of the expansion of
(1 + ax − a 2 x 2 )3 .
25. In the expansion of (1 + ax)n the coefficient of x is −12 and the

coefficient of x 2 is 60. Find the values of n and a.
26. The expansion of (1 + px)n begins 1 + 4x + 7x 2 + . . . .

Find the values of n and p.
27. Expand (x + y)4 .

If x + y = a and x y = b, show that x 2 + y 2 = a 2 − 2b
Express x 4 + y 4 in terms of a and b.
28. Given
p that n ≥ 4, show that the coefficient of x 2 in the expansion of
(1 + 2x)n is n(n − 1).
Find the value of n for which the coefficients of x 2 and x 4 are equal.
29. Find the expansion of (1 + x)4 .

p p
Hence, or otherwise, show that (1 + 2)4 = 17 + 12 2.
p p p
Find 17 + 12 2 in the form a + b 2.
p
30. Show that x = 81 is a solution of the equation 1 + 19x = 1 + 4x.
3
By writing the equation in the form (1 + 4x)3 = 1 + 19x, find the other
two solutions.
31. In the expansion of (1 + ax)(1 + bx)4 , where a, b 6= 0, the coefficient of x

is 10 and there is no x 3 term. Find the values of a and b.

32. The function f (n) is defined by

f (n) = (1 + 2n 2 )4
where n is an integer.
By expanding the bracket and selecting a suitable value for n show that
994 − 1 is divisible by 392.
By factorising show that 994 − 1 is also divisible by 49 and that its last
two digits are zero.
¡n+1¢ 1
33. Show that = n(n + 1).
n−1 2
¡n+1¢
Find the positive value of n for which = 36.
n−1
34. By substituting a suitable value of x in the expansion of (1 + x)4 , show
that Ã ! Ã ! Ã ! Ã ! Ã !
4 4 4 4 4
+ + + + = 24
0 1 2 3 4
Write down the value of
Ã ! Ã ! Ã ! Ã ! Ã !
4 4 4 4 4
− + − +
0 1 2 3 4
35. When a fair coins tossed N times the probability that it lands on a Head
¡ N ¢ ¡ 1 ¢10
exactly M times is .
M 2
Find the probability of obtaining exactly 3 Heads from 10 tosses of the
coin.
36. The probability of obtaining exactly r sixes from n throws of a 6-sided

¡n ¢ ¡ 1 ¢r ¡ 5 ¢n−r
unweighted dice is .
r 6 6
Find the probability of obtaining exactly 3 sixes from 8 throws of the
dice.

Solutions 33
1. 1.0829 [1 + 0.08 + 0.00|28 + 0.00|00|56]
2. 0.9861 [1 − 0.01|4 + 0.00|00|84 − 0.00|00|00|28]
3. (a) 1 + 18x + 144x 2 + 672x 3 (b) 1 − 5x + 10x 2 − 10x 3

(c) 1 + 18x + 135x 2 + 540x 3 (d) 1 − 4x + 7x 2 − 7x 3
(e) 1 + 40x + 600x 2 + 4000x 3
4. 32.8080 [32 + 0.8 + 0.00|8 + 0.00|00|4]
5. (a) 81 + 108x + 54x 2 + 12x 3 + x 4 (b) 81 − 108x + 54x 2 − 12x 3 + x 4

(c) 32 + 240x + 720x 2 + 1080x 3 + 810x 4 + 243x 5
(d) a 6 + 12a 5 x + 60a 4 x 2 + 160a 3 x 3 + 240a 2 x 4 + 192ax 5 + 64x 6
6. (a) 1.0721 [1 + 0.07 + 0.00|21 + 0.00|00|35]

(b) 1.0202 [1 + 0.02 + 0.00|01|8 + 0.00|00|00|96]
(c) 0.7214
[1 − 0.32 + 0.04|48 − 0.00|35|84 + 0.00|01|79 − 0.00|00|05|73|44]
In part (c), knowing how many terms we need is not so easy to see.
Adding them one by one we get
0.68, 0.72|48, 0.72|12|16, 0.72|13|95|2, 0.72|13|89|46|56
7. (i) 1 + 12x + 60x 2 + 160x 3 + 240x 4 + 192x 5 + 64x 6

(ii) 1 − 12x + 60x 2 − 160x 3 + 240x 4 − 192x 5 + 64x 6
(iii) 2.0481 [2 + 0.04|8 + 0.00|00|76|8 + 0.00|00|00|00|82]
8. 1 + 10x + 40x 2 + 80x 3 + 80x 4 + 32x 5 , 1 − 10x + 40x 2 − 80x 3 + 80x 4 − 32x 5
1
x = ±p [2x 4 + x 2 − 1 = (2x 2 − 1)(x 2 + 1) = 0, or substitute y = x 2 ]
2
9. 1 + 21x + 180x 2 + 800x 3
10. (a) 1 + 11x + 48x 2 (b) 1 + 19x + 140x 2 (c) 1 − 6x + 17x 2

(d) 1 + 5x + 11x 2 (e) 3 − 20x + 52x 2 (f ) 16 + 16x
11. (1+4x +6x 2 +. . . )(1+12x +60x 2 +. . .) = 1+(12+4)x +(60+48+6)x 2 +. . .

3 + 12x 2 + 8x 3 = 1 + 9x 2 ⇒ x 2 (8x + 3) = 0]
12. x = 0 or − 8 [1 +
6x 6x
+
13. (1 + x)5 (1 − x)4 = (1 + x)[(1 + x)(1 − x)]4 = (1 + x)(1 − x 2 )4 =.

1 + x − 4x 2 − 4x 3 + 6x 4 + 6x 5 − 4x 6 − 4x 7 + x 8 + x 9
[(1 + x)(1 − 4x 2 + 6x 4 − 4x 6 + x 8 )]
14. (1 + x 2 )(1 − x 2 ) = (1 + x 2 )(1 + x)(1 − x)

Expression is (1 − x 4 )3 = 1 − 3x 4 + 3x 8 − x 12
15. a = −2, b = −6. Coefficient of x 3 is 112. [(1 + (ax + bx 2 ))4 =

1 + 4(ax + bx 2 ) + 6(a 2 x 2 + 2abx 3 + . . .) + 4(a 3 x 3 + . . .) =
1 + 4ax + (4b + 6a 2 )x 2 + (12ab + 4a 3 )x 3 + . . .
⇒ 4a = −8 and 4b + 24 = 0. Coefft of x 3 is 12(−2)(−6) + 4(−2)3 .]
16. (a) a 6 + 6a 5 b + 15a 4 b 2 (b) x 8 − 8x 7 y + 28x 6 y 2 (c) p 4 + 8p 3 q + 24p 2 q 2

6.5.4 ¡ 1 ¢3
17. 20 [ 1.2.3 x 3 x
]
³ ´2
4.3 3
18. Third term in the expansion is 1.2 (x 2 )2 x2
= 54
4.3 ¡ ¢2
19. 24x 2 [3rd term is 1.2 x 4 − x2 = 24x 2 ]
20. (a 4 + 4a 3 b + 6a 2 b 2 + 4ab 3 + b 4 ) − (a 4 − 4a 3 b + 6a 2 b 2 − 4ab 3 + b 4 ) =

2(4a 3 b + 4ab 3 ) = 8ab(a 2 + b 2 ).
p p
112 3 [Put a = 2, b = 3 into 8ab(a 2 + b 2 )]
21. n = 8. [1 + 2nx + n(n−1)

1.2
(2x)2 + n(n−1)(n−2)
1.2.3
(2x)3 + . . .
8n(n−1)(n−2)
⇒ 6 = 4 × 4n(n−1)
2 ⇒ n − 2 = 6]
22. n = 5 [(2 + x)n = 2n + n2n−1 x + n(n−1)

2 2n−2 x 2 + . . .
n2n−1 =
⇒ n(n − 1)2n−3 ⇒ 22 .
2n−3 2n−3
= (n − 1)]
23. 1 + 12x + 78x 2 [1 + 6(2x + 3x 2 ) + 15(2x + . . .)2 + . . .]
24. 1 + 3(ax + bx 2 ) + 3(ax + bx 2 )2 + (ax + bx 2 )3 =

1 + 3ax + 3bx 2 + 3a 2 x 2 + 6abx 3 + a 3 x 3 + · · · = 1 + 3ax + 3(b + a 2 )x 2 + . . .
The x 3 coefficient is a(6b + a 2 )
1 + 3ax − 5a 3 x 3 [Put b = −a 2 the expansion]
25. a = −2, n = 6 [na = −12, n(n − 1)a 2 = 120. Substitute a = − 12

n
]
1
26. n = 8, p = 2 [np = 4, n(n − 1)p 2 = 14. Substitute p = n4 ]
27. (x + y)4 = x 4 + 4x 3 y + 6x 2 y 2 + 4x y 3 + y 4
a 2 = (x + y)2 = x 2 + y 2 + 2b
x 4 + y 4 = a 4 + 2b 2 − 4a 2 b [(x + y)4 = x 4 + y 4 + 4x y(x 2 + y 2 ) + 6(x y)2 ]
n(n−1) p 2
28. Coefft of x 2 is 2
( 2) = n(n − 1).
n = 5 [n(n − 1) = n(n−1)(n−2)(n−3)
1.2.3.4
4x 4 ⇒ (n − 2)(n − 3) = n 2 − 5n + 6 = 6]
29. (1 + x)4 = 1 + 4x + 6x 2 + 4x 3 + x 4
p p p p
Therefore (1 + 2)4 = 1 + 4 2 + 6(2) + 4(2 2) + 4 = 17 + 12 2
p p p q p p
3 + 2 2 [ 17 + 12 2 = (1 + 2)4 = (1 + 2)2 ]
q
3 27 3
30. l.h.s = 8
= 2 = r.h.s.
7
Other solutions are x = 0 and x = − 8 [1 + 3(4x) + 3(16x 2 ) + 64x 3 =
1 + 19x ⇒ x(64x 2 + 48x − 7) = x(8x − 1)(8x + 7) = 0]
31. a = −2, b = 3 [1 + (a + 4b)x + 2b(2a + 3b)x 2 + 2b 2 (3a + 2b)x 3 . . .

⇒ a + 4b = 10 and a + 2b = 0]
32. f (n) = 1 + 8n 2 + 24n 4 + 32n 6 + 16n 8 .

Putting n = 7: 994 − 1 = 8[72 + 3(74 ) + 4(76 ) + 2(78 )]
= 8 × 49[1 + 3(72 ) + 4(74 ) + 2(76 )]
994 − 1 = (992 + 1)(992 − 1) = (992 + 1)(99 − 1)(99 + 1), so 100 is a factor.

33. Ã !
n +1 (n + 1)! (n + 1)n(n − 1)! 1
= = = n(n + 1)
n −1 (n − 1)!2! (n − 1)!2 2
n = 8 [n(n + 1) = 72 ⇒ (n + 9)(n − 8) = 0]
34. Substitute x = 1.
Then substitute x = −1 so the sum is 0.
15
35.
128
36. 0.104 (3SF)


• Binomial coefficients
Expanding expressions like (a + b)n for values of n > 2 without
multiplying out lots brackets. Pascal’s triangle.
• Expanding (1 + x)n
Formula and simple problems. For example finding the coefft of x 3 in
the expansion of (1 − 2x)5 .
Using substitution to simplify the expansion e.g. X = −2x.
• Binomial notation for ¡

coefficients
n
¢
Shorthand notation: and n C r . Definition using the factorial
r
function.
• Expanding (y + x)n
Formula and problems. For example
finding the first 4 terms in the expansion of (4 + x)(2 − 3x)6 ,
or the x 3 coefficient in the expansion of (1 − x + 3x 2 )4 ,
1 6
or to find the constant term in the expansion of x 2 + 2x
¡ ¢
• Approximation using the binomial expansion

For example estimating (1.04)5 or (0.996)7 correct to 4 decimal places
without using a calculator.

Chapter 10
CALCULUS: DIFFERENTIATION
AND INTEGRATION
Overview
10.1 Introduction. What is calculus? The value of y versus how rapidly y is
changing. The gradient at a point on a curve.
10.2 Differentiation. The gradient function, or derived function or

dy
derivative of y = f (x) = ax n is nax n−1 = f 0 (x) = d x .
Differentiation from first principles. Application to functions and curves.
p 1
Differentiating non-polynomial functions such as 4 x + x 3 .
Increasing and decreasing functions.
Tangent and Normal to a curve.
Stationary values of a function: turning points (local maximum or local
minimum points) and points of inflexions.
d2y
The second derivative f 00 (x) = d x . Its use for distinguishing types stationary
value. Practical applications.
Z
ax n+1
10.3 Indefinite integration. ax n d x = n+1 + c (n 6= −1)
Conversion to and from index form, for example

2 − 13 2 8+x 5
x = p , = 8x −2 + x 3
3 33x x2
10.4 Definite integration and areas. Evaluation areas.

Theory (optional). Odd and even functions. Harder area examples.
10-1
CHAPTER 10. CALCULUS: DIFFERENTIATION AND INTEGRATION
10.1 Introduction
You are going to love this chapter. Developed in the 17th century, in England
by Isaac Newton and in Germany by Gottfried Wilhelm von Leibniz at
around the same time, calculus is without doubt one of the richest and most
fascinating areas of Mathematics. Apart from being wonderfully elegant, it
also happens to be one of the most useful areas of Maths.
What makes it is it so useful? The answer is that it applies to just about any
situation in which quantities change: the speed of an accelerating car, the
temperature of the water in a tank, rising prices, the distance from the Sun of
a orbiting comet. The list of applications is endless. Later in the chapter we’ll
show you a couple of practical examples.
We can think of it this way: we know that a lot of Maths is about the
relationship between variables. For a certain value of x, what’s the value of
y? If we have a formula for this relationship we can plot a graph, and y will
be the height of the curve above a particular value of x.
Differential calculus, or differentiation, answers a different question: at
each point along the curve
not what is the value of y

but
how quickly does y change?
Or put another way

how steep is the curve?
At its simplest, differentiating is about calculating how steep it is at any point

on a curve.
We shall also be looking at integral calculus which is essentially differential
calculus in reverse.
But first differentiation and we start by asking ourselves exactly we mean by

steepness.
This is easy to answer if the relationship between x and y is linear, so its
graph is a straight line. In this case the steepness—which is the same
everywhere—is the gradient of the line.
We need a way of defining gradient at points on a curve.

For the gradient at a point P on a curve, imagine drawing the tangent to the
curve at P . Then
the gradient of the curve at P = the gradient of the tangent at P .
y
6
tangent at P
Z
Z
~
Z

•
P

- x
Let’s get a feel for how the gradient changes at different points on the curve
below.
y
6
gradient = −0.5
gradient = 1.2 H
HH?
•
@ H
H
H
@
R •
gradient = 0

?
•
- x
0 1 2 3 4 5 6
Moving along the curve, starting at x = 0, y increases and the gradient starts
off positive. As we move from x = 1 to x = 2, y is still increasing but more
slowly. In other words the gradient is still positive but its getting smaller.
In the vicinity of x = 3 the gradient is negative, i.e. y is decreasing. There will
have been a point somewhere between x = 2 and x = 3 where the gradient
was zero. The curve levelled off at that point, reaching a “high" point.
After a stretch where the gradient is negative, we pass through x = 5 where
the gradient switches to positive again, having been zero again at x = 5.
The important thing to appreciate is that the value of y and the value of the
gradient at the same point are quite separate: one can be high and the other
low, and vice versa. They can have the same sign or the opposite sign.
One is about “how big?" the other is about “how is it changing?"

We could take a stab at finding what the gradient is at a particular point

using a graph, a pencil and a ruler. Clearly not a very reliable approach.
Calculus allows us to calculate it exactly.
The process for doing this is called differentiation. It gives us a formula for
the gradient anywhere along the curve; the formula is known as the
derivative function, the derived function, or simply the gradient function.
In the next section we’ll show you how to differentiate any polynomial1
function.
The important thing to remember is this:
if the equation of a function tells us the height of the curve at any point,
its derivative function tells us the gradient at that point.
For the function y = f (x), it’s conventional to write the derivative function as
f 0 (x).
y = f (x)
6
f 0 (3) = −0.5
10
H ?
HH
•
f 0 (1) = 1.2

H
H
H
HHj•
f 0 (5) = 0
5

f (1) = 6
f (3) = 9
?
•
f (5) = 2
- x
0 1 3 5
Notice how f (x) and f 0 (x) are not related in any simple way: for example we
could have a large f (x) and a small f 0 (x) or vice versa; one could be positive
the other negative, and so on.
1
We’ll need a range of different techniques if we are to differentiate any function.

10.2 Differentiation
How do we construct the derivative function f 0 (x) from f (x)? It’s
surprisingly simple. We’ll start by explaining how to construct it (for any
polynomial function). In the next section we’ll show you how Newton and
Lebniz derived the method.
For our purpose here each term of the function will be of the form ax n ,
where a and n are constants. All we do is convert each term into a new term,
using the transformation rule
A
!
A
ax n → nax n−1 A
A
In words we
multiply by the power and reduce the power by 1
The only exception to this rule is constant terms: they simply disappear.
c → 0
Typically the function we are differentiating will consist of a number of

terms. The derivative function is the sum of these new terms.
For example, take the function
y = f (x) = x 2 − 4x + 7.
Differentiating each term:

x 2 → 2x 1
−4x 1 → −4x 0 = −4
+7 → 0
so f 0 (x) = 2x − 4.

We can sketch this function by completing the square: f (x) = (x − 2)2 + 3.

This tells us there is a minimum point at (2, 3).
y = f (x) = x 2 − 4x + 7
6
10
5 @
I
@
g r ad i ent = 2
4
- x
−1 0 1 2 3 4 5
Let’s look at one point on the curve, say x = 3,

f (3) = x 2 − 4x + 7 = (3)2 − 4(3) + 7 = 4 and
f 0 (3) = 2x − 4 = 2(3) − 4 = 2.
The height of the curve at x = 3 is 4 and the gradient at that point is 2.
We can turn this round and ask the question where on the curve is the
gradient equal to −3, for example?
It will be where 2x − 4 = −3, i.e. where x = 12 .
This answer doesn’t look unreasonable. We can see from the graph that the
gradient anywhere to the left of x = 2 is negative.
We can use calculus to find where the lowest point on the curve is. It’s where
is the gradient zero, or 2x − 4 = 0, i.e. at x = 2. We already knew this; it’s
something we can determine for any quadratic by completing the square.
The power of calculus is that it allows us to find maximum and minimum
points for any function.
QQ
1. Without doing any further calculations write down, for the function
f (x) = x 2 − 4x + 7
(a) the value of f 0 (2), and
(b) the values of f (1) and f 0 (1).
2. Where on the curve is the gradient equal to 6?
(The answers are in the footnote2 .
2
1. (a) At the minimum point the gradiet is zero so f 0 (2) = 0. (b) Being a quadratic the
curve is symmetrical about x = 2, so f (1) = f (3) = 4, and f 0 (3) = − f 0 (1) = −2.
2. Where 2x − 4 = 6, i.e. at x = 5.

Example 1
A function is defined by
y = 2x 3 + x 2 − 8x − 1.
Find
(a) the gradient of the curve at (i) x = 2, (ii) x = −1 and (iii) x = 32 .
(b) the values of x for which the gradient of the curve is (i) zero, (ii) −8.
(c) the equation of the tangent to the curve at x = 0.
Solution
(a) The derivative function of y = f (x) is
f 0 (x) = 6x 2 + 2x − 8.
(i) f 0 (2) = 20
(ii) f 0 (−1) = −4
(iii) f 0 ( 32 ) = 6( 94 ) + 3 − 8 = 27
2
− 5 = 17
2
.
(b) (i) The gradient is zero where 6x 2 + 2x − 8 = 0, i.e.

2(3x 2 + x − 4) = 2(x − 1)(3x + 4) = 0,
whose solutions are x = 1 and x = − 43 .
(ii) We need to solve the equation f 0 (x) = −8, i.e.
6x 2 + 2x = 2x(3x + 1) = 0.
The points are at x = 0 and x = − 31 .
(c) The gradient f 0 (0) = −8. We can also see that f (0) = −1.
We want the equation of the line with gradient −8 that passes through
(0, −1), i.e.
y − (−1) = −8(x − 0)
y = −8x − 1

Here is another function:

f (x) = x 3 − 6x 2 .
Differentiating this one we have
x 3 → 3x 2
−6x 2 → −12x 1 = −12x
so
f 0 (x) = 3x 2 − 12x.
We are going to sketch f (x), and f 0 (x), to see the relationship between the
two curves.
f (x) is a cubic, which we can write as y = x 2 (x − 6), so the curve cuts the
x-axis a x = 6 and touches it at the origin. The coefficient of x 3 is positive,
and that tells us which way up it is. SEE PAGE.
The gradient of f (x) is zero where f 0 (x) = 3x(x − 4) = 0. One solution of this
equation is x = 0. This is no surprise: there is a repeated root at x = 0, so the
curve touches the x-axis at the origin. The other point with zero gradient is
at x = 4.
y = f (x)
6
The gradient of f (x) is

positive when x < 0 and
when x > 4,
the gradient is negative
- x
0 4 6 when 0 < x < 4.
The gradient is zero at

x = 0 and x = 4.
y = f 0 (x)
6 So y = f 0 (x) is above the
y-axis when x < 0
and when x > 4;
it is below the x-axis
when 0 < x < 4.
- x
0 4 6
The curve crosses the
axes at x = 0 and x = 4.
We can also see, for example, that as we approach x = 0 from the left, the
gradient of f (x) is positive but decreasing in value. That is reflected in the
f 0 (x) curve: for the same values of x the height of this curve is falling.

10.2.1 Differentiation from first principles

You will be expected not just to differentiate functions but understand how
the theory works, how Newton and Leibniz cracked the centuries-old
problem of calculating rates of change whatever the relationship between
the variables.
First we’re going to illustrate the underlying principle using a simple

numerical example. After that we’ll generalise what we’ve seen to give us
something much more useful. That won’t give a complete explanation—that
requires some slightly more advanced maths, but it’s enough to lift the veil.
In the last section we looked at the function f (x) = x 2 − 4x + 7 and used some
rules to show that the gradient at x = 3 is 2. We’ll show that the gradient is 2
without using those rules, i.e. from first principles.
y = f (x) = x 2 − 4x + 7
6
10
5
P
4 = f (3)
-x
−1 0 1 2 3 4 5
We are looking closely at the curve

Q near P. We can think of the gradient
of the chord from P to Q as a first

stab at the gradient of the tangent at
P, because Q is at x = 3.01.

The height of Q above the x-axis is
f (3.01) = (3.01)2 −4(3.01)+7 = 4.0201

f (3.01) = 4.0201
QR
So grad of PQ = PR
= 0.0201
0.01
= 2.01.

P R Now let’s move Q even closer to P , to

say x = 3.001, the gradient of PQ now
f (3) = 4

turns out to be 2.001.
There’s no prize for quessing what
the gradient would be if Q were at
x = 3.0001. It’s 2.0001.
3 3.01
The gradient of PQ is 2 + the difference between the x-values of P and Q.
The final step is to say this: let that difference tend to zero, i.e. Q → P .
The chord PQ morphs into the tangent at P and its gradient is clearly 2.

Now we can see how to tackle the problem, the next step would be to work
out a formula for the gradient at any point on the curve. In fact what we’re
going to do is find the derivative function for
f (x) = x 2
and show how we can use this to differentiate any quadratic function.
Part of the fun of differential calculus will be extending this to polynomials
in general, and from there to any function at all. Some of this you will have to
take on trust for now.
Generalising what we did before, the x-coordinates of P and Q are x and

x + h, and their heights above the x-axis are x 2 and (x + h)2 , since the curve is
y = x 2 . The points P and Q are close together, i.e. the distance h is small.
The length of QR is
(x + h)2 − x 2 = 2xh + h 2 ,
so the gradient of PQ is
QR 2 2
PR
= 2xh+h
h
= 2xh
h
+ hh = 2x + h.
Q
What we have is a formula for the
gradient of any chord on the curve
y = x 2.

For example, if the chord is between
(x + h)2

x = 3 and x = 7, h = 4 and the
gradient of the chord is 2 × 3 + 4 = 10.

P R For our purposes we want Q to be

a point close to P . Then, as before,
we let Q get closer and closer, until

x2
eventually h = 0.
When h = 0, P and Q are at the
same point and the gradient of 2x +
x x +h
h becomes 2x. The gradient of the
tangent at P is 2x.
We say that 2x is the limit of 2x + h as
h → 0 (h tends to zero).
But this is a formula for the gradient at any point on y = f (x) = x 2 , so it is the
derivative function of f (x):
f 0 (x) = 2x.
We say x 2 differentiates to 2x.
We said that we could use this as the basis for differentiating any quadratic
function. For that we need to be able to differentiate ax 2 , where a is any
constant. You may want to have a go at repeating what we have just done,
but with ax 2 instead of x 2 . You should get the answer 2ax. The calculation is
on the next page.

If the function is y = f (x) = ax 2 ,
QR = f (x + h) − f (x) = a(x + h)2 − ax 2

= a(x 2 + 2hx + h 2 ) − ax 2
= 2ahx + ah 2
and the gradient of PQ is
QR 2ahx + ah 2
=
PR h
2ahx ah 2
= +
h h
= 2ax + ah
As before, if we let h → 0, Q → P and the gradient → 2ax.

The way we normally write this (because it is more concise) is
lim (2ax + ah) = 2ax

h→0
We say the limit as h tends to zero is 2ax, i.e. this is differential of ax 2 .
If you want a challenge, find what ax 3 differentiates to. (Solution on the next
page.)

Using the same notation and labelling as before, for y = f (x) = ax 3 ,
QR = f (x + h) − f (x) = a(x + h)3 − ax 3

= a(x 3 + 3hx 2 + 3h 2 x + h 3 ) − ax 3
= a(3hx 2 + 3h 2 x + h 3 )
and the gradient of PQ is
QR 3ahx 2 + 3ah 2 x + ah 3
=
PR h
2
3ahx 3ah 2 x ah 3
= + +
h h h
2 2
= 3ax + 3ahx + ah
As before, if we let h → 0, Q → P and the gradient → 3ax 2 , i.e.
lim (3ax 2 + 3ahx) = 3ax 2

h→0
ax 3 differentiates to 3ax 2 .

Derivative of ax n
We will show that its derivative is anx n−1 .
For this we will need the binomial expansion we learned about in Chapter 9.
n n−1 n(n−1) n−2 2 n(n−1)(n−1) n−3 3
(a + b)n = a n + a b+ a b + a b +...
1 1.2 1.2.3
If it hasn’t already occurred to you it will quickly be clear that we only need
the first few terms. (The a in this expansion is obviously different from the a
in ax n . We are simply using the expansion to remind us of the pattern.)
Using the same notation and labelling as before, for y = f (x) = ax n ,
QR = f (x + h) − f (x)
= a(x + h)n − ax n
n n−1 n(n − 1) n−2 2 n(n − 1)(n − 1) n−3 3
= a(x n + x h+ x h + x h + . . .) − ax n
1 1.2 1.2.3
1 1
= anx n−1 h + an(n − 1)x n−2 h 2 + an(n − 1)x n−3 h 3 . . .
2 6
When writing an expression for the gradient of PQ, knowing that we shall be
dividing by h and letting h → 0 we can safely ignore every term from the h 3
and beyond.
QR anx n−1 h + 21 an(n − 1)x n−2 h 2 + . . .
=
PR h
1
= anx n−1 + an(n − 1)x n−2 h + . . .
2
As before, we let h → 0, so Q → P .
1
lim [anx n−1 + an(n − 1)x n−2 h] = anx n−1
h→0 2
ax n differentiates to anx n−1 .

10.2.2 Differentiating a polynomial function

If you chose not to follow the explanation on the previous page take it on
trust that if n ≥ 1,
any terms of the form ax n differentiates to anx n−1
For the case n = 1 the rule works: ax differentiates to 1ax 0 = a. This makes
sense because the function y = f (x) = ax is straight line whose gradient at
every point is b, i.e. f 0 (x) = a. So ax differentiates to a.
What about differentiating a constant term like y = c? We could write it as
cx 0 giving a derivative of 0 × x −1 = 0. A simpler way to think of it is to
consider the function y = f (x) = c. This a a “horizontal" straight line, i.e. its
gradient its zero, that is
when we differentiate a constant term it disappears
We have taken it as read that we can differentiate a sum of terms by
differentiating each term individually.
For example if
3
f (x) = 5x 4 − 2x 3 + x 2 − x + 13
2
f 0 (x) = 20x 3 − 6x 2 + 3x − 1
In a polynomial every n will be an integer. In fact the differentiation rule still

applies if n is not an integer, and it doesn’t even need to be > 0. But for the
moment it will be.
We’ll see plenty of examples of this very shortly.

Exercises 34
In each case find from first principles the gradient of the function f (x) at the
given point.
1. f (x) = x 2 + 5x at x = 2
2. f (x) = 1 − x 2 at x = −1
3. f (x) = 3x 2 − 4x + 11 at x = 3
4. f (x) = k(x 2 + 1) at x = 1
5. f (x) = (x − 1)2 at x = 1
1 1 1
6. f (x) = at x = 2, by writing − as a single fraction.
x 2+h 2
p p
7. f (x) = x at x = 9, given that the expansion of a + h begins
1 1 3
a 2 + 12 a − 2 h − 81 a − 2 h 2 + . . .

Solutions 34
1. 9
2. 2
3. 14
4. 2k
5. 0
6. − 41
1
7. 6

dy
10.2.3 dx
notation
This is a good place to introduce an alternative notation used in
differentiation.
If you haven’t been following the optional (blue) text, skip to where the text
turns to black.
QR
When Q is close to P , we can think of the gradient PR
as
a small increase in y
a small increase in x
where the increase in y determined by how much x is increased by, and of
course what the function is. The shorthand notation3 for this is
δy
δx
It’s only a notation issue: δx = h.
But there is a good reason for using it. As Q → P , δx → 0 and δy → 0 and the
δy
ratio δx → a specific value—the gradient of the tangent at P . We call this
dy
.
dx
δy
This is a rather odd entity: it’s what δx becomes when numerator and
denominator both become zero.
It’s odd because we can’t have a fraction 00 , but the fraction can be
meaningful when both δx → 0 and δy → 0 in the way we have described.
Using this notation the explanation of how we differentiate x 2 looks like this
The length of QR is δy, so

Q δy = (x + δx)2 − x 2 = 2xδx + δx 2 ,
so the gradient of PQ is
δy 2 2
+ δx
QR
= δx = 2xδx+δx = 2xδx

δx δx δx

PR
= 2x + δx.
(x + δx)2

As before we let Q get closer and
closer to P so that δx eventually
P R
becomes zero. The gradient becomes
2x and this is then the gradient of the
dy
curve at P , which we write as d x .
x2
x x + δx
If this all seems rather esoteric, don’t worry.
3 ∆y
In some textbooks it is written ∆x .
(In the Greek alphabet δ is a small d and ∆ is a capital d .)

Just remember
y = f (x)
dy
= f 0 (x).
dx
For example, if y = f (x) = x 3 + 5x − 1
f 0 (x) = 3x 2 + 5
dy
or we can write it = 3x 2 + 5.
dx
They are different notations for exactly the same thing; they are
interchangeable.
dy
For example,we can say: when x = 1, d x = 8.
It’s slightly more concise to say: f 0 (1) = 8.

A word of warning.
We said when differentiating an expression, we take it one term at a time:

we differentiate each term and add the answers together. If the expression is
something like
y = x 2 (4x + 5)
we need to convert it into the sum of a series of terms, each of the form ax n ,
where a is a constant. So in this case, multiplying out the bracket we have
y = 4x 3 + 5x 2
which differentiates to
dy
= 12x 2 + 10x,
dx
which we may want to write as
dy
= 2x(6x + 5).
dx
What we can’t do is differentiate x 2 , differentiate (4x + 5) and multiply the

answers together:
x 2 differentiates to 2x and 4x + 5 differentiates to 4. Their product is 8x. This
is not the correct answer.

10.2.4 Differentiating harder polynomial functions

We know that we can differentiate any polynomial function using the
principle that
any terms of the form ax n differentiates to anx n−1 ,
provided n ≥ 1. If n = 0, i.e. the term is a constant, if differentiates to zero.

We can differentiate the terms one by one.
For example, if
y = 2x 5 − 7x 4 + x 3 − 11x + 1
dy
= 10x 4 − 28x 3 + 3x 2 − 11.
dx
To differentiate an expression we will sometimes have to write it in a
different way.
Example 2
Differentiate each of these functions:
(a) y = 3x(x 3 − 1) (b) y = (2x + 5)2
(c) y = 13 x 6 − x (d) y = 12 (x + 1)2

3x 2 +1
(e) y = 4
You may want to try differentiating these yourself before looking at the
answers on the next page.

Solutions
dy
(a) y = 3x 4 − 3x, so d x = 12x 3 − 3.
dy
(b) Expanding the bracket, y = 4x 2 + 20x + 25, so d x = 8x + 20.
dy
(c) 13 × 6 = 13 × 61 = 2, so d x = 2x 5 − 1.
dy
(d) 21 (x + 1)2 = 21 (x 2 + 2x + 1) = 12 x 2 + x + 12 , so d x = x + 1.
(e) We need to write the function as the sum of terms of the form ax n .
3x 2 1
y = +
4 4
3 2 1
= x +
4 4
dy 3
= 2× x
dx 4
3
= x
2
3 3 2
If you are not sure why 2× 4 = 2 think of the 2 as 1 and multiply the two
fractions4 .
4
If you thought it should be 68 you are confusing it with equivalent fractions, which is
writing the same fraction in a different form. In this case by multiplying both top and
bottom by 2. What we want to do is double the size of 34 and should expect the answer to
be 1 21 .

QQ
1. Differentiate the following functions

(i) y = 3x 3 + 5x 2 − 2x + 11
(ii) y = 1 + x + x 2
(iii) y = x 2 (x 2 + 3)
(iv) y = (3x + 1)2
(v) y = 21 x(x 3 − 8)
7x+2x 2
(vi) y = 5
2. The function f (x) is defined by

f (x) = x 4 − 2x 2 + 5
(a) Find the gradient of the curve y = f (x) at x = 2 and at x = −1.
(b) For what values of x is the gradient of the curve equal to zero?
3. Find the equation of the tangent to the curve

y = x 3 − 8x + 9
at x = 2.
5
1. (i) 9x 2 + 10x − 2 (ii) 1 + 2x (iii) 4x 3 + 6x (iv) 18x + 4 (v) 2x 3 − 4. (y = 21 x 4 − 4x)
(vi) 75 + 54 x
2. (a) f 0 (x) = 4x 3 − 4x. f 0 (2) = 24, f 0 (−1) = 0
(b) x = −1, 0, +1. f (x) = 4x(x 2 − 1) = 4x(x + 1)(x − 1).
dy dy
3. Tangent is y = 4x − 7. d x = 3x 2 − 8. On the curve, when x = 2, y = 1 and dx = 12 − 8 = 4.
Equation of tangent is y − 1 = 4(x − 2)

10.2.5 Differentiating non-polynomial functions

In this section we are going to learn about differentiating functions whose
terms are powers of x that are not positive integers. Functions like
p 1
y =4 x + 3.
x
We won’t be covering all non-polynomial functions, far from it, but it’s an
important step towards it.
We said that
ax n differentiates to anx n−1 for any n ≥ 1.
In fact
ax n differentiates to anx n−1 for any value of n.
n can take values < 1, including negative ones, and it can be a fraction6 .
The principle remains the same but there is more manipulation involved, so
we need to spend a little time on it.
Very briefy, what you need to remember about indices is as follows:
Negative indices
1 1 1
a −n = , so for example 2−3 = 3 = .
an 2 8
Fractional indices
m p p 4 p
a n = ( n a)m or
n
a m , so for example 8 3 = ( 8)4 = 24 = 16.
3
1 1 1
The index could be a negative fraction. For example 9− 2 = 1 = 3.
92
For more about negative and fractional indices see Chapter 2.

If you find you need to brush up on fractions see Chapter 4.
p
6
In fact it can be any real number, including irrational ones e.g. x 2 or x π .
We also know that n = 0 is a special case: constant terms differentiate to zero.

Example 3
Differentiate the function

4
f (x) = .
x2
Solution
4
we can write as 4x −2 , so
x2
f 0 (x) = −8 x −3
1
= −8.
x3
8
= −
x3
All these three answers are correct. The last one is likely to be the most
useful.
Example 4
3
Find the gradient of the function y = px at x = 4.
Solution
We can write the function as

1
y = 3x − 2
which differentiates to
dy 1 3
= −3. x − 2
dx 2
3 3
= − . x− 2
2
3 1
= − . 3
2 x2
3
= − 3
2x 2
3 p dy 3
When x = 4, x 2 = ( 4)3 = 8 and d x = − 16 .

Example 5

p 1
y =4 x+ 3
x
and find its gradient when x = 1.
Solution
We can write the function

1
y = 4x 2 + x −3
so
dy 1 1 1
= 2 x − 2 − 3x −4 = 2. − 3.
dx x
1
2 x4
2 3
= −
x
1
2 x4
2 3
= p − 4.
x x
As you get more fluent you’ll find yourself missing out the middle stages. But
don’t hurry this; there’s a lot going on.
dy
When y = 1, d x = 2 − 3 = −1.
Example 6

3x 3 + 2
y= .
x
Solution
We can’t differentiate this as it stands, but we can if we write it as
3x 3 2
y= + = 3x 2 + 2x −1 .
x x
dy
= 6x − 2x −2
dx
2
= 6x −
x2
2 1
Notice that 2x −2 = x 2 , not 2x 2 . The index (−2) is only attached to the x.

Here are some more example. If we are going to be substituting in a value of

x, it is easiest to write the expression in a form that doesn’t include negative
or fractional indices.
f (x) → f 0 (x)
1 1
= x −1 → −x −2 =−
x x2
1 2
= x −2 → −2x −3 =−
x2 x3
p 1 1 1 1 1 1
x = x2 → 2
x−2 =
2 x 21
= p
2 x
p 1 1 2 1 1 1
3
x = x3 → x−3 =
3 x 32
= p
3
3 3 ( x)2
p1 1 1 p1
1 3
x
= x−2 → − 21 x − 2 =−
2 x 23
=
2 ( x)3

What we are seeing is that whilst the process of differentiating is no more

complicated than it is with polynomials, we need to take great care with the
execution when there are negative and fractional indices involved. This juicy
example is as hard as you’d be expected to tackle at this level.
Example 7
Show that the derivative function of

4x + x 3
f (x) = p
3 x
is 4 + 5x 2
f 0 (x) = p .
6 x
Solution
4x x3
y = 1
+ 1
3x 2 3x 2
4 1 1 5
= x2 + x2
3 3
dy 2 −1 5 3
= x 2 + x2
dx 3 6
3
2 5x 2
= p +
3 x 6
4 5x 2
= p + p
6 x 6 x
4 + 5x 2
= p
6 x

Exercises 35
1. Differentiate the following functions. Simplify your answers where
possible.
(a) y = x 5 + 2x 3 (b) y = 100x (c) y = 31 x 6
(d) y = x + x 2 + x 4 (e) y = −2 (f) y = x 100 + 99
(g) y = 32 x 3 (h) y = 18 x 4 (i) y = x + 12 x 2 + 31 x 3 + 14 x 4
x6
(j) y = −(3x 3 − x) (k) y = 3 (l) y = (2x + 5)2
p p
(m) y = x 2 (1 − x 3 ) (n) y = ( x + 2)( x − 2)
dy
2. For these functions find d x . Simplify your answers and express them in
a form that does not include negative indices.
1
(a) y = 2x 4 (b) y = 5x −2 (c) y = 6x 3
4
(d) y = 3 (e) y = 3x (f) y = x
p 1
(g) y = 2x −1 (h) y = 8 x (i) y = x 3
4 3 1
(j) y = 6 x 3 (k) y = px (l) y = 2x 2
3
(m) y = x 2 + x −2 + x42 (o) y = x 2 x + x1
¡ ¢
(n) y = x
1
¢2 ¢2
x + x1
¡
(p) y = x 2 (1 + x + x 2 ) (r) y = 2x − x1
¡
(q) y =
3. Differentiate each of these functions.

1 5 p
(a) y = 3x (b) y = 4x 2 (c) y = x 2 x
´3
p
³
1 1
(d) y = 2px (e) y = ( x + 1)2 (f) y = px
p p p
q
3 4 1
(g) y = 3( x) (h) y = x(2 − x) (i) y = x3
dy
4. In each case find d x .
x 2 +4x 3 3x 4 −x 2 x+5x 3
(a) y = (b) y = (c) y =
x x2 x2
p
2x 2 −4 6x+ x 8x 5 −6x 3
(d) y = (e) y = p (f) y =
x x 2x 2
3 2 5 3 p
2x 2 +x 3 x 4 −4x 4 x+x 2
(g) y = (h) y = 1 (i) y = p
3
x x2 x
³p
p p
´
(j) y = 2x x+ p1 (k) ( x − 3)( x + 5) (l) y =(1 + px )2
1
x
5. In each case find the gradient of the curve at the point specified.
(a) y = x 3 + 2x at the point where x = 1.
1
(b) y = x at the point (2, 21 ).
(a) y = 2x 2 − 5x + 7 at the point (−1, 14).

Solutions 35
1. (a) 5x 4 + 6x 2 (b) 100 (c) 2x 5
(d) 1 + 2x + 4x 3 (e) 0 (f ) 100x 99
(g) 2x 2 (h) 21 x 3 (i) 1 + x + x 2 + x 3
(j) 1 − 9x 2 (k) 2x 5 (l) 8x + 20
(m) 2x − 5x 4 (n) 1 [y = x − 4]
10 2
2. (a) 8x 3 (b) − x 3 (c) 2
x3
4
(d) 0 (e) 3 (f ) − x 2
2 4 3
(g) − x 2 (h) px (i) − x 4
p
3 3 1
(j) y = 8 x (k) − 3 (l) − x 3
2x 2
2 3
(m) 2x− x 3 (n) − x 2 − x83 [y = 3x −1 + 4x −2 ]
2
(o) 3x 2 + 1 [y = x 3 + x] (p) − x 3 − x12 [y = x −2 + x −1 + 1]
2 2
(q) 2x− x 3 [y = x 2 + 2 + x −2 ] (r) 8x− x 3 [y = 4x 2 − 4 + x −2 ]
1 5
3. (a) − 3x 2 [y = 13 x −1 ] (b) − 2x 3 [y = 54 x −2 ]
3 5 1 1
(c) 5
2 x2 [y = x 2 ] (d) − 3 [y = 21 x − 2 ]
4x 2
1 1 3 1 3
(e) 1+ px [y = x + 2x 2 + 1] (f) − 5 [y = (x − 2 )3 = x − 2 ]
2x 2
p 4 1 1
(g) 4 3 x [y = 3x 3 ] (h) px −1 [y = 2x 2 − x]
3
(i) − 5 [same as part (f )]
2x 2
4. (a) 1 + 8x [y = x + 4x 2 ] (b) 6x [y = 3x 2 − 1]
1 1 4 4
(c) − x 2 + 5 [y = x + 5x] (d) 2 + x 2 [y = 2x − x ]
3 1
(e) px [y = 6x 2 + 1] (f ) 3(4x 2 − 1) [y = 4x 3 − 3x]
1 1 1 1 3 1 3 1
(g) 1 − 4 [y = 2x 2 + x − 3 ] (h) 1 − 3 [y = x 4 −4x 4 ]
x2 3x 3 4x 4 x4
1 2 1 5 p 1 3 1
(i) 5 + 53 x 3 [y = x 6 +x 3 ] (j) 3 x + px [y = 2x 2 + 2x 2 ]
6x 6
1 1 1 1 1
(k) 1 + px [y = x + 2x 2 − 15] (l) − 3 − x2 [y = 1 + 2x − 2 + x −1 ]
x2
dy
5. (a) gradient = 5. [ d x = 3x 2 + 2]
1 dy
(b) gradient = − 4 [ d x = − x12 ]
dy
(c) gradient = −9 [ d x = 4x − 5 ]

10.2.6 Increasing and decreasing functions

As the name suggests a function is increasing over an interval I , if for every
x ∈ I , f 0 (x) ≥ 0.
A function is decreasing over an interval I , if for every x ∈ I , f 0 (x) ≤ 0.
For example y = 2x + 1 is an increasing function for all values of x. Another
would be y = x 3 , although strictly speaking y = 2x + 1 is a strictly increasing
function, because f 0 (x) 6= 0 anywhere, whereas with y = x 3 , f 0 (0) = 07 .
Generally speaking a function will be increasing over certain values of x and
decreasing over others. Take the function below
y = x 2 (3 − x)
x
−1 0 1 2 3
The function is decreasing for x ≤ 0. For x ≥ 0 it increases as far as its turning

point and thereafter decreases.
dy
The turning point is where dx
= 0, i.e. where 6x − 3x 2 = 3x(2 − x) = 0.
The two turning points are at x = 0 and x = 2.
We can now complete the picture: the function is decreasing for x ≤ 0 and
x ≥ 2; it is increasing for 0 ≤ x ≤ 2.
Another notation for a ≤ x ≤ b is [a, b], with square brackets. We use round
brackets (a, b) to indicate the region a < x < b. If equality applies at one end
but not the other we would write [a, b) or (a, b].
For these functions write down where they are increasing and where they are
decreasing. The answers are in the footnote8 .
y
y
6 6
5 - x
−4 0 4
(2, 1)
-x
7
If a function is increasing but not strictly increasing it is sometimes referred to as being
non-decreasing.
8
Quadratic. Decreasing for x ≤ 2, increasing for x ≥ 2
Quartic. Decreasing for x ≤ −4, 0 ≤ x ≤ 4, increasing for −4 ≤ x ≤ 0, x ≥ 4.

1
Sometimes we have to a bit careful. The function y = may appear to
x
decreasing everywhere.
1 The function is undefined at

y=x x = 0 so it does not have a
6 gradient therea .
So the function is decreasing
for x < 0 and x > 0.
-x
a
If anything it has an infinite
gradient.
If we have a sketch of the function and know where its turning points are we
can simply write down where the function is increasing and where it is
decreasing. Suppose we don’t.
Example 8
Find where the function f (x) = x 3 − 3x 2 − 9x + 7 is increasing and where it is
decreasing.
Solution We want to know where f 0 (x) ≥ 0 and where f 0 (x) ≤ 0.
f 0 (x) = 3x 2 − 6x − 9
= 3(x 2 − 2x − 3)
= 3(x − 3)(x + 1)
So we want to know where (x + 3)(x − 1) ≥ 0 and where (x + 3)(x − 1) ≤ 0.

In Chapter 7 we saw how to solve quadratic inequalities using a sketch of the
function.
y
y = (x + 1)(x − 3)
1 In this case we want to know for
y = (x + 1)(x − 3) where y ≥ 0 and
x
−1 0 1 2 3 where y ≤ 0.
−1 From the sketch therefore
−2 f (x) = x 3 − 3x 2 − 9x + 7 is
increasing for x ≤ −1 and x ≥ 3.
−3 It is decreasing for −1 ≤ x ≤ 3.
−4

Example 9
16
Find where the function f (x) = x + is increasing and where it is
x
decreasing.
Solution
We need to find where f 0 (x) ≥ 0 and where f 0 (x) ≤ 0
f (x) = x + 16x −1
f 0 (x) = 1 − 16x −2
16
= 1− 2
x
f 0 (x) ≥ 0 when x 2 ≥ 16 i.e. when x ≤ −4 and x ≥ 4.
f 0 (x) ≤ 0 when x 2 ≤ 16 i.e. when −4 ≤ x ≤ 4.
Example 10
Find in terms of the constant p the values of x for which the function
f (x) = x 3 (p − x) is decreasing.
Solution
f (x) = px 3 − x 4
f 0 (x) = 3px 2 − 4x 3
= x 2 (3p − 4x)
3p
Since x 2 ≥ 0 f 0 (x) ≤ 0 when 3p − 4x ≤ 0, i.e. when x ≥
4
Example 11
Find the values of x for which the function f (x) = x 2 (x 2 + 6) is increasing.
Solution
f (x) = x 4 + 6x 2
f 0 (x) = 4x 3 + 12x
= 4x(x 2 + 3)
Since (x 2 + 3) ≥ 0 f 0 (x) ≥ 0 when x ≥ 0.
Sometimes you will be asked to show that a functions is increasing or

decreasing for all values of x.

Example 12
1. Show that these function are increasing for all values of x
(a) f (x) = 2x 3 + 7x − 1 (b) f (x) = x 3 + 3x 2 + 3x − 5.
2. Show that this function f (x) = 10 − 5x − x 3 is decreasing for all values of x.
Solution
1. (a) f 0 (x) = 6x 2 + 7 ≥ 7 > 0 for all x.
(b) f 0 (x) = 3x 2 + 6x + 3 = 3(x + 1)2 ≥ 0 for all x.
2. f 0 (x) = −5 − 3x 2 ≤ −5 < 0 for all x
Example 13
Show that the function f (x) = x 3 − 4x 2 + 15x is increasing for all values of x.
Solution
f 0 (x) = 3x 2 − 8x + 15.
This quadratic doesn’t factorise. The positive coefficient of x tells us it is a u-
shaped curve and it intersects the y-axis at 15. So we need to know whether
or not it has any roots. If it does there will be values x for which f 0 (x) < 0, i.e.
it will be decreasing.
We can check this by calculating the discriminant9 .
The discriminant is (−8)2 −4(3)(15) = −116. Since this is negative the equation
has no roots. The curve of f 0 (x) does not cross the x-axis. f (x) is an increasing
function for all values of x.
9
For the equation ax 2 + bx + c, the discriminant is b 2 − 4ac.

Exercises 36
1. For each of these functions find the values of x for which it is an

increasing function:
(a) 3x 2 − 12x − 7 (b) 4x − 8x 2 (c) x 3 − 9x 2 + 15x + 2
(d) 12x − x 3 (e) x 4 + 2x 2 (f) x 5 + 5x 3
2. For each of these functions find the values of x for which it is a

decreasing function:
(a) 1 + 6x − x 2 (b) (2x − 3)2 (c) x 3 − 18x 2 + 108x − 77
(d) 2x 3 + 15x 2 (e) 8x 3 − 3x 4 (f) 3x 5 − 10x 3 + 15x
3. Find the values of x for which f (x) is an increasing function, if f (x) is:
4 1 p x 2 +1 p
(a) x + (b) x 2 − 2 (c) 6 x − x (x > 0) (d) (e) ( x − 1)2 (x > 0)
x x x
4. For the function f (x) = x 3 + 3x 2 + kx + 1 find the values of k such that
f (x) is increasing for all values of x.
5. If f (x) = x(λ + x − x 2 ), for what values of λ is f (x) decreasing for all

values of x.

Solutions 36
1. (a) x ≥ 2 (b) x ≤ 41 (c) x ≤ 1 and x ≥ 5 (d) −2 ≤ x ≤ 2 (e) x ≥ 0 (f ) All x
2. (a) x ≥ 3 (b) x ≤ 32 (c) x ≤ 6 (d) −5 ≤ x ≤ 0 (e) x ≥ 2 (f ) No values of x
3. (a) x ≤ −2 and x ≥ 2 (b) x ≥ 0 (c) 0 < x ≤ 9 (d) x ≤ −1 and x ≥ 1 (e) x ≥ 1
4. k ≥ 3 [ f 0 (x) = 3x 2 +6x +k. This is always ≥ 0 if the discriminant is ≤ 0]

1
5. λ ≤ − 3 [Similar reasoning as Question 4]

10.2.7 Using differentiation

Before the end of this section we’ll be solving some real life problems that are
impossible without calculus.
The first step towards this is approaching gradient calculations from the
other end, that is finding the points on a graph which have a particular
gradient. Here is an example.
Example 8
Find the coordinates of the points on the curve y = f (x) at which the
gradient is −7, where
f (x) = x 3 − 4x 2 − 3x + 2.
Where is the gradient zero?
Solution
We’ll need the derivative function:

dy
= f 0 (x) = 3x 2 − 8x − 3.
dx
The gradient is −7 when
3x 2 − 8x − 3 = −7
3x 2 − 8x − 4 = 0
(3x − 2)(x − 2) = 0
2
i.e. when x = 2 and 3 .
2 40
The coordinates10 of the two points are (2, −12) and ( 3 , − 27 ).
For the gradient to be zero

3x 2 − 8x − 3 = 0
(3x + 1)(x − 3) = 0
1
i.e. when x = 3 and − 3 .
1 68
The coordinates11 of the two points are (3, −16) and (− 3 , 27 ).
We call these turning points.
10
The y-coordinates are f (2) and f ( 23 ).
11
The y-coordinates are f (3) and f (− 13 ).

10.2.8 Tangents, Normals and Turning Points

Tangents and Normals
In this section we’ll be constructing the equations of a lot of straight lines. If
you want a reminder about how we do this, see Chapter 5. (If it’s just the
formulae you need you’ll find them on page 5-27.)
Example 9
y
Find the equation of the 6
tangent to the curve
y = 3x 2 − 5x + 1
at the point (2, 3). •

(2, 3)

Find where the tangent
crosses the two coordinate

axes.
-x
Solution
dy
dx
= 6x − 5.
dy
When x = 2, d x = 12 − 5 = 7. The gradient of the tangent is 7.
The line passes through (2, 3), so its equation is
y − 3 = 7(x − 2)
y − 3 = 7x − 14
y = 7x − 11.
This line crosses the y-axis at (0, −11) and the x-axis at ( 11
7 , 0).

A tangent is a straight line that touches a curve,
a normal is a straight line that crosses a curve at right angles,
so it is perpendicular to the tangent.
In Chapter 5 we saw that if we have a line, it’s easy to find the gradient of any
line perpendicular to it: the product of their gradients is −1 (see page 5-34).
Example 10
y
Find the equation of the 6
normal to the curve @
@
y = x 3 − 4x 2 + 7x + 2
@
@• P
@
at the point P , whose x- @
coordinate is 1. @
@
-x
Solution
When x = 1, y = 1 − 4 + 7 + 2 = 6, so P is the point (1, 6).

dy
dx
= 3x 2 − 8x + 7.
dy
When x = 1, d x = 3 − 8 + 7 = 2.
1
The gradient of the tangent at P is 2, so the gradient of the normal is − 2 .
The equation of the normal at P is
1
y − 6 = − (x − 1)
2
2y − 12 = −(x − 1) = −x + 1
2y = 13 − x.
or 2y + x = 13.
In a question like this we don’t really need a diagram at all. And if we draw
one it doesn’t have to be accurate.

Example 11
Show that the point P (2, 8) lies on the curve y = 6x − x 2 .

(a) Sketch the curve and mark the point P .
dy
(b) Determine d x and use it to find where tangent at P crosses the
coordinate axes.
(c) Find the equation of the normal at P .
(d) The normal to the curve at P intesects the curve again at Q. Find the
coordinates of Q.
Solution
x = 2, y = 8 satisfies the equation y = 6x − x 2 .

(a) The curve crosses the x-axis where x(6 − x) = 0, i.e. at x = 0 and 6. Since
the coefficient of x 3 is negative the curve looks like this.
y
6

H
HH
H
HH
P (2, 8)
•H
HH
HH
HH
H
HH
HQ
HH
H

0 - x
6

dy
(b) d x = 6 − 2x, so the gradient at P is 6 − 4 = 2.
The tangent at P is y − 8 = 2(x − 2), which simplifies to y = 2x + 4,
which intersests the axes at (0, 4) and (−2, 0).
1
(c) The normal at P is perpendicular to the tangent so its gradient is − 2 , so
1
its equation is y − 8 = 2 (x − 2), which simplifies to 2y + x = 18.
(d) Solving the equation of the normal and the curve simultaneously:
2(6x − x 2 ) + x = 18
12x − 2x 2 + x = 18
0 = 2x 2 − 13x + 18
(x − 2)(2x − 9) = 0
9
x = 2 or x =
2
The solution x = 2 is hardly surprising!
Substituting x = 29 into the equation of the curve y = 27 9 27
4 , so Q is ( 2 , 4 ).

Example 12
The function f (x) is defined by
x2 + 3
f (x) = p , where x > 0.
x
(i) Show that

3(x 2 − 1)
f 0 (x) =3
.
2x 2
(ii) Show that the tangent to the curve y = f (x) at A(9, 4) passes through the
origin O.
Solution
(i) First we need to write f (x)as two deperate terms in index form:
x2 3
f (x) = 1
+ 1
x 2 x2
3 1
= x 2 + 3x − 2
3 1 3 −3
f 0 (x) = x2 − x 2
2 2
3
3x 2 3
= − 1
2 2x 2
3x 2 3
= 1
− 1
2x 2 2x 2
3(x 2 − 1)
= 3
2x 2
3(8) 4
(ii) The gradient of the curve at A is f 0 (9) = 2(27) = 9 .
But this is the gradient of the line O A, so the tangent passes through the
origin. (We don’t need to find the equation of the tangent.)

Example 13
The tangent to the curve y = x 2 (x + a) at (−1, b) is parallel to the straight line

5x + y = 2, where a and b are constants.
Find the values of a and b.
Solution
A sketch is not necessary in this case.

Rearranging the equation of the line: y = −5x + 2, the gradient of the tangent
we are looking for is −5.
dy
The curve is y = x 3 + ax 2 , so its gradient function is d x = 3x 2 + 2ax.//[6pt] At
dy
x = −1, d x = 3 − 2a, which we know to be −5. Hence a = 4.
The function is therefore y = x 2 (x + 4), so b = (−1)2 (−1 + 4) = 3.

Example 14
The diagram shows the function

1 y
y =x+ .
x 6
(a) Find the equation of the B
normal to the curve at the point
A(2, 52 ).
(b) Find the coordinates of the

point B where the normal cuts the
curve again. A
(c) Find the smallest value y can

take, when x > 0. -x
[It’s an interesting curve, isn’t it? It

has two branches. Can you see that
for very small values of x it is going
to have a graph very close to that
of y = x1 , and for very large values
of x it will get increasing close to
y = x. The line y = x is therefore an
asymptote.]
Solution
dy
(a) y = x + x −1 so d x = 1 − x −2 .
The gradient when x = 2 is 1 − 14 = 43 , so the gradient of the normal at A is − 43 ,
and its equation is
5 4
y− = − (x − 2)
2 3
6y − 15 = −8(x − 2)
6y = 31 − 8x
6
which crosses the curve where 6x+ x = 31 − 8x.
This simplifies to 14x 2 − 31x + 6 = 0.
We know one of the roots of this equation must be x = 2, so factorising is
easy: (x − 2)(14x − 3) = 0.
3 205
The coordinates of B are ( 14 , 42 ).
dy 1
(c) For values of x > 0, y is smallest when d x = 1− x 2 = 0, i.e. x = 1, and the
smallest value of y is 2.
This leads us naturally onto our last topic in this introduction to

differentiation.

10.2.9 Stationary values and turning points

Graphically, a stationary value is any point where the gradient is zero. It
could be a “peak"—where it stops going up and starts to go down, a
“trough"—where it stops going down and start going up, or where it
momentarily stops gong up (or down) and then continues in the same
direction.
A turning point is a stationary value for which the direction of the curve
changes. We call such a point a local maximum or local minimum. “Local"
because there may be other points on the curve that are above the local
maximum or below the local minimum. We reserve the terms global
maximum and global minimum for such points. We often drop the “local"
since it is normally implied.
The third type of stationary value or stationary point we call a point of
inflexion12 . The simplest example of this is y = x 3 , where the gradient is
positive on both sides of the origin and zero at the origin.
• • •
MAXIMUM MINIMUM POINT OF INFLEXION
Strictly speaking we should call the first two as a local maximum and a local
minimum, because there may be larger / smaller values elsewhere on the
curve.
What all three have in common is that
at a turning point the gradient is zero.
So
dy
to find turning points we solve the equation f 0 (x) = d x = 0.
In practical problems, finding a turning point is often about making an

optimal decision. For example what to spend on marketing to maximise
profits, or a design choice that minimises manufacturing costs.
In each case, provided we can formulate the problem as a mathematical
function13 , differentiating it and equating the answer to zero will give us
what we need.
We’ll meet some practical examples later.
12
Later we’ll see that, while a turning point may be a point of inflexion, a point of inflexion
isn’t necessarily a turning point. To explain this clearly we’ll need a bit more calculus.
13
In real life the decision is likely to depend on a number of factors, in which case the
formula would contain more than one variable. At this level we confine ourselves to the
single variable case.

Example 15
Find the coordinates of the turning points on the curve
y = 2x 3 − 3x 2 − 12x + 5.
Sketch the curve and use it to determine which turning point is a minimum
and which is a maximum. (It is not necessary to mark where the curve
crosses the x-axis.)
Solution
dy
The turning point are where d x = 6x 2 − 6x − 12 = 0.
6(x 2 − x − 2) = 0
x2 − x − 2 = 0
(x + 1)(x − 2) = 0
x = −1 or x = 2
when x = −1, y = 12 and when x = 2, y = −15.

The turning points are at (−1, 11) and (2, −15).
The coefficient of x 3 is positive so the curve is this way up:
and it passes through (0, 5).
y
6
(−1, 11)
- x
(2, −15)
From the sketch (−1, 11) is a (local) maximum and (2, −15) is a (local)
minimum.

Example 16
(i) Show that the line y + 6x + 2 = 0 intersects the curve y = x 3 − 9x at (−2, 10).
Show also that it touches the curve at (1, −8).
(ii) Draw a sketch showing where the curve cross the axes, and where the line
and curve meet.
(iii) Find the turning points of the curve and mark them on your sketch.
Solution
(i) Notice the question doesn’t say “find where the line intersects the curve”.
All we have to do is verify what we are told, which we can easily do:
x = −2, y = 10 satisfies both y + 6x + 2 = 0 and y = x 3 − 9x. So the line
intersects the curve at (−2, 10).
The values x = 1, y = −8 also satisify both equations. If the line is to touch
the curve at this point, the gradient of the line must be the same as the
gradient of the curve at (1, −8).
The equation of the line can be written y = −6x − 2, so its gradient is −6.
dy dy
For the curve d x = 3x 2 − 9, so when x = 1, d x = 3 − 9 = −6.
The line touches the curve at (1, −8).
(ii) Try sketching the curve for yourself. If you want a reminder about
sketching polynomials see the first section of Chapter 8.

We have a cubic curve whose roots are where

x 3 − 9x = x(x 2 − 9) = x(x + 3)(x − 3) = 0
i.e. at x = 0, 3 and −3.
The x 3 coefficient is positive which tells us which way up it is.
y
6
A(−2, 10) •
- x
−3 0 3
B (1, −8)
•
•
(iii) The turning points are where the gradient is zero.

dy dy
dx
= 3x 2 − 9, so d x = 0 where
3x 2 − 9 = 0
x2 − 3 = 0
p
x = ± 3
p p
When x = − 3, y = x(x 2 − 9) = −6 3 ≈ −10.39.
p p
When x = + 3, y = x(x 2 − 9) = +6 3 ≈ +10.39.
Notice that the maximum and minimum points are not midway between the
roots (they are not at x = ±1.5).
From the positions of the roots and turning points we can see that the curve
has 180◦ rotational symmetry14 about the origin.
Notice too that, although no sketch has to be very accurate, if it’s too rough it
will be hard to see what is going on. In this case B and the minimum point
are close but not together.
14
This is true for all cubic curves, although not generally about the origin.

d2y
10.2.10 Second derivative f 00 (x) or d x2
Calculus is beautiful. It provides such concise, elegant solutions to so many
otherwise intractable problems. We shall see about one here: deciding
whether a stationary points (where the gradient is zero) is a turning point (a
local minimum or a local maximum) or a point of inflexion. You may wonder
why that is a problem. That’s because for all the examples we have looked at
we’ve known what the curve looks like; our yes tell us. In practice we often
won’t so we need a way of deciding that doesn’t rely on a sketch of the curve.
Take the function f (x) = 31 − 18x 2 + 8x 3 − x 4
Attempting to see what the curve looks like would be very tedious because it
doesn’t factorise. We would have to plot points.
Lets see if we can find where the turning point are:
f 0 (x) = −36x + 24x 2 − 4x 3
= −4x(x 2 − 6x + 9)
= −4x(x − 3)2
There are turning points at x = 0 and x = 3.
Before we learn how to distinguish the type of stationary value simply, using
calculus, there is a crude numerical way of finding out.
If there is a local minimum at x = a, f 0 (a) = 0 but just to the left of a we
would expect to find that the gradient positive, and just to the right of a it
would be negative.
Similarly if there is a local minimum at x = a, f 0 (a) = 0 but just to the left of a
we would expect to find that the gradient negative, and just to the right of a
it would be positive.
At a point of inflection x = a, the gradient would be positive on both sides or
negative on both sides.
In our example we will see what the gradients are a distance 0.1 either side of
the stationary value, given that f 0 (x) = −4x(x − 3)2 .
a f 0 (a − 0.1) f 0 (a) f 0 (a + 0.1)

0 −4(−0.1)(−3.1)2 > 0 0 −4(0.1)(−2.9)2 < 0 local maximum
3 −4(2.9)(−0.1)2 < 0 0 −4(3.1)(0.1)2 < 0 point of inflection
Notice that we don’t have to find the values, just see what the sign will be.
This may seem a perfectly reasonable way of distinguishing stationary
values. Not to a mathematician. Always after absolute rigour, she would say
“how do we know the curve doesn’t do something unexpected within ±0.1 of
the stationary point?". In general we don’t of course. Remember we are not
just talking about nice smooth polynomial curves; we could be dealing with
any kind of function, for example ones with asymptotes.

In the sketches below, the top row shows a (local) maximum, minimum and
point of inflexion. On each of them we have marked where the gradient of
f (x) is positive and where it is negative (with a + sign or a − sign).
At a maximum f 0 (x) (the gradient) goes from + to 0 to −, at a minimum it
goes from − to 0 to + and at a point of inflexion it goes from + to 0 and back
to +. (A point of inflexion could equally go from − to 0 and back to −.)
The row below plots the behaviour of f 0 (x). For a maximum f 0 (x) goes from
so + to 0 to −, and so on.
No look at the gradient of the f 0 (x) function at the stationary value: for a
maximum it is negative, for a minimum it is positive and for a point of
inflection it is zero.
f (x)
f (x) f (x)
+
0 0
+ − − + +
f 0 (x) f 0 (x) f 0 (x)
MAXIMUM MINIMUM POINT OF INFLEXION
We can distinguish between the different types by differentiating f 0 (x) and

finding whether it is positive, negative of zero at the stationary point.
The differential or derived function for f 0 (x) we write as f 00 (x).
dy d2y
Or in notation, we write15 . d x 2 .
dx
To summarise, if there is stationary value at P , for which x = a
P is a local maximum if f "(a) < 0

A
P is a local minimum if f "(a) > 0
!
A
A
P is a point of inflexion if f "(a) = 0 A
15
dy d
The reason we write it like this is because we can think of as y.
dx dx
dy d d d 2
¡ ¢
Then the derivative of is y or y.
dx dx dx dx

In Example 15 we found, using a sketch that the function

y = 2x 3 − 3x 2 − 12x + 5 has a (local) maximum at x = −1 and (local) minimum
and x = 2.
y = f (x) = 2x 3 − 3x 2 − 12x + 5
dy
= f 0 (x) = 6x 2 − 6x − 12 = 6(x + 1)(x − 2)
dx
d2y
= f 00 (x) = 12x − 6 = 6(2x − 1)
d x2
f 00 (−1) = 6(−2 − 1) < 0 ⇒ maximum
f 00 (2) = 6(4 − 1) > 0 ⇒ minimum
At the start if this section we used a crude numerical method to distinguish

the stationary values of the function f (x) = 31 − 18x 2 + 8x 3 − x 4 . Lets use the
second derivative method to confirm those results.
y = f (x) = 31 − 18x 2 + 8x 3 − x 4
dy
= f 0 (x) = −36x + 24x 2 − 4x 3
dx
= −4x(x 2 − 6x + 9)
= −4x(x − 3)2 turning points at x = 0 and x = 3
2
d y
2
= f 00 (x) = −36 + 48x − 12x 2 = −12(x 2 − 4x + 3)
dx
= −12(x − 1)(x − 3)
f 00 (0) = −36 < 0 ⇒ maximum
f 00 (3) = 0 ⇒ point of inflexion

Example 17
Find the position and nature of the stationary values on the curve
3
y= + 16x 3 (x 6= 0).
x
Sketch the curve,
Solution
We switch to f (x) notation.
f (x) = 3x −1 + 16x 3
f 0 (x) = −3x −2 + 48x 2
3(16x 4 − 1)
=
x2
3(4x 2 + 1)(4x 2 − 1) 1
= stationary values at x = ±
x2 2
f 00 (x) = 6x −3 + 96x
6 + 96x 4
= the sign of which depends only on the denominator
x3
µ ¶
00 1 1
f − < 0 ⇒ maximum at x = −
2 2
µ ¶
1 1
f 00 > 0 ⇒ minimum at x =
2 2
f − 12 = −8 and f¡ 12 ¢= 8, so we have a local maximum at − 12 , −8 and a

¡ ¢ ¡ ¢ ¡ ¢
local minimum at 21 , 8 .
Plotting first the positions of the stationary values, we can also see that x = 0
is an asymptote and y → ∞ as x → ±∞.
3+16x 4
The curve does not cross the x axis anywhere because y = which
x
4
can never be zero since 3 + 16x can never be zero.
y
20
10
¡1 ¢
2
,8
0 x
−1 −0.5 0.5 1
¡ 1 ¢
− 2 , −8
−10
−20

Example 18
A solid cylinder has a radius r m and volume 2000 m3 .
Show that its total surface area A is given by
4000
A = 2πr 2 + .
r
Find the radius for which the cylinder has the minimum surface area.
Solution
We need to introduce a new variable. Let the height of the cylinder be h.
2000
The volume is therefore: πr 2 h = 2000, i.e. h = .
πr 2
A = 2πr 2 + 2πr h
2000
= 2πr 2 + 2πr ×
πr 2
4000
= 2πr 2 +
r
In this case A is a function of r , so for the minimum surface area we need to
dA
find where = 0
dr A = 2πr 2 + 4000r −1
dA
= 4πr − 4000r −2
dr
1000
= 0 when r 3 =
π
10
i.e. r = p
3
π
You may be wondering how we could assume this gives the minimum value
not the maximum value for for the surface area. The answer is that there is
no upper limit the value of A. If we look at the formula for A, A → ∞ as
r → 0; we can make it as large as we wish.
d2A 8000
Alternatively we can find the second derivative: = 4π+ 3 > 0 for all
dr 2 r
values of r (because r > 0). The stationary value must therefore be a
minimum point.

Example 19
Sketch the gradient function for each of the curves below. The dotted lines
mark the positions of the stationary points. For the curve shown in (b) the
x-axis is an asymptote.
(a) (b)
f (x) f (x)
x x
−3 −2 −1 0 1 2 −2 −1 0 1 2 3 4
Solution
(a) There are stationary points at (b) There are turning points at
x = −2 and x = 1, so the gradient x = −1 and x = 1, so the gradient
function meets the x-axis at these function meets the x-axis at these
values. values.
Because the gradient is positive Because its is an asymptote, as

either side of x = −2 (the point x → ∞, the f (x) curve gets closer
of inflexion), the gradient function to the x-axis so f 0 (x) → 0.
touches the x-axis there.
f 0 (x) f 0 (x)
x x
−3 −2 −1 0 1 2 −2 −1 0 1 2 3 4

Example 20
The function y = f (x) is defined by
µp ¶2
x −1
f (x) = (x > 0)
x
1 p
Show that f 0 (x) = − 3 (x − 3 x + 2)
x
Hence find the coordinates of the stationary points and determine their
nature.
Solution p
x −2 x +1
f (x) =
x2
3
= x −1 − 2x − 2 + x −2
5
f 0 (x) = −x −2 + 3x − 2 − 2x −3
1
−x + 3x 2 − 2
=
x3
p
−(x − 3 x + 2) −(X 2 − 3X + 2) p
= = where X = x
x3 X6
−(X − 2)(X − 1)
=
X6
p p
−( x − 2)( x − 1)
=
x3
= 0 when x = 1 or x = 4
µ ¶
1
i.e. at (1, 0) and 4,
16
15 − 7
f 00 (x) = 2x −3 − x 2 + 6x −4
2
1
f 00 (1) = ⇒ minimum at (1, 0)
2
µ ¶
00 1 1
f (4) = − ⇒ maximum at 4,
256 16

QQ
1. For the following curves find the coordinates of any point for which the
gradient is zero.
In each case establish whether the point is a local maximum, a local
minimum or a point of inflexion.
(a) y = 2x 2 + 6x (b) y = 11 + x − x 2 (c) y = x 3 − x 2 − x − 5
1 16
(d) y = x(x 2 − 3x + 3) (e) y = x + (f) y = x 2 +
x x
p x+1
(g) y = 4 x − x (h) y = x 4 − 8x 2 (i) p
x
2. Sketch the gradient function for each of these curves, indicating any
points where the gradient is zero. The answers are over the page.
(a) (b)
f (x)
f (x)
x
−3 0 3
(c) (d)
f (x) f (x)
x
−1 1 x
1
16
(a) (− 32 , − 92 ) min (b) ( 12 , − 45 1 130
4 ) max (c) (− 3 , − 27 ) max, (1, −6) min
(d) (1, 1) pt of inflexion (e) (1, 2) min, (−1, −2) max (f ) (2, 12) min
(g) (4, 4) max (h) (0, 0) max, (−2, −16) min, (2, −16) min (i) (1, 2) min

(a) (b)
f 0 (x)
f 0 (x)
x x
−3 0 3
(c) (d)
f 0 (x) f 0 (x)
x
−1 1 x
1

A delightful little theorem

We are going to show you that a very simple idea can be surprisingly useful.
We introduce it here because it gives us a method of finding the (local)
minimum point of the function
1
y =x+
x
that doesn’t require calculus.
The simple idea is this: for any number X , it must follow that X 2 ≥ 0.
Furthermore X 2 = 0 only if X = 0. Boring, eh? Too obvious? Well let’s look at
an example.
Suppose a and b are any two positive numbers.

p p
Let X = a − b. It follows that
p p
( a − b)2 ≥ 0
p
a + b − 2 ab ≥ 0
p
a + b ≥ 2 ab
a +b p
≥ ab
2
The left hand side is the arithmetic mean of a and b. The right hand side we
call the geometric mean.
We can check this out with a few pairs of (positive) numbers.
a+b p
For example if a = 1 and b = 9, = 5 and 1 × 9 = 3.
2
a+b p p p
little theorem also tells us that and ab are only equal, when a = b,
2
i.e. when a = b.
2a p 2
We can easily check this algebraically. Putting b = a we have = a .
2
In fact the theorem extends to any set of positive numbers (although is not
so easily proved in the general case):
a1 + a2 p
≥ a1 a2
2
a1 + a2 + a3 p
3
≥ a1 a2 a3
3
a1 + a2 + a3 + a4 p
4
≥ a1 a2 a3 a4
4
and so on. The two sides are equal only when all the numbers are equal.

We said we’d use the theorem to find, without using calculus, the minimum
value for positive x, of the function
1
y =x+ .
x
(See Example 12.)
p 1
Putting X = x−p
x
¶2
p
µ
1
x−p ≥ 0
x
1
x −2+ ≥ 0
x
1
x+ ≥ 2 for all values of x > 0.
x
We also know that X 2 = 0 only when X = 0.

1 x−1
That is to say x + is equal to 2 only when p = 0, i.e, when x = 1.
x x
This is precisely what we established in part (c) of Example 12: for values of
1
x > 0 the minimum value of x + is 2.
x

10.2.11 Practical examples

Inflation
Here is a practical example that may help to make it all a bit more real. Take
this statement.
“They say inflation is falling but it seems to me that prices are still
going up.”
Does this seem like a reasonable observation? It isn’t.

In function terms we have a variable—let’s say the Retail Price Index
(RPI)—changing over time. Let’s call this variable p and plot its value over
time t (in months, say). The graph might look something like this.
p

6

"

"

"
"
"
"
"
"
"
"
"
"
"
"
- t
The Retail Price Index p is represented by the height of the graph at a

particular time t .
Inflation is represented by the rate at which price are increasing:
dp
inflation =.
dt
In other words inflation is shown by the gradient of the graph.
In our example the gradient is falling over time, even though the RPI
continues to rise.

Kinematics
Another example, this time to do with physical motion. First some
conventions.
We normally describe distance by saying how far something has moved from
its starting point, and in which direction. This is referred to as its
displacement, and is measured using the variable s. A variable that captures
an amount and a direction is called a vector.
We use the term velocity, rather than speed, when describing how fast
something is moving for the same reason: velocity is also a vector and to
describe it we need to know its direction as well as its size. We generally use
the variable v.
If the velocity is changing then it has an acceleration, denoted by a.
Even if something is moving in a straight line, direction is important,
because it can end up either to the right or the left of its starting point, and it
can be travelling from left to right or from right to left.
Calculus comes in because
velocity is the rate at which displacement is increasing with time
or in normal language speed is the rate at which distance is changing:
ds
v= .
dt
If the velocity is changing over time, its acceleration is the rate at which it is
increasing:
dv
a= .
dt
By convention we normally regard displacement, velocity and acceleration
to the right as positive but it’s only a convention.
− +
We can think of a negative acceleration as a deceleration.

However it’s usually more helpful to think of it as an acceleration in
the negative direction.

Suppose an object P moves along a line in such a way that its displacement
in metres from the point O t seconds after the start of the motion is given by
s = 40t − 5t 2 = 5t (8 − t ).
P
@ - s
@
O
So for example, after 2 seconds it is 60 m to the right of O.
The graph opposite shows its sm

displacement from O at time t . 6
After 8 secs it is back to where it
started (s = 0), after which it is
to the left of O (s < 0).
Can you see that it stopped and
started going in the opposite
- t sec s
direction after 4 secs, when it 0 4 8
was 80 m from O?
Differentiating s, we have:
v ms −1
ds
v= = 40 − 10t
6
dt
40 Q
the units of which would be Q
Q
metres per second, which we Q
Q
- t sec s
write as m s −1 .
Q
0 4 QQ 8
This is consistent with the Q
Q
object stopping when t = 4,
Q
QQ
and moving in the opposite
direction (v < 0).
The velocity is changing uniformly (it reduces by the same amount every
second). In other words the acceleration is constant—and negative because
it is going down. We could call it a deceleration but that doesn’t really help.
Why? Well, after t = 4 the object moves at an increasing speed in the negative
direction. Is that a negative acceleration or a decceleration? It’s easiest just to
talk about positive or negative acceleration.
We said that acceleration is the rate of change of velocity:
dv
a= = −10.
dt
Once again the Maths confirms what we knew, a constant negative
acceleration. The units are metres per second per second, written m s −2 .
Every second the velocity reduces by 10 metres per second.
Initially it is travelling at 8 ms −1 . After 4 secs its velocity is zero and after a
further 4 seconds it is again traveling at 8 ms −1 but in the opposite direction.

Optimising
Finding the cheapest, the most effective, the best in some clearly defined
way, is one way calculus can sometimes help us. Provided we can express
mathematically what we are trying to optimise in terms of the relevant
factors. Here is an example.
A farmer has 60 m of fencing with which to make three sides of a rectangular

enclosure for some sheep, using an existing wall for the fourth side. What
should the dimensions of the enclosure be to provide the maximise possible
area for the sheep?
Let’s suppose that the width of the enclosure is x, as shown in the diagram.
WALL
x x
60 − 2x
The length of the third side of the enclosure is 60 − 2x, so its area—let’s call it
A—is given by
A = x(60 − 2x).
When making statements like this it can confuse things if we put in the units:
we may have a variable called m. The units are implied; we re-introduce them
when stating the answer.
dA
For a minimum or a maximum value for A, we must have d x = 0
A = 60x − 2x 2
dA
= 60 − 4x
dx
60 − 4x = 0
x = 15
The optimum enclosure is a 15 m by 30 m rectangle, whose area is 450 m 2 .

Because the function we are trying to optimise happens to be a quadratic, we

could arrive at the result without calculus. Completing the square:
A = 60x − 2x 2 = −2[x 2 − 30x]

= −2[(x − 15)2 − 225]
= 450 − 2(x − 15)2
(x − 15)2 ≥ 0 whatever the value of x, and it equals 0 when x = 15.

So A ≤ 450, and it takes its greatest value, 450, when x = 15.
The graph of A looks like this.
A
6
450
- x
0 15 30
Of course x can’t be negative, nor can A, so the parts of the curve that lie below
the x-axis have not practical meaning.

Exercises 37
1. Find the equation of the tangent to the curve y = x 2 +3x −1 at the point
(2, 9).
2. Find the equation of the tangent to the curve y = x(4x − 1) at the point
(−1, 5).
8
3. (a) Find the equation of the normal to the curve y = x at the point (2, 4).
(b) Find where the normal crosses the two coordinate axes.
4. Find the equation of the tangent to the curve y = x 2 + x − 4 at the point

where the curve cuts the y-axis.
5. The curve y = 2x 2 − x 3 meets y

the x-axis at the origin and at 6
P.
Find the x-coordinate of P .
Find also the equation of the P -x
normal to the curve through P . 0
6. Find the values of x for which the gradient of the curve

y = 2x 3 − 9x 2 + 12x − 7 is zero. Hence find the equations of the tangents
to the curve that are parallel to the x-axis.
7. (a) Find the equations of the two tangents to the curve

y = x 2 − 9 at the points where the curve cuts the x-axis.
(b) Find where the two tangents intersect.
8. Find the equation of the tangent to each of the following curves at the
point indicated. Express your answer in a form that does not involve
any fractions.
8
(a) y = x 2 at (2, 2) (b) y = (x + 1)(x + 3) at (1, 8)
x 2 +2 p
(c) y = x at x = 1 (d) y = 8 x at (4, 16)
4
(e) y = x− px at x = 4 (f ) x y = 6 at x = 2
9. (i) Find the coordinates of the points where the curve y = x 2 − 2x − 8

crosses the x-axis and hence draw a sketch of the curve.
(ii) Show that the gradients of the tangents to the curve at these points
are −6 and 6.
(iii) Find the equations of the two tangents, in the form y = mx + c.
(iv) Show that the two tangents intersect at the point (1, −18).
10. The equation of the normal to the curve y = x 2 + 4x − 1 at (−1, −4) is

2y + x + c = 0. Find the value of c.
Find the coordinates of the point where the normal crosses the curve
again.

11. Show that the equation of the normal to the curve y = x 2 − 6x + 10 at

(4, 2) is x + 2y − 8 = 0.
Find the coordinates of the point where the normal intersect the curve
again.
12. A curve has equation y = 2x 3 + 6x 2 − 12x + 7.

(a) Find the equation of the tangent to the curve at (1, 3) in the form
y = mx + c.
(b) show that the equation of the normal to the curve at (1, 3) is
x + 6y − 19 = 0.
13. Find the equation of the normal to each of the following curves at the
point indicated. Express your answer in a form that does not involve
any fractions.
(a) y = 4x 3 − 3x 2 at (1, 1) (b) y = (2x − 3)2 at (2, 1)
4 p
(c) y = +x 2 at x = −2 (d) y = x at (4, 2)
x
27 2+x
(e) y = 2 at x = 3 (f ) y = p at x = 4
x x
14. Find the coordinates of the two points on the curve
y = x 3 − 3x 2 − 4x + 5 at which the gradient is +5.
Show that the gradient is −7 at exactly one point on the curve.
15. The gradient of the curve y = px 2 − 4 at (3, q) is 12. Find the values of p
and q.
16. The tangent to the curve y = x 3 + ax at (2, b) is parallel to the line y =

10x − 3. Find the values of a and b.
4
17. (a) Find the equation of the tangent to the curve y = at P (4, 1),
x
and the coordinates of the point A where the tangent crosses the
y-axis.
(b) Find the equation of the normal to the same curve and the
coordinates of the point B where it crosses the y-axis.
(c) Find the area of the triangle P AB .
(d) Find the coordinates of the point where the normal to the curve
meets the curve again.
18. Find the values of a and b such that the function f (x) = x 2 − 4x + 5 can
be expressed as f (x) = (x + a)2 + b. Sketch the curve y = f (x).
The line y = 2x crosses the curve y = f (x) at P and Q. Show that the
tangents to the curve at P and Q meet on the line x = 3.
p p
19. The curve y = x 4 − 8x 2 meets the x-axis at (0, 0), (a 2, 0) and (b 2, 0),
where a < b.
(a) Find the values of a and b.
(b) Find also the coordinates of the turning points and hence sketch
the curve.

20. The curve of the function y = ax 2 + bx + c passes through (1, 7) and has
a turning point at (3, 11). Find a, b and c.
128
21. Find the coordinates of the turning point on the curve y = x 2 + p .
x
Determine whether it is a maximum or a minimum.
1
22. Show that the function y = x 2 − 2 x 6= 0 is increasing for x > 0 and
x
decreasing for x < 0.
23. Find the coordinates of the stationary points on the curve

y = (x − 4)(x − 1)2 . From a sketch determine whether each of them is a
maximum or a minimum point.
24. Find the coordinates of the turning points for the function y = x 4 −4x 3 −
d2y
8x 2 + 7. By finding or from a sketch distinguish the nature of each
d x2
turning point.
1 2
25. Show that the function f (x) = + 3 (x 6= 0) is decreasing for all non-
x x
zero values of x.
x 2 +8
26. Find the coordinates of the turning points on the curve y = ,
p p x
expressing you answers in the form (a 2, b 2).
Determine the nature of each turning point.
27. Show that x −1 p

p = x − 1.
x +1
Hence find the coordinates of the stationary point of the function
x 2 −x
f (x) = p .
x+1
28. For what values of a is the function f (x) = x 3 + ax(x + 1) increasing for
all values of x.
29. Find in terms of a the coordinates of the stationary points of the

¡ a ¢ ¡ x ¢2
function y = 16 − .
x a
Show that it is a minimum point whatever the value of a.
30. Show that f (x) = x n + nx, where n ≥ 2, has no stationary values if n is

odd. Find the coordinates of the stationary point if n is even and show
that it will always be a minimum point.
p 1 dy 1
31. If y = x + p (x > 0) show that can be written x k (x − 1).
x d x 2
d2y
Find the value of k. Show that = 0 when x = 3.
d x2
2
32. If x and h are in the appropriate units sin(x +h) ≈ sin x +h cos x − h2 sin x
for small values of h. Show that the derivative of sin x is cos x.
2
33. For small values of h, e h ≈ 1 + h + h2 .
If f (x) = e x , by writing e x+h as e x e h show that f 0 (x) = e x .
Find from first principles the derivative of f (x) = e ax where a is a
constant.

34. A man has a piece of wood 8 m long with which he intends to make a
football goal for his small son. It will have two (equal) vertical pieces
and one piece across the top. To make it as easy as possible for his son
he wants the area of the goal to be as large as possible.
If the height of each vertical piece is x m and the area of the goal is
y m 2 , write down an expression for y in terms of x. (You may ignore
the thickness of the wood.)
What should the height of the goal be to maximise its area?
35. A company finds that its annual profit depends on how much it spends
on marketing. If it spends £x on marketing, it makes a profit £P (both
measured in thousands), where
P = 25x 2 (x − 15).
What should the company spend on marketing to maximise its annual
profit?
36. A rocket is fired from the ground vertically into the air. Its height y, after
t seconds, is given by
y = 60t − 5t 2 .
(a) What is the height of the rocket after 2 seconds?
(b) Find a formula for the velocity of the rocket after t seconds.
(c) After how long does the rocket reach its maximum height?
(d) What is the greatest height the rocket reaches?
37. The diagram shows a rectangle ABC D. The parallelogram PQRS is

such that AP = BQ = C R = DS = x.
D R C
A x
x A
A
SA A
A A
A AQ
A
A x
x A
A P B
(a) If AB = 10 and BC = 6, find in terms of x the area of PQRS.

(b) Find the value of x for which the area of PQRS is as small as possible,
and find that area.
(c) Repeat part (b) if AB = a and BC = b.
38. When we buy goods they often lose their value over time. We call this
depreciation and the loss in value the depreciation cost.
3,200
A particular car costing £20, 000 is expected to depreciate by £ x in
year x. As the car gets older it also becomes more expensive to
maintain, costing £ 200x to maintain in year x.
Find the combined cost of depreciation and maintenance in years 1
and 2?
The owner wants to sell the car before the combined cost of
depreciation and maintenance starts rising. Use calculus to determine
in which year should she sell the car?

39. A sheep enclosure is to be erected alongside a section of wall (marked

in blue) that contains a 4 m gate (marked in red), as shown in the
diagram.
2m
4m
xm
ym
(i) The enclosure is to be made using 22 m of fencing. If it is y m long

and x m from the wall at its widest point, show that y = 24 − 2x.
(ii) Show that the area A of the enclosure is given by the formula
A = −2x 2 + 24x − 8.
(iii) By differentiating A, find what x and y should be if the enclosure is
to have the greatest possible area.
40. The curve y = x 2 (x + λ) crosses the x-axis at the origin and at P , where
λ is a positive constant.
(i) Sketch the curve.
(ii) Find, in terms of λ, the equation of the tangent to the curve at P .
The tangent meets the curve again at Q. Find, in terms of λ, the
coordinates of Q.
p k
41. (a) Show that the gradient of the curve y = k x at x = 1 is 2 .
(b) Hence show that the equation of the tangent to the curve at x = 1 is
2y = k(x + 1).
(c) Show that the tangent passes through (−1, 0) for all values of k.
(d) Find the value of k for which the tangent also passes through (3, −1).
42. (i) Show that the line y = x touches the curve y = x(x − 1)(3 − x) at the
point (2, 2).
(ii) On the same diagram sketch the line and the curve.
(iii) Hence explain why the curve has a stationary point at (a, b), where
2 < a < 3 and b > 2.
43. Find the x-coordinates of all the points where the line y = 5 meets the
curve 4
y = x2 + 2 .
x
Find also the coordinates of the turning points of the curve. Verify that
they are both local minima and sketch the curve.
Without any further differentiation find the coordinates of the turning
points of the curve
2 2
µ ¶
y = x+ .
x

44. The functions f (x) and g (x) are defined thus
f (x) = x 2 and g (x) = 3x 2 − x 3 .
(a) Find the coordinates of the two points where the curves y = f (x)
and y = g (x) intersect.
(b) Verify that both these points are stationary points on the curve
y = g (x). Hence sketch the two curves in the same graph.
(c) Solve the equation
f 0 (x) = g 0 (x).
Interpret the solution in terms of the gradients of the two curves.
45. (a) Show that (x − 1)2 (x + 2) = x 3 − 3x + 2.

(b) Find the equation of the tangent to the curve y = 2x 3 at x = 1.
(c) Find the coordinates of the point where the tangent meets the
curve again.
46. The diagram shows part of the y

curve y = f (x), for x > 0, where
6
x2 + 4
f (x) = .
x
(a) Find the x-coordinates of
A and B , where the line y = 5
y =5
crosses the curve.
A B
(b) Show that
4
f 0 (x) = 1 − . C
x2
- x
(c) The tangents at A and B meet at C . Find coordinates of C .

(d) What is the smallest value of f (x) for x > 0?
47. The graph shows the curve y

p 6
y =2 x
A
The line x = 4 intersects the
curve at A.
-x
B 0 4 C
(i) Show that the equation of the tangent to the curve at A is

2y = x + 4,
(ii) Find the equation of the normal to the curve at A.
(iii) The tangent crosses the x-axis at B and the normal crosses it at C .
Find the area of the triangle ABC .

48. The graph shows the curve

y
16
2
y =x + . 6
x
The
p curve crosses the x-axis at
3
(a 2, 0), where a is an integer.
(a) Find the value of a.
(b) Find the coordinates of the
turning point of the curve.
(c) Show that the equation of
-x
the tangent to the curve at
(4, 20) is
y = 7x − 8.
(d) Use the graph to write
down how many solutions
there are to the equation:
16
x2 + = 7x − 6.
x

Solutions 37
dy
1. y = 7x − 5 [ d x = 2x + 3 = 7 when x = 2.]
dy
2. y = −9x − 4 [ d x = 8x − 1 = −9 when x = −1.]
dy
3. (a) 2y = x + 6 [dx = − x82 = −2 when x = 2.]
(b) (0, 3), (−6, 0)
dy
4. y = x −4 [y = −4, d x = 2x + 1 = 1 when x = 0.]
dy
5. x = 2, 4y = x − 2 [ d x = 4x − 3x 2 = −4. Grad of normal = 14 .]
dy
6. x = 1 and 2. y = −2, y = −3 [ d x = 6(x 2 − 3x + 2).]
dy
7. (a) y = 6(x − 3), y = −6(x + 3). [Cuts at x = ±3. d x = 2x = ±6.]
(b) (0, −18) [By symmetry they meet on the y-axis]
dy
8. (a) y = 6 − 2x [dx = − 16
x3
]
dy
(b) y = 6x + 2 [ d x = 2x + 4]
dy 2
(c) y = 4 − x [ d x = 1− x 2 ]
dy
(d) y = 2x + 8 [dx = p4x ]
dy 2
(e) 4y = 5x − 12 [ d x = 1+ 3 ]
x2
dy 6
(f) 2y = 12 − 3x [ d x = − x2 ]
dy
9. (i) (4, 0), (−2, 0) [(ii) d x = 2x − 2]
(iii) y = 6x − 24, y = −6x − 12 [(iv) e.g. add the two equations]
y
6
- x
−2 4
−8
1
10. c =9 [The normal is y + 4 = − 2 (x + 1)]
(− 27 , − 11
4
) [They intersect where 2x 2 + 9x + 7 = 0.]
1
11. ( 32 , 13
4
) [Normal is y − 2 = − 2 (x − 4), 2x 2 − 11x + 12 = 0]
1
12. (a) y = 6x − 3 [(b) Normal is y − 3 = − 6 (x − 1)]

dy
13. (a) 6y + x = 7 [ d x = 12x 2 − 6x]
dy
(a) 4y + x = 6 [ d x = 8x − 12]
dy 4
(c) 5y = x + 12 [ d x = − x 2 +2x]
dy 1
(d) y + 4x = 18 [ d x = 2p x ]
dy 54
(e) 2y = x + 3 [ d x = − x3 ]
dy 1 1
(f) y = −8x − 35 [dx = − 3 + 1 ]
x2 2x 2
14. (−1, 5), (3, −7) [Two solutions of 3(x 2 − 2x − 3) = 0. Only one to (x − 1)2 = 0]
dy
15. p = 2, q = 14 [ d x = 2px ⇒ 12 = 6p. y = 2x 2 − 4.]
dy
16. a = −2, b = 4 [ d x = 3x 2 + a ⇒ 12 + a = 10. y = x 2 − 2x.]
dy 4 1 1
17. (a) 4y + x = 8, (0, 2) [ d x = − x 2 = − 4 . y − 2 = − 4 (x − 0)]
(b) y = 4x − 15, (0, −15) [y + 15 = 4(x − 0)]
1
(c) 34 sq units. [ 2 AB × 4]
4
(d) (− 41 , −16) [4x − 15 = x ⇒ 4x 2 − 15x − 4 = (x − 4)(4x + 1) = 0]
18. a = −2, b = 1
[x 2 − 4x + 5 = 2x ⇒ (x − 1)(x − 5) = 0. P (1, 2), Q(5, 10)]
y dy
[ d x = 2x − 4. y = 4 − 2x and y = 6x − 20]
6
P
(2, 1)
-x
p p
19. (a) a = −2, b = 2 [x 2 (x 2 − 8) ⇒ x = 0 or x = ± 8 = ±2 2]
dy
(b) (0, 0), (−2, −16), (2, −16) [ d x = 4x(x 2 − 4)]
y
6
p p - x
−2 2 −2 0 2 2 2
−32
20. a− = −1, b = 6 and c = 2

21. (4, 80) min
dy ³
1
´
22. = 2x 1 + x4
which has the same sign as x.
dx

23. (1, 0) max, (3, −4) min [coefft of x 3 positive]

24. (−1, 4) min,(0, 7)max,(4, −121)min
1
25. f 0 (x) = − 4 (x 2 + 6) < 0 for all x 6= 0
x
p p p p
26. (2 2, 4 2) min, and (−2 2, −4 2) max.
p p
x − 1 = ( x − 1)( x + 1). 49 , − 274
¡ ¢
27. min
28. 0 ≤ a ≤ 3 [Discriminant of f 0 (x) must be ≤ 0, i.e. a(a − 3) ≤ 0]
d2y 2
29. (2a, 4). When x = 2a, =
d x2 a2
30. x n−1 + 1 = 0 has no solutions as n − 1 is even. (−1, 1 − n).
f 00 (−1) = n(n − 1)(−1)n−2 ≥ 0 since n ≥ 2 and n − 2 is even.
2
3 d y 1
31. k= − . = (−x + 3)
2 d x 2 4x 52
h 2
sin(x+h)−sin x h cos x− 2 sin x
32. ≈ → cos x as h→0
h h
h2
³ ´
x
f (x+h)− f (x) e h+ 2
33. ≈ → e x as h → 0
h h
f (x + h) = e a(x+h) = e ax e ah , so
a2h2
³ ´
ax
f (x+h)− f (x) e ah+ 2
≈ → ae ax as h → 0
h h
dy
34. y = 8 − 2x. Height = 2 m [A = 8x − 2x 2 , d x = 8 − 4x = 0]
dP
35. £10, 000 [ d x = 75x 2 − 750x ⇒ x = 10 (ignoring solution x = 0)] 36. (a) 100 m
dy
(b) v = 60 − 10t [v = d t ]
dy
(c) 6 secs. [Maximum height when v = 0, i.e. d t = 0]
(d) 180 m. [Value of y when t = 6]
37. (a) A = 60 − 16x + 2x 2 [Total area - area of 4 triangles = 60 − x(10 − x) − x(6 − x)]
dA
(b) x = 4, Area = 28 sq units. [ d x = −16 + 4x = 0. Or by completing the square]
1 dA
(c) x = 4 (a + b) [A = ab − x(a + b) + 2x 2 . d x = 4x − (a + b)]
1 (a+b)2 (a+b)2 8ab−2(a+b)2 +(a+b)2
Area = 8 (6ab − a 2 − b 2 ) [A = ab− 4
+2 16 = 8
]
3200
38. £3, 400, £2, 000 [ x + 200x]
dy 3200
Year 4 [ d x = − x 2 + 200 = 0]
39. (i) (ii) [x + y + (x − 2) = 22, A = x y − 8 = x(24 − 2x) − 8]

dA
(iii) x = 6, y = 12 [ d x = −4x + 24]
dy
40. (ii) y = λ2 (x + λ), (λ, 2λ3 ) [ d x = 3x 2 + 2λx = 3λ2 − 2λ2 = λ2 when x = −λ
then x 2 (x + λ) = λ2 (x + λ) ⇒ x = −λ or x = ±λ]
dy k k
41. (a) (b) (c) [(a) d x = 2px (b) y − k = 2 (x − 1) (c) y = 0 ⇒ (x − 1) = 0]
1
(d) k = − 2 [2(−1) = 4k]

dy
42. (i) [y = −x 3 + 4x 2 − 3x, d x = −3x 2 + 8x − 3 = 1 when x = 2
and (2, 2) also lies on y = x, which has gradient 1.]
(ii)
y
6
(2, 2)
- x
0 1 3
(iii) max point is between

x = 2 and x = 3,
and above (2, 2).
43. x = 1, −1, 2, −2 [x 4 − 5x 2 + 4 = 0, or (X − 4)(X − 1) = 0, where X = x 2 ]

p p dy 8
( 2, 4), (− 2, 4) [ d x = 2x− x 3 ⇒ x 4 = 4]
d2y
= 2 + 24
d x2 x2
for all x
p p 4
( 2, 8), (− 2, 8) [y = x 2 + x 4 +4. Translation +4 parallel to y-axis]
y
20
p 10 p
(− 2, 4) ( 2, 4)
0 x
−5 −4 −3 −2 −1 1 2 3 4 5
44. (a) Curves cross at (2, 4) [ f (x) = g (x) where x 2 = 3x 2 − x 3 ⇒ x 2 (2 − x) = 0]

and touch at the (0, 0) [g (x) = x 2 (3 − x) = 0. Crosses at x = 3, touches at x = 0]
y
6
(2, 4)
- x
0 3
(b) Stationary pts at x = 0, 2 [g 0 (x) = 3x(2 − x) = 0]

4
(c) x = 0, x = 3 [2x = 6x − 3x 2 ⇒ x(3 − 4x) = 0]
The curves have the same
4
gradient at x = 0 and x = 3

45. (a) [(x + 2)(x 2 − 2x + 1) = x(x 2 − 2x + 1) + 2(x 2 − 2x + 1)]

dy
(b) y = 6x − 4 [ d x = 6x, so y − 2 = 6(x − 1)]
(c) (−2, −16) [2x 3 = 6x − 4 ⇒ 2(x 3 − 3x + 2) = 2(x − 1)2 (x + 2) = 0]
46. (a) (1, 5), (4, 5) [x 2 + 4 = 5x]

x2 4
(b) [ f (x) = x + x = x + 4x −1 ]
8 16 3
(c) ( 5 , 5 ) [tgt at A: y − 5 = −3(x − 1), at B : y − 5 = 4 (x − 4)]
(d) 4 [ f 0 (x) = 0 ⇒ x = ±2 and x > 0]
dy 1 1
47. (i) [ d x = px , so y − 4 = 2 (x − 4)]
(ii) y + 2x = 12 [y − 4 = −2(x − 4)]
1
(iii) 20 sq units. [B (−4, 0), C (6, 0). Area = 2 BC × 4]
p 3
p3
p3
48. (a) a = −2 [x = − 16 = − 8 × 2]
dy 16
(b) (2, 12) [ d x = 2x− x 2 = 0 when 2x 3 = 16]
dy
(c) [when x = 4, d x = 7, so y − 20 = 7(x − 4)]
(d) 3 solutions [y = 7x − 6 is parallel to the tangent, but “above" it]

10.3 Indefinite Integration

We said at the start of this chapter that integral calculus, or integration, is
simply the reverse process to differentiation. We shall discover that it allows
us to find the area under a graph, something that up to now we could only
do if it consisted of straight lines or circles.
In case you are reading this not having done any differentiating for a while,
let’s remind ourselves of the differentiation process.
For our purpose here, each term of the function will be of the form ax n ,
where a and n are constants. To differentiate we convert each term into a
new term, using the transformation rule
ax n → nax n−1
In words we
multiply by the power and reduce the power by 1
The only exception to this rule is constant terms: they simply disappear.
c → 0
So the reverse process.

dy
Suppose d x = 6x 2 . This would be the result if we differentiated y = 2x 3 .
it would also be the result if we differetiated y = 2x 3 + 5, or y = 2x 3 − 99 12 , or

indeed any function of the form y = 3x 3 + c, where c is any constant.
The notation for this process, the integration sign, is an elongated S symbol,
and we say “the integral of 6x 2 with the respect to x is 3x 3 + c”, and write it
Z
6x 2 d x = 3x 2 + c.
The “dx” is the “respect to x” part. It tells us which variable to use. In this
case x is the only variable, but if the function contained more than one, we
would need to know which one to use. Why the “d ” in “d x”? We will see why
putting d x at the end makes sense when we come on to talk about areas. For
a reason that will also become clear when we talk about areas, what we have
just described is called an indefinite integral.

In general, to integrate a function, we take each term and use the

transformation rule
ax n+1
ax n → (n 6= −1)
n+1
In words we
increase the power by 1 and divide by the new power
Notice that the rule breaks down if n = −1, an issue that is addressed in later
modules.
We can see how to integrate a power of x but how do we integrate a constant
term a?
The answer is we can write it as ax 0 and then apply the same rule:
ax 1
a = ax 0 → → ax
1
So the integral of a is ax.
Using the integration notation:
ax n+1
Z
ax n d x = +c (n 6= −1)
n+1

Some examples. Notice how little complications are added as we go down

the list:
Z
3 d x = 3x + c
1 1 5
Z Z
1 5
(x + 5) d x = x + d x = x2 + x + c
2 2 2 4 2
ax 2
Z
(ax + b) d x = + bx + c (where a and b are constants)
2
3x 2 3 x 2 3x 2
Z
3 2 3
3x d x = +c or x + c, since x 2 = × =
2 2 2 2 1 2
12x 3
Z
12x 2 + 5 d x = 4x 3 + 5x + c since = 4x 3
3
Z Z
2x(x − 1) d x = 2x 2 − 2x d x (multiplying out the bracket)
2x 3
= − x2 + c
3
Z Z
(2x + 3)2 d x = 4x 2 + 12x + 9 d x
4x 3
= + 6x 2 + 9x + c
3
2x 3
= − x2 + c
3
2x 3 − 3 1 3 3
Z Z
dx = x − dx
4 2 4
1 4 3 x4 − 6
= x − x + c or +c
8 4 8
You’ll find it’s important to be able to switch naturally between expressions

3x 2 3
like and x 2 .
2 2
In the last example we get the second version of the answer by putting two
fractions (in this case algebraic fractions) over a common denominator.
In neither case is one way of writing it “better” than the other but depending
on the context you may find one clearer than the other.

QQ
Find the following indefinite integrals (the answers are in the footnote17 .)
Z Z Z
(a) 9x 2 d x (b) 4x + 5 d x (c) 7 d x
Z Z Z
4
(d) x dx (e) 3x(x + 1) d x (f) (x − 1)2 d x
µ ¶2
1 3 1
Z Z Z
2 2
(g) 3 − x dx (h) (x + 1) d x (i) x x + dx
2 x
x 2 + x2
Z Z Z
2
(j) 2πx + 1 d x (k) 6x + d x (l) dx
4 3
17
(a) 3x 3 + c (b) 2x 2 + 5x + c (c) 7x + c (d) 51 x 5 + c
3x 2 x3 1
(e) x 3 + 2 +c (f ) 3 − x2 + x + c (g) 3x − 3 x3 + c (h) 81 x 4 + 1
2x +c
x5 2x 3
(i) 5 + 3 +x +c (j) πx 2 + x + c (k) 32 x + 1
9 x3 + c

We sometimes have a bit of work to do to express it as a series of terms of the

form ax n . You will see some examples below.
Z Z
1
dx = x −2 d x
x2
x −1
= +c
−1
1
= − +c
x
p
Z Z
1
x dx = x 2 dx
3
x2
= 3 +c
2
3
2x 2 2 23
= + c or x +c
3 3
2 p 3 1
= x x +c since x 2 = x 1 × x 2
3
³ ´
x3 + 2
Z Z
x + 2x −2 d x
¡ ¢
dx =
x2
x 2 2x −1
= + +c
2 −1
x2 2
= − +c
2 x
Although calculus is regarded as an advanced area of Maths, for many

students it is the algebra that often goes along with it that causes them the
most trouble. This is mostly to do with fractions, so time for a quick reminder
of some important ideas.

Fractions within fractions

For this reason we make no apology for first making sure you are
comfortable with manipulating fractions within fractions, like in the
example
3 12
1 16
In Chapter 4 (see page 4-20) we describe a quick way of dealing with this
kind of thing, and explain why it works. I call it the hammer and boot rule.
Here is a reminder. We start with this example:
3
5 3 3 1 3
=÷2 = × =
2 5 5 2 5×2
We’ve deliberately not finished the calculation because it’s the pattern we are
after, not the answer. What has happened? We imagine hitting the top
fraction 53 with a hammer, and think of it pushing the 5 down to join (and
multiply) the 2 underneath, like this:
3
3
5 3
↓ ==
2 5 × 2 10
Now let’s see what happens if the fraction within the fraction is on the bottom.
As before:
2 5 2 7 2×7
5
= 2÷ = × =
7
7 1 5 5
In this case we imagine booting the 7 up onto the top.
-2 14
5 = 15
7
When we have one fraction divided by another, we hammer and boot.

2 x 1
7 2 × 5 10 2 4x a 3
3
= = , 1
= = 2x, 2a
=
5
7 × 3 21 4
2 3
2a 2
If we add indices into the mix it’s not hard to see how the manipulation can
quickly become less than trivial. Remember when integrating (or
differentiating) each term needs to be of the form ax n .

More examples:
2 2
x3 3x 3 3 2
2 = or x 3
3
2 2
k 2 6 k 3k 2
x ×6 = × x2 = x
4 1 4 2
1 1 1 1 1
p = . p = x− 2
3 x 3 x 3
1 2 p
µ ¶
1 1 5 1 3
x x + p = x2 + x2
2 x 2 2
µ ¶2
4
x − = x 2 − 8 + 16x −2
x
px
6− 3 x 1 2
= − = 3x − 3 − 12 x 3
23x x
1
3 2x
1
3
QQ
Write each of these expressions as a series of terms in the form ax n , where a

and n are constants. The answers are in the footnote18 .
4 1
4x − 2
x3 2 2
¡ x ¢2
(a) 4 (b) (c) x × 10 (d) 3x ×
3
− 12 5 2
¶2
p
µ ¶ µ
4 3 1 2
(e) (f ) p (g) 2x x − p (h) x +
x3 2 x x x
p
1 + x2 (1 + x)2
µ ¶
− 12 2 1 1p
− 2x
(i) (j) x x + 2 (k) (l)
3x 2 x x x3
4 1 1 3 1
18
(a) 34 x 3 (b) −2x − 2 (c) 4x 2 (d) 34 x 3 (e) 4x −3 (f) 32 x − 2 (g) 2x 2 − 2x 2
3 5 1 1 5
(h) x 2 + 4 + 4x −2 (i) 31 x −2 + 1
3 (j) x 2 + x − 2 (k) x − 2 − 2x 2 (l) x −3 + 2x − 2 + x −2

Example 21
Find the following indefinite integrals.

p
Z
(i) ( x)3 d x
Z µ ¶2
1
(ii) 2x − dx
x
p
(x x − 4)2
Z
(iii) dx
x3
Solution
p p
Z Z
3 1
(i) ( x)3 d x = x 2 d x, since ( x)3 = (x 2 )3
5 5
x2 2x 2
= 5 +c = +c
2
5
Z µ ¶2
1
Z
(ii) 2x − dx = 4x 2 − 4 + x −2 d x
x
4x 3 1 x −1 1
= − 4x − + c, since −1 = − x
3 x
p p
(x x − 4)2 x 3 − 8x x + 16
Z Z
(iii) dx = dx
x3 x3
Z
3
= 1 − 8x − 2 + 16x −3 d x
1
= x + 16x − 2 − 8x −2 + c
16 8
= x + px − x 2 + c
Satisfy yourself that the working is correct in these examples19 . Never

underestimate how much care and attention you need to give to
manipulating and simplifying expressions. It is where a very high proportion
of errors creep in.
1
19 p 2 2 8x − 2 3 −1 16x −2
In (iii) (x x) = x × x = x , − 1 = +16x 2 and = −8x −2
−2 −2

10.3.1 Converting to and from index form

To carry out integration (and differentiation) you will need to express every20
term in the form ax n .
This may involve introducing fractional indices or negative indices. For
p 1 1
example we write 3 x as x 3 , and x 2 as x −2 .
Having integrated, however, it is often best to revert to roots and reciprocals,
as we have in parts (ii) and (iii) of Example 13. This is simply because we are
more familiar with them in that form.
5
This is not always the case,
p however. In part (i) we left x 2 in index form,
5
p 5
rather than convert it to x or ( x) .
5 1 p
Because x 2 = x 2 × x 2 we could have chosen to write it as x 2 x.
3 1p
Similarly, an expression like 12 x − 2 we could write as .
2x x
The point is: consider alternative ways of writing expressions involving

indices and choose the one you think is going to be most helpful.
Most of the time we switch to index form for differentiating or integtrating,

and switch back again at the end. Here is a typical example:
6 1
Find the gradient of the curve y = px − x at the point where x = 4.
1
y = 6x − 2 − x −1
dy 3
= − 3x − 2 + x −2
dx
3 1
= − p 3 + 2
( x) x
dy 3 1
When x = 4, = − p + 2
dx ( 4) 3 4
3 1
= − +
8 16
5
= −
16
20
Later modules cover the treatment of functions like sin x, 2x and log x which can’t be
expressed in this form.

QQ
Integrate these functions. (You will recognise them from the last exercise
and you can use your results from there as the first step in performing the
integrals.)
In each case convert your answer, if it seems appropriate, to a form that
doesn’t involve negative or fractional indices.
¶2
p
µ ¶ µ
4 3 1 2
(i) 3 (ii) p (iii) 2x x−p (iv) x +
x 2 x x x
p
1 + x2 (1 + x)2
µ ¶
− 12 2 1 1 − 2x
(v) (vi) x x + 2 (vii) p (viii)
3x 2 x x x3
p p
21 2 p 4x 2 x x x
(i) − +c (ii) 3 x + c (iii)
− +c
x2 5 3
p
x3 4 1 x 2x 2 x 2
(iv) + 4x − + c (v) − + +c (vi) − p +c
3 x 3x 3 5 3x x
p 4 p 1 4 1
(vii) 2 x − x x + c (viii) − + p − +c
3 2x 2 3x x x

Example 22
dy
A function y = f (x) is such that such that d x = 6x 2 − 4.
Find the equation of its curve given that it passes through the point (2, 11).
Solution
Reversing the process of differentiating is integrating, so y = 2x 3 − 4x + c.

Since the curve must pass through (2, 11), we have 11 = 16 − 8 + c, i.e. c = 3.
The equation of the curve is y = 2x 3 − 4x + 3.
Let’s see what this looks like graphically.

We can think of y = 2x 3 − 4x + c as a family of curves that differ only by a
constant.
For example if we start with y = 2x 3 − 4x and translate this a distance +2
unit in the y direction we have y = 2x 3 − 4x + 2; translate it a distance −3 in
the y direction and we have y = 2x 3 − 4x − 3, and so on.
We’ve illustrated this on the graph below. Notice that for any given value of x,
the gradient is the same for all the curves: it is given by the formula
dy
dx
= 6x 2 − 4.
dy
For example if x = 1, d x = 6 − 4 = 2.
y
6 x =1
y = 2x 3 − 4x + 2 2
- x
−2 −1 0 1 2
y = 2x 3 − 4x
−3
y = 2x 3 − 4x − 3

QQ
In each case find y in terms of x. The answers are in the footnote22 .
dy
(a) = 4x + 3, given that y = 9 when x = 1
dx
dy
(b) = 6x 2 − 8x + 1, given that y = 0 when x = 2
dx
d y 18
(c) = , given that y = 4 when x = 3
d x x3
dy p
(d) = 12 x, given that y = 20 when x = 4
dx
22 9 p
(a) y = 2x 2 + 3x + 4 (b) y = 2x 3 − 4x 2 + x − 2 (c) y = − 2 + 5 (d) y = 8x x − 44
x

Example 23
dy
The gradient of a curve is given by d x = ax + b.
The curve passes through (1, 3). It also passes through (2, 19), where its
gradient is 23.
Find the equation of the curve.
Solution
The curve is y = 21 ax 2 + bx + c.
We have three constants to find: a, b and c. We also have three pieces of
information about the curve, each giving us an equation connecting them.
a
Since the curve must pass through (1, 3): 3 = 2 + b + c.
Similarly 19 = 2a + 2b + c.
And from the gradient equation: 23 = 2a + b.
There are many ways to solve these simultaneous equations. The easiest is
probably to use the gradient equation to substitute b into the other two:
b = 23 − 2a
a
so 3 = + 23 − 2a + c
2
and 19 = 2a + 2(23 − 2a) + c
i.e. 3a = 40 + 2c
and 2a = 27 + c
Hence a = 14, b = −5 and c = 1
The equation of the curve is y = 7x 2 − 5x + 1.

Example 24
x2
Write p in the form x k where k is a fraction.
x
dy x 2 +1
The gradient of a curve at (x, y) is given by = p (x > 0).
dx x
Given that the curve passes through (4, 9), show that its equation can be
written
p
y = a x(x 2 − 5) + b (x > 0),
where a and b are constants.
Solution
x2 1 3
p = x2 × x− 2 = x 2 .
x
dy 3 1
Hence = x 2 + x− 2
dx
2 5 1
therefore y = 5 x 2 − 2x 2 + c.
Since the curve passes through (4, 9), we have 9 = 25 .32 − 4 + c,

i.e. c = 15 .
2 5 1
The equation is y = 5 x 2 − 2x 2 + 15 .
To express the equation in a form that doesn’t use fractional indices we will
adopt the method we used in the first part of the question, in reverse:
5 1 p
x 2 = x 2 × x 2 = x 2 × x.
Don’t be afraid to take your time over this next bit of algebra. Manipulation
that involves fractions does not come naturally to most people. It will take
you time to become fluent – and eventually be able to miss out steps.
Getting it right is the main thing.
The equation is
2 2p p 1
y = x x −2 x +
5 5
2 2p 10 p 1
= x x − x +
5 5 5
1 p p 1
= (2x 2 x − 10 x) +
5 5
2p 1
= x(x 2 − 5) +
5 5
which is in the required form with a = 52 and b = 15 .

dy
The requirement x > 0 follows from the restriction in the definition of dx
,
which would have undefined values if x were permitted to be ≤ 0.

Exercises 38
1. Integrate with respect to x:

(a) 3x 2 (b) 4x (c) 12x 3 (d) 7
4 2
(e) −10x (f) x + 3 (g) 2 − x (h) x 3 + 3x
(i) 2(4x + 1) (j) −x (k) x 3 + x 2 + x + 1 (l) 12 x 2
2. Find the following indefinite integrals:

Z Z Z Z
3
(a) 8x d x (b) (6x 2 − 2x) d x (c) 4
x dx (d) 3dx
1
Z Z Z Z
(e) dx (f) x2 d x (g) (2x 5 + 1) d x (h) (4 + x 3 ) d x
Z 2
x
Z Z Z
2 2
(i) x(3x + 1) d x (j) 2x(x − 4) d x (k) x (1 − x ) d x (l) dx
Z ³4 ´3
4x x
Z Z Z
(m) (1 + x)2 d x (n) dx (0) (2x)3 d x (p) dx
5 2
3. Integrate each if these with respect to x:
(a) x −2 (b) x −3 + x (c) 2x −6 (d) 21 x −2
1 2
(e) 8x −5 (f) x −4 + x −7 (g) 6x 2 (h) x 3
1 1 1 1 1
(i) x − 2 (j) 20x 4 (k) 12x − 3 (l) x 2 − x − 2
4. Write each of your answers to question 3 in a form that does not involve
negative of fractional indices.
5. Using negative or fractional indices, write each of these expressions in

a form suitable for integration.
2 1
−3
3 = 2x
For example p .
x
1 5 p p
4
(a) 3 (b) (c) 3 x (d) x
x x2
4 p 2 1
(e) p (f ) x 3 (g) (h) p
x 7x 2 2 x
p p
³ ´
1 p1
2−x ¡ 1
¢
(i) p + x
x
(j) x 1− x
(k) 3
x
(l) x 1+ x4
p
p x3 − 1 x +x 2
1 +p (3 x+1)2
(m) x(1 + x 2 ) (n) (o) (p) p
2x 3 x x
6. Integrate each of the expressions in question 5, leaving your answer in a
form that does not involve negative of fractional indices23 . For example
4 p 1 p
x 3 = ( 3 x)4 , or it may be better to write it as x × x 3 = x 3 x.
dy
7. If = 6x + 5, find and expression for y in terms of x, if y = 1 when
dy
x = −2.
dV
8. V is a function of t and = 2(t + 1). Find V in terms of t , if V = 9
dt
when t = 2.
9. Given that f 0 (x) = 1 + x 2 and f (1) = 3, find f (0).

23
Don’t underestimate the amount of careful algebra required to get the answers into this
form. Be prepared to spend some time on them. There is no “right” way to express the answer
where indices are involved; the important thing is to be able to switch freely between the
alternatives, to suit the context.

10. In each case find an expression for y in terms of x, if

dy
(a) = (3x + 1)(x − 1), given that y = 7 when x = 2
dy
dy 1
(b) = 8x+ p , given that y = 71 when x = 4
dy x
dy 1
(c) = 1− 2 , given that y = 3 when x = 1.
dy x
3x + 1 p
Z
11. Show that p d x = x(x + 1) + c.
2 x
Z 2
x −1 x 2 +1
12. Show that d x = + c.
x2 x
1 2 3x 4 −6x 2 −1
Z µ ¶
13. Show that 3x − dx = + c.
x x
p x −1
Z
p x−1
14. Show that = x − 1. Hence find p d x.
x+1 x +1
dy
15. The gradient function of a curve is = 3x 2 + 2.
dx
Find the equation of the curve, given that it passes through (1, 8).
dy
16. The gradient of a curve is given by = πx + 1, and the curve passes
dx
through (2, 2π + 3). Find its equation.
dy
17. The point (2, 16) lies on a curve whose gradient is given by = 8x −4.
dx
Find the equation of the curve and the coordinates of the points where
it crosses the x-axis.
dy
18. The gradient of a curve is = ax + b. FInd the equation of the curve,
dx
given that its gradient at x = 0 is 1, and it passes through the points (1, 2)
and (−2, 8).
19. Find the function f (x), given that f 0 (x) = px + q, f 0 (2) = 0, f (1) = −1
and f (−1) = 7.
dy A
20. A curve crosses the x-axis at (4, 0) and its gradient is given by =p .
dx x
¡1 20
¢
Find its equation if it also passes through 9 , − 3 .
Write down the smallest value y can take.
dy
21. (a) The gradient of a curve C is = k(1 − x 2 ), where k is a constant.
dx
Find the value of k and the equation of the curve if its gradient at (2, 0)
is −18.
(b) Using the Factor Therem, or otherwise, show that the equation can
be written y = 2(2 − x)(x + 1)2 .
(c) Sketch C.
dy a b
22. The gradient of a curve C is given by = + .
d x x2 x3
The x-axis is an asymptote of C, and it passes the points (1, −2) and
(−1, 4). Find the equation of C and show that it crosses the x-axis where
1
x = 3 . [Hint: what would x be where C is a asymptotic to the x-axis?]

Solutions 38
1. (a) x 3 (b) 2x 2 (c) 3x 4 (d) 7x

(e) −2x 5 (f ) 21 x 2 + 3x (g) 2x − 31 x 3 (h) 41 x 4 + 23 x 2
(i) 4x 2 + 2x (j) − 21 x 2 (k) 14 x 4 + 31 x 3 + 21 x 2 + x (l) 16 x 3
2. (a) 2x 4 + c (b) 2x 3 − x 2 + c (c) 51 x 5 + c (d) 3x + c
(e) 14 x 2 + c (f) 13 x 3 + c (g) 13 x 6 + x + c (h) 4x + 41 x 4 + c
(i) x 3 + 21 x 2 + c (j) 32 x 3 − 4x 2 + c (k) 13 x 3 − 15 x 5 + c (l) 81 x 2 + c
(m) x + x 2 + 13 x 3 + c (n) 25 x 2 + c (o) 2x 4 + c (p) 1 4

32
x +c
3. (a) −x −1 (b) − 21 x −2 + 12 x 2 (c) − 25 x −5 (d) − 21 x −1

3 5
(e) −2x −4 (f) − 13 x −3 − 16 x −6 (g) 4x 2 (h) 53 x 3
1 5 2 3 1
(i) 2x 2 (j) 16x 4 (k) 18x 3 (l) 32 x 2 − 2x 2
1 1 x 4 +1 2
4. (a) − x (b) − 2x 2 + 12 x 2 or 2x 2 (c) − − 5x 5 1
(d) − 2x
2 1 1+2x 3 p p p
(e) − x 4 (f) − 3x1 3 − 6x 6 or − 6x 6 (g) 4( x)3 or 4 x 3 (h) 53 ( 3 x)5
p p p p
(i) 2 x (j) 16( 4 x)5 (k) 18( 3 x)2 (l) 23 x(x − 3)
1 1
5. (a) x −3 (b) 5x −2 (c) 3x 2 (d) x 4
1 3 1
(e) 4x − 2 (f) x 2 (g) 72 x −2 (h) 21 x − 2
1 1 1
(i) x − 2 + x 2 (j) x 2 − 1 (k) 2x −3 − x −2 (l) x + x −3
1 5 1 1 3 1 1
1
(m) x 2 + x 2 (n) 2
− 21 x −3 (o) x − 2 + x 2 + x 2 (p) 9x 2 + 6 + x − 2
x −2 1 5x −1 5
6. (a) −2 = − 2x 2 (b) −1 = − x
5
3 p p 4x 4 4 p 4 p
(c) 2x 2 = 2( x)3 = 2x x (d) 5 = 5 ( 4 x)5 = 5 x 4 x
5
1 p 2x 2 2 p
(e) 8x 2 = 8 x (f) 5 = 5 ( x)5
2x −1 2 1 p
(g) −7 = − 7x (h) x 2 = x
1 2 3 2p 2 3 1 p
(i) 2x 2 + 3 x 2 = 3 x(3 + x) (j) 3 x 2 − x = 3 x(2 x − 3)
x −1 x2 x −2 x4 − 1
(k) −x −2 + x −1 = x 2 (l) 2 + −2 = 2x 2
3 7
2x 2 2x 2 2 p x x −2 2x 3 + 1
(m) 3 + 7 = 21 x x(7 + 3x 2 ) (n) 2 + 4 = 4x 2
3 5
1
2x 2 2x 2 3 1
(o) 2x + 3 + 5
2 (p) 6x 2 + 6x + 2x 2
2p p p
= 15 x(15 + 5x + 3x 2 ) = 2 x(3x + 3 x + 1)
7. y = 3x 2 + 5x − 1 [y = 3x 2 + 5x + c and 1 = 12 − 10 + c ⇒ c = −1]
8. V = (t + 1)2 [V = t 2 + 2t + c and 9 = 4 + 4 + c ⇒ V = t 2 + 2t + 1]
1 1 1 5
9. f (−1) = 3 [ f (x) = x+ 3 x 3 + c. f (1) = 3 = 1+ 3 +c ⇒ c = f (0) = 3 ]

dy
10. (a) y = x 3 − x 2 − x + 5 [ d x = 3x 2 − 2x − 1 and 7 = 8 − 4 − 2 + c ⇒ c = 5]
p dy 1 1
(b) y = 4x 2 + 2 x + 3 [ d x = 8x + x − 2 ⇒ y = 4x 2 + 2x 2 + c and 71 = 64 + 4 + c]
1 dy
(c) y = x+ x +1 [ d x = 1 − x −2 ⇒ y = x + x −1 + c and 3 = 1 + 1 + c]
Z µ ¶
3 1 1 −1 3 1 p p
11. x + x
2 2 dx = x 2 +x 2 +c = x x + x +c
Z 2 2
−2 1
d x = x + x −1 + c = 1+ x +c
¡ ¢
12. 1−x
Z
¡ 2 1
9x − 6 + x −2 d x = 3x 3 − 6x − x + c
¢
13.
p p
x − 1 (x − 1)( x − 1) (x −X1)(
XX
x − 1)
14. p = p p =
X
x+1 ( x + 1)( x − 1) x−1
X XX
Z
1 2 3
(x 2 − 1) d x = 3 x 2 − x + c.
15. y = x 3 + 2x + 5 [y = x 3 + 2x + c and 8 = 1 + 2 + c]
1 1
16. y = 2 πx 2 + x + 1 [y = 2 πx 2 + x + c and 2π + 3 = 2π + 2 + c ⇒ c = 1]
17. y = 4x 2 − 4x − 8. (−1, 0) and (2, 0). [4(x + 1)(x − 2) = 0 ⇒ x = −1 or x = 2]

dy
18. y = 3x 2 + x − 2 [At x = 0, d x = 1, so b = 1.
1
Then 2 = 2 a + 1 + c and 8 = 2a − 2 + c. e.g. eliminate c.]
1
19. f (x) = x 2 − 4x + 2 [ f (2) = 0 ⇒ q = −2p, f (x) = 2 px 2 − 2px + c
3 5
so −1 = − 2 p + c and 7 = 2 p + c. Eliminating c, p = 2, q = −4 and c = 2.
Notice that solving the three simultaneous equations is made simpler
by substituting q = −2p right at the start.]
p p
20. y = 4 x − 8. Minimum y = −8. [y = 2A x + c. We have 0 = 4A + c and
20 2A p p
− 3 = 3 + c ⇒ a = 2, c = −8. For min y: x ≥ 0 so 4 x ≥ 0 ⇒ y ≥ −8]
21. (a) y = 6x − 2x 3 + 4. [−18 = k(−3) ⇒ k = 6 ⇒ y = 6x − 2x 3 + c

Then 0 = 6(2) − 2(23 ) + c ⇒ c = 4]
(b) If y = f (x) we know that f (2) = 0 so (x − 2) is a factor.
By polynomial division: y = (x − 2)(−2x 3 − 4x − 2) = −(x − 2)(x 2 + 2x + 1)
Alternatively we could simply multiply out (2 − x)(x + 1)2 .
(c) y
6
-x
−1 2
1 3
22. y = x 2 − x , which passes through ( 13 , 0).
a b
[y = − x − 2x 2 +c. C asymptotic to the x-axis ⇒ y → 0 as x → ±∞ ⇒ c = 0.
b b
−2 = −a − 2 and 4 = a − 2 ⇒ a = 3 and b = −2. Notice that being an
asymptote the the x-axis does not prevent C from cutting it elsewhere.]

10.4 Definite Integration and Areas

Integration is differentiation in reverse. But so what? Why might we want to
integrate a function?
The answer is that integration has a great many applications, in the field of
science especially. One that is very easy to appreciate is finding areas. Up to
now we have only been able to find the area of a shape if every side is either a
straight line or a circular arc. Suppose one of the sides is some other curve?
Here is an example. To find the area enclosed between the curve
2
f (x) = x + 1, the x-axis and the lines x = 1 and x = 2. It is the area shaded
blue in the diagram below.
y
6 f (x) = x 2 + 1
x
−1 0 1 2 3
First we will show how to calculate it, and then explain what is going on.
Z 2
Shaded area = (x 2 + 1) d x.
1
R
The numbers attached to the symbol mean that we are restricting x to
values between 1 and 2.
We call this a definite integral and the two values of x the lower limit and
the upper limit of the integral:
Z 2 · 3 ¸2
x
· ¸ · ¸
2 8 1 7 10 1
(x + 1) d x. = +x = +2 − +1 = +1 = = 3
1 3 1 3 3 3 3 3
Can you see what we have done? Having integrated the function we have
found the value of the expression at each of the limits and subtracted the
answers - starting with the upper limit.
(Notice that with a definite integral, for a reason I’ll explain later, we don’t
need to include a constant of integration c.)
Let’s see if the answer looks about right. f (1) = 2 and f (2) = 5, so the area of
the trapezium with the same vertices would be 1 × 12 × (2 + 5) = 72 = 3 12 .
This is very close to the exact area, found by integration. And because the
curve “bend inwards” the exact area is slightly less, as we would expect.

Before we talk more about finding areas, a little practice performing definite
integrals. But first a couple of comments. The first is a warning to be careful
about signs. For example:
Z 1 i1
(−1)3
h h i
x3
x2 d x =
£1¤ £1¤ £ −1 ¤ 2
3 −1
= 3
− 3
= 3
− 3
= 3
−1
It is all too easy to end up with the answer 0, clearly not the correct answer
when we look at a sketch:
y
2 y = x2
0 x
−2 −1 1 2
We can think of the integral as made up of two integrals:

Z 1 Z 0 Z 1
2 2
x dx = x dx + x2 d x
−1 −1 0
each (by symmetry) having the value 31 .
The second comment is that any factors inside the integral can be taken
outside (and vice versa obviously). The same is true of factors inside or
outside the square bracket. This device helps to keep the working as tidy as
possible. An example, first without the simplification:
Z 2 4x + 1
Z 2 4 1 ¤2
x + d x = 32 x 2 + 13 x 1 = 38 + 23 − 32 + 13 = 7
£ £ ¤ £ ¤
dx = 3
1 3 1 3 3
Now with:
Z 2 4x + 1
Z 2 ¤2 ³ ´
1 1 1
[8 + 2] − [2 + 1] = 37
£ 2
dx = 3 4x + 1 d x = 3 2x + x 1 = 3
1 3 1
Another example:
Z 2 Z 2
4 2
h 3 i ¤2
2
(5x 2 − x 3 ) d x = 2 5x3 − x4 2
20x 3 − 3x 4 1
£
2x (5 − x) d x = 2 = 12
1 1 1
...by pulling any fractions out of the bracket, the evaluation inside the
bracket becomes easier—and therefore less prone to error:
³ ´
= 6 [160 − 48] − [20 − 3] = 95
1
6
Note that it is only constant factors that can moved through the integral sign;
it would not make sense to say something like this:
Z 2 Z 2
2 2
2x (5 − x) d x = 2x (5 − x) d x !!
1 1

QQ
Evaluate each of these definite integrals. In some cases it will be helpful to

take out factors in the way that we have just seen. (The answers are in the
footnote.24 )
Z 2 Z 3 Z 2
2
(a) 8x d x (b) 9x d x (c) x3 d x
1 1 −1
2 2 1−x 2 x2
Z Z Z
(d) (1 − x) d x (e) dx (f) dx
1 1 3 0 2
Z 2 3
Z 2 4
Z 4 p
(g) x dx (h) dx (i) x dx
0 5 1 x3 1
2 2 2 µ ¶2
1
Z Z Z
2
(j) x(2 + x) d x (k) (3x − 1) d x (l) x+ dx
−1 0 −1 x
15
24
(a) 12 (b) 78 (c) 4 (d) − 12 (e) − 61 (f) 4
3 (g) 6
5 (h) 3
2 (i) 14
3 (j) 6 (k) 14 (l) 15
2

We haven’t explained how we know what area we are going to get when we
perform a definite integral between two values of x? The answer is: the area
looks like a “curtain” hanging from the curve down to the x-axis. The sides of
the “curtain” are always vertical and the bottom always lies along the x-axis.
QQ
(i) Which of the shaded areas below fit the above definition of an integral
between two values of x? For those that do, write down what the values are,
i.e. the limits of the integral. (The answers are in the footnote25 .)
(a) y (b)
y
x2
4 y= 4
+1
y = 4−x 2 3
3
2
2
1
1
0 x
−1 1 2 3
0 x
1 2 3
(c) y (d ) y
4
3
3
2 y = x2 + 1
2 y = x3 + 2
1
1
0 x
−1 1 2
0 x
−1 −1 1 2 3
−1
(e) (f ) y
y
p y = 4 − x42
2 y= x 3
1 2
0 x 1
1 2 3 4 5
x
0 1 2 3
(ii) Find all the shaded areas. Where you cannot do it directly with an
integral, find it indirectly. (The answers are on the next page.)
25
(a) 1 to 2, (d) 0 to 1. None of the others are the direct result of an integration between
two values of x, although you will see they can all be found using integration.

If any of the answers to (i) don’t make sense to you, imagine (or sketch)
where the “curtain” would fall. Starting on the x-axis at the two values of x
you have chosen, the sides of the “curtain” are the vertical lines from these
two points up to the curve. If the resulting “curtain” corresponds exactly to
the area we want, the definite integral – between those two values of x –
gives that area. This is what we find in examples (a) and (d).
If we can’t find the area directly in that way, we can think of it as the
difference of two other areas, one of which we find by integration.
Answers to (ii) (in square units):

¸2 ·
x3
Z 2 · ¸ · ¸
5 2 8 1 5
(a) (4 − x ) d x = 4x − = 8− − 4− =
3 1 3 1 3 3 3
¸2 ·
x2
2 · 3
x
µ ¶ ¸
2 8 8
Z
(b) +1 dx = +x = + 2 − [0] =
3 0 4 12 0 12 3
[Subtract the area of the triangle with vertices (0, 0), (2, 0) and (2, 2) from this.]
¸1
1 x3
· · ¸
2 1 4
Z
2
(c) (x + 1) d x = +x = + 1 − [0] =
3 0 3 0 3 3
[Subtract this area from that of the rectangle between (0, 0) and (1, 2).]
¸1
1 x4
· · ¸
9 1
Z
3
(d) (x + 2) d x = + 2x = + 2 − [0]
4 0 4 0 4
4 · ¸4 · ¸ · ¸
2 2 3 2 2 14
Z
1
(e) x dx = x 2
2 = .8 − .1 =
3 1 3 1 3 3 3
[Subtract from this the area of the rectangle between (1, 0) and (4, 1).]
Z 2 ¤2
(4 − 4x −2 ) d x = 4x + 4x −1 1 = [8 + 2] − [4 + 4] = 2
£
(f) 4
1
[Subtract this from the area of the rectangle between (0, 0) and (2, 3).]
In all the above graphs both axes have been drawn to the same scale. This is so
you can see that the areas look about right. From now onwards, as with curve
sketching, we may use a different scale on each axis.
But before going on, an explanation of why integrating a function gives the
area under it’s curve, and why – with a definite integral – we don’t need the
constant of integration. You won’t be examined on either of these so you can
skip the next section if you are happy to take them on trust.

10.4.1 Definite integration and area: the theory

When we described the theory behind differentiation we used the notation
δx and δy, standing for a small increment (or increase) in x or y. Now let’s
suppose the area under the curve, from the y-axis up to the point P is A.
y
y + δy
δA
y
A
x x + δx
We are going to see how much new area would be created if we were to go as
far as Q, rather than stop at P . We will call this extra area δA. How big it is
will depend on δx, y (the height of P ) and y + δy, the height of Q.
We can’t find δA exactly, but a trapezium is a good approximation to the
shaded area, and this estimate will get even better when we let P and Q get
closer to each other, i.e. as δx → 0. (≈ means “is approximately equal to”.)
1 1
δA ≈ δx [y + (y + δy)] = yδx + δxδy
2 2
δA 1
If we divide through by δx we have ≈ y+ δy.
δx 2
If we now move Q closer and closer to P , until eventually it’s at the same
point as P , δy = 0 and the trapezium area is no longer an approximations.
We can then write
dA
=y
dx
If all seems very abstract, let’s look at the example we started with: y = x 2 + 1.
dA
= x 2 + 1.
dx
A is going to be a formula for the area, which tells us what A is for different
values of x. When we differentiate that formula we get x 2 + 1, so to find the
formula we integrate x 2 + 1: x3
A= +x +c
3
Since A is the area starting at the y-axis26 , A = 0 when x = 0, so c = 0.
To find the area under the curve between, say 1 and 2, we find the area as far
23 13
as 2, which is 3 + 2, and subtract the area as far as 1, which is 3 + 1. This is
exactly what we did a few pages back:
Z 2 · 3 ¸2 · 3 ¸ · 3
x
¸
2 2 1 10
(x + 1) d x = +x = +2 − +1 =
1 3 1 3 3 3
26
Because we are always subtracting two areas it doesn’t matter where we measure A from.

10.4.2 Finding areas: harder examples

All that remains is to look at the types of question you might expect in an
exam, which mostly involve finding an area formed by a curve intersected by
one or more lines, or by another curve.
Example 25
Find the coordinates of the points where the line y = x + 1 intersects the
curve y = (x + 1)(x − 3).
Hence find the area enclosed by the line and the curve.
Solution
The equation of the curve can be written y = 3 + 2x − x 2 , so to find where the
line crosses it
x + 1 = 3 + 2x − x 2
x2 − x − 2 = 0
(x + 1)(x − 2) = 0
x = −1 or x = 2
The points of intersection are P (−1, 0) and Q(2, 3).
The curve cuts the x-axis at x = −1 and x = 3; it also passes through (0.3).
The coefficient of x 2 is negative, so the area looks like this
y The definite integral

y = (x + 1)(3 − x) Z 2
4 (3 + 2x − x 2 ) d x
−1
3 Q
gives the area under the curve
between x = −1 and x = 2.
2
The (shaded) area we are
1 looking for is therefore this
integral minus the area of the
P R 1
0 x triange PQR, whose area is 4 2 .
−1 1 2 3
−1
¸2
2 x3
Z ·
2 2
(3 + 2x − x ) d x = 3x + x −
−1 3 −1
· ¸ · ¸
8 (−1)
= 6+4− − (−3) + (+1) −
3 3
= 9
1
The shaded area is therefore 4 2 , the same as the area of triangle PQR.
Notice how we didn’t try to do too much in our head in evaluating the integral.
When you have something like this get into the habit of writing down one or
more intermediate stages, even if they seem very trivial.
h i
The final fraction, for example, is − − (−1)(−1)(−1)
3
= − 13

So far every area we have looked at has been above the x-axis. When it is, all
the relevant y-values are positive so the area will be positive. If an area is
below the x-axis, the y-values are negative and the area, mathematically
speaking, will be negative.
Here is a simple example: the area between the curve y = 4x(x − 1)(2 − x) and
the x-axis. We’ll start with the area between x = 0 and x = 1.
The equation can be written y = −4x 3 + 12x 2 − 8x and the curve looks like
this27 : Z 1
y
(−4x 3 + 12x 2 − 8x) d x
0
y = 4x(x − 1)(2 − x) £ 4 ¤1
2 = −x + 4x 3 − 4x 2 0
= [−1 + 4 − 4] − [0]
0 x = −1
−1 1 2 3
The actual area is +1 of course,
−2 because an area can’t be negative.
This raises the question of what to do if part of the area is above the x-axis
and part of it is below. The answer is we have to calculate the two parts
separately, as we see if we try to find the area between 0 and 2. It is
y £ 4 ¤2
−x + 4x 3 − 4x 2 0
= [−16 + 32 − 16] − [0]

2
= 0
This should come as no surprise,

0 x because the curve has rotational
−1 1 2 3
symmetry about (1, 0). The two
parts of the shaded area must be
−2 equal, but have opposite signs.
Mathematically, when integrating between x = 0 and x = 2, an area of −1 and

an area of +1 combine to give an area of zero.
Z 2
Write down the value of 4x(x − 1)(2 − x) d x.
1
By sketching the curve y = x 3 , write down, for any constant k, the value of
Z +k
x 3 d x.
−k
27
The roots (where it crosses the x-axis) are at 0, 1 and 2. The coefficient of x 3 is negative,
which tells us which way up it is.

The answers to the first question is of course +1.

The cubic curve y = f (x) = x 3 has rotational symmetry about the origin: for
any value x = a, f (−a) = − f (a), so the integral is 0, whatever the value of k.
A function like y = f (x) = x 3 , for which f (−a) = − f (a) is called an odd

function, and has rotational symmetry about the origin.
A function like y = f (x) = x 2 , for which f (−a) = f (a), is called an even
function, and has reflective symmetry in the line x = 0 (i.e. the y-axis).
From a quick sketch of any odd function and any even function, we can see
that
Z +k
if f (x) is an odd function f (x) d x = 0
−k
Z +k Z +k
if f (x) is an even function f (x) d x = 2 f (x) d x
−k 0
This is most easily seen graphically.

Turning again to the curve y = x 3 , as an example of an odd function:
y
the symmetry means the two areas
y = x3
on either side of the y-axis are equal
in size but opposite in sign, so the
total mathematical area = 0.
x Z k
−k 0 k
x 3 d x = 0.
−k
In the case of an even function, the symmetry allows us to convert the

integral into an integral which has a lower limit of 0, which will invariably
simplify the calculation.
y
Using y = x 2 as an example
Z k Z 0 Z k
2 2
x dx = x dx + x2 d x
−k −k 0 y = x2
Z k
= 2 x2 d x
0
x
−k 0 k

Example 26
Sketch the curve y = x 2 − 4.
Find the area between the curve, the two coordinate axes and the line x = 3.
Solution
The roots (where y = 0) are at x = ±2 and the y-intercept is at y = −4.
y
y = x2 − 4
6
2
B
0 x
−2 −1 1 2 3 4
−2 A
−4
Because part of the area is above the x-axis and part of it is below, we need to
find those areas separately. We can’t simply integrate between x = 0 and
x = 3.
Z 2 · 3 ¸2
2 x
A= (x − 4) d x = − 4x
0 3 0
· ¸
8
= − 8 − [0]
3
16
= − < 0
3
¸3
3 x3
Z ·
2
B= (x − 4) d x = − 4x
2 3 2
· ¸
8
= [9 − 12] − −8
3
7
> 0 =
3
16 7 23
The total area is therefore28 |A| + |B | = + = 3.
3 3
28
|x| is the modulus of x, i.e. it’s magnitiude ignoring the sign. We read it “mod x”.

QQ
Write down the values of these integrals (the answers are in the footnote29 ):
Z 3 Z 2
2
(a) (x − 4) d x (b) (x 2 − 4) d x
0 −2
Z 2 Z 3
2
(c) (4 − x ) d x (d) (3x 2 − 12) d x
0 2
29
(a) −3 [A + B ] (b) − 32
3 [2A] (c) + 16
3 [−A] (d) 7 [3B ]

Example 27
Find the area enclosed between the curve y = x 2 + 2, the y-axis and the line
y + x = 4.
Solution
The line crosses the curve where 4 − x = x 2 + 2. This is easily solved to give
x = 1 or x = −2.
The solution we are after must be x = 1.
The line passes through (0, 4) and (4, 0) and intersects the curve at (1, 3).
The sketch looks like this, with the area we are looking for shaded:
y
0 x
−1 1 2 3 4
The way to start a question like this is to ask yourself: what area would I get if
I integrated between two possible values of x, for example between 0 and 4,
or between 0 and 1. The “curtain” idea tells us which area each integral gives
us.
We are not necessarily looking for an integral that gives us the area directly,
but one that we could use to find the area.
It quickly becomes clear that the integral from 0 to 1 will do the trick,
because it gives us the area underneath the curve, and the shaded area can
be found by subtracting that area from the area of a trapezium.
Z 1 · 3 ¸1
2 x
(x + 2) d x = + 2x
0 3 0
· ¸
1
= + 2 − [0]
3
7
=
1 7 3
Trapezium area = 1× 2 (3 + 4) = 2
7 7 7 1
Shaded area = 2 − 3 = 6 = 1 6
If the answer doesn’t “look right”, it may be because the scales on the axes are
different. It is always a good idea to try a check whether we are in the right
ball park. In this case the area of the triangle with the same vertices as the
shaded area, i.e. (0, 2), (1, 3) and (0, 4), is exactly 1, very close to the shaded
area (but slightly smaller, as expected).

Example 28
2
Find the area enclosed between the curve y = , the x-axis, and
x2
(a) the lines x = 1 and x = 4,
(b) the line x = 1.
Solution
(a) The sketch looks like this (the curve is drawn accurately – isn’t it
beautiful!):
y The area is
Z 4
2x −2 d x
3 1
¤4
−2x −1 1
£
=
· ¸4
−2
2 =
x 1
· ¸
1
= − − [−2]
2
1 3
=
2
0 x
1 2 3 4 5
(b) If we take away the line x = 4 we have an area that extends to infinity.
Surely the answer must be infinite? Let’s see what happens when we do the
integration Z ∞
¤∞
2x −2 d x = −2x −1 1
£
1
· ¸∞
−2
=
x 1
· ¸
2
= − − [−2]
∞
1
= 2 because = 0
∞
Even though the region is infinitely long, its area is finite. One way to think of
it is this: as we go further and further to the right, the region gets thinner
faster than it gets longer.

Example 29
Find the equation of the tangent to the curve y = 9 − x 2 at (1, 8).
Find area enclosed by the curve, the tangent and the two axes.
Solution
The gradient of the curve is given by
dy
= −2x
dx
so at x = 1 the tangent has gradient −2, and its equation is
y − 8 = −2(x − 1)
i.e. y = 10 − 2x
which crosses the axes at (0, 10) and (5, 0). The curve crosses the axes where
y = 9 and where x = ±3. The area we are looking for is shaded.
y
10
9
(1, 8)
0 x
−1 1 2 3 4 5 6
If we integrate the function between 0 and 3, and subtract that from the area
of the triangle with vertices (0, 0), (5, 0) and (0, 10), we will have the area we
need.
¸3
x3
Z 3 ·
2
(9 − x ) d x = 9x −
0 3 0
= [27 − 9] − [0]
= 18
The area of the triangle is 25, so the shaded area is 7.

Example 30
p p
The graph30 shows the two curves y = x and y = 2 − x.
p
Find the coordinates of Q and R, where the curve y = 2 − x crosses the axes,
and the coordinates of P , where the two curves meet.
Hence show that the area of the region enclosed by the two curves and the
x-axis is the same as the area of the triangle OP R.
y
p
y= x
Q
x
O R
p
y = 2− x
Solution
p
For the curve y = 2 − x
p
x = 0 ⇒ y = 2, so Q is (0, 2). And y = 0 when x = 2 ⇒ x = 4. R is (4, 0).
p p p
The curves cross where 2 − x = x, i.e. where 2 = 2 x ⇒ x = y = 1.
The area of triangle OP R is therefore 2.
p
Between x = 0 and x = 1 the region has y = x as a boundary; between x = 1
p
and x = 4 it has y = 2 − x. The area is therefore
2 3 1 2 3 4
Z 1 Z 4³ · ¸ · ¸
1 1
´
x 2 dx + 2−x 2 dx = x 2 + 2x − x 2
0 1 3 0 3 1
· ¸ · ¸ · ¸
2 16 2
= − [0] + 8 − − 2−
3 3 3
= 2
the same as the area of triangle OP R.
30 p
To see why the curves look like this, start with y = x.
p
This is the positive (top) half of the curve y 2 = x, because we can think of it as y = ± x.
y 2 = x will have the same shape as y = x 2 but with the axes swapped over. Can you see
p
this? The bottom half of y 2 = x is y = − x. Adding 2 to this function moves it up 2 units,
p
so its equation is y = − x + 2, which is the other curve in the diagram.

Exercises 39
1. Find the area enclosed by the curve y = x 2 + 2, the two coordinate axes and
the line x = 3.
2. Find the area between the curve y + x 3 , the x-axis and the lines x = 1 and
x = 2.
3. Sketch the curve y = x(2 − x) and find the area between the curve and the
x-axis.
Write down the area between the curve y = 3x(2 − x) and the x-axis.
4. Use integration to find the area between the line x + y = 3 and the two
coordinate axes.
Verify your answer using the formula for the area of a triangle.
5. Finding where the curve y = x 2 − 4x + 3 crosses the two coordinate axes.

Hence find the area between the curve and the x-axis.
4
6. Sketch the graph of the function y = x 2 and find the area between the
curve and the lines x = 2 and x = 4.
Find also the area between the curve and the x-axis for all x ≥ 2. Use this
Z ∞
4
result towrite down the value of 2
d x.
4 x
7. Find the area enclosed by the curve y = 3x 2 − 18x + 2 and the x-axis.
8. Sketch the curve y = x 2 (3 − x). Find the area bounded by the curve and the
x-axis.
9. Draw a sketch showing the line x = 2 crossing the x-axis at P , the curve
y = x 2 at Q and the curve y = 2x 2 at R. If O is the origin, show that the areas
OPQ and OQR (where each area is bound by at least one of the curves) are
equal. See if you can prove it without doing any integration.
Z 4
10. Evaluate the integral |2 − x| d x.
0
Z 2
11. Evaluate the integral (4 − x 2 ) d x.
0
Sketch the curves y = 4 − x 2 and y = x 2 − 4 and use the symmetries to write
down the area enclosed between the two curves.
12. Find where the line y = 6 intersects the curve y = x 2 −4x +9. Find the area
of the region between the line and the curve.
13. Find the area between the curve y = x(6 − x) cut off by the line y = 8.
14. Find where the line y = x intersects the curve y = x 2 − 2x + 2. Hence find
the area between the line and the curve.
15. Find the area enclosed between the curve y = x(5 − x) and the line y = 2x.
16. Find where the line y = x + 2 crosses the curve y = x 2 − 4x + 6. Hence find
the area enclosed by the line and the curve.

Solutions 39
³Z 2 Z 2
15 4
x(2 − x) d x = 3 × 34
¢
1. 15 2. 4 3. 3, 4 3x(2 − x) d x = 3
1 1
y
2
y = x(2 − x)
1
0 x
−1 1 2
−1
9
4. 2 5. (0, 3), (1, 0) and (3, 0). Area = − 43
³£ ¤ ´
4
6. Area = 1 − x4 2 = [−1] − [−2]
y For all x ≥ 2, area = 2
6 £ 4 ¤∞
− x 2 = 0 − [−2]
4
³Integral = 1 ´
Equal to the area beyond x = 4
2
0 x
1 2 3 4 5
³ h i4 ¤4 ´
x3
− 3x 2 + 8x = x 3 − 9x 2 + 24x 2
£
7. −4 y = 3(x − 2)(x − 4). Area = 3 3 2
8. y
Area = 27 4
4 3
³h i ´
x 3 − x4
0
x
−1 0 1 2 3
9. Z 2 Z 2
2
y Area OP R = 2x d x = 2 x2 d x
y = 2x 2 0 0
R = 2× Area OPQ
Q y = x2
x
O P
y
10. Answer: 4 2
x
0 1 2 3 4

16 64
11. Integral (shaded area) = 3
Area enclosed by the two curves = 3
y
0 x
−2 −1 1 2
−2
−4
³ ´
12. (1, 6) and (3, 6) x 2 − 4x + 9 = 6 ⇒ (x − 1)(x − 3) = 0
³Z 3 ´
Area = 3 4
(x 2 − 4x + 9) d x
1
13. Area = 34
¡
We can see from sketches that the configuration can be transformed into that
of the previous question by¢ two translations and a reflection. This explains
why the areas are the same.
1
14. Area = 6
y Intersection at (1, 2) and (2, 2).
Area of trapezium = 23 .
Z 2
2 (x 2 − 2x + 2) d x = 43
1
3 4
Overlap area = 2
− 3
1
x
−1 0 1 2
9
15. Area = 2 y
Z 3 h i3 (3, 6)
2 5x 2 x3 27
(5x − x ) d x = 2
− 3 0
= 2 y = x(5 − x)
0
Area of triangle = 9.
27 9
Shaded area = 2
−9 = 2
x
0 3 5
9
16. Area = 2
y Intersection where
y = x 2 − 4x + 6 = x + 2 ⇒
6 x = 1, y = 3 or x = 4, y = 6
y = x +2 Area beneath curve
h 3 i4
= x3 − 2x 2 + 6x = 9
1
2 Trapezium area
3 27
= 2 (3 + 6) = 2
x 27
0 1 2 3 4 5 Area required = 2 −9


• Introduction. What is calculus? The value of y versus how rapidly y is
changing. The gradient at a point on a curve.
• Differentiation. The gradient function, derived function or derivative

dy
of y = f (x) = ax n is nax n−1 = f 0 (x) = d x .
• Differentiation from first principles using the binomial expansion and

the idea of limits.
p 1
• Differentiating non-polynomial functions such as 4 x + x 3 .
• The values of x for which a function is increasing or decreasing.
• Tangent and Normal to a curve.
• Stationary values of a function, where the gradient is zero. These are

turning points (i.e. a local maximum or local minimum points) and
points of inflexions.
d2y
• The second derivative f 00 (x) = d x . Its use for distinguishing types
stationary value.
To summarise, if there is stationary value at P , for which x = a, then P
is
a local maximum if f "(a) < 0
a local minimum if f "(a) > 0
a point of inflexion if f "(a) = 0
Z
ax n+1
• Indefinite integration. ax n d x = n+1 + c (n 6= −1)
Conversion to and from index form, for example

2 − 13 2 8+x 5
x = p , = 8x −2 + x 3
3 33x x2
• Definite integration and areas. Finding areas: under curves and

between lines / curves.
Theory of integration (optional). Odd and even functions. Harder area
examples.

Chapter 11
EXPONENTIAL EQUATIONS AND

LOGARITHMS
Overview
Preview of the index rules underpinning the whole chapter.
11.1 Exponential equations. Where the solution can be found by

inspection, or after some simple manipulation. For example
3x = 81, 252x = 5, 16x = 8x+2 .
11.1.1 Quadratic Exponential equations. Equations containing exponents
that have an underlying quadratic structure, for example
32x − 3x − 6 = 0, 16x − 3(4x ) − 4 = 0, 9x = 36 − 3x+2 .
11.2 Exponential graphs. Sketching them. For example y = 2x and y = 2−x .
11.3 The function f (x) = e x What is special about the number e ?
Gradient function of e x and e ax .
11.4 Transforming exponential functions. Translating, stretching,
reflecting. Effect on intercepts with the axes and on the asymptote.
11.5 Logarithms. Definition: log A Y = X ⇔ Y = A X .
Hence log A A = 1, log A 1 = 0. Solving equations e.g. log3 x = 4, y = log10 1000
Logarithm Laws.
log X + log Y = log(X Y )
X
µ ¶
log X − log Y = log
Y
log X N = N log X
Solving exponential equations with logs e.g. 2x + 10, 22y+5 = 0.5, 4x = 121−x
1
Simplifying expressions e.g. 3 log 2 + log 5 − 2 log 4, ln e 2 , 2 log pq − log 3p 2 .
11.6 Exponential modelling. Example of practical problems e.g. population
growth, radioactive decay.
11.7 Non-linear relationships. Using logs to convert functions of the form
y = ax n or y = ab x to a linear relationship.
11-1
CHAPTER 11. EXPONENTIAL EQUATIONS AND LOGARITHMS
First a reminder of all the index rules, including negative and fractional
indices. These provide the basis for everything we’ll be doing with
exponential equations and logarithms.
a m × a n = a m+n
am
= a m−n
an
(ab)n = a n b n
¡ a ¢n a n
= n
b b
(a m )n = a mn
1
a −n =
an
m p p
a n = ( n a)m or n a m

11.1 Exponential equations

Before we get started on this topic, some useful numbers. The bases you are
most likely to meet are 2, 3, 4, 5, 10, etc and you will be expected to recognise
the powers in the table below:
n 0 1 2 3 4 5 6 7 8 9 10
2n 1 2 4 8 16 32 64 128 256 512 1024
3n 1 3 9 27 81 243 729
5n 1 5 25 125 625
Commonly occurring powers of 4 can be deduced from the powers of 2. For

example 45 = (22 )5 = 210 = 1024 and 93 = 36 = 729.
So what’s an exponential equation? It’s one in which the variable appears in
the index (or exponent). Some simple examples:
10x = 1000 Solution: x = 3

1
32 y = 2 Solution: y = 5
1
3z = 81 Solution: z = −4
x+3
2 = 256 Solution: x = 5
We can solve these equations by inspection, because we recognise the

numbers on the right hand side, and understand the laws of indices (see
page 2-2).
Let’s look a bit more closely at how this works; it will gives us a clue as to how
we can solve equations that are not so straight forward. Take the last
example 2x+3 = 256 we can write as 2x+3 = 28 . Because both sides are written
in exponent form and both bases are the same, their exponents must be the
same: x + 3 = 8, so x = 5. Here is another example.
Example 1 101−2x = 0.001
Solution
Since 0.001 = 10−3 , we have
101−2x = 10−3
∴ 1 − 2x = −3
4 = 2x
x = 2
Substitute x = 2 into the original equation to satisfy yourself that it’s correct.
We were able to express the right hand side in exponential form, using the
same base as the left hand side. Then all we had to do was equate the
exponents and solve the resulting equation.

Suppose both sides of the equation are already in index (or exponent) form
but the bases are different? You may remember in Chapter 2 we talked about
decomposing bases—breaking them down into factors. We can use this
technique here.
Example 2 54x = 25x−1
Solution
Since 25 = 52 ,
54x = (52 )x−1 = 52x−2
∴ 4x = 2x − 2
2x = −2
x = −1
We could have chosen to make the bases on each side1 25. It is generally
easier to convert to the smaller base.
Example 3 32 y = 8 y−1
Solution
In Example 2 we converted both sides to the same base. We could do the
5
same here. We could write 32 as 8 3 , and then proceed as before:
5
(8 3 ) y = 8 y−1
5y
8 3 = 8 y−1
5y
∴ = y −1
3
5y = 3y − 3
3 = 2y
3
y = −
2
The decomposition method is usually simpler, converting both bases to the
smallest possible value, in this case 2.
(25 ) y = (23 ) y−1

25y = 23y−3
5y = 3y − 3
3
y = −
2
1
The solution would then be
1
((25) 2 )4x = 25x−1
(25)2x = 25x−1
2x = x −1
x = −1

We end with an example that involves solving a quadratic equation rather

than a linear one.
2
Example 4 9x = 272x
Solution.
2 2
The expression 9x may look ambiguous to you. Does it mean 9(x ) or (9x )2 ?
It isn’t because, as an index, the 2 only applies to what it’s attached to, in this
case the x. We’d need to put in a bracket if we wanted it to apply to the whole
of 9x , i.e. (9x )2 .
Decomposing each base as we did in Example 3, we have
2
(32 )x = (33 )2x
2
32x = 36x
2x 2 = 6x
2x 2 − 6x = 0
2x(x − 3) = 0
∴ x = 0 or x = 3

11.1.1 Quadratic Exponential Equations

We have seen that the first thing to check with an equation that has a variable
in an exponent is whether or not it is of the form A B = C D . If it is, we may
be able to use decomposition to write it so that the bases are the same. You’ll
find out what to do if the bases can’t be made the same in C2.
Until then, if the equation doesn’t fit this pattern, the substitution technique
will always work. The examples we saw of this in the previous section were of
simultaneous exponential equations. Here we will use substitution to solve
exponential equations that have an underlying quadratic structure.
Example 5 22x − 2x − 12 = 0
Solution
The trick is to regard 22x as (2x )2 .
(We’ll be using this trick repeatedly.) Now we can write the equation as
(2x )2 − 2x − 12 = 0
Substituting2 y = 2x , the underlying quadratic structure emerges:
y 2 − y − 12 = 0
(y − 4)(y + 3) = 0
∴ y = 4 or y = −3
x
2 = 4 or 2x = −3
We appear to have two solutions. One of them, 2x = 4, yields the solution

x = 2, but the other, 2x = −3, has no (real) solution. This is because raising 2
to any power will give a positive number. (Making the power negative simply
turns it into a fraction.) So there is only one solution: x = 2. Check for
yourself that it’s a correct solution.
A key feature of this type of problem is that the quadratic structure can be
disguised in different ways. Look at this equation:
4x − 2x − 12 = 0
Can you see that

4x = (22 )x = 22x = (2x )2
so it is exactly the same equation! An examiner could make it appear even
less like a quadratic by writing it like this:
4x = 2x + 12
Notice that we can’t solve this by equating exponents; although tantalisingly

close, it is not of the form A B = C D .
2
Any variable will do. We could equally well use, say X = 2x .

Example 6 72x + 7x = 2
Solution
As before, we write 72x as (7x )2 and substitute y = 7x , giving
(7x )2 + 7x = 2
2
y +y = 2
2
y + y −2 = 0
(y + 2)(y − 1) = 0
∴ y = −2 or y = 1
x
that is 7 = −2 or 7x = 1
As we saw in the previous example, since 7x is always > 0, the only solution
comes from 7x = 1, that is x = 0. (A quick check tells us that x = 0 does
indeed satisfy the equation.)
Example 7 9t − 10(3t ) + 9 = 0
Solution
This example looks more complicated, but the clue is in the presence of a 3t
and a 9t , which can be written as (3t )2 . We use substitution. The choice of
variable is ours; all we are doing in effect is giving 3t a label. Putting x = 3t
(3t )2 − 10(3t ) + 9 = 0
x 2 − 10x + 9 = 0
(x − 1)(x − 9) = 0
∴ x = 1 or x = 9
and the solutions are given by 3t = 1 and 3t = 9, that is t = 0 or t = 2.
Example 8 22x = 2x+1 + 8
Solution
The clue that it is a quadratic is in the 22x term, which is (2x )2 . But we would
hope to see a 2x term; instead there is a 2x+1 term. This is a common
disguise: fortunately we can write 2x+1 = 2(2x ) and now we can see that the
equation is just like the previous example. Rearranging and substituting
y = 2x we have
(2x )2 − 2(2x ) − 8 = 0
y 2 − 2y − 8 = 0
(y − 4)(y + 2) = 0
Now we have had some practice, we can see that the factor (y + 2) doesn’t
provide a solution. The only one comes from y − 4 = 2x − 4 = 0, that is x = 2.

Example 9 2x − 12(2−x ) = 1
Solution
In this example the disguise is even more subtle. Notice the negative index in
the second term. With very few exceptions it’s safe to assume that whenever
we have a negative index it will help to convert it into fraction form. In this
case it changes the equation to
1
2x − 12 =1
2x
Now, substituting y = 2x , we have3
12
y− = 1.
y
Multiplying through by y and rearranging in the usual way, we have
y 2 − y − 12 = 0
You may recognise this as the exactly the same as the equation we derived in
Example 5. So the answer is x = 2.
Check.
2x − 12(2−x ) = 22 − 12(2−2 ) = 4 − 12 × 14 = 4 − 3 = 1.
Example 10 3x − 32−x = 8
Solution
In this example the disguise is even heavier. The negative index hidden within
1
the second term is the clue, because 3−x = x . We can re-write the second
3
term:
9
32−x = 32 3−x = x
3
and the equation is now
9
3x − x = 8
3
a form that you’ll recognise from the previous example. Putting y = 3x and
rearranging we have:
y 2 − 8y − 9 = 0
Solve this for yourself and hence find the roots of the original equation (the
answer is in the footnote4 ).
You don’t need to remember all these variations; you just need to know what
to look out for, which comes with practice.
3
The second term is 12 1 12
1 × 2x = 2x .
4
(y − 9)(y + 1) = 0 ⇒ y = 3 = 9 or y = 3x = −1. The only root is x = 2.
x

Exercises 40
1. Solve each of these exponential equations by inspection.

1
(a) 3x = 81 (b) 1000 y = 10 (c) 4x = 16 (d) 8 y = 32
(e) 25 y = 125 (f ) 10z = 0.1 (g) 9a = 1 (h) 64k = 18
¡ 3 ¢t
(i) 400x = 20 (j) 400x = 0.05 (k) 16 y = 0.125 (l) 2 = 0.6̇
2. Solve each of these equations.
(a) 102x = 1000 (b) 36−x = 9 (c) 27 y = 23+y (d) 52x − 5x = 0
2
(e) 43x+1 = 16 (f ) 5 y = 53y (g) 81x+1 = 9 (h) 2a+1 = 0.5
y x
(i) 92x = 3 (j) 8 3 =2 (k) 7 5 = 49 (l) 10−t = 0.01
3. Solve each of these equations.

(a) 16x = 8x+2 (b) 1002y = 1000 y+1 (c) 125x = 52−x
¡ 1 ¢x 1
(d) 272x = 81x−2 (e) 2 = 41−x (f) 16 y = 2 y
x
(g) 9(3x ) = 27 (h) 9(3x ) = 1 (i) 4 3 = 2x+1
4 2 2
(j) 16 x = 512x (k) 2x = 512 (l) 2x = 16x
2 2
(m) 3 y = 81 (n) 27 y = (3 y ) y (o) 5x −1
=1
4. Solve these equations.(This question is optional.)

3 1 2 2
(a) 2t = 256 (b) 10−y = 0.0001 (c) 4x = −2
2
2 ax 1
(d) 3 y = (3−y )2 (e) a 2x = a 3 (f) 8x = 2x+2
1 2 2 27x x
(g) 8x = 2x +2
(h) 9x = 3 (i) 5(x ) = (5x )x
5. Solve these simultaneous equations.
(a) 5x+y = 125 (b) 9x = 27 y

2 y−x = 32 x+y = 5
(c) 3x+2y = 3 (d ) 10 y+1 = 1000x
2x+y = 8 53x−2y = 625
(e) 2x + 5 y = 13 ( f ) 4x+1 + 2(3 y ) = 19
2x+1 − 5 y = 11 4x+21 + 3 y+1 = 31
(g ) 9x × 3 y = 81 (h) (2x ) y = 64
2x+1 − 5 y = 11 102y−x = 10, 000
x
(i ) 27 y = 9 (j) 7x − 6 y = 6
2x+7 = 161y 2(7x ) − 3(6 y ) = 11
6. In each case find the value of k.

(a) 2n+2 + 2n+1 = k.2n (b) 5(3n+1 ) + 3n+2 = k.3n
(c) 16n − 42n = k.4n (d) 2n − 2n−1 = k.2n−1

7. Solve these exponential equations.

(a) 32y − 3 y − 6 = 0 (b) 4x − 9(2x ) + 8 = 0
(c) 52x + 5x = 2 (d) 9t = 3t + 72
(e) 49 y − 8(7 y ) + 7 = 0 (f ) 16x − 3(4x ) − 4 = 0
(g) 42x = 4 + 3(4x ) (h) 22x + 2x+1 − 8 = 0
(i) 81 y − 9 y − 6 = 0 (j) 64x = 12 + 8x
(k) 25 y + 5 y+1 = 6 (l) 9x = 36 − 3x+2
2 1
(m) 4x + 4x =3 (n) 2x + 2x =2
x −x x
(o) 7 + 7 =2 (p) 4 + 8(4−x ) = 6
(q) 3 y = 10 − 9(3−y ) (r) 6 y+1 + 6 y = 42
(s) 3x + 31−x = 4 (t) 4x+1 + 4x−1 = 17
(u) 5t = 5t −1 + 20 (v) 2x − 23−x = 7
(w) 2x − 2x−3 = 7

Solutions 40
1. (a) x = 4 (b) y = 13 (c) x = −2 (d) y = 35

(e) y = 32 (f) z = −1 (g) a = 0 (h) k = − 12
(i) x = 21 (j) x = − 12 (k) y = − 34 (l) t = −1
2. (a) x = 23 (b) x = 4 (c) y = 12

(d) 0. Write the equation as 52x = 5x , or as 5x (5x − 1) = 0.
(e) x = 13 (f) y = 0 or y = 3, because y 2 − 3y = 0.
(g) x = − 12 (h) a = −2 (i) x = 41
(j) y = 1 (k) x = 10 (l) t = 2
3. (a) x = 6 (b) y = 3 (c) x = 12

(d) x = −4 (e) x = 2 (f) y = ±2
(g) x = 1 (h) x = −1 (i) x = −3
(j) x = ± 34 (k) x = ±3 (l) x = 0 or x = 4
(m) y = ±2 (n) y = 0 or y = 3 (o) x = ±1
4. (a) t = −2 (b) y = ±2
(c) No solution. a x is always positive. (d) y = 0 or y = −2
(e) x = −1 or x = 3 (f) x = 0 or x = 2
(g) x = − 21 (h) x = −1 or x = −2 (i) x = 1
2
or x = 1
5. Alongside the solution is the underlying pair of linear equations,

together with any substitutions.
(a) x = −1, y =4 x + y = 3, y −x =5
(b) x = 3, y =2 2x = 3y, x+y =5
(c) x = 5, y = −2 x + y = 3, x + 2y = 1
(d) x = 1, y =2 y + 1 = 3x , 3y − 2x = 2
(e) x = 3, y =1 X + Y = 13 , 2X − Y = 11
X = 8, Y =5 X = 2x , Y = 5y
(f) x = −1, y =2 4X + 2Y = 19 , 16X + 3Y = 31
X = 14 , Y =9 X =4 , x
Y = 3y
(g) x = 32 , y =1 4x = 3y + 3 , 2x + y = 4
(h) x = 2, y =3 xy = 6 , 2y − x = 4
or x = −6, y = −1 (x 2 + 4x − 12 = 0 or y 2 − 2y − 3 = 0)
(i) x = −1, y = − 32 3x
2y = 2 or 3x = 4y , x + 7 = −4y
(j) x = 1, y =0 X −Y = 6 , 2X − 3Y = 11
X = 7, Y =1 X = 7x , Y = 6y

6. (a) k = 6. [4(2n ) + 2(2n )] (b) k = 24. [15(3n ) + 9(3n )]

(c) k = 0. [16n = (42 )n ] (d) k = 1. [2(2n−1 ) − 2n−1 ]
7. The underlying quadratic equation is given in the bracket.

(a) y = 1 [x 2 − x − 6 = 0] (b) x = 0 or 3 [y 2 − 9y + 8 = 0]
(c) x = 0 [y 2 + y − 2 = 0] (d) t = 2 [y 2 − y − 72 = 0]
(e) y = 0 or 1 [x 2 − 8x + 7 = 0] (f ) x = 1 [y 2 − 3y − 4 = 0]
(g) x = 1 [Same as part(f)] (h) x = 1 [y 2 + 2y − 8 = 0]
(i) y = 12 [x 2 − x − 6 = 0] (j) x = 23 [y 2 − y − 12 = 0]
(k) y = 0 [x 2 + 5x − 6 = 0] (l) x = 1 [y 2 + 9y − 36 = 0]
1
(m) x = 0 or 2
[y 2 − 3y + 2 = 0] (n) x = 0 [y 2 − 2y + 1 = 0]
(o) x = 0 [Same as part (n)] (p) y = 0 or 2 [x 2 − 10x + 9 = 0]
(q) x = 12 or 1 [y 2 − 6y + 8 = 0] (r) y = 1 [6 y (6 + 1) = 42]
(s) x = 0 or 1 [y 2 − 4y + 3 = 0] (t) x = 1 [4x (4 + 14 ) = 17]
(u) t = 2 [5t (1 − 51 ) = 20] (v) x = 3 [y 2 − 7y − 8 = 0]
(w) x = 3 [2x (1 − 18 ) = 7]

11.2 Exponential graphs

An exponential function5 is one that can be written
y = ax + b
where a and b are constants. The variable x only appears in an exponent, or

index. In the examples we’ll be looking at here b = 0.
An example would be y = 2x .
We met this function in Chapter 2 (Indices and Surds).
You’ll find a summary of the rules of indices on page 2-48.
x 0 1 2 3 4 5 6 7 8 9 10
2x 1 2 4 8 16 32 64 128 256 512 1024
Although this starts off looking not very different from y = x 2 , by the time we
reach x = 10 they differ by a factor of about 10. We’ve drawn below the two
curves up to this point. The further we go the more dramatically they diverge:
for example 152 = 225 whereas 215 is approximately 33, 000.
We can perhaps see why the term exponential growth has entered ordinary
language to describe a very rapid rate of increase.
5
If you Google Exponential Function, you will see a different definition. The term
has in practice been co-opted for the specific function y = e x , where e is Euler’s
number. The reason for this, and what Euler’s number is, will become clear if you
go on to study A2 Maths.

What about negative values of x? In Chapter 2 we saw that a negative index

indicates a fraction:
1
a −x = x .
1 a
So for example 2−3 = 23 .
The table below covers the range x = −5 to +5
x −5 −4 −3 −2 −1 0 1 2 3 4 5
1 1 1 1 1
2x 32 16 8 4 2 1 2 4 8 16 32
If we keep to low values of x, we can draw the curve fairly accurately:
y = 2x
30
25
20
15
10
x
−3 −2 −1 0 1 2 3 4 5
Outside this range it becomes difficult to draw: for large positive values of x
the graph is almost indistinguishable from a vertical line.
For large negative values x the curve will appear to be virtually horizontal. In
fact of course it is getting closer and closer to the x-axis—the curve is
asymptotic to the line y = 0.

The number we are raising to a power, in this case 2, is called the base.
(Confusingly “base" has different mathematical meanings in different
contexts.) If we use a bigger base, the function increases even more rapidly.
What happens if the base is < 1? For example
µ ¶x
1
y= .
2
Using the rules of indices, we can rewrite the function:
µ ¶x
1 1x 1
= x = x = 2−x .
2 2 2
The table of values for this function, in the range x = −5 to +5, is exactly the
same as before except that all the values have been “reflected" in the middle
value (20 ).
x −5 −4 −3 −2 −1 0 1 2 3 4 5
1 1 1 1 1
2x 32 16 8 4 2 1 2 4 8 16 32
The curve looks like this ¡ 1 ¢x

y = 2−x = 2
15
10
x
−4 −3 −2 −1 0 1 2 3 4
As you may have guessed, this produces the curve for y = 2x reflected in the
y-axis.
We’ll be picking this idea up again in the next section. We’ll see that being
able to transform functions in various ways is a powerful aid to sketching
curves.
You will recognise y = 2x as an increasing function, so gradient functions
must be positive everywhere. Similarly y = 2−x as a decreasing function, so
its gradient function must be negative everywhere. We shall return to this in
the coming pages: the question of how to differentiate these functions.

Earlier in the chapter we solved lots of exponential equations. In every case

we solved them exactly; the answers were integers or a rational numbers6 .
What about an equation like this
2x = 13
The solution lies somewhere between x = 3 and x = 4.
The answer is x ≈ 3.70. If you check this using your calculator you will find
that 23.70 ≈ 12.996. But what does this mean?
37 p
10
23.7 = 2 10 = ( 2)37
We can see that it does have a meaning using what we understand about
fractional indices.
Later in the chapter we will see how to solve such equations, using
logarithms. For now the best we can do is the estimate it using a graph. The
value in this is not its practical use but to see that an equation such as
2x = 13 makes sense.
y = 2x
20
15
13
10
x
−3 −2 −1 0 1 2 3 4
6
A fraction formed by one integer divided by another.

11.3 The function y = e x

The letter e stands for a very special number, rather like the number π.
π ≈ 3.141593, e ≈ 2.718282
They are both irrational7 numbers, so we can never write down their exact
values.
They are numbers that have a significance that is built into the way the
universe is constructed. In the case of π, it is the ratio of the circumference
of a circle to its diameter; we are about to see what that significant is for e.
We start by looking at the graphs of y = 2x and y = 3x , because
y = (2.718282)x must in some sense lie between them.
y = 2x y = 3x
30 30
25 25
20 20
15 15
10 10
5 5
x x
−3 −2 −1 0 1 2 3 4 −3 −2 −1 0 1 2 3 4
As expected y = 3x increases much more rapidly that y = 2x . At least it does

for positive values of x; for negative values of x the reverse is true, although it
is hard to see this clearly on the graphs, so we have drawn a close up for each
curve on the same graph.
y = 3x
8
y = 2x
x
−2 −1 0 1 2
7
An irrational number if one that cannot be written as one integer divided by another.

The next step is to look at the gradient function8 for each of them. We don’t
yet have the tools for differentiating them so you will have to take this on
trust.
Function Gradient function

f (x) = 2x f 0 (x) ≈ 0.693 × 2x
f (x) = 3x f 0 (x) ≈ 1.099 × 3x
This time we will compare each function with its gradient function. You see
something interesting which point the way the the function y = e x . To make
it clearer we are focussing in on the lower part of each curve.
This time the blue curve is f (x) and the red curve is the gradient function
f 0 (x).
f (x) = 2x f (x) = 2x f (x) = 3x f 0 (x) = 1.099 × 3x
6 6
f (x) = 3x
f 0 (x) = 0.693 × 2x
4 4
2 2
x x
−2 −1 0 1 2 −2 −1 0 1 2
Notice that the gradient function 2x is below the function itself, whereas the
opposite is true of the function 3x , where the gradient function is above
it–and very close to it because the factor 1.099 is close to 1.
There must be a function therefore, of the form f (x) = k x for which
f 0 (x) = f (x), where 2 < k < 3. This number is e.
A
x 0 x
If f (x) = e , then f (x) = e
!
A
A
A
It is this property that makes the number e so special, and why it is woven
into the fabric of the universe in the same way that π is.
But before we see how, we need to learn how to differentiate the function
y = e ax where a is a constant.
8
Many students are tempted to differentiate 2x by saying it is x2x−1 . This is wrong: the
rule that works for x n only applies if the index (i.e. n) is a constant.

Differentiating y = e ax
You don’t need to now why, but to the derivative of y = e ax is the same
function multiplied by a. (a is a constant.)
A
ax 0 ax
If f (x) = e , then f (x) = ae
!
A
A
A
We already know that when we multiply a function by a constant, that

constant remains when we differentiate it. Therefore if k is a constant and
f (x) = ke ax , then f 0 (x) = kae ax . Some examples
f (x) f’(x)
4e 2x 8e 2x
1 1
6e 2 x + 2 3e 2 x
2e −x − x 2 −2(e −x + x)
1 + αe −αx −α2 e −αx
e kx + e −kx k(e x − e −x )
1
= e −2x −2e −2x
e 2x
e 5x −1 e 5x 1
= e 3x − e −2x 3e 3x
+ 2e −2x 2x − 2x
e 2x e e
(e 3x + 2)2 6e 3x (e 3x + 2) e 6x + 4e 3x + 4
p 3 3x 1 3
e 3x 2
e2 (e 3x ) 2 = e 2 x

Why e is such a special number

We are now in a position to see why the exponential function is build into
the way the universe works.
Consider a variable that has the property that its rate of change is
proportional to the value of the variable.
For example an infectious disease spreading: if there are N infected people,
dN
its rate of spreading i.e. d t is proportion to N .
Radioactive decay has a similar feature. A mass m of radioactive material
dm
turns into radiation, at a rate proportional to the mass remaining: d t is
proportional to −m (negative because the mass is reducing over time).
dy
We can generalise this using the relationship ∝ y, i.e.
dx
dy
= k y for some constant k.
dx
We call this a differential equation for obvious reasons.
Now consider the function y = e kx . This is a solution of the equation, because
dy
= ke kx = k y.
dx
Check for yourself that y = C e kx also satisfies the equation, where C is any
constant. This is the equivalent of the constant of integration that we met in
Chapter 10.
dy
So the general solution of the equation = k y is y = C e kx .
dx

Example 11
(a) Find the gradient of the curve y = 3e 2x at x = −1, x = 0 and x = 2.
(b) Write down the gradient at x = 0 of the curves y = 1 + 3e 2x and y = 3e −2x .
(c) Find the equation of the tangent to the curve y = 3e 2x at x = 2 and
determine the coordinates of where the line meets the axes.
Solution
dy
(a) If y = 3e 2x , = 6e 2x ,
dx
6
So the gradients at x = −1 is 6e −2 = ,
e2
at x = 0 it is 6 and at x = 2 it is 6e 4 .
(b) At x = 0 the gradient of y = 1 + 3e 2x is 6, because the curve is a translation
of y = e 2x parallel to the y-axis.
y = 3e −2x is the reflection of y = 3e 2x in x = 0 so its gradient at x = 0 is −6.
(c) When x = 2, y = 3e 4 , so the equation of the tangent is y − 3e 4 = 6e 4 (x − 2)
i.e. y = 6e 4 x − 9e 4 or y = 3e 4 (2x − 3)
3
When x = 0, y = −9e 4 , when y = 0, x = .
2
The tangent therefore meets the axes at (0, −9e 4 ) and 23 , 0
¡ ¢
QQ
1. Differentiate each of these functions (the answers are in the footnote 9 ).

1
(a) e 4x (b) e −x (c) e kx (d) 2e 5x (e) 8e 2 x
1 2 e x +2
(f) e x (e x + 1) (g) e 2x (h) e x (i) e x (j) (e 2x − 1)2
2. Find the equation the tangent to the curve y = e 3x − 2 at x = 1.
3. Show that
1
1. (a) 4e 4x (b) −e −x (c) ke kx (d) 10e 5x (e) 4e 2 x
9
−2 2
(f) e x (2e x + 1) (g) 2x (h) 2e −x = x (i) −2e −x (j) 4e 2x (e 2x − 1)
e e
2. y = 3e 3 x − 2e 3 − 2

11.4 Transforming exponential functions

In Chapter 8 we examined the transformation of a number of functions. We
complete that work here looking at how translations, stretches and
reflections apply to the exponential function.
Example 12
The curve C is the function y = 2x .

When C is stetched by a scale factor 2 parallel to the y-axis it becomes C 1 .
When C is translated a distance −1 parallel to the x-axis it becomes C 2 .
(a) Sketch C and C 1 on the same graph.
(b) Explain why the curve C 2 is identical to C 1 .
Solution
(a) The curve C is in black; C 1 is in blue. We have labelled some points to

help us contruct what the curves look like.
C1
y = 2x
6
10
C
9
8 • (2, 8)
4
×2
7
5
6
4 • (2, 4)
2•
1•
-x
−3 −2 −1 0 1 2 3 4
(b) If C represents the function y = f (x) = 2x then C 2 will represent the

function
y = f (x + 1)
= 2x+1
= 2 x × 21
= 2 f (x)
which is the curve C 1 .

It makes sense algebraically. What about geometrically? Visualise translating
C 1 to the left and satisfy yourself that the curve C 2 looks the same as C 1 . There
is a graph showing this on the next page.

C2
y = 2x
6
10
C
9
5
C 1
4 •(1, 4) • (2, 4)
2• • (1, 2)
1
-x
−3 −2 −1 0 1 2 3 4

Transforming y = e x
Owing to the importance of the function e x we will look more closely at
transformations of it. We know that e x → ∞ as x → +∞, and it → 0 as
x → −∞. We also known that the x-axis is an asymptote. Remembering what
we learned about curve sketching in Chapter 8, here are some examples.
We start with the curve10 y = e x . They all pass through (0, 1) because a 0 = 1.
Identify each of the transformations. The answers are at the bottom of the
next page.
(i) y = ex (ii) y = ex + 2
•
(1, e + 2)
• 3
(1, e)
2
1
x x
−2 −1 0 1 2 −2 −1 0 1 2
(iii) y = e −x (iv) y = e −x − 1
(−1, e)•
1
x x
−2 −1 0 1 2 −2 −1 0 1 2
−1
(v) y = −e −x (vi) y = 3 − e −x
3
2
x x
−2 −1 0 1 2 −2 −1 0 1 2
−1
Think of it as
y = −e −x + 3
10
All curves of the form y = a x look very similar when sketched (a is a constant); it is only
the scale that changes.

(vii) y = ex (viii) y = 2e x
•
(1, 2e)
•
(1, e)
2
1
x x
−2 −1 0 1 2 −2 −1 0 1 2
(ix) y = e x+1 = e.e x (x) y = e 2x
• (0, e) •1
( 2 , e)
1
x x
−2 −1 0 1 2 −2 −1 0 1 2
(xii) y = e −2x
x+1
(xi) y =e −2
•
• (0, e − 2) (− 12 , e)
x
−2 −1 0 1 2 1
x
−2 −2 −1 0 1 2
Answers
¡0¢
We use the vector notation to describe translations, e.g. −3 is −3 units
parallel to the y-axis.
(ii) trans 02 (iii) refl in y-axis (iv) refl in y-axis and transl −1
¡¢ ¡0¢
(v) refl in x-axis and refl in y-axis (vi) refl in x-axis and transl 03
¡¢
(viii) Stretch s.f. 2 ∥ y-axis (ix) transl −1

¡ ¢
or stretch s.f. e ∥ y-axis
1
¡−1¢0 ¡0¢
(x) stretch s.f. 2 ∥ x-axis (xi) transl 0 and transl −2
(xii) stretch s.f. 21 ∥ x-axis and refl in y-axis.

QQ
1. In each case sketch on the same graph the two functions given. Mark
where each curve meets the y-axis and any asymptotes. State the
transformation from the first function to the second function.
1
(a) e x and 3e x (b) e x and e x − 1 (c) e −x and 2e −x (d) e x and e 2 x
2
(e) e x and e x−2 (f) −e x and 2 − e x
(g) 2e x and e x (h) e 2x and e 2x−3
2. The sketches below are of the function y = ae bx + c. Find a possible
equation for each function, where b will either be +1 or −1.
(a) y y
(b)
x
−2 −1 0 1 2
x −1
−2 −1 0 1 2
−2
(c) y (d) y
1
x
−2 −1 0 1 2
e −3•
x
−2 −1 0 1 2
(e) y (f) y
3
2
4
x
2 −2 −1 0 1 2
x
−2 −1 0 1 2
The answers are on the next page.

Solutions
1.(a) y = 3e x (b)
y = ex y = ex
y = ex − 1
1 1
x x
−2 −1 0 1 2 −2 −1 0 1 2
−1
¡0¢
Stretch s.f. 3 // y-axis Translation −1
(c) (d)
y = 2e −x
y = ex
1
y = e −x y = e 2x
2
1 1
x x
−2 −1 0 1 2 −2 −1 0 1 2
Stretch s.f. 2 // y-axis Stretch s.f. 2 // x-axis

(e) (f)
y = ex 2
1
x
−2 −1 0 1 2
−1 y = 2 − ex
y = e x−2
³ ´ y = −e x
1 0, e12 • (2, 1)
x
−2 −1 0 1 2
¡2¢ ¡0 ¢
Translation 0
Translation 2
(g) (h)
y = e 2x
y = 2e −x y = 2e x y = e 2x−3
2
³ ´
1
0, e 3 ¡3 ¢
1 • 2,1
x x
−2 −1 0 1 2 −2 −1 0 1 2
−1
3
Reflection in y-axis Stretch s.f. 2 // x-axis
2.(a) 3e −x (b) e x − 2 (c) e x−3 (d) 1 − e −x (e) 2e −x + 2 (f ) 3 − e x

11.5 Logarithms
By the end of this page you will know what a logarithm is, but for the
moment bear with me. The story starts with an equation we solved earlier in
the chapter
10x = 1000 and 2 y = 32.
We call these exponential equations because the unknown is in the exponent
(another word for power or index).
The solutions, x = 3 and y = 5, are whole numbers (integers). We may be able
go a little further and solve an equation like this
9z = 27
because 3 p
9 2 = ( 9)3 = 27.
3
The solution in this case, z = 2 , is a fraction (rational number).
We call this solving by inspection. It relies on our recognising familiar powers
of certain numbers, in other words we have to be a bit lucky.
What about equations like this?
10x = 1047 and 2 y = 93 and 3.7z = 2.
No luck here. For equations like this we need logarithms.

The easiest way to understand the concept of a logarithm is to see that these
two statement are equivalent:
log10 1000 = 3 and 103 = 1000.
We can think of the (base 10) logarithm of a number as the answer

to the question:
what power of 10 gives us this number?
We have seen that y = e x is a very important function, so we will often want

to solve an equation like
e x = 4 or e 2t +1 = 0.6.
From the definition of logs: x = loge 4 and 2t + 1 = loge 0.6

Because base e logs occur so often, rather than loge 4 or loge 0.6 we would
write
ln 4 or ln 0.6
(ln is short for natural logarithm). Your calculator will have a ln button. The
inverse of which is the e x function, because in the same way
We can think of the natural logarithm (base e) of a number as the

answer to the question:
what power of e gives us this number?

If follows, for example, that
log10 100, 000 = 5 and log10 1 = 0.
And slightly more interesting
log10 0.01 = −2
because
1
10−2 = .
100
Some example using the natural log:
p 1
ln e = and ln 1 = 0.
2
A log can have any base, so for example
log2 64 = 6 and log3 81 = 4

because 26 = 64 and 34 = 81.
Taking this a stage further
1 1 p
3
log3 27 = because 27 3 = 27 = 3
3
5 5 p
and log4 32 = because 4 2 = ( 4)5 = 32
2
In general, a logarithm (to any base) can be defined by the statement:
A
X
log A Y = X ⇐⇒ Y =A
!
A
A
A

Before moving on to solving equations using logs, let’s make sure you are
clear about what a logarithm is.
QQ
Without using a calculator find the values of these logs (the answers are in
the footnote 11 ).
(a) log2 32 (b) log9 81 (c) log7 17 1

(d) log10 1000
p 1
(e) log13 1 (f) log2 2 (g) log3 3 (h) ln e 2
11 1
(a) 5 (b) 2 (c) −1 (d) −3 (e) 0 (f ) 2 (g) 1 (h) −2

11.5.1 Logarithms and exponential equations

Let’s see how this helps us to solve equations like the ones above.
For obvious reasons, base 10 logs are used for calculations12 . By convention
when using base 10 we can drop the suffix and simply write
log 1000 = 3.
Use the log button on your calculator to check that log 1000 = 3.
Now use your calculator to find log 1047. The answer is 3.0199 (to 4 DP). From
our definition of what a log is, this means that
103.0199 ≈ 1047.
We have solved the equation 10x = 1047.

(An answer slightly bigger than 3 is not surpising, given that 103 = 1000.)
Now use the 10x button to confirm that 103.0199 gets us back to
(approximately) 1047.
We say that the 10x function is the inverse of the log x function, and vice
versa; each one reverses what the other does.
This is all very well, provided the base is 10. It doesn’t help us with an
equation like
2 y = 93.
There is no log2 button on a calculator. To overcome this problem we are
going to have to learn more about the log function.
log A A = 1 A
!
A
log A 1 = 0 A
A
it is easy to see why, because A 1 = A and A 0 = 1.
12
Later in the course you will discover that an even more important base than 10 is the base
e. This is a special number, rather like π. It occurs naturally in the physical world—in such
areas as radioactive decay. The value of e is approximately 2.7183—a strange number to have
as a base for logarithms you might think.
So important is it that, like π, it earns its own buttons on a calculator: the e button and the
loge function which is almost always written ln (standing for natural logarithm).

11.5.2 The fundamental laws of logarithms

These govern how we combine them and manipulate them13 . We’ll state
them first and then talk about them.
For any base of logs

X
µ ¶
A
Y
!
A
A
N A
log X = N log X
To see why, all we have to do is go back to how a log is defined. We’ll use an
example to demonstrate:
8 × 16 = 128 or 23 × 24 = 27 .
From the definition of what a log is:

3 = log2 8, 4 = log2 16, and 7 = log2 128
so log2 8 + log2 16 = log2 (8 × 16)
3 + 4 = 7
Logs are, in effect, indices: we add them when multiplying numbers and
subtract when dividing them.
By extension we could have more than two number multiplied together, for
example
log 6 + log 0.5 + log 7 = log(6 × 0.5 × 7) = log 21.
The third law is now easy to demonstrate:
3 log 2 = log 2 + log 2 + log 2 = log(2 × 2 × 2) = log 23

a particular case of N log X = log X N .
You won’t be required in your exams to prove the laws. But you will need to
commit the laws to memory and be fluent at applying them. This is very
important because they are not intuitive. Later we will look at some the
common mistakes students make as a result.
13
There are some more laws but, for your syllabus, you won’t need them yet.

Some more examples of applying the log laws:

µ ¶
1
log + log 14 = log 7
2
log 1.8 − log 0.3 = log 6
log 0.257 = 7 log 0.25
µ ¶
1
log = − log 5
5
This last example always comes as a bit of a surprise. We get to the same
result a different way:
µ ¶
1
log = log 1 − log 5
5
= 0 − log 5
because log 1 = 0 in any base.

Sometimes we want to go the other way, as it were. For example
µ ¶
2
log = log 2 − log 3
3
1
log 25 = log 5
2
Time for some algebraic examples:
log 3a − log a = log 3

log pq + log 3p 2 = log 3p 3 q
µ 2¶
3 2 y
log x y − log x y = log = 2 log y − log x
x
And an example using a specific base:
log2 8x = log2 8 + log2 x = 3 log2 2 + log2 x = 3 + log2 x.

I cannot emphasis enough that, until you are extremely familiar with them,
you will have to tread very carefully when applying the log laws. You need to
follow them slavishly because you won’t be able to rely on your intuition.

Before you try some examples for yourself there are a couple of very common
pitfalls to watch out for. The commonest is to assume that something like
log 3 + log 2 is equal to log 5
rather than log 6.

Here is another. Look at this expression
log 3a 2 .
This could be written 2 log 3a. Right?

Look again. Can you see why this is wrong? If not, there is an explanation on
the next page.
The main purpose of the following questions is not so much about getting
the “right” answer (sometimes there is more than one). It’s really about
getting used to manipulating logs correctly.
Don’t be surprised if initially you make lots of mistakes; that’s how you learn.
QQ
In each case use the log laws to rewrite these expressions. (Some possible
answers are on the next page.)
1. (a) log 30 − log 6 (b) log 0.2 + log 8 (c) log 710
(e) log2 18 (f) 12 log 9

¡ ¢
(d) 2 log 6 − log 9
p
(g) log10 100
¡ ¢
3
(h) log3 3 (i) 4 log 5 − log 125
2. (a) log 10x − log 5x (b) log 2ab + log ab 2 (c) 2 log y 3
³ ´ ¡y¢
(d) log b 5 − 3 log b (e) log xy + log x (f) log K − log K12
(g) log 2x (h) log 52y + log 5 y

Why is log 3a 2 not equal to 2 log 3a?
The appropriate law is: log X N = N log X .

If the expression were log(3a)2 we could write it as 2 log 3a (putting X = 3a
and N = 2).
In our case only the x is squared.
Solutions to the question on the previous page.
1. (a) log 5 (b) log 1.6

(c) 10 log 7 (d) log 4 [log 62 − log 9 = log( 36
9
)]
1 1
(e) −3 [2−3 = 8 ] (f) log 3 [log 9 2 ]
1 p
(g) 2 − log10 3 [log10 100 = 2] (h) 21 [3 2 = 3]
(i) log 5 [log 125 = 3 log 5]
2. (a) log 2 (b) log 2a 2 b 3

(c) 6 log y or log y 6 (d) 2 log b or log b 2
(e) 0 [log 1] (f ) 3 log K or log K 3 [log K −2 = −2 log K ]
(g) x log 2 (h) 3y log 5 [2y log 5 + y log 5]

11.5.3 Solving exponential equations using logarithms

We are now in a position to solve the types of equation we started this
section with. To do so we simply take the log of each side of the equation14 .
We can choose any base but typically we use base 10.
Example 13
Solve these equations, giving your answer to 4 decimal places.

(i) 10x = 1047
(ii) 2a = 51
(iii) 72y+3 = 0.5
Solutions
(i) We have already solved this but we solve it again here to illustrate the
approach of “taking logs”.
10x = 1047
x
log10 10 = log10 1047
i.e. x log10 10 = x = log10 1047
= 3.0199 (to 4 DP)
(ii)
2a = 51
log10 2a = log10 51
a log10 2 = log10 51
log10 51
a =
log10 2
a = 5.6724 (to 4 DP)
A quick check reassures us that this is in the right ball park: 25 = 32 and
26 = 64, so 25.6724 will be somewhere between the two.
For the time being, when using base 10 logs we are going to drop the suffix.
This is a common convention: if no base is specified it means base 10.
(iii)
72y+3 = 0.5
log 72y+3 = log 0.5
(2y + 3) log 7 = log 0.5
log 0.5
2y + 3 =
log 7
= −0.356207
−0.356207 − 3
y =
2
= −1.6781 (to 4 DP)
Notice that we only rounded to 4 DP at the last stage. We carried a little more
precision in the intermediate calculations.
14
This is valid because an equation remains true provided we do the same thing to both
sides.

Example 14
Solve these equations, giving your answer to 3 significant figures.

(i) 4x = 121−x
(ii) e 2x+3 = 10
Solutions
(i) As before we are going to take logs on both sides. Although this has the
desired effect of getting rid of the exponents, the linear equation we get is
more awkward looking. Try not to be put off by this.
4x = 121−x
x log 4 = (1 − x) log 12
0.60206x = 1.07918(1 − x)
= 1.07918 − 1.07918x
1.68124x = 1.07918
1.07918
x = = 0.64190
1.68124
= 0.642 (to 3 SF)
(ii) A straightforward example of the use of natural logs.
e 2x+3 = 10
2x + 3 = ln 10
= 2.3026
x = −0.3487
= −0.349 (3 SF)

QQ
Solve these equations, where necessary giving your answer to 3 significant

figures (the answers are in the footnote 15 ).
(a) 3x = 174 (b) 4x = 1.9 (c) 11x = 13 (d) e x+1 = 0.1
(e) 84x−1 = 2 (f ) 0.2x = 6 (g) 52x−1 = 2x (h) 3x+2 = 43x
15 1
(a) 4.20 (b) 0.423 (c) −0.458 (d) −3.30 (e) 3 (f) −1.11 (g) 0.637 (h) 0.718
(e) is the only part that can be done without using a calculator.
8 = 23 so l.h.s. is (23 )4x−1 = 23(4x−1) . So 3(4x − 1) = 1.
If you had difficullty with (g) or (h) here is part of the solution to part (h)
(x + 2) log 3 = 3x log 4 ⇒ 2 log 3 = x(3 log 4 − 2 log 3) = x(log 43 − log 32 ) = x log 64
9

11.5.4 Simplifying expressions involving logs

Not all equations involving the use of logs are as simple as these. Sometimes
the equation itself will contain a log function. We need to be able to
manipulate, rearrange and simplify expressions containing logs.
Example 15
Express each of these as a single logarithm.

(i) 3 log 2 + log 5 − 2 log 4
(ii) log 2 + log(3x − 1) − 2 log(x + 1)
Solve the equation log(6x − 2) − 2 log(x + 1) = 0.
Solutions
(i) 3 log 2 + log 5 − 2 log 4 = log 23 + log 5 − log 42

= log(8 × 5) − log 16
40
= log
16
5
= log
2
(ii) log 2 + log(3x − 1) − 2 log(x + 1) = log(6x − 2) − log(x + 1)2
6x − 2
= log
(x + 1)2
The equation log(6x − 2) − 2 log(x + 1) = 0 can be written
6x − 2
log = 0
(x + 1)2
6x − 2
⇒ = 1
(x + 1)2
because log 1 = 0 (in any base). Hence
6x − 2 = x 2 + 2x + 1
0 = x 2 − 4x + 3
(x − 1)(x − 3) = 0
⇒ x = 1 or x = 3
You may like to substitute x = 1 and x = 3 into the equation

log(6x − 2) − 2 log(x + 1) = 0 to satisfy yourself that they are solutions.

Example 16
ab 2
(a) Express log pc in terms of log a, log b and log c.
x6
(b) If log x = 43 log y express log y in terms of log y.
Solutions
(a) ab 2 p
log p = log a + log b 2 − log c
c
1
= log a + 2 log b − log c
2
(b) x6
log = log x 6 − log y
y
= 6 log x − log y
6 4
= × log y − log y = 8 log y − log y
1 3
= 7 log y

QQ
1. Write each of these expressions as a single logarithm (the answers are in

the footnote 16 ).
(a) log 3 + log 7 (b) 2 log 6 − log 3 (c) log 3 + 5 log 2 − log 4
(d) 12 log 9 − log 5 (e) 3 log a + log ab − log b 2 (f ) 31 log x 12
(g) k log 4 + log 4 (h) log(x + y) + log(x − y) (i) log(x + 3) − log(2x − 1)
2. Write this expression in terms of log x and log y

x
log 5 + 3 log x y.
y
The next two pages are for those of you who feel confident with logs and want
a bit of a challenge. If you don’t, you can skip them; they are not necessary for
tackling the rest of the chapter.
4
1. (a) log 21 (b) log 12 (c) log 24 (d) log 53 (e) log ab
16
x+3
(f ) log x 4 (g) log 4k+1 (h) log(x 2 − y 2 ) (i) log
2x−1
x x4
2. 4 log x − 2 log y [log 5 + log x 3 y 3 = log 2 = log x 4 − log y 2 ]
y y

Common errors
If you are comfortable with the log rules and want to test yourself, here are
some examples illustrating some of the commonest errors:
log 2 + log 5 6= log 7
log 8a − log 2a 6= log 6a
log 5x 2 6= 2 log 5x
log 6
6= log 2
log 3
log(x + 4) 6= log x + log 4
The log rules can be applied to the first three examples: they could be
written log 10, log 4 and log 5 + 2 log x respectively.
The last two examples appeal to what we might call wishful thinking:
imagining that there is an alternative way of writing them. There isn’t17 . If
you are unconvinced, look back at the rules.
The following list of statements is designed to try and trip you up. Many of
them are incorrect. Some are correct, although perhaps not obviously so. See
if you tell which is which (the answers are over the page).
1. log 3 + 3 log 2 = log 24

2. log a + log a 2 + log a 3 = 6 log a
3. log 3y 2 − log y 2 = log 2y 2
4. log ba + log ba = 0
log x 3
5. log x
= log x 2
1 p p
6. 2
log x + 21 log y = log x + log y
7. If log X + log Y = 2 then X + Y = 2
¡ ¢3
8. log 2 = 3 log 2
log 16
9. 2
= log 8
log 16
10. 2
= log 4
11. log 8x 2 = 2 log 8x
17
At least not as envisaged here.

Answers to the questions on the previous page.
1. log 3 + 3 log 2 = log 3 + log 8 = log 24. This is correct.
2. log a + log a 2 + log a 3 = log a 6 = 6 log a. This is correct.

3y 2
3. Incorrect. log 3y 2 − log y 2 = log y
= log 3.
4. Correct. log ba + log ba = log 1 = 0
log x 3
5. log x cannot be simplified
1 p p p
6. Incorrect. 2
log x + 21 log y = log x + log y = log x y or 12 log x y
7. Incorrect. It assumes that log(X + Y ) = log X + log Y .

¡ ¢3
8. log 2 cannot be simplified
log 16
9. The statement 2
= log 8 is incorrect.
log 16 1
10. This one is correct. 2
is the same as 12 log 16 = log 16 2 = log 4
11. Incorrect. log 8x 2 = log 8 + 2 log x

11.5.5 More problems involving logs

Here we will be drawing together everything we have learned about logs.
Armed with these tools we will be able to crack some very interesting
problems, including some from apparently unrelated topic areas.
Example 17
1
log4 (x + 3) − log4 x =
2
Solutions
x +3 1
log4 =
x 2
x +3 1
= 42 = 2
x
x + 3 = 2x
x = 3
Example 18
Write down the value of log8 2.

Hence solve the equation log7 x = 6 log8 2
Solution
1
83 = 2 ⇒ log8 2 = 13
The equation becomes
log7 x = 2
x = 72
= 49

Example 19
Solve the equations 32x = 8(3x ).
Solutions
The key here is to see that 32x = (3x )2 .
This is a quadratic in 3x . Putting y = 3x , and remembering that we collect all
the terms of a quadratic onto one side, we have
y 2 − 8y = 0
y(y − 8) = 0
y = 0 or y = 8
i.e. 3x = 0 or 3x = 8
3x = 0 has no solution. We will explain why in a moment.

So the only solution is to the equation 3x = 8, which we know how to solve
(e.g. see Example 18).
So why does 3x = 0 not have a solution? Whatever power we raise 3 (or

indeed any number) to, we never get zero. This is confirmed by looking at
the graph of such a function (the graph of y = 2x in on page 11-14). The
curve approaches the x-axis but never reaches it.
Example 20
Solve these equations

(a) e 2x − 6e x + 5 = 0 (b) e x − 1 = 2e −x (c) e x−1 (e 3x + 1)2 = 0.
Solutions
(a) We can either use a substitution e.g. X = e x , or simply regard the equation
as a quadratic in e x . Either way we end up with
(e x − 5)(e x − 1) = 0, which has two solutions, arising from e x = 1 and e x = 5.
i.e. x = 0 or x = ln 5 = 1.61 (3 SF).
(b) Multiplying through by e x and rearranging the terms we have
e 2x − e x − 2 = 0, which factorises: (e x − 2)(e x + 1).
But e x > 0 for all values of x so e x + 1 = 0 has no (real) solution.
The only solution is from e x − 2 = 0, i.e. x = ln 2 = 0.693 (3SF).
(c) This is something of a trick question.
First we can say that either e x−1 = 0 or (e 3x + 1)2 = 0.
We know that the exponential function is never negative: e x−1 > 0 whatever
the exponent. The curve of e x−1 is e x translated +1 parallel to the x-axis.
For the same reason e 3x + 1 > 1 so this is never zero either. The equation has
no solutions. You would not be set a question like this in an exam.

Example 21
Find p and q, given that

p = 3q and log3 p + log3 q = 2
Solution
The equation log3 p + log3 q = 2 can be written log3 pq = 2
i.e. pq = 32 = 9.
Substituting p = 3q into this equation we have
3q 2 = 9
q2 = 3
p p
q = 3 or − 3
p p
and p = 3 3 or − 3 3
Notice that there are exactly two solutions to the equation.

p p p p
p = 3, q = 3 3 or p = − 3, q = −3 3.
p p
Mixing and matching doesn’t work (e.g. p = 3 with q = −3 3).
Example 22
Given that
4 log x − log x y + log 3x 2 = 0
Find y in terms of x.
Solution
The way the question is worded implies that we must get rid of logs
altogether, which we are going to do using two slightly different methods.
Method 1 log x 4 − log x y + log 3x 2 = 0

x 4 × 3x 2
log = 0
xy
3x 5
⇒ = 1 [A 0 = 1 for any A]
y
y = 3x 5
Method 2 log x 4 + log 3x 2 = log x y

log 3x 6 = log x y
x y = 3x 6
y = 3x 5

Exercises 41
Questions 1 to 5 involve transforming curves. To refresh your memory on this

see Section 6, or for a more extensive description go to Section 4 of Chapter 8.
1. Find the equation of the curve that results from stretching y = 5x by a

factor 25 parallel to the y-axis. Express your answer in the form y = 5a .
2. Show that 4x = 22x . Hence, or otherwise, describe the transformation

from the curve y = 2x into the curve y = 4x .
3. (a) Write down the coordinates of the point A where the curve y = 2x
crosses the y-axis.
The the point B (k, 8) lies on the curve. Write down the value of k.
(b) The curve y = 2x is transformed into the curve y = 2x − 1. Describe
the transformation geometrically.
The points A 0 and B 0 are the images of A and B under this
transformation. Write down their coordinates.
(c) Repeat part (b) but where the new curve is y = 2x−1 .
4. The curve C has equation y = 3x .

¡ p ¢
Verify that the point P 21 , 3 lies on C .
C 1 is the curve y = 2(3x ) and C 2 is the curve y = 32x .
Describe the transformations that turns (i) C into C 1 , and (ii) C into C 2 .
Write down the image P under each of these transformations.
4 e 2x −3
5. Solve the equations (a) e x − 5e −x = 4 (b) e x = 2 − (c) = 4
ex ex
6. Show that the curves y = 2x and y = 2x 2 meet at x = 1. By sketching the
curves show that there is one other solution at x = a, where −1 < a < 0.
7. On the same graph sketch the curves y = e x −1 and y = 2e −x . Show that

they intersect when x = ln 2 and find the y-coordinate at that point.
8. Find the coordinates of the point where the two curves y = 2+4e −x and
y = e x − 1 meet. Sketch the two curves on the same graph.
9. (i) Write down the equation of the reflection of y = 10x in the line
(a) y = 0, (b) x = 0.
(ii) How many solutions are there to these equations:
(a) 10x = 792, (b) 10x + 1 = 0
(iii) Show that y = 333x + 1 cuts the curve y = 10x at x = 0 and x = 3.
Hence write down the solutions to the equation 333x = 10x − 1.
10. Find where the curve y = 8x intersects the curve y = 641−x .
11. By sketching the curve y = 2−x , show that there is exactly one solution
to the equation 12 x = 2−x . Use trial and error to find that solution.
12. Without using a calculator find the value of each of these logs:
1
(a) log2 16 (b) log7 49 (c) log9 3 (d) log10 100 (e) log5 0.2
p 1
(f)log6 6 (g) log4 4 (h) log20 1 (i) loga a 4 (j) log2 p
3
2

13. Complete each of these statements:

(i) logä 32 = 5 (ii) log100 10 = ä (iii) loga a = ä (iv) log3 ä = 2
1 1
(v) log8 2 = ä (vi) log16 1 = ä (vii) logä 3 = 4
(viii) logp p3
=ä
14. Express each of these as a single logarithm, e.g. 2 log 10 + log 5 = log 500.
(a) log 5 + log 4 (b) log 18 − log 3 (c) 3 log 2 (d) log 7 + 3 log 2
(e) log 3 + log 4 + log 5 (f ) 4 log 4 − 6 log 2 (g) 3 log 4 + log 8 − 5 log 2
15. Simplify these expressions, using the log laws. There may be more than
one answer. The purpose of the exercise is to become competent at
applying the laws correctly, rather than arriving at a particular result.
(i) log 6x − log 2x (ii) log 2a + log 5a (iii) 3 log x 2
(iv) 12 log 16 (v) 2 log y + log y 3 (vi) log x2 + log x2
¡ ¢ ¡ ¢
p
(vii) log X 2 − log X1 (viii) log 10x + log 102x (ix) log x + 32 log x
16. Which of these statements is FALSE? Think very carefully about them
because they relate to some very common errors.
(a) log(x + y) = log x + log y (b) 2 log a + 3 log a = 5 log a
3
log x
(c) = log x 2 (d) log PQ 2 = 2 log PQ
log x
1
(e) log(x 2 + 2x + 1) = 2 log(x + 1) (f ) − log X = log X
(g) loge e n = n (h) (log x)2 = 2 log x
17. Solve these equations, giving your answers correct to 3 significant

figures.
(i) 2x = 30 (ii) 7x = 50 (iii) 5x = 0.3 (iv) 1.1x = 2
2 1
(v) 3x+1 = 100 (vi) 102x−1 = 200 (vii) 3x = 17 (viii) 2 x = 3
18. Solve these equations, giving your answers correct to 3 significant

figures.
x
(i) 7x−1 = 32x (ii) 2x+1 = 34−x (iii) 0.1x = 21−2x (iv) 4x+2 = 3 2
19. Write each of these expressions in the form log X .

(i) log 2a + log a 3 (ii) 6 log x + log x (iii) log b 2 − 4 log b
(iv) log(x 2 + 2x) − log x (v) log(x 2 + 2x + 1) − log(x + 1)
p log x 6
(vii) log yx2 + 2 log x y (viii) 32 log x + log x (viii) 2
20. Write down the value of log2 8. Hence solve the equation log5 x = log2 8.
21. Solve the equation log3 (x + 1) − log3 x = 2.
22. Solve the equations

(a) 52x − 3(5x ) = 0, and (b) 4x = 7(2x ).
23. Given that p = logq 32, express each of these in terms of p:

(a) logq 2, and (b) logq 16q.
24. If log5 x = 2p and log5 y = p − 1, express each of these in terms of p:

¡x ¢
(a) log5 (x y) (b) log5 (x 3 ) (c) log5 5
25. Solve the equation log2 (x − 5) + log2 x = 3.

26. ab = 16 and log4 b − log4 a = 3. Find the possible values of a and b.
27. Find y in terms of x if log 3x − log x 2 y + 4 log x = 0.
28. What is the smallest integer n for which 2n > 10, 000?
29. Find the equation the tangent to the curve y − 1 + e −x at x = −1.
30. Show that 2x = 3 y can be written as y = kx where k is a constant to be

found in exact form.
31. Show that y = αe −2x is a solution of the differential equation

dy
+ 2y = 0.
dx
If y = 3 when x = 1, find α and hence write down the relationship
between x and y.
32. Find the turning point on the curve y = e x + 2e −x , and determine

whether it is a maximum or a minimum point.

Solutions 41
1. y = 5x+2 . [y = 52 × 5x ]
1
2. 4x = (22 )x = 22x . Stretch scale factor 2
parallel to the x-axis.
3. (a) (0, 1), k = 3. (b) Translation −1 parallel to the y-axis. A 0 (0, 0), B 0 (3, 7).
(b) (1, 1), (4, 8).
1 p p ¢ ¡1 p ¢
4. 3 2 = 3. (i) Stretch sf 2 // y-axis 12 , 2 3 (ii) Stretch sf 1
¡
2 // x-axis 4, 3 .
5. (a) ln 5 (b) ln 2 (c) 0 and ln 3
6. 21 = 2(1)2
y
When x = −1, e x = 0.368,
and 2(−1)2 − 2 > 0.368, so
they meet where x > −1.
1•
x
−2 −1 0 1 2
7.
y 1 1
If x = ln 2, e x = 2 and e x = 2 ,
so e x − 1 = 2e −x = 1.
The curves meet at (ln 2, 1).
• (ln 2, 1)
x
−2 −1 0 1 2
8.
2 + 4e −x = ex − 1
y
0 = e 2x − 3e x − 4
= (e x − 4)(e x + 1) 6
ex = 4 since e x + 1 6= 0
x = ln 4
and y = e x − 1 = 4 − 1
• (ln 4, 3)
y = 3
x
−2 −1 0 1 2

1
9. (i)(a) y = −10x , (b) y = 10−x or 10x (ii)(a) 1 (b) none. [from a sketch]
(iii) (0, 1) and (3, 1000) satisfy both equations.
Rearranging we have 10x = 33x + 1, so solutions are x = 0 and x = 3.
¡2 2 1
[(23 )x = (26 )1−x ⇒ 23x = 26(x−1) ⇒ x = 2(x − 1) and 8 3 = 61 3 = 4]
¢
10. 3,4
11. x = 1 [The curve is sketched in Section 5. Any line with a positive

gradient will intersect the curve exactly once.]
1 1
12. (a) 4 (b) 2 (c) 2
(d) −2 (e) −1 (f ) 2
(g) 1 (h) 0 (i) 4 (j) − 13
1 1
13. (i) 2 (ii) 2
(iii) 1 (iv) 9 ( v) 3
(v) 81 (vi) −3
14. (a) log 20 (b) log 6 (c) log 8 (d) log 42 (e) log 60 (f) log 4 log 16
15. (i) log 3 (ii) log 10a 2 (iii) 6 log x or log x 6

(iv) log 4 or 2 log 2 (v) log y 5 or 5 log y (vi) 0 [= log 1]
3
(vii) log X or 3 log X (viii) 3x log 10 (ix) 2 log x
16. (a), (c) (these two obey no Rule), (d) [it is not log(PQ)2 ] (h) [No Rule].
For the correct ones, part (e) is log(x + 1)2 and in (g) loge e = 1
17. (i) 4.91 (ii) 2.01 (iii) −0.748 (iv) 7.27

(v) 3.19 (vi) 1.65 (vii) ±1.61 (viii) 1.58
18. (i) −7.74 [x(log 7 − 2 log 3) = log 7] (ii) 2.07 [x(log 2 + log 3) = log 81
2
]
(iii) −0.756 [x(log 0.1 + 2 log 2 = log 2)] (iv) −3.31 [x log 16
3
= −4 log 4]
x 2 +2x
19. (i) log 2a 4 (ii) log x 7 (iii) log b −2 (iv) log(x + 2) [ x
= x + 2]
(x+1)2
(v) log(x + 1) [ x+1 = x + 1] (vi) log x 3 [log x − log y 2 + log x 2 y 2 ]
20. log2 8 = 3, x = 125 [x = 53 ]

1 x+1
21. x = 8 [ x = 32 ]
22. (a) x = 2.05 (b) x = 2.81 [4x = (22 )x = (2x )2 ]
p 4p
23. (a) 5 [p = logq 25 ] (b) 5 + 1 [logq 24 + logq q]
24. (a) 3p − 1 (b) 6p (c) 2p − 1
25. x = 6 or x = −1 [x(x − 5) = 23 ]
b
26. a = 21 , b = 32 or a = − 12 , b = −32 [ a = 64 then substitute]
3x
27. y = 3x 3 [ x 2 y ×x 4 = 1]
28. n = 14 [2n = 10, 000 ⇒ n = 13.28]
29. y = 1 − ex [gradient −e −x , so grad is −e. Equation y − (1 + e) = −e(x + 1)]
log 2
30. k = log 3
31. y = 3e 2−2x [3 = αe −2 so α = 3e 2 ⇒ y = 3e 2 e −2x ]

p p 1 p 2
32. (ln 2, 2 2) [e x − 2e −x = 0 so e 2x = 2 i.e. 2x = ln 2, x = ln 2 2 . y = 2+ p ]
2
d2y
Minimim [ d x 2 = e x + 2e −x > 0 for all x]

11.6 Exponential modelling

The exponential function e x crops up in a wide range of practical areas, from
radioactive decay to population studies and epidemiology. The reason for
this is that in many of these area the rate of increase or decrease of a variable
is proportional to the the value of the variable at that point.
For example in an epidemic the number of people being infected by a
disease is proportional to the number of people already infected. In
radioactive decay mass is lost as it turns into radiation; the rate at which this
happens is proportional to the mass that remains. Mathematically we can
express this relationship as
dy
= ky
dx
where the constant k > 0 for increasing variables and k < 0 for decreasing
variables.
We saw in Section 2 that the solution of this equation has the form y = ae kx
(where a is a constant). Hence the exponential function crops up in all these
situations.
Example 23
The number of people infected with a disease is modelled by the formula
N = 350 − 200e −0.5t
Where N is the number of people infected t years after the disease is first
detected.
(a) How many people with the disease were initially detected?
(b) What is the long term prediction of the number of people with the
disease18 .
(c) After how long (to the nearest month) will the number have increased by
half?
(d) Sketch the graph of N against t .
Solution
(a) When t = 0, N = 150
(b) As t → ∞, N → 250
(c) 225 = 350 − 200e −0.5t
200e −0.5t = 125
e −0.5t = 0.625
−0.5t = ln 0.25 = −0.4700
t = 0.94 years
= 11.28 ≈ 11 months (to the nearest month)
N
(d)
250
150
t
18
In practice other factors will mean the relationship is more complex, leading to a long
term reduction of numbers

Example 24
When carbon-14 decays, its mass after T years is given by
m = m 0 e −λT
where m 0 is the mass ay time T = 0 and λ is constant equal to 0.000121.

(a) Find what proportion of the mass has decayed after 1000 years.
(b) How many years does it take for the initial mass to be reduced by 10%?
(c) Find the half-life of carbon-14.
dm
(d) Show that = −λm.
dT
(e) Sketch the curve of m against T , marking where the half-life is.
Solution
m
(a) m = m 0 e −0.121 , so = 0.8860,
m0
i.e. 11.46% (3 SF) has decayed.
(b) 0.9m 0 = m 0 e −λT ,
so e −λT = 0.9 or −λT = ln 0.9
−0.000121T = −0.10536, or T = 870.75 years.
(c) The half-life is the time it takes for the mass to reduce by half.
This identical to the calculation in (b) replacing 0.9 with 0.5,
so −0.000121T = ln 0.5 = −0.6931.
The half-life (to the nearest year) is 5, 728 years.
dm
(d) = −λm 0 e −λT = −λm
dT
(e)
m
m0
1
m
2 0
0 T years
5, 728

Exercises 42
1. The population of a country is modelled using the formula

t
P = 500 + 100e 40
where P is measured in thousands and t is the time in years since 2000.
(a) State the population in the year 2000.
(b) Use the model to predict the population in the year 2020, to the
nearest 1, 000.
(c) Explain why it not be sensible to use this model to predict the
population in 2500.
2. The value of a car is modelled by the formula

t
V = 30, 000e − 10
where V its its value in £ after t years.

(a) State the value of car when new.
(b) Find its value to the nearest £100 after 4 years.
(c) How long does it take, to the nearest year, for its value to halve?
3. Starting at the beginning of January the number of rabbits n in a local

population after m months is modelled by the formula
n = 150e 0.08m
(a) Use this model to estimate the number of rabbits (i) at the end of
March, (ii) after 1 year.
(b) Show that at after 18 months the rabbit population is expected to be
increasing by approximately 41 rabbits per month.
(c) Give reasons why the model may not be valid for large enough values
of t .
4. A microbiologist researching a new strain of bacteria puts 100 in a Petri

dish to divide. After 8 hours he returns to find 500 bacteria in the dish.
Bacteria are known to divide in such a way that their number can be
modelled using the formula
N = N0 e kt
where N0 is the initial number and N is the number after t hours.

(a) Find the value of k for this strain of bacteria correct to 2 decimal
places.
(b) How many Bactria are there after 24 hours, to the nearest 50?
(c) How many hours will it take for number of bacteria to reach 100, 000?

Solutions 42
1. (a) 600, 000 (b) 665, 000 (c) The model will almost certainly need
updating by then.
t
2. (a) 30, 000 (b) 20, 100 (c) 7 years [e − 10 = 12 ⇒ t = 10 ln 2]
3. (a) (i) 191, (ii) 392

(b) Rate of increase = ddnt = 0.08(150)e 0.08×18 = 50.65 /yr.
(c) Over time the factors affecting population growth, like weather and
disease, will change.
4. (a) 0.20 (b) 12, 150 (c) 35 (to the nearest hour)

11.7 Logs and non-linear data

In many practical situations, variables are connected by a non-linear
relationship. The problem we are seeking to solve is this. We have some data,
i.e. values of y and corresponding values of x. If the relationship was a linear
one we could plot the point on a graph, fit an approximate straight line to it
and so estimate what the relationship is between x and y.
If the relationship is not linear we are not going to get a straight line and
estimating what the is relationship from a curve is very difficult.
In this section we will be looking at two that take the form
y = ax n and y = ab x .
We will see that by transforming the data using logs, we can convert them into
linear relationships and so estimate the parameters (i.e. a and n in the first
type, and a and b in the second).
Relations of the type y = ax n

The theory is quite simple. If we take logs (we generally use base 10) on both
sides of the equation we have
y = ax n
log y = log a + log x n
or log y = log a + n log x
Now we define two new variables: X = log x and Y = log y. In terms of the
new variables the relationship is
Y = n X + log a.
If we compare this will the equation of a straight line y = mx + c, the gradient
of the line we obtain is n and the y-intercept is log a. If we measure these
from our graph, we will know n and we able to find a.
Suppose, having plotted the points and find n = 2.3 and log a = 1.75. It follows
that a = 5.75 (to 3 SF) and the relationship between x and y is
y ≈ 5.75x 2.3 .
Relations of the type y = ab x

If we take logs on both sides of the equation we have
y = ab x
log y = log a + x log b
Now we define two variables: Y = log y. In terms of the new variable the
relationship is
Y = x log b + log a.
If we compare this will the equation of a straight line y = mx + c, the gradient
of the line we obtain is log b and the y-intercept is log a. If we measure these
from our graph, we will know b and we able to find a.
Suppose, having plotted the points and find log a = 0.48 and log b = 0.18. It
follows that a = 3.01 and b = 1.51 (to 3 SF) so the relationship between x and
y is
y ≈ 3(1.5)x x .

Example 25
Data are collected on two variables P and T as shown in the table below. It is
suspected that the variables are related by a relationship of the form P = aT n
for some constants a and n. By transforming the variable P and plotting
points on a graph estimate the values of a and n.
T 2.1 6.2.7 3.1 4.0 4.6

P 5.10 9.57 13.51 25.60 36.28
Solution
Taking logs we have log P = log a + n log T .
If we plot log P against log T we should get a straight line. Its gradient will be
n and its y-intercept log a.
The values of log T and log P are
log T 0.322 0.431 0.491 0.602 0.663

log P 0.708 0.981 1.131 1,408 1.560
log P
x
1.5
x
1.0 x
0.5
0 log T years
0.2 0.4 0.6
From the graph log a = −0.1, so a = 100.1 = 0.794 ≈ 0.8

Gradient n = 2.51 ≈ 2.5
The relationship is P ≈ 0.8T 2.5

Example 26
Data are collected on two variables y and x as shown in the table below
known to be connected by a relationship of the form y = ab x for some
constants a and b. By transforming the variable P and plotting points on a
graph estimate the values of a and n.
x 1.2 1.7 1.9 2.3 2.5

y 31.9 27.6 26.1 23.2 21.9
Solution
Taking logs we have log y = log a + x log b.
If we plot log y against x we should get a straight line. Its gradient will be
log b and its y-intercept log a.
The values of log y are
log y 1.504 1.441 1.417 1.365 1.341
log y
1.5 x
x x
x x
0.5
0 x
0.5 1 1.5 2 2.5
From the graph log a = 1.65, so a = 100.165 = 44.67 ≈ 45

Gradient = log b = −0.127, so b = 0.746 ≈ 0.75
The relationship is y ≈ 45(0.75)x

Exercises 43
The tables below give observed data points corresponding to a relationship
between two variables of the form given. In each case, transform the variables
as necessary to convert the relationship into a linear one, and by plotting the
points on a graph estimate the parameters a and n or a and b.
When constructing a graph you will get more accurate answers if you are able
to restrict each of the scales to be close to the values you are plotting.
In some cases n will be a negative number. Similarly when the relationship is
of the form y = ab x it may actually turn out to be of the form y = ab −x .
1. y = ax n x 1 3 6 7 10
y 4.50 3.74 8.59 10.33 15.85
2. Y = ax n x 0.82 0.96 1.10 1.14 1.30

Y 10.48 20.97 38.18 44.67 79.62
3. V = at n t 0.35 0.41 0.48 0.77

V 14.51 10.41 7.47 2.77
4. P = aq n q 2 3 4 5 6
P 62.9 34.3 22.3 15.9 12.1
5. y = ab x x 0.5 1.0 1.5 2.0 2.5

y 4.29 5.76 7.73 10.37 13.91
6. T = ab v v 0.3 0.7 1.1 1.9 2.2

T 0.66 0.49 0.36 0.20 0.16
7. S = ab t t 1 2 3 4 5
S 13.0 16.25 20.31 23.39 31.74
8. F = ab x x 1.0 2.6 2.8 3.3 4.1

F 242 261 264 270 281
Solutions 43
All the answers are approximate, because plotting points and fitting a line will
never be exact. Just check that your answers are reasonably close to those given
here.
1. y ≈ 4.5x 1.2 2. Y ≈ 25x 4.4 3. V ≈ 1.6t −2.1 4. P ≈ 178q −1.5

5. y ≈ 3.2(1.8)x 6. T ≈ 0.82(2.1)−v 7. S ≈ 10.4(0.8)−t 8. F ≈ 230(1.05)x


Reminder of the index rules underpinning the whole chapter.
• Exponential equations. Where the solution can be found by

inspection, or after some simple manipulation. For example
3x = 81, 252x = 5, 16x = 8x+2 .
• Quadratic Exponential equations. Equations containing exponents

that have an underlying quadratic structure, for example
32x − 3x − 6 = 0, 16x − 3(4x ) − 4 = 0, 9x = 36 − 3x+2
solved by substitution.
• Exponential graphs. Sketching them. For example y = 2x and y = 2−x .
• The function f (x) = e x What is special about the number e ?

f 0 (x) = e x . If f (x) = e ax , f 0 (x) = e ax .
• Transforming exponential functions. Translating, stretching,

reflecting. Effect on intercepts with the axes and on the asymptote.
e.g. f (x) = e x transforms to f (x + 2) = e x+2 by a translation,
or as e 2 f (x) = (e 2 )e x by a stretch.
• Logarithms. Definition: log A Y = X ⇔ Y = A X .

Hence log A A = 1, log A 1 = 0.
Solving equations e.g. log3 x = 4, y = log10 1000
• Logarithm Laws
X
µ ¶
Y
log X N = N log X
• Exponential equations using logs e.g. 2x + 10, 22y+5 = 0.5, 4x = 121−x
• Simplifying log expressions

1
e.g. 3 log 2 + log 5 − 2 log 4, ln e 2 , 2 log pq − log 3p 2 .
• Exponential modelling. Example of practical problems e.g. population

growth, radioactive decay.
• Non-linear relationships. Using logs to convert functions of the form

y = ax n or y = ab x to a linear relationship.

Chapter 12
TRIGONOMETRY
Overview
Introduction to trigonometrical functions.
12.1 SOHCAHTOA. A reminder. Different ways of applying it.
12.2 Two important trig relationships.

sin A
tan A = cos A , sin x = cos(90◦ − x) and cos x = sin(90◦ − x)
12.3 Re-defining trig functions: for angles > 90◦ .
12.4 Trig functions for any angle, including outside 0◦ < x < 360◦ .
12.5 The sin, cos and tan of 0◦ and 90◦ .

sin 0◦ = cos 90◦ = 0, sin 90◦ = cos 0◦ = 1, tan 0◦ = 0, tan 90◦ = ∞
12.6 The Sine, Cosine and Tan curves. Symmetries.

Using the graphs to find values for angles outside 0◦ ≤ x ≤ 90◦ .
12.7 Sine Rule. Applications.

a b c sin A sin B sinC
= = or = =
sin A sin B sinC a b c
12.8 Cosine Rule. When information means cannot begin with sine rule.
a 2 = b 2 + c 2 − 2bc cos A,
b 2 = c 2 + a 2 − 2c a cos B ,
c 2 = b 2 + a 2 − 2ba cosC
12.9 More trig equations. Including using the curves to solve.

The important identity sin2 x + cos2 x ≡ 1.
12.10 Area of a triangle formula. A = 12 ab sinC = 12 bc sin A = 12 c a sin B
12.11 More on trig identities. Manipulation. Proving new identities.
12.12 Other types of trig problem. Example:

If sin θ = 54 and θ is obtuse, find the exact values cos θ and tan θ.
12-1
CHAPTER 12. TRIGONOMETRY
Trigonometrical functions
Introduction
You have already met the three trigonometrical functions we’ll be exploring
here: sine, cosine and tangent. We almost always use the abbreviations you
see on your calculator: sin, cos and tan, and refer to them collectively as
“trig" functions. In later modules we’ll meet more of them, but this is all we
need at present.
Of all the functions you will ever meet it is hard to exaggerate how
fundamental these functions are. Without them we couldn’t describe waves,
and we wouldn’t be able to understand light, sound, electricity, magnetism,
radio, TV, lasers, and so on, right down to the structure of matter itself. In
short trig functions play a central role in our understanding of everything
from cosmology to quantum mechanics.
They are without doubt extremely fascinating functions; they are also the
most challenging functions A Level Maths students meet.
But wait a minute, I hear you say, what’s so difficult about SOHCAHTOA and
all that? The answer is that we need to extend our understanding of trig
functions to any angle. For example what could sin 200◦ , or sin(−50◦ )
possibly mean?
So we are going to start by re-defining the trig functions, in a way that allows
for any angle. Although this is the obvious place to explain it, you won’t need
to understand the thinking behind it until C2, so you may choose to skip the
next three pages and come back to them later.

12.1 SOHCAHTOA
Skip over this section if you are completely familiar with how the trig ratios
are defined using a right angled triange and SOHCAHTOA1 :
Opposi t e Ad j acent Opposi t e

Sin = , Cos = , Tan =
Hy pot eneuse Hy pot eneuse Ad j acent
The hypoteneuse is the side opposite the right angle.

For SOHCAHTOA purposes what we call the “opposite” side is the one
opposite the relevant angle (x in the diagram below).
The remaining side we call the “adjacent”2 side.
"opposi t e 00 t he
r i g ht ang l eH u se
e ne
ot
H
j
H
p
Hy Opposi t e
x
Ad j acent
It is a common convention to label the angles of a triangle with capital

letters, and the corresponding opposite angle with the same lower case letter.
We often refer to these as trigonometrical ratios, or simply trig ratios. If we

were, for example, to double the size of a triangle, the ratios wouldn’t change.
1
If you are familiar with the Hokkien Chinese dialect you may choose to remember it as
TOA-CAH-SOH, which apparently means ’big-footed woman’.
2
Along with the hypoteneuse it is adjacent to, or next to, the angle we are considering.

Applying SOHCAHTOA
You will probably be familiar with using SOHCAHTOA, to find a length or an
angle in a right angled triangle. We’re going to look at a simple example.
Please bear with me because although the type of question may be very
familiar, the way we are going to use SOHCAHTOA may not be.
We can write the ratios using little triangles like this:
O A O
S H C H T A
And the point of writing it like this? We can see immediately how to find any
one of the variables from the other two.
For example in the sine relationship:
O
S H
O O O
S= = H =
H S S×H
Example 1
In the right angled triangle shown, find x.
Solution
For the 65◦ angle the 4 cm side is the opposite side,

and x is the hypoteneuse, so we need the sine ratio:
x cm 4
4 cm sin 65◦ =
x
which we can rearrange to find x.
65◦ O
Or more directly, using = H,
S
4 4
x = ◦
=
sin 65 0.9063
= 4.4135
Answer: x = 4.41 cm (3SF)

QQ
Using the direct method described on the previous page, find the values
marked on each triangle. Give your answers correct to 3 SF.
(a) (b)
2
10 a
b
30◦ 40◦
(c) (d)
4.5
◦
24
c 8
50◦
d
(e) (f)
x
15
y 8
0.61
0.94 z
f
3
(a) a = 5.00 (b) b = 3.11 (c) c = 2.00 (d) d = 12.4
(e) x = 57.0◦ , y = 33.0◦ [y = 90◦ − x] (f ) z = 32.2◦ , f = 12.7 [Pythagoras or trig]

12.2 Two important trig relationships

The sine function and the cosine function are related in various ways—some
very beautiful, as we’ll see later in the chapter when we look at their curves.
But for now there are two that are particularly fundamental to trigonometry.
The first connects the tan function with sin and cos.
sin A
A
tan A =
!
A
cos A
A
A
We can see why this is:

B
a b a c
sin A = cos A = tan A =
c c b a
a
sin A c a c a
∴ = = × = = tan A
cos A b c b b A C
c
b
The second connects the sin and cos functions.
sin x = cos(90◦ − x) A
!
A
cos x = sin(90◦ − x)
A
A
90◦ − x
c
b
x
a
The right angled triangle above makes these relationships pretty self-evident,
but they are nonetheless worth remembering as you will use them a lot:
b a
sin x = = cos(90◦ − x) and cos x = = cos(90◦ − x)
c c

Two important triangles

Before we introduce a very and importanrt way of thinking about the trig
ratios, there are two triangles that are so important that it’s worth getting
familiar with them.
The first is the right angled triangle with angles 45◦ , 45◦ and 90◦ .
We can think of it as a square cut in half along a diagonal.
The trig ratios for this triangle depend on the shape of the triangle only, not
its particular dimensions, so we may as well
p choose one formed from a unit
square. By Pythagoras the hypoteneuse is 2.
Using SOHCAHTOA
π
◦ 1 4
sin 45 = p
2 p
2 1
1
cos 45◦ = p
2
π
tan 45◦ = 1 4
1
The second triangle is one with angles 30◦ , 60◦ and 90◦ .
We can think of this as an equilateral triangle cut in half along a line of
symmetry. Again, for the trig ratios we can choose any lengths, so we’ll make
each side 2 units long to avoid fractions when we cut one of them half.
p
Pythagoras tells us that the height of the triangle is 3, the other two sides
being 1 and 2.
Now

p
π

3
sin 60◦ = = cos 30◦ 6
2

1
cos 60◦ sin 30◦

= =
2 2

p

3
p
tan 60◦ = 3

π

3
1
These angles, and the trig ratios that go with them, crop up so often that you
are strongly advised to memorise them.
For example if you saw the equation sin x = 0.5, would you reach for your
calculator? Or would you recognise immediately that sin 30◦ = 0.5?
In an exam, when starting a trig question, it’s a good idea to make a little
sketch of each of these triangles.

12.3 Re-defining the trig functions

Imagine a Water Mill, or a model of one, powered by a water wheel. The
wheel protrudes from the wall of the mill, as shown in the picture. To keep
things simple we’ll take the radius of the wheel to be 1 unit and its centre to
be where the wall and the water level meet.
y
6
• P • P (x, y)
w al l
1
θ -x
w at er l evel
As the wheel turns the point P can be at position on the circumference of the
wheel, including inside the mill and below the water level. If the radius to P
makes an angle θ 4 with the horizontal, we define sin θ and cos θ as follows:
cos θ = the distance of P from the wall;

sin θ = the height of P above the water level.
Picking out the triangle we can see that
x y
cos θ = and sin θ = .
1 1
Therefore
cos θ = x
sin θ = y
P (x, y) This will allow us to define sin θ and cos θ

for any angle, including angles bigger than
360◦ and negative angles—despite the fact
that this doesn’t make any sense from the
1
y SOHCAHTOA perspective.
θ
x
To define tan θ for any angle, we simply note that

y sin θ
tan θ = = .
x cos θ
So we now have a way of defining sin θ, cos θ and tan θ for any angle.
4
This is the greek letter theta, and is often used when denoting an angle.

Let’s explore the implications of this and see how we can easily find the
values of a trig function for any angle. You may think this is unnecessary:
could we not just use a calculator? It will soon become clear that there are
many situations when this won’t work, not least when the angle in question
is algebraic!
We start in the range 0◦ to 90◦ , in what’s called the first quadrant. Notice that
we measure angles in an anti-clockwise direction, starting from the positive
x-direction. This is the universally adopted convention.
From here we move to the other quadrants by recognising four related
angles. These are the angles formed by taking an angle in the first quadrant
and reflecting it repeatedly in both axes.
We illustrate this using 30◦ . The four related angles are 30◦ , 150◦ , 210◦ and
330◦ .
We call the angle in the first quadrant, 30◦ in this case, the principal value.
y y
6 6
•
30◦ 180◦ − 30◦ •
150◦
30◦ -x -x
y y
6 6
210◦ 330◦
-x -x
180◦ + 30◦ • • 360◦ − 30◦

Think of the four angles as if they were points on a circle. This is much easier
than trying to label them in usual way, using arcs.
We’ll continue to use a circle of radius 1 unit. The radius can be anything we
like because trig functions are ratios of lengths.
y y
6 6
• 30◦ 150•◦
1 1
-x -x
y y
6 6
-x -x
1 1
◦•
210 • 330◦

Our calculator tells us that (to 3 SF)
cos 30◦ = 0.866 and sin 30◦ = 0.500.
So now we can put in the value of x and y, in other words the value of sin θ
and cos θ, for each of the related angles. Not surprisingly all the sin values
and all the cos valued differ only in their sign.
y y
6 6
• 30◦ 150•◦
1 1
y = +0.5 +0.5
-x -x
x = +0.866 −0.866
cos 30◦ = + 0.866 cos 150◦ = − 0.866

sin 30◦ = + 0.5 sin 150◦ = + 0.5
y y
6 6
−0.866 +0.866
-x -x
−0.5 −0.5
1 1
210◦• • 330◦
cos 210◦ = − 0.866 cos 330◦ = + 0.866
sin 210◦ = − 0.5 sin 330◦ = − 0.5

Remembering how the signs change

There’s a neat way of summarising what the
signs of sin, cos and tan are in each quadrant.
Q2 Q1
To make discussion easier the quadrants are
labelled Q1, Q2, Q3 and Q4, starting where
x and y are both positive and moving anti- Q3 Q4
clockwise.
Then the diagram below tells us which functions are POSITIVE in each
quadrant.
We saw on the previous page that they are all positive in Q1. In the other
quadrants only one of them is positive.
SI N ALL
S A A
!
A
A
T C A
T AN COS
Some people use a mnemonic to help them remember it, like All Science
Teachers Calculate, or you can make up your own. We’ll simply refer to it as
the CAST diagram.
The diagram neatly captures all the information we need about where each
of the three functions is positive and where it is negative, i.e.
SI N COS T AN
+ + − + − +
− − − + + −
We’ll be using this when solving trig equations. But first we need to see what
happens if an angle is outside the range 0◦ to 360◦ . The reason for this will
soon become clear.

12.4 Angles outside the range 0◦ to 360◦

Angles greater than 360◦
From here it is a small step to define the values sin θ and cos θ for any angle.
As the wheel rotates, any point on it will keep coming back to the same
position, so for example
sin 30◦ = sin 390◦ = sin 750◦ = . . . . . . = 0.5

y
6
• 30◦ , 390◦ , 750◦ , ....
30◦ -x
Negative angles
In the same way we can think of negative angles as the wheel turning in the
opposite (clockwise) direction, so for example
sin(−30◦ ) = sin 330◦ , sin(−150◦ ) = sin 210◦ etc.
y
6
210◦
-x
+210◦ • • +330◦
−150◦ −30◦
We are not saying that these angles are all the same. Just that the values
of their trig functions are the same.

Example 2
Find all the solutions of the equation

sin x = 0.5
◦ ◦
for which 0 ≤ x < 360 .
Solution
Method 1
Referring to the 30◦ , 60◦ , 90◦ triangle, sin 30◦ = 12 .
So we immediately have the principal solution.
y
6
• •
1
30◦ -x
In the range 0◦ to 360◦ there are two points on the unit circle5 which have a
y-coordinate = 0.5.
One is 30◦ . The other is its reflection in the y-axis, i.e. at 180◦ − 30◦ = 150◦ .
The solutions are x = 30◦ and x = 150◦ .
You may want to check on your calculator that sin 150◦ = 0.5.
Method 2
The CAST picture tells us

S A that sin is positive in Q1 + +
an Q2
T C − −
So the two angles where sin x = +0.5 are

x = 30◦ and x = 180◦ − 30◦ = 150◦ .
In Method 1 we can see why the solutions are what they are.
Method 2 is more mechanical, but quicker.
5
Circle of radius 1 cm.

Example 3

cos x = −0.225
◦ ◦
for which 0 ≤ x < 360 .
Solution
Method 1
The princpal value, using a calculator is cos−1 0.225 = 77◦ .

y
6
77◦ -x
We need the points related to 77◦ for which the x−coordinate is negative.
These are at 180◦ − 77◦ and 180◦ + 77◦ , i.e. 103◦ and 257◦ .
The solutions of the equation cos x = −0.225 are x = 103◦ and 257◦ .
You might like to check that these are correct with your calculator.
Method 2
The CAST diagram tells

S A us that cos is negative in − +
Q2 an Q3
T C − +
From the calculator6 cos−1 (−0.225) = 103◦ , an angle in Q2.

Since 103◦ = 180◦ − 77◦ , the principal value is 77◦ , and the two solutions are
x = 180◦ ± 77◦ , i.e. x = 103◦ or 257◦ .
6
Some calculators and textbooks will call this arcsin.

Example 4
tan x = −2
for which −180◦ < x ≤ 180◦ .
Solution
This time we will only use the CAST method.
The CAST diagram tells

S A us that tan is negative in − +
Q2 an Q4.
T C + −
The range is specified as −180◦ ≤ x ≤ 180◦ . This is just a different way of

describing the points on the circle. Instead of getting to them by going
anti-clockwise from 0◦ to 360◦ , we get to half of them by going 180◦ anti-
clockwise, and the other half by going 180◦ clockwise.
A calculator7 gives us one of them: tan−1 (−2) = −63.43◦ .

The principal value is +63.43◦ (its reflection in the x-axis).
So (to 3SF) the two solutions in the range −180◦ ≤ x ≤ 180◦ are
x = −63.4◦ and 180◦ − 63.4◦ , i.e. x = −63.4◦ and 116.6◦ .
7
Some calculators and textbooks will call this arctan.

Example 5

p
3
sin x = −
2
for which 0◦ ≤ x < 360◦ .
Solution
We don’t need to use our calculator,
because we know that
p 30◦
◦ 3
sin 60 = . p
2 2 3
60◦
1
The principal angle is sin 60◦
S A The sin function is + +

negative in Q3 an Q4.
T C − −
So the two solutions are

x = 180◦ + 60◦ and 360◦ − 60◦ , i.e. x = 240◦ and 300◦ .
³ p ´
3
Had you used your calulator to find sin−1 − 2
you would have got the
answer −60◦ , which can be described at +300◦ .

QQ
Solve these equations for angles in the range 0◦ ≤ x ≤ 360◦ , giving your
answers to the nearest degree.
(a) cos x = 0.5

(b) sin x = 0.34
(c) cos x = −0.707
(d) sin x = −0.26.
1
(e) tan x = − p .
3
We’ve confined ourselves to angles between 0◦ and 360◦ . You might wonder
why we would ever be interested in angles outside this range.
You’ll soon have the answer to this question, and in anticipation of needing
them the next section shows how we deal with such angles.
8
(a) x = 60◦ , 300◦ [cos positive in Q1 and Q4 ⇒ x = 60◦ and 360◦ − 60◦ ]
(b) x = 20◦ , 160◦ [sin positive in Q1 and Q2 ⇒ x = 20◦ and 180◦ − 20◦ ]
(c) x = 135◦ , 225◦ [cos negative in Q2 and Q3 ⇒ x = 180◦ − 45◦ and 180◦ + 45◦ ]
(d) x = 195◦ , 345◦ [sin negative in Q3 and Q4 ⇒ 180◦ + 15◦ and 360◦ − 15◦ ]
(e) x = 150◦ , 330◦ [tan negative in Q2 and Q4 ⇒ x = 180◦ − 30◦ and 360◦ − 30◦ ]

12.5 Trig values for 0◦ and 90◦

Sine and Cosine functions
The CAST diagram tells us where trig functions are > 0 or < 0. But where are
they = 0?
We’re going to explore this using a simple diagram. We start with a right
angled triangle that has a small angle θ.
θ
c
? b
a
b a
sin θ = and cos θ =
c c
Now imagine making the angle θ smaller and smaller.

b a
The closer it gets to zero, the closer gets to 0 and the closer gets to 1.
c c
In mathematical language
As θ → 0◦ , sin θ → 0 and cos θ → 1,

which tells us that sin 0◦ = 0 and cos 0◦ = 1.
Now we do the same thing again, but this time making θ approach 90◦ .
a b
sin θ = and cos θ =
c c
As θ gets closer to 90◦ ,

a b
gets closer to 1 and gets closer to 0.
c c
In mathematical language
As θ → 90◦ , sin θ → 1 and cos θ → 0,

which tells us that sin 90◦ = 1 and cos 90◦ = 0.
c a
To summarise
sin 0◦ = 0 cos 0◦ = 1 A
!
A
sin 90◦ = 1 cos 90◦ = 0
A
A
θ
J
^
J
b
If you forget any of these, a quick sketch of a long thin triangle will tell you
what you need to know.

Tangent function
We could find the values of tan 0◦ and tan 90◦ using a long thin triangle, as we
did with sin and cos. A more direct way is to use the fact that
sin θ
tan θ = cos θ
But first a reminder about the notion of infinity. It simple means a value that
is not limited—or to use mathematical jargon unbounded—so it isn’t a
number at all. Nevertheless it’s useful to have a symbol for it, i.e. ∞.
If we divide 1 by a small number, e.g. 0.0001 we get a large result: 10, 000. If
we take this to the extreme of dividing by 0, the result we call ∞.
In the same way, if we divide 1 by a large number the result is a small
number. Taking this to the extreme means dividing by ∞. Strictly speaking
this is meaningless, of course, but in the sense that the result must be
immeasurably small, we say
1 1
=∞ and =0
0 ∞
This is discussed in more detail on page 1-9.
So now sin 0◦ 0
tan 0◦ = ◦
= = 0
cos 0 1
and
sin 90◦ 1
tan 90◦ = = = ∞
cos 90◦ 0
tan 0◦ = 0 A
!
A
tan 90◦ = ∞
A
A

We suggested that the CAST approach doesn’t tell us where trig functions are
= 0.
Actually for the sine and cosine functions it does. In fact it points to all the
angles where the functions are zero.
The function is zero where its value crosses from positive to negative, or
from negative to positive. We’ve marked these on the diagrams below with a
0.
SI N COS
0
+ + − +
0 0
− − − +
0
You’ll see that this line of argument doesn’t work for the tan function (for
example tan 90◦ 6= 0).
T AN
?
− +
0 0
+ −
?
It’s to do with the fact that tan 90◦ = ∞. Exactly what’s going on will become
clear in the next section when we look at the graphs of each of the functions.

12.6 The sine, cosine and tan curves

Now that we have a way of defining the sine and cosine functions for any
value of the angle, we can see what the curves look like.
y = sin x
6
JJ
- x
−180◦ −90 ◦
0 ◦
90 ◦
180 ◦
270 ◦
360 ◦
450 ◦
540 ◦
630 ◦
720◦
−1
y = cos x
6
-x
−180◦ −90◦ 0◦ 90◦ 180◦ 270◦ 360◦ 450◦ 540◦ 630◦ 720◦
−1
Each curve has cyclical symmetry, repeating every 360◦ . We say it has a
period of 360◦ .
The functions have numerous other symmetries too: a vertical line of
reflective symmetry every 180◦ , and rotational symmetry 9 of order 2 at every
point where they cross the x-axis.
The functions oscillate, between a maximum value of +1, and a minimum
value of −1.
9
If the entire curve is rotated through 180◦ about the point where it crosses the
x-axis, it looks exactly the same.

We know that
sin x
tan x =
cos x
so at each value of x, we can calculate tan x by dividing the value of sin x by

the value of cos x.
We don’t need too many points to see what the curve must look like.
0
When x = 0◦ , tan x = 1 = 0. When x = 45◦ , tan x = 1.
1 −1
When x = 90◦ , tan x = 0 = ∞; when x = 270◦ , tan x = 0 = −∞.
Since sin x and cos x are both cyclical, or periodic, this pattern repeats.
y = tan x
6
- x
0◦ 90◦ 180◦ 270◦ 360◦
−1
Like sin x and cos x, the curve has rotational symmetry of order 2 about every
root. Unlike sin x and cos x it has no lines of symmetry.
The lines x = 90◦ and x = 270◦ are two vertical asymptotes of the curve (the
curve gets closer and closer to them as x gets closer to 90◦ and 270◦ , but never
reaches them). The diagram shows only two of the asymptotes; there are of
course an infinite number, since they repeat every 180◦ in both directions.

Let’s take a closer look at the sine function, concentrating on one complete
period. If we understand this we understand everything about the sine
curve. Because the cosine curve behaves in the same way, we will also
understand everything about that curve—and by implication the tan curve
too.
y = sin x
6
1
2

- x
0◦
30◦ 90◦ 150◦ 180◦ 210◦ 270◦ 330◦ 360◦

− 12
−1
We know sin 30◦ = 0.5.

From the reflective symmetry about 90◦ , it’s clear that
sin 150◦ = sin 30◦ = 0.5.
For angles beyond 180◦ , rotational symmetry about 180◦ tells us that
sin 210◦ = sin 330◦ = −0.5.

QQ
Do not use a calculator for these questions.
1. Given that cos 60◦ = 0.5, use the graph of y = cos x to write down the
values of
(a) cos 120◦ (b) cos 240◦ (c) cos 420◦ (d) cos(−60◦ ).
2. Correct to 2 SF sin 70◦ = 0.94. By any method write down the values of
(a) sin 110◦ (b) sin 250◦ (c) sin(−70)◦ (d) sin 790◦ .
3. We know that tan 45◦ = 1. By referring to the graph of y = tan x write down
the solutions of the equation
tan 45◦ = −1
for values of x in the range 0◦ ≤ x ≤ 360◦ .
We are going to digress for a while, to look at the Sine Rule, the Cosine Rule and
bearings. If you are already very familiar with these, or want to go straight on
to more interesting trig equations, jump to page 12-49.
10
1. (a) −0.5 (b) −0.5 (c) 0.5 (d) 0.5
2. (a) 0.9 (b) −0.9 (c) −0.9 (d) 0.9
3. x = 135◦ , 255◦

12.7 The Sine Rule

We know how to “solve” a right angled triangle—to find any side or angle,
using SOHCAHTOA. What if the triangle doesn’t contain a right angle? Two
additional rules: the Sine Rule and the Cosine Rule will enable us to solve
the triangle in every case.
These rules appear at this point in the chapter because (a) they often involve
angles greater than 90◦ , and (b) they sometimes give rise to trig equations
that have more than one solution. Both of these we are now equipped to deal
with.
First the Sine Rule, which states that in a triangle the ratio between the sine
of any angle and the length of the side opposite that angle is always the
same.
B
c
A
a
For the triangle above, we can express this in two ways:
a b c
A
= =
!
A
sin A sin B sinC A
A
or
A
sin A sin B sinC
!
A
= =
a b c
A
A
They are both telling us the same thing, of course, so we don’t need both. It’s
just that when finding the length of a side it’s slightly quicker to use the first
one; when finding an angle it’s slightly quicker to use the second one.
Each one is really three relationships rolled into one. For example the second
version is:
sin A sin B sin B sinC sinC sin A
= and = and =
a b b c c a
the one we use depending on what we already know and what we’re trying to
find out.

Example 6
b = 35◦ and A BC
In the triangle ABC, BC = 8 cm, C AB b = 50◦ .
Find the length of AC , giving your answer correct to 3 SF.
Solution
x cm
8 cm
35◦
A
50◦
B
We don’t know B Cb A nor do we know the length c (AB), so we use
a b
=
sin A sin B
8 x
=
sin 35◦ sin 50◦
8 sin 50◦
= x
sin 35◦
x = 10.6844
= 10.7 (3 SF )
If the question doesn’t specify a level of precision, it’s usual to asume 3 SF.
The answer is x = 10.7 cm.

The mechanics of the calculation

Two things to notice about the calculation above.
8
First, when we needed to multiply by sin 50◦ ,
sin 35◦
8 sin 50◦
we simply multiplied the numerator by sin 50◦ to give .
sin 35◦
If the reason for this isn’t immediately clear, take a simple example:
2 4
2× = .
5 5
If you were momentarily tempted to multiply top and bottom of the fraction
by 2, you were probably confusing the process with equivalent fractions.
Second, we waited till the end before reaching for our calculator. We didn’t
introduce intermediate values, and use these to calculate the final answer.
Let’s see what happens if we do so, working to 4 SF because the final answer
needs to be correct to 3 SF.
8 x
=
sin 35◦ sin 50◦
8 x
=
0.5736 0.7660
x
13.95 =
0.7660
x = 0.7760 × 13.95
x = 10.8252
= 10.8 (3 SF )
We don’t get the answer correct to 3 SF (i.e. x = 10.7).

The lessons we draw from this:
1. Avoid approximations as far as possible;

if necessary use your calulator memory.
2. Where this is not practical do the working

to several more SFs than the answer needs.

Example 7
b is acute11 .
b = 48◦ and P QR
In the triangle PQR, P R = 2 m, RQ = 3.5 m, R PQ
(i) Find the angle P QR,
b correct to 3 SF.
(ii) Hence calculate the length of PQ.
Solution
The triangle looks like this.
2m 3.5 m
48◦
P Q
(i) We don’t know Q RP

b nor do we know the length PQ, so we use
sin P sinQ
=
p q
sin 48◦ sinQ
=
3.5 2
2 sin 48◦
sinQ =
3.5
= 0.42465
Qb = 25.1◦ (3 SF )
(ii) To find PQ we need Q RP

b . But this we can easily find:
R = 180◦ − 48◦ − 25.1◦ = 106.9◦ .
Then PQ p
=
sin R sin P
PQ 3.5
=
sin 106.9◦ sin 48◦
3.5 × sin 106.9◦
PQ =
sin 48◦
= 4.5063
Length of PQ = 4.51 m (3 SF )
In part (ii) we could have used P R and Qb to find PQ. It was safer to use QR
and Pb, because it didn’t rely on an answer we had calculated.
11
Acute means an angle in quadrant Q1, obtuse in Q2 and reflex in Q3 or Q4.

In Example 7 we were told that Qb is an acute angle. In the next example we

see what the implications are of lifting this restriction.
Example 8
b = 30◦ .
In the triangle ABC, AC=9, BC =7 and C AB
C
9 7
30◦
A B
(i) Find the two possible values for the angle A BC

b .
(ii) Draw a sketch showing the two possible solutions, labelling the two
possible positions for B as B 1 and B 2 .
(iii) Find the distance B 1 B 2 .
Solution
sin B sin A
(i) =
b a
sin B sin 30◦
=
9 7
9 × 0.5
sin B =
7
Principal value: B = 40◦ (almost exactly)
The other solution (in Q2) is
180◦ − 40◦ = 140◦
(ii)
C
9
7 7
30◦ 140◦ 40◦

A
B2 N B1
(iii) The triangle C B 1 B 2 is isosceles (C B 1 = C B 2 ).

So N , the the foot of the perpendicular from C , bisects B 1 B 2 .
N B 1 = 7 × cos 40◦
B 1 B 2 = 2 × 7 × cos 40◦
= 10.725
= 10.7 (3 SF )

In these examples we have been give three pieces of information about a

triangle (two sides and an angle or two angles and a side) and asked to find a
fourth.
However the Sine rule doesn’t always work: we may need to use the Cosine
Rule. Before we see why, look at this example.
Example 9
In the triangle X Y Z , Y Z = 6 cm, X Yb Z = 110◦ and Y Zb X = 45◦ .

Find the length of X Y .
Y
6 cm
Z 110◦
45◦
X
Solution
The Sine Rule tells us that

6 XY YZ
= ◦
= .
sin X sin 45 sin 110◦
Whichever pair of fractions we take there will be two unknowns.
In this case the problem is easily solved, because

X = 180◦ − 110◦ − 45◦ = 25◦
so 6 XY
◦
= ,
sin 25 sin 45◦
from which we can find X Y .

QQ
These questions are to be answered using the Sine Rule.
1. In the triangles below find the lengths marked, giving your answers
correct to 3 SF.
12 4.6 81◦
a
42◦
◦
25◦ 110
b
2. In these triangles find the angles marked, giving your answers correct to 1
DP. The angles x and y are acute.
6.0
◦ z
51
1.9
1.3 8.2
x
37 ◦ y
3. (a) In the triangle below find the two possible values for the angle α, giving
your answers correct to 1 DP.
(b) If α is obtuse, find the length c correct to 2DP.
1.1
30◦
0.6
c
α
12
1. a = 5.40 b = 6.79
2. x = 24.3◦ y = 34.7◦ z = 94.3◦
3. α = 66.4◦ or 113.6◦ c = 0.71

When the Sine Rule fails
There are two scenarios where Sine Rule doesn’t help us:
(a) (b)
C
60◦ C
7 10 6 5
A c B A B
9
10 7 c 5 6 9
= = = =
sin A sin B sin 60◦ sin A sin B sinC
Just as in Example 9, every pair of fractions contains two unknowns.

However, unlike Example 9, there is no way of getting round the problem.
This is because each of them falls into a specific category:
In (a) we know two sides, and the angle between those two sides.
In (b) we know all three sides.
In these cases—and only in these cases—we need to use the Cosine Rule.
In all the other cases the Sine Rule will work.

And the good news is that
we only ever have to use the Cosine Rule once when solving a triangle.
The information we get from one application of the Cosine Rule will always
be sufficient to find anything else we need to know about the triangle—using
the Sine Rule or standard trigonometry.

12.8 The Cosine Rule

There are three versions of the Cosine Rule, each one involving all three sides
and just one of the angles.
B
c
A
a
C
For this triangle (where a is opposite the angle A, etc)
a 2 = b 2 + c 2 − 2bc cos A
A
2 2 2
b = c + a − 2c a cos B !
A
A
A
c 2 = a 2 + b 2 − 2ab cosC
You’ll notice the formulae are not dissimilar to Pythagoras theorem—more

about that shortly. But first a word about using them.
Remember how each one is constructed—its “anatomy” if you will.
Each formula involves one angle.
The formula gives the length of the side opposite that angle.
The two sides on the right hand side are the ones that form that angle.
Start with Pythagoras’ Theorem, and....
...subtract twice the product of those sides times the cosine of the angle.

Two pages back we saw that there are 2 scenarios where the Sine Rule doesn’t
help us. Let’s see how the Cosine Rule solves such problems.
(a) C
60◦
7 10
A c B
c2 = 72 + 102 − 2(7)(10) cos 60◦

= 149 − 140(0.5)
= 149 − 70 ← NOT 9(0.5) (a surprisingly common mistake)
= 79
c = 8.89 (3 SF)
(b)
C
6 5
A B
9
92 = 52 + 62 − 2(5)(6) cosC
81 = 61 − 60 cosC ← NOT 1 × cosC
60 cosC = −20
cosC = −0.33333
C = 109.5◦ (1 DP)

PURE MATHS: AS LEVEL GUIDE
Pythagoras as a special case of the Cosine Rule

Acute angle Right angle Obtuse angle
A A
A
c 4 c 4
c 4
B 100◦ C
◦
80 C B C 3
B 3
3
12-36
c 2 = a 2 + b 2 − 2ab cosC c 2 = a 2 + b 2 − 2ab cosC c 2 = a 2 + b 2 − 2ab cosC
= 32 + 42 − 24 cos 80◦ = 32 + 42 − 24 cos 90◦ = 32 + 42 − 24 cos 100◦

= 9 + 16 − 4.16756 = 9 + 16 − 0 = 9 + 16 + 4.16755
= 20.83244 = 25 = 29.16756
c = 4.56 (3 SF) c =5 c = 5.40 (3 SF)
Where C < 90◦ , c < 5, because Where C = 90◦ , cos 90◦ = 0, so Where C > 90◦ , c > 5, because although
we subtract 2ab cosC from a 2 + b 2 . c 2 = a2 + b2 we are still subtracting 2ab cosC ,
C is now in Q2, so cosC is negative
i.e. Pythagoras’ Theorem. ⇒ −2ab cosC is positive
QQ
These questions are to be answered using the Cosine Rule.
1. In the triangles below find the lengths marked, giving your answers
correct to 3 SF.
5 4.6 105◦ 3.7

a
30◦ b
2
2. In these triangles find the angles marked, giving your answers correct to 1
DP.
5.6
y
7
5 8.1
12.8
x◦
8
13
1. a = 3.42 b = 6.61
2. x = 60◦ y = 137.5◦

SOHCAHTOA Reminder
Since we are going to be seeing quite a few right-angled triangles in the
coming pages, a reminder about the neat way to apply SOHCAHTOA.
We can write the ratios using little triangles like this:
O A O
S H C H T A
And the point of writing it like this? From these we can find any one of the
variables from the other two.
For example in the case of the sine relationship:
O
S H
O O O
S= = H =
H S S×H

Proof of the Sine Rule and Cosine Rule

Sine Rule
C
b h
Using SOHCAHTOA we can express h in terms of A and b, or in terms of B

and a. The two expressions must we equal.
h = b sin A and h = a sin B
so
a sin B = b sin A
a b
=
sin A sin B
If we drop the perpendicular from another vertex, say A, we get
b c
= .
sin B sinC
Therefore a b c
= =
sin A sin B sinC

Cosine Rule
The proof starts the same way as the proof for the Sine Rule, with a
perpedicular which divided the side C A into two segments.
Simple trigonometry tells us that the lengths of these segments are a cosC
and b − a cosC (because C A = b). We’ll mark these on the diagram.
a h
b − a cos C
C
a cos C
As before we we’ll be writing down two different expressions for h (or in this
case h 2 ), which must therefore be equal. By Pythagoras:
h 2 = a 2 − (a cosC )2 and h 2 = c 2 − (b − a cosC )2 .

So
a 2 − (a cosC )2 = c 2 − (b − a cosC )2
a 2 − a 2 cos2 C = c 2 − (b 2 + a 2 cos2 C − 2ab cosC )
a 2 − a 2 cos2
C = c 2 − b 2 − a 2 cos2
XX XX
XX X XX C + 2ab cosC
X
c 2 = a 2 + b 2 − 2ab cosC
The other two versions of the formula arise if we drop the perpendicular from
A or C .

Exercises 44
Questions 9,10, 19 and 20 involve bearings. Chapter 13 shows how vectors

can sometimes be used to great effect to solve such problems.
1. Show that in these triangles, x = 3.66 and y = 1.92 (to 3 SF).
6 y 105◦ 0.8
x
55◦
◦
30◦ 125
[If you are not sure how to find y, there is a hint in the footnote14 .]
2. Show that in these triangles α = 44.8◦ , β = 30.3◦ and γ = 23.3◦ (correct

to 1DP), where α, β and γ are acute angles.
α, β and γ are the Greek letters alpha, beta and gamma.
9.6
◦
5 51
3 14.8
25◦
α β
B
0.31
γ A
110◦ 0.24
C
[If you are not sure how to find γ, there is a hint in the footnote15 .]
3. Show that in these triangles, correct to 3 SF, a = 2.05, b = 20.1, and that
correct to 1 DP, θ = 52.6◦ , φ = 127.7◦ and δ = 32.5◦ .
θ, φ and δ are the Greek letters theta, phi and delta.
30◦
100◦
3 11 15
4
C a b
1.9
θ φ
7 1.2
10
δ 2.8
A B
8
14
Find the third angle.
15
Find Bb first.

p
4. In the triangle ABC , Ab = 30◦ and Bb = 45◦ , show that AC = 2 BC .
5. In the triangle X Y Z , X Y = 3, Y Z = 5 and Z X = 7. Find Yb .
6. ABC is an isosceles triangle in which AC = BC = 7. The point P lies on

AB , with AP = 6 and P B = 4.
b =5.
(i) Explain why cosC AB 7
(ii) Find the length of C P .
1
(iii) Show that cosC PbB = − 5 .
7. Try to complete this question without using either the Sine Rule or the
Cosine Rule.
b = 30◦ and B Cb A = 40◦ .
In the triangle ABC , BC = 6, C AB
Find (i) the perpendicular distance from B to AC
(ii) the length of AB
(iii) the length of AC
(iv) the area of the triangle ABC .
8. PQRS is a parallelogram in which S R

b = 60◦ and P QS
PQ = 12, P S = 8, P SQ
60◦
b
is an acute angle.
8
Find (i) P QS,
b (ii) SQ (iii) P R.
Find the area of the parallelogram.
P Q
12
9. A ship starting at a port P travels to a point X which is 10 km due North
of P .
It then sails on a bearing 130◦ until it reaches a point Y , the bearing of
which from P is 070◦ .
Draw and label a diagram showing the movement of the ship.
(i) Find, correct to 2 DP, the distances X Y and P Y .
(ii) How far East of the port is the point Y ?
(iii) When the ship is at Y , what is the bearing of the port?
10. After flying for 60 km, an aircraft alters its bearing by an angle α and flies
for a further 100 km. It is then 120 km from where it started.
Find the angle α.
11. In the triangle PQR, R PQb = 30◦ and QR = 10. T a point on PQ such
that R TbQ = 50◦ and P T = 8.
R
Find (i) P R (ii) P QR.
b
10 Find the area of the

triangle PQR.
30◦ 50◦
P 8 T Q
p
12. In the triangle ABC , BC = 19, AB = x and AC is 2 units longer than
b = 60◦ . Find x.
AB . The angle C AB

13. The diagram shows five beams forming part of a bridge structure.
The beams AB and C D are horizontal, and B D
b A is a right angle.
D C
◦
35
65◦
A 1.8 m B
Find the length of B D and BC , and the angle A BC

b .
It is decided to add an extra beam, running between A and C . How long
will this new beam be?
14. The diagram shows the cross D

section of part of a building. 60◦ 5m
The walls AD and BC are
vertical. C
(i) Find the height of C above
6m
ground level.
p
(ii) Show that AC = 31 m.
(iii) Find the angle θ.
θ
A B
15. Z Without using a calculator,

find the two possible values
p
2 1 for the angle φ.
30◦ φ
X Y
16. Find the area of the triangle

and the length x, without 2 p
using a calculator. p 2+1 120◦ 2−
2 1
x
17.
C
6
(i) Write down the
D value of cos B AC
b .
x
4 (ii) Find x.
(iii) Find the exact
value of cos B Cb A.
A 3 B
18. In the triangle PQR, sinQ = 2 sin P and P R = 6.

(a) Find the length of QR.
(b) Write down the largest possible value that sinQ can take. Hence or
otherwise show that P < 30◦ .

19. In a desert rally, a car travels from its starting point P to the first
checkpoint which is at Q, 25 km from P on a bearing of 040◦ . From Q it
travels, on a bearing of 105◦ for a further 15 km to R, where it breaks
down. There is a recovery team based at P .
(a) When the car breaks down, how far north and east of P is it?
(b) How far does the recovery team need to travel to reach the car?
(c) What bearing does the recovery truck need to follow?
20. At 0900 hours a woman sets out from A, walking at a speed of 3km/h,
on a bearing of 120◦ . One hour later, a man set out from B , a point 5 km
due North of A. He walks on a bearing of 210◦ at a speed of 4km/h. At
1200 hours the woman is at C and the man is at D.
(a) Draw a diagram to show the positions of A, B , C and D, marking in
the lengths of AB , AC and B D.
(b) Find how far the man is from A at 1200 hours.
(c) Find the bearing of A from D.
(d) Find the angle D AC
b and hence, or otherwise, calculate the distance
DC .
21. Find, in terms of a, the

C
length of N B .
Hence, or otherwise,
2a
find the ratio of
area M ANC : area
30◦ M B NC .
A N B
The angle of elevation
22.
C of the top of a church
tower C is 28◦ from A
and 41◦ from B , whch
is 20 m closer to the
church than A.
Find (i) BC , (ii) the
28◦ 41◦ height of the tower.
A 20 m B Give your answers to
the nearest cm.
23. The diagram shows a triangle X Y Z enclosedp in a cuboid, whose sides

are of length 6, 4 and 2. Show that X Y = 2 10.
Find, in surd form, the lengths of Y Z and Z X . Hence, or otherwise, find
the angle X Yb Z .
Find the area of the triangle X Y Z .
Z
2
X
4
6 Y

24. In the diagram opposite, O is the

centre of a circle, radius 5, R is the S
midpoint of OP and OS = 3.
O
PQ is a tangent to the circle.
b = 30◦ .
(i) Show that O PQ
R
(ii) Find the length of QS.
Q
P
p
3
25. Show that in the triangle below, sin θ = 1− 2
θ
p
3+1
30◦
p
3−1
26. The curve y = 4 + 3x − x 2 crosses the x-axis at P (a, 0) and Q(b, 0), and
the y-axis at R(0, c), where a < b.
(a) Find the values of a, b and c.
(b) State the angle A BC
b .
(c) Find the angle ACbB .
p
5 2
(d) Show that the perpendicular distance from P to RQ is 2 .
27. The lengths of the sides of a triangle are a, a + d and a + 2d , where

d > 0. If the largest angle of the triangle is φ, show that
a − 3d
cos φ = .
2a
Describe the triangles for which d = 0.
Find, in terms of a, the value of d for which the triangle is right-angled.
Hence, or otherwise, show that for the sides of a right-angled triangle to
be in arithmetic progression, they must be in the ratio 3 : 4 : 5.
28. The internal angles of a triangle are A,

b Bb and Cb. Show that
(i) sin(A + B ) = sinC , and
(ii) cos(A + B ) + cosC = cos(B +C ) + cos A = cos(C + A) + cos B = 0.
29. We are familiar with right angled triangles the lengths of whose sides
are all integers. (For example 3, 4 and 5.)
Show that there is a triangle that is not right angled but has one angle
equal to 60◦ and sides of length 5, 7 and 8.

Solutions 44
1 1
4. Use Sine Rule, where sin Ab = 2 and sin Bb = p .
2
32 +52 −72
5. 120◦ [ cos Y = 2(3)(5) = − 12 ]
6. (i) 4ABC is isosceles, so divides C
into two right-angled triangles.
(ii) C P = 5 [Cosine Rule 4C AP ] (iii) 7 7
[Cosine Rule 4C P B ]
A 5 1P 4 B
C
◦
7. (i) 3.86 [6 sin 40 ]
NB 40◦
(ii) 7.71 [ sin 30◦ ] N
(iii) 11.28 [6 cos 40◦ + AB cos 30◦ ] 6
(iv) 21.74 [ 12 AC × N B ] 30◦
A
B
◦
8. (i) 35.3
(ii) 13.8 [Pb = 84.7◦ then Sine Rule]
(iii) 15.0 b = (35.3◦ + 60◦ ) then Cosine Rule]
[P QR
Area = 95.6 [2× area triangle PQS]
9.
N
(i) X Y = 10.85 km, P Y = 8.85 km [P Yb X = 60◦ ]
(ii) 8.31 km (or 8.32) [X Y sin 50◦ or P Y sin 70◦ ]
(iii) 250◦ [70◦ + 180◦ ]
X
50◦
10 km
Y
70◦
P
10. 62.7◦ [Actual directions not relevant. Use Cosine Rule to find 180◦ − α]
11. (i) 17.9 b = 20◦ ]

[P RT
(ii) 63.6◦ [Sine Rule triangle PQR]
b = 66.4◦ , so P RQ
(iii) 89.4 (or 89.3) [T RQ b = 86.4◦ ,
then 12 P R × RQ × sin 86.4◦ ]
12. x = 3 [x 2 + (x + 2)2 − 2x(x + 2) = 19 ⇒ (x + 5)(x − 3) = 0, x > 0]
13. B D = 1.63 m [1.8 sin 65◦ ]

BC = 1.20 m b = A BbD = 25◦ , then Sine Rule triangle BC D]
[B DC
b = 145◦
A BC b = 120◦ ]
[D BC
AC = 2.87 m [Cosine Rule triangle ABC ]
14. (i) 3.5 m [6 − 5 cos 60◦ ]

(ii) Cosine Rule triangle AC D.
(iii) 51.1◦ [By Sine Rule, knowing AC ]

sin φ
1 1
15. φ = 45◦ or 135◦ [ p = 2 ⇒ sin φ = p . φ = 45◦ or 180◦ − 45◦ ]
2 2
p p p p
7 3 3
16. Area = 4 . [(2 2 + 1)(2 2 − 1) = 7, sin 120◦ = sin 60◦ = 2 ]
1
x =5 [Cosine Rule, cos 120◦ = − cos 60◦ = − 2 ]
3
17. (i) 4 [B AC
b = B AD]
b
(ii) x = 8 [Cosine Rule]
31
(iii) cos B Cb A = 32 [Cosine Rule]
QR 6 6
18. QR = 3 [ sin P = sinQ = 2 sin P ]
Largest possible value of sinQ = 1.
1 1 1
sinQ ≤ 1, so sin P = 2 sinQ ≤ 2 . Since sin 30◦ = 2 , Q ≤ 30◦ .
19.
It is sometimes easiest to
N solve bearing questions using
Q standard trigonometry.
10
5
◦
15 k (a) 15.3 km, 30.6 km.

75◦ m
R [P X = sum of red lengths
km
= 25 sin 50◦ + 15 sin 75◦

25
X R is difference of blue ones

= 25 cos 50◦ − 15 cos 75◦ ]
40◦
α (b) 34.2 km [Pythagoras]
P X (c) 63.4◦ [α = 26.6◦ ]
20. N (a) When drawing your diagram make it big, to
allow plenty of room for measurements, angles
etc.
B 210◦ (b) 2.70 km [Cosine Rule triangle AB D]
(c) 97.7◦ [Sine Rule. θ = 67.7◦ ]
30◦ (d) 157.7◦ [97.7◦ + 60◦ ]
5 km
DC = 11.5 km
m
8k
[Cosine Rule (using AD) or the method of Q19]
30◦ A 120◦
θ 60◦ 9k
m
D
C
2a
21. N B = 3a [AB = sin 30◦ = 4a. Ab = 60◦ , so AN = 2a cos 60◦ = a]
Ratio = 1 : 3 [Triangles have same height, so ratio = AN : N B ]
22. (i) BC = 41.74 m [ACbB = 13◦ . Then Sine Rule triangle ABC ]
(ii) Height = 27.38 m [BC sin 41◦ ]
p p p p
23. X Y = 62 + 22 = 40 = 4 10.
p p
Y Z = 2 5, Z X = 2 13,
X Yb Z = 81.9◦ [Cosine Rule triangle X Y Z ]
Area = 14 square units. [Surprisingly perhaps, this is the exact answer.]

b = 90◦ (tangent is perp to radius), so sinO PQ OQ 1

24. (i) P QO b =
OP
=2
p p
3
(ii) QS = 7 [Cosine Rule on PQS, where PQ = 5 3, cosO PQ = 2 ,
b
b = cos 120◦ = − .] 1
or on QSO, where cosQ OS 2
1
25. sin θ 2
p p =
3−1 3+1
p p
3−1 ( 3 − 1)2
sin θ = p = p p
2( 3 + 1) 2( 3 + 1)( 3 − 1)
p p
3−2 3+1 2(2 − 3)
= =
2(3 − 1) 4
p
3
= 1−
2
26. y (a) a = −1, b = 4, c = 4.
6 [When x = 0, y = 4, so c = 4
4 R Then roots of x 2 − 3x − 4 = 0
N (x + 1)(x − 4) = 0, x = −1 or 4]
b = 45◦ [OQ = OR]
(b) P QR
b = 59.0◦
(c) P RQ
[Pythag PQ 2 = 17, RQ 2 = 32 then
P Q- x Cosine Rule]
−1 0 4
p p
1 5 2
(d) P N = PQ sin 45 = 5× = 2 or Area = 10 = 12 32 × P N
p ◦
2
27. a 2 + (a + d )2 − (a + 2d )2 a 2 − 2ad − 3d 2
cos φ = =
2a(a + d ) 2a(a + d )
(a + dX)(a − 3d )
XXX
a − 3d
= =
2aX(aX dX)
+X 2a
d = 0 ⇒ an equilateral triangle.
a
Right-angled if d = 3 [Largest angle is 90◦ , so cos φ = 0]
a 2a 3a 4a 5a
Sides are a, a+ 3 , a+ 3 , i.e. 3 , 3 , 3 . Ratio is 3 : 4 : 5.
28. (i) A + B = 180◦ −C , so sin(A + B ) = sin(180◦ −C ) = sinC

(ii) cos(A + B ) = cos(180◦ −C ) = − cosC , so cos(A + B ) + cosC = 0.
Similarly for B +C and C + A.
29. The 60◦ angle is between the sides of length 5 and 8.

12.9 More trigonometrical equations

12.9.1 The sine, cosine and tangent curves
Before solving some equations, a reminder about what the curves of these
functions look like. The sketches show where the functions take the values
−1, 0 and +1 (and in the case of the tan curve ∞).
y = sin x
6
1
1
2
J
J◦
◦ - x
−180 −90◦ 30◦ 90◦ 180◦ 270◦ 360
− 21
−1
y = cos x
6
1
1
2
- x
−180◦ −90◦ 0◦ 60◦ 90◦ 180◦ 270◦ 360◦
− 21
−1
y = tan x
6
- x
−180◦ −90◦ 0◦ 45◦ 90◦ 180◦ 270◦ 360◦
−1
Seeing how the functions switch from positive and negative, and where the
symmetries are, can also be used to find all the roots of an equation, once we
have the principal value.
We have illustrated this with principal values 30◦ , 60◦ and 45◦ respectively.

However, it is often easier way to find other solutions using the CAST
diagram, which tells us which trig function is POSITIVE in each of the
quadrants.
S i ne Al l
Tan C os
Often we are asked for solutions in the range 0◦ ≤ x ≤ 360◦ .

In this case, if P is the principal value, the candidates (possible solutions) are
P, 180 − P, 180 + P, 360 − P degrees
180 − P • • P
180 + P • • 360 − P
So if P = 30◦ , the other angles to consider will be 150◦ , 210◦ and 330◦ .
The CAST diagram tells us which of them is a valid solution.
Sometimes we are asked for solutions in a different range, for example

−180◦ ≤ x ≤ 180◦ .
In this case, if P is the principal value, the other candidates are
±P and ± (180 − P ) degrees
180 − P • • P
−(180 − P ) • • −P
So if P = 30◦ , the other angles to consider will be −30◦ , 150◦ , and −150◦ .
The CAST diagram tells us which of them is a valid solution.

12.9.2 Identities and the notion of equivalence

Look at these two statements:
x +1 = 7 and 2(x + 1) = 2x + 2.
They both express equality but there is a key difference.
x + 1 = 7 is an equation: it is only satisfied by one value of x;
2(x + 3) = 2x + 6 is an identity: it is satisfied by all values of x.
We say that the two sides of an identity are equivalent, and when we want to
emphasise this we can use the ≡ sign, rather than the = sign:
2(x + 3) ≡ 2x + 6.
You have already met this concept in arithmetic: an equivalent fraction is the
same fraction just written a different way, so strictly speaking we should use
the ≡ sign when we write:
2 4
≡ .
3 6
We met the difference of two squares identity in Chapter 3:
x 2 − y 2 ≡ (x + y)(x − y)
The term identity is related to the word “identical"— a stronger term than
“equal".
It is not essential to use the ≡ notation every time, in
fact we usually don’t. The key thing is to recognise that
with an equation we are looking for values that satisfies it,
whereas an identity gives us an alternative way of writing
something. Therein lies its power: it provides a different
way of looking at something.
Earlier in the chapter we met two identities fundamental to trigonometry:

A
sin A
!
A
tan A ≡ cos A A
A
and
A
sin x ≡ cos(90◦ − x)
!
A
A
cos x ≡ sin(90◦ − x) A

12.9.3 The identity sin2 x + cos2 x ≡ 1

This identity is central to much of the work ahead. It says that if we take the
sine of any angle and the cosine of the same angle and square each of them,
the two numbers always add up to 1.
The notation is important:
when we write sin2 x, we mean (sin x)2
similarly cos2 x means (cos x)2 , and tan2 x means (tan x)2 .
This may see an odd convention, but if we were to write sin x 2 , it would
mean that the angle x is being squared.
Two examples.
1. Let’s select an angle at random, say 23◦ . Working to 5DP:
sin 23◦ = 0.39073
cos 23◦ = 0.92050
sin2 23◦ + cos2 23◦ = (0.39073)2 + (0.92050)2
= 0.99999
The answer isn’t exactly 1, but this is because we only worked to 5DP.
2. If we choose 60◦ , we can use exact values of sin 60◦ and cos 60◦ , so we
should get an exact result: p
◦ 3
sin 60 =
2
1
cos 60◦ =
2
3 1
sin2 60◦ + cos2 60◦ = +
4 4
= 1
Proof that sin2 x + cos2 x ≡ 1

Take any triangle ABC with a right-angle at C . Suppose the angle at A is x.
By Pythagoras’ Theorem
a2 + b2 = c 2
Dividing through by c 2
a2 b2 c2
+ =
c2 c2 c2
c a ³ a ´2 µ b ¶2
+ = 1
c c
(sin x)2 + (cos x)2 = 1
x
b which we write as
sin2 x + cos2 x = 1

QQ
Verify the identity sin2 x + cos2 x = 1

(a) for any angle you care to choose,
(b) for x = 45◦ , using the exact (surd) form of sin 45◦ and cos 45◦ .
The proof for (b) is in the footnote16 .
³ ´2 ³ ´2
16
sin2 45◦ + cos2 45◦ = p1 + p1 =1
2 2

The identity tells us that if we know the sine of an angle we can work out17
what its cosine, and vice versa:
cos2 x ≡ 1 − sin2 x
A
!
A
A
sin2 x ≡ 1 − cos2 x A
It may seem that this is of limited value, because we have access to a

calculator. To begin to see why it’s useful here are two examples.
Example 10
Solve the equation

3 sin2 x − cos2 x = 2
finding all the solutions in the range 0◦ ≤ x ≤ 360◦ .
Solution
We can replace the cos2 x term with an expression involving sin x.
3 sin2 x − cos2 x = 3 sin2 x − (1 − sin2 x)

= 4 sin2 x − 1
so the equation is
4 sin2 x − 1 = 2
4 sin2 x = 3
3
sin2 x =
4
r p
3 3
sin x = ± = ±p
4 4
p
3
= ±
2
The principal value is 60◦ , and because sin x can be positive or negative, we
need the related angles in all four quadrants.
x = 60◦ , 120◦ , 240◦ or 300◦ .
We could have replaced the sin2 x term with an expression involving cos x.
As an exercise for yourself, solve the problem this way. The solution is over
the page.
17
Because we will be taking a square root, there will typically be two answers—a positive
value and a negative value. In view of what we now understand about angles greater than
90◦ , and the possibility of negative sines and cosines, this should not surprise us.

Example 10 (alternative solution)
Solve the equation

3 sin2 x − cos2 x = 2
finding all the solutions in the range 0◦ ≤ x ≤ 360◦ .
Solution
3 sin2 x − cos2 x = 3(1 − cos2 x) − cos2 x

= 3 − 4 cos2 x
so the equation is
3 − 4 cos2 x = 2
1 = 4 cos2 x
r
1 1
cos x = ± = ±p
4 4
1
= ±
2
the solutions of which are x = 60◦ , 120◦ , 240◦ 300◦ , as before.

Example 11
Solve the equation

2 sin2 x − cos x − 1 = 0
where 0◦ ≤ x ≤ 360◦ .
Solution
Just as in the previous example, we can use the identity sin2 x + cos2 x = 1 to
get rid of the sin term, leaving a quadratic18 in cos x:
2(1 − cos2 x) − cos x − 1 = 0
0 = 2 cos2 x + cos x − 1
(2 cos x − 1)(cos x + 1) = 0
1
cos x = or cos x = −1
2
i.e. x = 60◦ , 180◦ or 300◦
We have used CAST to identify the solution x = 360◦ − 60◦ = 300◦ .

It is worth reminding outselves that cos2 x is our way of writing (cos x)2 .
This avoids any confusion with cos x 2 (where it is the angle that is squared!).
Example 12
Solve the equation

5 sin2 x = sin x
where 0◦ ≤ x ≤ 360◦ .
Solution
You might be tempted to divide both sides by sin x.

This would give 5 sin x = 1, i.e. sin x = 0.2.
This will give us valid solutions, but not all of them. Why would it be wrong19
to divide by sin x? Because that assumes sin x 6= 0.
Instead we take the usual route for quadratics, moving everthing over to one
side:
5 sin2 x − sin x = 0
sin x(5 sin x − 1) = 0
1
sin x = 0 or sin x =
5
i.e. x = 0◦ , 11.5◦ , 168.9◦ or 180◦
We have used CAST to identify the solution x = 180◦ − 11.5◦ = 168.9◦ .
18
We could susbstitute y = cos x giving the quadratic equation 2y 2 + y − 1 = 0. This
factorises, to give y = 12 or −1, from which we can work out the values of x.
19
It is equivalent to solving an equation like 5x 2 = x by dividing by x. This would miss out
possible the solution x = 0.

Example 13
Solve the equation

cos2 x = 3 sin2 x
where 0◦ ≤ x ≤ 180◦ .
Solution
Method 1
Using the approach we used in the previous example
1
tan2 x = p
3
1
tan x = ± p
3
x = 30◦ or 180◦ − 30◦ = 150◦
Method 2A
We could use the identity sin2 x + cos2 x = 1
1 − sin2 x = 3 sin2 x
4 sin2 x = 1
1
sin x = ±
2
x = 30◦ or 180◦ − 30◦ = 150◦
Method 2B
Using the identity sin2 x + cos2 x = 1 the other way round
cos2 x = 3(1 − cos2 x)
= 3 − 3 cos2 x
4 cos2 x = 3
3
cos2 x = ±
4
r p
3 3
cos x = ± = ±
4 2
x = 30◦ or 180◦ − 30◦ = 150◦
Many trig equations involve using one of the identities to rewrite it. So far we
have only met two of these:
sin θ
tan θ = and sin2 θ + cos2 θ = 1
cos θ
In this next example we use both.
Later on you will meet many more identities, enabling you to solve more and
more complex equations.

The symbol θ, often used to denote an angle, is the greek letter theta.
Example 14
Solve the equation

tan θ sin θ = 2
where 0◦ ≤ θ ≤ 360◦ .
Solution
sin θ
× sin θ = 2
cos θ
sin2 θ = 2 cos θ
1 − cos2 θ = 2 cos θ
cos2 θ + 2 cos θ − 1 = 0 p
−2 ± 8 p
cos θ = = −1 ± 2
2
= 0.4142 or − 2.4142
The second of these is not a valid solution (the cosine function takes values
between −1 and +1), so
θ = 65.5◦ or 294.5◦
Try not to treat these as standard types of equation. Be prepared to try

different approaches until you find one that works. In the next equation there
is a cos2 x term, but it doesn’t help in this case to write it as 1 − sin2 x.

Example 15
Solve the equation

sin x cos x = 3 cos2 x
where −180◦ ≤ x ≤ 180◦ .
Solution
Notice that solutions must lie in the range −180◦ ≤ θ ≤ 180◦ .
This is just another way of describing every possible angle in a complete
circle. So for example, the angle 200◦ could equally well be described as the
angle −160◦ .
We have already seen that we mustn’t simply divide both sides by cos x.
Instead make one side of the equation = 0 and factorise.
cos x cos x − 3 cos2 x = 0

cos x (sin x − 3 cos x) = 0
cos x = 0 or sin x = 3 cos x
cos x = 0 or tan = 3
i.e. x = 71.6◦ , 90◦ , 251.6◦ , 270◦
using CAST and an understanding of what the cosine curve looks like.
One final step. Since −180◦ ≤ x ≤ 180◦ we need to state the solutions as
x = 71.6◦ , 90◦ , −108.4◦ , −90◦ .
+90◦ +90◦
• +71.6 ◦ • +71.6◦
• •
71.6◦ 71.6◦
≡
• •
+251.6◦ • ◦
−108.4◦ •
+270 −90◦
0◦ ≤ x ≤ 360◦ − 180◦ ≤ x ≤ +180◦

Example 16
Solve the equation

5 cos x + 2 sin x = 0
where 0◦ ≤ x ≤ 360◦ .
Solution
There is no question now of using sin2 x + cos2 x = 1 to get rid of the sin x
term or the cos x term, because the formula connects the (sin x)2 with
(cos x)2p
. If we wanted to go down this route, we would need to substitue
p
cos x = 1 − sin2 x or sin x = 1 − cos2 x. Not an atrractive prospect if we can
avoid it.
sin x
Fortunately we can use the identity tan x = cos x .
Rearranging and dividing20 both sides by cos x
5 = −2 tan x
tan x = −2.5
x = 150◦ or x = 330◦
A calculator is likely to give the value −68.2◦ (to 3SF). This is in the 4th
quadrant, which is consistent with the CAST diagram, which tells us that
tan x is negative in quadrants 2 and 4.
The principal angle is 68.2◦ , so the solutions are 180◦ − 68.2◦ = 111.8◦ and
360◦ − 68.2◦ = 291.8◦ .
20
You may be wondering why it is now OK to divide by cos x when in the previous example
we said it was wrong to divide by sin x. In that case it was because we can’t divide by zero and
we don’t know that sin x 6= 0. This is not a problem here because in this equation cos x = 0
would imply sin x = 0 as well, which happens for no value of x.

Example 17
p
The angle θ is such that cos θ = 2 − 1. Find the exact values of
(i) sin2 θ, and (ii) tan2 θ.
Solution
Because the question asks for anpexact answer we can’t use a calculator (or
tables). We must work21 with the 2.
p
(i) cos θ = 2−1
p
cos2 θ = ( 2 − 1)2
p
= 2−2 2+1
p
= 3−2 2
sin2 θ = 1 − cos2 x
p
= 1 − (3 − 2 2)
p
= 2 2−2
p
= 2( 2 − 1)
(ii) sin θ
tan θ =
cos θ
sin2 θ
tan2 θ =
cos2 θ
p
2( 2 − 1)
= p
3−2 2
This is the correct answer. However, in Chapter 2 we saw how fractions

involving surds can sometimes be simplified. (If you want a reminder about
how we do this, see page 2-65.)
p p p
2( 2 − 1) 2( 2 − 1)(3 + 2 2)
p = p p
3−2 2 (3 − 2 2)(3 + 2 2)
p p
2(3 2 − 2 2 − 3 + 4)
=
9−8
p
= 2( 2 + 1)
21
p
We saw in Chapter 1 that 2 is irrational, which means its exact value can never be
written down.

QQ
answers to the nearest degree. Try to avoid using a calculator if you can.
(a) 2 cos2 x + 5 cos x − 3 = 0

(b) sin2 x + 3 sin x + 1 = 0
(c) 2 sin x = 3 cos x
(d) 4 sin2 x = sin x cos x.
(e) tan2 x = tan x + 2.
(f) sin2 x + 3 cos2 x = 2
(g) 2 sin2 x = 5(cos x + 1).
(h) tan x = sin x cos x.
22 1
(a) x = 60◦ , 300◦ [cos x = 2 or −3 (no solutions)]
◦
(b) x = 210 , 270 , 330 ◦ ◦
[sin x = − 21 or −1]
(c) x = 56◦ , 236◦ [tan x = 1.5]
(d) x = 0 , 14 , 180◦ , 194◦ , 360◦
◦ ◦
[sin x = 0 or tan x = 14 ]
(e) x = 63◦ , 135◦ , 243◦ , 315◦ [tan x = 2 or −1]
◦ ◦
(f) x = 45 , 135 , 225 , 315 ◦ ◦
[tan x = ± p1 ]
2
(g) x = 180◦ [cos x = −1 or − 32 (no solutions)]
(h) x = 0◦ , 180◦ , 360◦ [sin x = sin x cos2 x ⇒ sin x(1 − sin2 x) = sin3 x = 0]

A warning
Take an expression like cos(x + 60◦ ).
cos(x + 60◦ ) is not the same as cos x + cos 60◦ .
We only have to look at an example. Suppose x = 30◦ .

p
3
cos(30◦ + 60◦ ) = cos 90◦ = 0, whereas cos 30◦ + cos 60◦ = 2 + 12 6= 0.
The trap is easy to fall into because it looks like we are multiplying “cos” by
(x + 60◦ ), in other words expanding a bracket.
Of course there is no such thing as a “cos”; cos is not a number so the idea of
mutiplying it by something doen’t make sense.
Here is an example of another common pitfall:
sin 2x is not the same as 2 sin x.
It might look as if we are simply reversing the order in which we multiply the
“sin” and the 2, but “sin” is not a number, either. In fact sin 2x means23
sin(2x).
Once again the example x = 30◦ makes the distinction clear:
p
3
sin(2 × 30◦ ) = sin 60◦ = 2
, whereas 2 sin 30◦ = 2 × 12 = 1.
This is not to say that sin 2x and 2 sin x won’t sometimes be equal. For
example if x = 0, they are. To be precise we should say the expressions are
not identical; they are not equivalent to each other; they are not the same for
every value of x. To make this absolutely clear, we can use the ≡ sign:
cos(x + 60◦ ) 6≡ cos x + cos 60◦

sin 2x 6≡ 2 sin x
Generalising:
sin(x + y) 6≡ sin x + sin y sin k x 6≡ k sin x
cos(x + y) 6≡ cos x + cos y cos k x 6≡ k cos x
tan(x + y) 6≡ tan x + tan y tan k x 6≡ k tan x
Depending on how much further you take your Maths, you will learn other
ways of writing cos(x + 60◦ ), sin 2x, etc but at present you won’t need one.
23
We would generally omit the bracket because we are unlikely to confuse it with (sin 2)×x.

QQ
In each case say whether the two expressions are identically equal, i.e. are
the same for all values of x. Take care, because the answer may not be as
obvious as it first seems.
For example sin(90◦ − x) 6≡ sin 90◦ − sin x.
But sin(90◦ − x) ≡ cos x (think of a right-angled triangle).
cos(x + 10◦ ) and cos x + cos 10◦
sin x and sin(180◦ − x)
tan 3x and 3 tan x
1 x
cos x and cos
2 2
sin x and sin(x + 720◦ )
cos(90◦ − x) and sin x
sin x − sin 45◦ and sin(x − 45◦ )
tan x and tan(180◦ + x)
24
cos(x + 10◦ ) 6≡ cos x + cos 10◦
sin x ≡ sin(180◦ − x) [remember the CAST diagram]
tan 3x 6≡ 3 tan x
1 x
cos x 6≡ cos
2 2
sin x ≡ sin(x + 720◦ )
cos(90◦ − x) ≡ sin x
◦
sin x − sin 45 6≡ sin(x − 45◦ )
tan x ≡ tan(180◦ + x) [remember the CAST diagram]

Example 18
Solve the equation

cos(x + 45◦ ) = 0.5
where 0◦ ≤ x ≤ 360◦ .
Solution
We have just seen that we can’t rewrite cos(x + 45◦ ) as cos x + cos 45◦ ,
so we have to work with the angle (x + 45◦ ).
One way to do this is to let x + 45◦ = y. Then the equation becomes
cos y = 0.5.
Solving this, we have y = 60◦ or y = 300◦ .
Since x = y − 45◦ , the solutions are x = 15◦ or y = 315◦ .
The next example looks very similar, but there is a twist.
Example 19
Solve the equation

sin(x + 45◦ ) = 0.5
where 0◦ ≤ x ≤ 360◦ .
Solution
Letting x + 45◦ = y we have

sin y = 0.5.
which gives y = 30◦ or y = 150◦ .
Since x = y − 45◦ , the solutions are x = −15◦ or x = 105◦ .
Since we are only interested in solutions in 0◦ ≤ x ≤ 360◦ we may think there

is only one solution: x = 105◦ . But we would be wrong
because if 0 ≤ x ≤ 360
then 45 ≤ y ≤ 360 + 45 = 405
In this range there is a solution y = 360◦ + 30◦ = 390◦ , i.e, x = 345◦ .

The complete solution to the equation is the two values x = 105◦ , 345◦ .
Another way of thinking about this is to say that the solution we discarded,
i.e. x = −15◦ , is really the same angle as x = 345◦ , but arrived at by going
anticlockwise rather than clockwise round the circle.
Missing one of the solutions may not seem a big deal, but in an exam if there is
one mark for each solution you don’t want to lose one of them. So let’s look at
another example.

Example 20
Solve the equation

cos 3x = 0.8 where 0◦ ≤ x ≤ 180◦
giving you answers correct to 3 significant figures.
Solution
First of all, notice that the values of x only need to go up to 180◦ .

However, if we let t = 3x, say
since 0◦ ≤ x ≤ 180◦
0◦ ≤ t ≤ 3 × 180◦ = 540◦ .
We now need to solve the equation cos t = 0.8, for values of t between 0◦ an
540◦ .
The principle solution (from our calculator) is t = 36.9◦ .
Where are the other solutions? we could use a sketch of the cosine curve.
cos t
6
1
•
- t
−180◦ 0◦ 36.9◦ 180◦ 360◦ 540◦ 720◦
−1
The principle solution is at the red dot, and dashed lines show the positions
of the other solutions (within the range 0◦ ≤ y ≤ 540◦ ).
Alternatively the CAST diagram tells us we must be in the 1st or 3rd
quadrants. But because we need to go up to 540◦ (= 360◦ + 180◦ ) it means
going one and a half time round.
So, by symmetry from the sketch or for applying CAST, the solutions are
t = 36.9◦ , t = (360 − 36.9)◦ = 323.1◦ , and t = (360 + 36.9)◦ = 396.9◦ ,
Since t = 3x, the solutions of our original equation cos 3x = 0.8 are
x = 13.3◦ , 107.7◦ and 132.3◦25 .
Reasuringly, all these values lie in the range 0◦ ≤ x ≤ 180◦ .
I have cheated a little here, doing all my working correct to 3 SF. This is
because it looks less messy, and I happened to know that it would still give me
the correct answers. You will need to do your working to 4 or 5 SF, to avoid
possible rounding errors.
25
Another approach could be to start with the solution x = 13.3◦ and sketch the curve y =
cos 3x to find the others. Most students (including me) find it much easier to stick with t and
only convert back to x at the very end.

Example 21
Solve the equation

tan(2x − 60◦ ) = −1
where 0◦ ≤ x ≤ 180◦ .
Solution
Letting 2x − 60◦ = y we have

tan y = −1.
Because 0 ≤ x ≤ 180
−60 ≤ y ≤ 360 − 60 = 330
In this range there are solution y = −45◦ and y = 135◦ ,

i.e. 2x − 60◦ = −45◦ or 135◦ .
The solutions of the equation are x = 7.5◦ and 97.5◦ .

QQ
answers to the nearest degree. None of these questions need a calculator.
(a) sin(x + 10◦ ) = 0.5.

(b) sin(x + 45◦ ) = 0.5.
p
(c) tan(x − 30◦ ) = 3
(d) sin 3x = 1
p
3
(e) cos 12 x = − 2
(f) tan(2x + 90◦ ) = 1.
You now have the tools you need to solve any trig equation you could come
across in an exam question. As always there are ways of disguising equations
so that the technique required may not be immediately apparent. We will
discuss some of these disguises later.
26
(a) x = 30◦ , 150◦ [10◦ ≤ x + 10◦ ≤ 360◦ + 10◦ , so x + 10◦ = 30◦ or 150◦ ]
(b) x = 105◦ , 345◦ [45◦ ≤ x + 45◦ ≤ 360◦ + 45◦ , so x + 45◦ = 150◦ or 360◦ + 30◦ ]
(c) x = 90◦ , 330◦ [−30◦ ≤ x − 30◦ ≤ 360◦ − 30◦ , so x − 30◦ = 60◦ or 360◦ − 60◦ ]
(d) x = 30◦ , 150◦ , 270◦ [0◦ ≤ 3x ◦ ≤ 3 × 360◦ , so 3x ◦ = 90◦ , 360◦ + 90◦ or 720◦ + 90◦ ]
(e) x = 300◦ [0◦ ≤ 12 x ◦ ≤ 180◦ , so 21 x ◦ = 180◦ − 30◦ ]
(f ) x = 67.5◦ , 157.5◦ , [90◦ ≤ 2x + 90◦ ≤ 720◦ + 90◦ ,
247.5◦ , 337.5◦ so 2x + 90◦ = 180◦ + 45◦ , 360◦ + 45◦ , 540◦ + 45◦ , 720◦ + 45◦ ,
or 2x = 135◦ , 315◦ , 495◦ , 675◦ ]

12.10 A formula for the area of a triangle

The new formula uses the lengths of two sides and the angle between them.
We can, of course, choose any pair of sides, so there are three variants.
1
Area =
2
ab sinC
A
1
!
A
Area =
2
bc sin A A
A
1
Area =
2
c a sin B
It’s very easy to see why the formula works.

A
b
h
C B
a
The height of the triangle h = b sinC .

1 1
Area = 2 a × h = 2 ab sinC .
Notice that the reasoning hold good, whether the perpendicular from A to
AB is inside or outside the triangle.
A
b
h
C a
B
How does this square with the familiar half base times height formula for the
area of a right angled triangle? If C = 90◦ , sinC = 1, and we have
1
Area = 2 ab.
C B
a

12.11 More on trigonometrical identities

So much of the business of doing Mathematics can be summed up in one
simple phrase: learning to write the things in a different way. And part of the
delight is suddenly seeing a way of doing so that opens the way to a solution.
Yes, there are some important concepts and techniques, but applying them
often boils down to seeing what’s in front of you slightly differently.
Quite often that means using an identity, like the difference of two squares:
a 2 − b 2 ≡ (a + b)(a − b).
Earlier in the chapter we met the trig identity
cos2 θ + sin2 θ ≡ 1
which we proved by thinking of Pythagoras’ Theorem in a different way. We
used this identity to solve some trig equations, but it has other uses. A
simple example would be to simplify an expression like this one:
3 sin2 x + 3 cos2 x = 3(sin2 x + cos2 x) = 3.
Often it means seeing that a cos2 term can be written as 1−sin2 , and similarly
with sin2 . For example
sin2 x 1 − cos2 x (1 + cos x)(1 − cos x)

= = = 1 + cos x
1 − cos x 1 − cos x 1 − cos x
In this case, by also using the difference of two squared identity as well, we
have been able to rewrite the expression without fractions - which is always
a good idea.
Here are some more examples. See if you can simplify them. Solutions are
over the page.
p
(a) 1 − sin2 x
(b) 4 − (2 cos x)2
(c) sin3 x cos x + cos3 x sin x
cos2 x
(d)
1−sin x
cos2 x− sin2 x
(e)
(cos x − sin x)2

Solutions.
p p
(a) 1 − sin2 x = cos2 x = sin x
(b) 4 − (2 cos x)2 = 4(1 − cos2 x) = 4 sin2 x
(c) sin3 x cos x + cos3 x sin x = sin x cos x(sin2 x + cos2 x) = sin x cos x
cos2 x (1+sin x)(1−sin x)
(d) 1−sin x
= 1−sin x
= 1 + sin x
cos2 x− sin2 x (cos x + sin x)(cos x − sin x) cos x + sin x
(e) (cos x − sin x)2
= (cos x − sin x)(cos x − sin x)
= cos x − sin x
1 + tan x
= 1 − tan x
In part (e) we get the alternative version by dividing (every term in) the
numerator and the denominator by cos x. Neither version is better than the
other; which one we choose would merely depend on the context.

Proving new trig identities

Sometimes you will be asked to show that two expressions are identical (i.e.
they are different ways of writing the same thing).
It is nornally easiest to start with the expression on the left hand side and
show that it can be turned into the one on the right. However, it is fine to
work from right to left if that looks easier to you—or even work from both
sides and meet in the middle.
In practice, when working with identities we normally drop the ≡ sign and
write =. But remember these are not equations—they don’t have a solution:
they work for all values.
Example 23
Prove the following identities:

(i) (1 − sin x)2 + (1 − cos x)2 = 3 − 2(sin x + cos x)
q
sin2 θ
(ii) = tan θ
1−sin2 θ
1
(iii) − cos θ = sin θ tan θ
cos θ
(iv) cos4 θ − sin4 θ = cos2 θ − sin2 θ
1 1
(v) tan θ+ =
tan θ sin θ cos θ
Solutions
(i) Expanding the l.h.s. gives 1−2 sin θ+sin2 θ+1−2 cos θ+cos2 θ. Then collect
terms.
sin2 θ
(ii) Under the square root we have = tan2 θ.
cos2 θ
(iii) Putting both terms overt the same denominator we have
1−cos2 θ sin2 θ sin θ sin θ
= = ×
cos θ cos θ 1 cos θ
(iv) cos4 θ = (cos2 θ)2 . Doing the same with sin4 θ we have
(cos2 θ)2 − (sin2 θ)2 = (cos2 θ + sin2 θ)(cos2 θ − sin2 θ)
by the difference of two squares27
sin θ 1 cos θ
(v) tan θ = , so = , and we have
cos θ tan θ sin θ
sin θ cos θ sin2 θ cos2 θ sin2 θ+cos2 θ
+ = + =
cos θ sin θ sin θ cos θ sin θ cos θ sin θ cos θ
27 2
a − b 2 = (a + b)(a − b) where a = cos2 θ and b = sin2 θ.

QQ
Prove these identities.
(a) (1 + cos x)2 + sin2 x ≡ 2(1 + cos x)

1
(b) −1 ≡ tan2 x
cos2 x
p
(c) tan x 1 − sin2 x ≡ sin x
sin2 x 1+cos x
(d) ≡ .
(1−cos x)2 1−cos x
(e) (1 + 2 cos x)2 + (1 + 2 sin x)2 ≡ 6 + 4(cos x + sin x).
(f) sin3 x + cos x sin2 x + sin x cos2 x + cos3 x ≡ cos x + sin x
(g) 1 − cos2 (90◦ − x) ≡ cos2 x.
q
sin x
(h) 1 ≡ tan x
sin x −sin x
28
(a) Expand and collect terms.
1 1−cos2 x sin2 x
(b) −1 = =
cos2 x cos2 x cos2 x
sin x
(c) l.h.s is × cos x
cos x
(d) sin2 x = 1 − cos2 x = 1 − (cos x)2 = (1 + cos x)(1 − cos x)
(e) Expand and collect terms.
(f ) l.h.s is sin2 x(sin x + cos x) + cos2 x(sin x + cos x) = (sin x + cos x)(sin2 x + cos2 x)
(g) cos( π2 − x) = cos(90◦ − x) = sin x
1 1−sin2 x cos2 x sin x sin x

(h) − sin x = = . So under the square root we have ×
sin x sin x sin x 1 cos2 x

12.12 Other types of trig problem

We try to use a calculator only when necessary. For example suppose we are
5
told sin θ = and are asked to find tan θ.
13
A calculator tells us that it’s about 0.4167. Drawing right angled triangle
containing the angle θ will give us the exact value.
By Pythagoras
x 2 = 132 − 52
13
= 144
5
θ x = 12
x 5
tan θ =
12
We could have approached it a different way. We could have used the

identity connecting sin2 θ and cos2 θ to find cos θ and worked out tan θ from
that.
µ ¶2
2 5
cos θ = 1 −
13
132 − 52
=
132
144
=
169
12
cos θ =
13
sin θ 5 13
tan θ = = ×
cos θ 13 12
5
=
12
I think you’ll agree that drawing a triangle is easier!
But we have glossed over something: in the working above we have taken the
positive square root only when finding cos θ. And in the triangle method we
have assumed θ is an acute angle. It need not be.
We’ll explore these issues further, and one or two other applications, using
some examples.

Example 22
sin θ = 45 , where θ is obtuse.

Find the exact values of cos θ and tan θ,
Solution
Since θ is obtuse29 we can’t draw a triangle containing θ. However we can

draw one containing the principal angle associated with θ.
Let’s call this angle θ 0 .
θ 0 sits in a 3 − 4 − 5 triangle (no need to use Pythagoras).
5 3
cos θ 0 =
4 5
3
θ0 tan θ 0 =
3 4
But the cosine and the tangent of an obtuse angle are both negative, so
cos θ = − 35 , tan θ = − 34 .
Example 23
tan α = − 23 , where .
1
Show that cos α + sin α = p ,
13
Solution
Since α > 90◦ we can’t draw a triangle containing α.

So we use α0 , the principal angle associated with α.
p
Pythagoas tells us that the hypoteneuse is 13.
p
0 3 13
cos α = p
13 2
2 α0
sin α0 = p
13 3
Since tan α < 0, α must be between 270◦ and 360◦ .

In this quadrant cos α > 0 and sin α < 0, so
3 2 1
cos α + sin α = + p −p =p
13 13 13
29
An obtuse angle is one that lies between 90◦ and 180◦ .
A reflex angle is one that lies between 180◦ and 360◦ .

Example 24
Given that sin x = p and cos x = 2p, where p > 0 and 0◦ ≤ x ≤ 360◦ ,
Find the exact value of p, and the possible values of x correct to 3 significant
figures.
Solution
1
p 2 + (2p)2 = 5p 2 = 1, so p = p .
5
1
sin x = p
5
x = 26.565◦ or 180◦ − 26.565◦ = 153.435◦
= 26.6◦ or 153.4◦
Example 25
If k sin θ = 7 and k cos θ = 24, find the possible values of k.
Solution
7 24 72 242
sin θ = and cos θ = , so 2 + 2 = 1
k k k k
k 2 = 49 + 576 = 625
k = ± 25
Example 26
Find x and y, if p
3
sin(2x + y) =
2
1
and cos 3y = p ,
2
where x and y are both acute angles.
Solution
2x + y = 60◦
3y = 45◦
so y = 15◦
and 2x = 60◦ − 15◦ = 45◦
x = 22.5◦

Example 27
In this question x is restricted to the values 0◦ ≤ x ≤ 360◦ .

(a) Sketch the graph of y = cos x.
(b) Solve the equation cos x = 0.3.
(c) Hence solve the inequality cos x < 0.3.
Solution
(a) y = cos x
6
1
0.3
- x
0◦ 90◦ 180◦ 270◦ 360◦
−1
(b) cos x = 0.3

x = 72.5◦ or 360◦ − 72.5◦ = 287.5◦
(c) From the sketch cos x < 0.3 when 72.5◦ < x < 287.5◦

QQ
1. In each case find the exact value (do not use a calculator).
4
(a) tan x = 3 and x is acute. Find cos x.
2
(b) sin x = 5 and x is obtuse. Find tan x.
7
(c) cos x = − 25 and x is obtuse. Find sin x.
1
(d) cos x = p and x is reflex. Find tan x.
17
2. If 2 cos θ = x and 3 sin θ = 2x, find x.

3. If sin α = 3a and cos α = 2a, find a and the value of tan α.
1
4. Find x and y, if sin(x + 2y) = 1 and cos 3x = 2 , where x and y are both
acute.
p 1
5. If tan(α + β) = 3 and sin(2α − β) = p , find α and β, where α and β are
2
both acute angles. Give your answers in degrees.
6. (i) Sketch the curve y = sin x for 0◦ ≤ x ≤ 360◦ .
For values of x in this range
(ii) Write down the solutions of the equation 2 sin x = 1.
(ii) Solve the inequality 2 sin x ≤ 1.
30 3 2 24
1. (a) (b) − p (c) (d) −4
5 21 25
6
2. x = ±
5
1 2 sin α 2a
3. a = ± p tan α = [By Pythagoras, or by dividing: tan α = = ]
13 3 cos α 3a
4. x = 20◦ , y = 35◦ [x + 2y = 90◦ , 3x = 60◦ ]
5. α = 35◦ , β = 25◦ [α + β = 60◦ , 2α − β = 45◦ ]

1
6. (ii) x = 30◦ , 150◦ [sin x = ] (iii) 0◦ ≤ x ≤ 30◦ and 150◦ ≤ x ≤ 360◦
2

Exercises 45
1. Without using a calculator, write down the exact value of each of these:
(i) cos 60◦ (ii) sin 150◦ (iii) tan(−135◦ )
(iv) cos 210◦ (v) sin 810◦ (vi) cos 3600◦
2. Solve the following equations, where 0 ≤ x ≤ 360◦ .

p
(i) sin x = 12 (ii) cos x = p1 (iii) tan x = − 3 (iv) sin x = 0
2
(v) tan x = −1 (vi) cos x = − 12
3. Show that all these expressions are identical

p
sin θ cos2 θ, tan θ cos3 θ, sin θ − sin3 θ, cos4 θ − cos6 θ
4. Solve these equations for x, where 0◦ ≤ x ≤ 360◦ .

p
(i) 2 cos x + 3 = 0
(ii) cos x tan x = 1
(iii) cos x − 2 cos2 x = 0
(iv) 2 sin2 x = 2 − cos x
(v) 2 sin x = 5 cos x
(vi) tan x = 2 sin x
2 sin x
(vii) = cos x
tan x+1
sin x+2 cos x
(viii) =1
2 sin x−cos x
(ix) 4 sin2 x = 1
5. Solve these equations for x, where 0◦ ≤ x ≤ 360◦ .

(a) 2 cos2 x + 5 cos x − 3 = 0
(b) 1 + sin x = 2 cos2 x
(c) tan x sin x = 1
4
(d) sin x = +3
sin x
(e) tan x = 2 cos x
1
(f) 3 cos x = 2 +
cos x
(g) (2 sin x + 1)(sin x + 1) = 3
2
(h) cos x + 1 =
4−3 cos x
6. Find the exact solutions of these equations, for 0◦ ≤ x ≤ 360◦ .
(a) cos4 x − sin4 x = 2 sin2 x.
cos x
(b) tan x = p .
2−sin x
1 1
(c) + = 8.
1+sin x 1−sin x
7. Given that 0◦ ≤ x ≤ 90◦ and 0◦ ≤ y ≤ 90◦ , solve the equations
tan(y + x) = 1
p
3
sin 2x =
2

8. Given that 0◦ ≤ x ≤ 90◦ and 0◦ ≤ y ≤ 90◦ , solve the equations

1
cos(x − 2y) = p
2
x + y = 90◦
p
9. Show that x = 2 − 1 is a solution of the equation
2x
= 1.
1 − x2
Hence, or otherwise, find a value of θ for which
2 tan θ
= 1.
1 − tan2 θ
10. Solve the equation (1 + sin θ)2 = 4 cos2 θ, where −180◦ ≤ θ ≤ 180◦ .
11. The function f (θ) = 2 cos2 θ − sin2 θ − sin θ.

Find the values of a, b and c such that
f (θ) = a sin2 θ + b sin θ + c.
Hence solve the equation f (θ) = 2, for 0◦ ≤ θ ≤ 360◦ , giving your answer
to the nearest 0.1◦ .
12. (a) Solve these equations for 0◦ ≤ x ≤ 360◦ .

(i) tan2 x = 3.
(ii) cos x − π6 = 12 .
¡ ¢
(iii) sin 2x = 1.
¢ p
(iv) tan x + π4 = 3.
¡
(b) Write down your answers to part (a) if the solutions are to be in the
range −180◦ ≤ x ≤ 180◦ .
13. In each case find the exact value (do not use a calculator).
3
(a) tan x = 4 and x is acute. Find sin x.
2
(b) sin x = 5 and x is obtuse. Find tan x.
12
(c) cos x = − 13 and x is obtuse. Find sin x.
1
(d) cos x = − p and x is reflex. Find tan x.
5
q
3
(e) sin x = 7
and x is acute. Find cos x.
p
a 4−a 2
14. If cos x = 2 , and x is acute, draw a triange to show that tan α = a
.
1
15. If tan α = p, and α is obtuse, show that cos α = − p .
1+p 2
16. If 3 sin θ = 2a and 2 cos θ = a, find a.

p
17. If α is acute, cos α = k and sin α = 3 k, find k and the value of α.
p p
18. (a) Show that (4 − 2)2 = 2(9 − 4 2).
p p
4+ 2 4− 2
(b) The acute angle x is such that cos x = 6 . Show that sin x = 6 .

19. Find x and y such that p

4 cos x + 2 3 sin y = −1
p
and cos x − 3 sin y = 2
where 0 ≤ x, y ≤ 360◦ .
20. Find the maximum and minimum values of these functions of θ.

(i) f (θ) = 2 + 3 cos θ
(ii) f (θ) = 1 − sin θ
(iii) f (θ) = (2 cos θ + 1)2
1
(iv) f (θ) =
2+sin θ
(v) f (θ) = 5 − 3 cos(θ + 30◦ )
4
(vi) f (θ) =
7−3 sin 2θ
21. Show that for all x

2 sin x 6
= 2− .
3 + sin x 3 + sin x
2 sin x
Hence, or otherwise, show that the minimum value of is −1.
3+sin x
Find the maximum value.
22. The disgram shows a square ABC D of side a resting on a horizontal

plane and rotating about the vertex A.
Show that P B 2 +QD 2 is a constant.

C
Show also that the gradient of the
line
DB is
tan θ − 1
B
D tan θ + 1
a
Hence, or otherwise, find the value of
1
θ θ for which the gradient of B D is 4 .
Q A P
2t
23. If 0 < t < 1, θ is acute and tan θ = , find cos θ in terms of t .
1−t 2
What is the value of θ for which t = 1?
24. The diagram shows a circle

with centre O and radius a. Q
is a point on the circumference
and OP = 2a. The line OQ Q
makes an acute angle θ with
θ
the line OP . P a a O
(a) Show that
PQ 2 = a 2 (5 + 4 cos θ).
p
(b) For what values of θ is PQ = 7a?
(c) Explain why a ≤ PQ ≤ 3a.

25. The diagram shows an isosceles triangle

C
ABC in which AC = BC = x. AB is also
the diameter of a semicircle.
(a) If C ÂB = α, find the area of the
x
triangle ABC in terms of x and α.
(b) Show that the area of the triangle
α ABC and the area of the semicircle are
A O B
equal to if
π
tan α = .
2
26. If x 2 + 2x + k = 0, where k ≤ 1, find x in terms of k.

Show that the equation
sin2 θ + 2 sin θ + k = 0
has solutions provided −3 ≤ k ≤ 1.

Solutions 45
p
1 1 3
1. (i) 2 (ii) 2 (iii) −1 (iv) − 2 (v) +1 (vi) 0
2. (i) 30◦ , 150◦ (ii) 45◦ , 225◦ (iii) 120◦ , 300◦ (iv) 0, 180◦ , 360◦
(v) 135◦ , 315◦ (vi) 120◦ , 240◦
sin θ
3. tan θ cos3 θ = cos θ cos3 θ = sin θ cos2 θ; sin θ(1 − sin2 θ) = sin θ cos2 θ;
p p p
cos4 θ(1 − cos2 θ) = cos4 θ 1 − cos2 θ = cos2 θ sin θ
4. (i) 150◦ , 210◦ (ii) 90◦ [sin x = 1]

(iii) 60◦ , 90◦ , 270◦ , 300◦ [cos x = 0 or 12 ] (iv) see (iii)
(v) 68.2◦ , 248.2◦ [tan x = 25 ]
(vi) 0◦ , 60◦ , 180◦ , 300◦ , 360◦ [sin x = 0 or cos x = 12 ]
(vii) 45◦ , 225◦ [2 sin x = sin x + cos x ⇒ tan x = 1]
(viii) 71.6◦ , 251.6◦ [3 cos x = sin x ⇒ tan x = 3]
(ix) 30◦ , 150◦ , 210◦ , 330◦ [(2 sin x + 1)(2 sin x − 1) = 0 ⇒ sin x = ± 12 ]
5. (a) 60◦ , 300◦ [(2 cos x − 1)(cos x + 3) = 0 ⇒ cos x = 12 ]

(b) 90◦ , 210◦ , 330◦ [(2 sin x + 1)(sin x − 1) = 0 ⇒ sin x = − 12 or 1]
p
5
(c) 51.8◦ , 308.2◦ [cos2 x + cos x + 1 =
p
0 ⇒ cos x = −1+2
5
ignore cos x = −1−2 < −1]
(d) 270◦ [(sin x − 4)(sin x + 1) = 0 ⇒ sin x = −1]
p
17
(e) 51.3◦ , 128.7◦ [2 sin2 x + sin x − 2p= 0 ⇒ sin x = −1+4
17
ignore sin x = −1−4 < −1]
(f) 0◦ , 109.5◦ , 250.5◦ , 360◦ [(3 cos x + 1)(cos x − 1) = 0 ⇒ cos x = − 13 or 1]
(g) 30◦ , 150◦ [(2 sin x − 1)(sin x + 2) = 0 ⇒ sin x = 12 ]
(h) 0◦ , 131.8◦ , 228.2◦ , 360◦ [(3 cos x + 2)(cos x − 1) = 0 ⇒ cos x = − 23 or 1]
6. (a) 30◦ , 150◦ , 210◦ , 330◦ [(cos2 x + sin2 x)(cos2 x − sin2 x) = 2 sin2 x
⇒ cos2 x = 3 sin2 x ⇒ tan x = ± p1 ]
3
p 2
(b) 45◦ , 135◦ [sin x( 2 − sin x) = cos x
p
⇒ 2 sin x = 1 ⇒ sin x = p1 ]
2
2
◦ ◦
(c) 60 , 120 , 240 , 300 ◦ ◦
[ = 8 ⇒ 4 cos2 x = 1 ⇒ cos x = ± 12 ]
1−sin2 x
7. x = 30◦ , y = 15◦ [x + y = 45◦ , 2x = 60◦ ]
8. x = 75◦ , y = 15◦ [x − 2y = 45◦ , x + y = 90◦ ]

p p p
9. ( 2 − 1)2 = 3 − 2 2, so denominator = −2 + 2 2 = numerator.
π p
x = 22.5◦ = 8 radians [tan x = 2 − 1]
10. −90◦ , 36.9◦ , 143.1◦ [(5 sin θ − 3)(sin θ + 1) = 0 ⇒ sin θ = 35 or −1]
11. a = −3, b = −1, c = 2

θ = 0◦ , 180◦ , 199.5◦ , 340.5◦ , 360◦ [sin θ(3 sin θ +1) = 0 ⇒ sin θ = 0 or − 13 ]

12. (a) (i) 60◦ , 120◦ , 240◦ , 300◦ (b) (i) ± 60◦ , ± 120◦
p
[tan x = ± 3]
(ii) 90◦ , 330◦ (ii) 90◦ , − 30◦

[y = x − 30◦ and cos y = 21 , where −30◦ ≤ y ≤ 360◦ − 30◦
⇒ y = 60◦ , 360◦ − 60◦ ]
(iii) 45◦ , 225◦ (iii) 45◦ , − 135◦

[y = 2x and sin y = 1, where 0◦ ≤ y ≤ 720◦
⇒ y = 90◦ , 360◦ + 90◦ ]
(iv) 15◦ , 195◦ (iv) 15◦ , − 165◦

p
[y = x + 45◦ and tan y = 3, where 45◦ ≤ y ≤ 360◦ + 45◦
⇒ y = 60◦ , 180◦ + 60◦ ]
13. Draw a triangle containing the angle x or, if x is not acute, containing
the principal value associated with it x 0 . Q indicates which quadrant
the ange x lies in.
3
(a) sin x = 5 [x in Q1]
2 2
(b) tan x = − p [tan x 0 = p and x is in Q2, where tan x < 0]
21 21
5 5
(c) sin x = 13 [sin x 0 = 13 and x is in Q2, where sin x > 0]
(d) tan x = 2 [tan x 0 = 2 and x is in Q3, where tan x > 0]
p
2 3
(e) cos x = p [ x in Q1 and sin x = p ]
7 7
14. Pythagoras with side a and hypotenuse 2.
15. Triangle for α0 has side p and hypotenuse 1. In Q2 cos α < 0.

2 2
6 25a 2
16. a = 5 [ 4a9 + a4 = 1 ⇒ 36
= 1]
1
17. k = 2 , α = 60◦ [α is acute so k 6= − 12 since. Could find α first by
dividing p
sin α 3k
p
the equations: cos α = k ⇒ tan α = 3]
p p p
18. (a) (4 − 2)2 = 16 − 8 2 + 2 = 2(9 − 4 2).
p p p
By Pythagoras a 2 + (4 + 2)2 = 36 ⇒ a 2 = 18 − 8 2 = (4 − 2)2
19. x = 60◦ or 300◦ , y = 210◦ or 330◦

p
3
[Let cos x = a and sin y = b. Solve for a and b: a = 12 , b = − 2 ]
20. (i) [−1, 5] (ii) [0, 2] (iii)[1, 9] (iv) [ 13 , 1] (v) [2, 8] (vi) [ 25 , 1]
3 6 6 1 1
21. 2 ≤ 3+sin x ≤ 3 ⇒ −1 ≤ 2− 3+sin x ≤ 2 . Maximum is 2 .
22. P B 2 +QD 2 = a 2 sin2 θ + a 2 cos2 θ = a 2

P B −QD a sin θ−a cos θ
Gradient = Q A+AP = a sin θ+a cos θ .
Cancel the a’s, divide (every term) top and bottom by cos x.
θ = 59.0◦ [4(tan θ − 1) = tan θ + 1 ⇒ tan θ = 53 ]

23. x 2 = (1 − t 2 )2 + (2t )2
= 1 − 2t 2 + t 4 + 4t 2
x
= t 4 + 2t 2 + 1
2t
= (t 2 + 1)2
θ so x = +(t 2 + 1) [x > 0]
1− t2 1− t2
cos θ =
1+ t2
When t = 1, tan θ = ∞, so θ = π2 . Notice that cos θ = 1+0
1−0
= 1 ⇒ θ = π2
24. (a) By Cosine Rule PQ 2 = OP 2 + OQ 2 − 2(OP )(OQ) cos P ÔQ

P ÔQ = 180◦ − θ, so cos P ÔQ = − cos θ. PQ 2 = 4a 2 + a 2 + 2(2a)(a) cos θ
(b) θ = 60◦ [7a 2 = 5a 2 + 4a 2 cos θ ⇒ cos θ = 12 ]
(c) By geometry or by trigonometry, max when θ = 0, min when θ = 180◦
25. (a) Area of triangle ABC = x 2 sin α cos α [base = 2x cos α, height
= x sin α]
(b) Area of semicircle = 21 π(x cos α)2 = 21 πx 2 cos2 α
sin α
So, sin α
cos
α = 12 π cos2 α [cos α 6= 0] ⇒ cos α
= 12 π
p p
26. x = −1 ± 1 − k, so sin θ = −1 ± 1 − k.
p
sin θ = −1 when k = 1. sin θ = +1 when 1 − k = 2, i.e. when k = −3.


• SOHCAHTOA. A reminder. Different ways of applying it.
• Two important trig relationships.

sin A
tan A = cos A , sin x = cos(90◦ − x), cos x = sin(90◦ − x)
• Re-defining trig functions for angles > 90◦ .
• Trig functions for any angle, including outside 0◦ < x < 360◦ .
• The sin, cos and tan of 0◦ and 90◦ .

sin 0◦ = cos 90◦ = 0, sin 90◦ = cos 0◦ = 1, tan 0◦ = 0, tan 90◦ = ∞
• The Sine, Cosine and Tan curves. Symmetries.

Using the graphs to find values for angles outside 0◦ ≤ x ≤ 90◦ .
• Sine Rule. Applications.

a b c sin A sin B sinC
= = or = =
sin A sin B sinC a b c
Cannot apply Sine Rule if our three data points are one from each
fraction. Need to apply Cosine Rule first.
• Cosine Rule.
a 2 = b 2 + c 2 − 2bc cos A,
b 2 = c 2 + a 2 − 2c a cos B ,
c 2 = b 2 + a 2 − 2ba cosC
cos X > 0 if X is acute, cos X < 0 if X is obtuse.
• More trig equations. Including using the curves to solve.

The important identity sin2 x + cos2 x ≡ 1. Examples:
2 sin x = 3 cos x, 3 cos2 = sin2 x, 2 sin2 x = cos x, 2 cos2 x − sin x − 1 = 0
p
Compound angles: sin 3x = 0.5, cos(x + 30◦ ) = 3
All solutions to be found for 0◦ ≤ x ≤ 360◦ or −180◦ ≤ x ≤ 180◦ .
• Area of a triangle formula. A = 12 ab sinC = 12 bc sin A = 21 c a sin B
• More on trig identities. Manipulation. Proving new identities.
• Other types of trig problem. Example:

If sin θ = 54 and θ is obtuse, find the exact values cos θ and tan θ.

Chapter 13
VECTORS
Overview
13.1 What are vectors? Vector quantity compared to a scalar quantity.
Scalar multiple of a vector. Parallel vectors. Triangle law of vector addition.
¡13.2
4
¢ Column 9
¢ For¡example
¡ 5 ¢ ¡vectors 2
¢ ¡ 6 ¢ ¡−10¢
+ = and 4 −3 =
1 −3 −2 −3 −5 3
13.3 i, j notation
(4 i + j) + (5 i - 3 j) = 9 i - 2 j. Find 2a - 3b if a = 2 i + 5 j and b = i - 3 j
¡a ¢ ¡ c ¢
Independence of components = ⇒ a = c and b = d .
b d
p
13.4 Magnitude of a vector |x i + y j| = x 2 + y 2
p p
Examples |3 i - 4 j| = 5, |7 i + 7 j| = 7 2, |λ i + 2λ j| = 5λ
13.5 Direction of a vector Examples

2 i + 3 j makes an angle θ above the direction of i, where tan θ = 32
-4 i - 4 j makes an angle 135◦ below the direction of i [135◦ = 180◦ − 45◦ ]
Components of a vector given its magnitude and direction.
13.6 Position vectors Fixing the starting point ¡ 2of¢a vector as the original
coordinates to specify a position, in which case ≡ 2 i - 3 j ≡ (2, −3)
−3
Using vectors to solve coordinate geometry problems.
13.7 Solving geometry problems using vectors For example

B Given the ratios BT : T A and AS : SO
find the ratio in which P divides OT and
T B S.
P
O S A
13.8 Modelling with vectors Applied to displacements (bearing problems),
forces, velocities and accelerations.
Bearing problems also involve the use of Sine Rule and Cosine Rule.
Combining forces to obtain a resultant force. The formula F = ma .
Formula connecting velocity, acceleration and time: v = u + at .
13-1
CHAPTER 13. VECTORS
13.1 What are vectors?

Vectors are a relatively new addition to the mathematical toolkit, being
invented in the late 19th century to express in a concise way the Laws of
Electromagnetism newly discovered by the brilliant Scottish Physicist James
Clerk Maxwell. Their usefulness should be come apparent to you during the
course of this chapter as you find that they offer an elegant alternative in
certain circumstances to coordinate geometry.
We start with some basic, beginning with the notion of ratios as applied to
distances. Suppose the point C divides line AB such that AC : C B = 5 : 2
A C B
BC AB
Write down the value of (i) AC : AB (ii) (iii)
AB AC
The answers are in the footnote 1 .
A vector is any quantity that has size or magnitude and direction. For
example a displacement or a translation, where we say not just how far but
in what direction. Anther example is a force, which is not fully described
unless we say not just how big it is but also the direction it acts in. A quantity
that only has size is called a scalar, for example the number a children in a
family.
We are very used to combining scalar quantities; here we are going to be
learning about combining vector quantities. To see why this is very different
we only have to think of combining two displacements (translations), such
as one of 4 km due North followed by one of 3 km due East. The result is not
a displacement of 7 km.
A diagram enables us to see how to combine them.
Q 3 R Pythagoras tells us that the magnitude

of the sum of the two vectors is 5,
its direction we can easily find too by
trigonometry.
4 5
This is called the triangle law of vector
addition.
To denote a vector books use either a bold letter, or a line segment with an
arrow over the top indicating the direction. Since you cannot write in a
boldface type the equivalent is a letter with a line underneath it.
Q b R a + b = c which you will write as a + b = c
Or
a c
−−→ −−→ −→
PQ + QR = P R
P
1 2 7
(i)5 : 7 (ii) (iii)
7 5

CHAPTER 13. VECTORS
Later we shall learn how vectors can be used to specify positions, but for now
assume that a vector floats freely: if it has the same magnitude and the same
direction it is the same vector.
Also by thinking of vectors as displacements we can see that a + a = 2a,
vector twice as long as a but in the same direction. In fact multiplying by
any scalar quantity will simply scale the vector by the same quantity.
The vector which has the same magnitude as a but whose direction is
−→ −→
reversed is −a . If a is AB in arrow notation, −a = B A.
B
2a
a − 12 a
a −a
A
−−→ −−→
In general if λ is any scalar, then the vectors PQ and λ PQ are parallel. It
follows that all these vectors are parallel
2a + 3b, 4a + 6b, -2a - 3b, a + 32 b
because each one is a scalar multiple of all the same vector,
e.g. a + 23 b = 12 (2a + 3b)
We can solve simple problems like this.

To find the value of µ if 6p + 5q is parallel to 9p + µq
9 3 3 15
The scalar multiple must be = , so µ = 5× = .
6 2 2 2
QQ
Which of these vectors is parallel to a - 2b:

(a) 2a - 4b (b) 14a + 7b (c) 2b - a (d) 2a - b (e) b - 12 a
2
(a) (c) and (e).

CHAPTER 13. VECTORS
Example 1
The diagram shows a parallelogram PQRS. The points X and Y lie at he

midpoints of QR and P S respectively.
S R
Y X
P Q
−−→ −−→
If PQ =a, and P Y =b, find in terms of a and b
−→ −→ −−→ −→
(a) P S (b) P R (c) Y Q (d) S X
Solution
−→
(a) P S = 2b
−→
(b) P R =a +2b
−−→ −−→ −−→
(c) Y Q = Y P + PQ =-b + a or a - b
−→ −−→
(d) S X = Y Q = a - b
Example 2
The diagram shows a triangle ABC . M is the midpoint of BC and T is the

point on AC such AT : T C = 2 : 1.
C
T•
•M
A B
−→ −→
If AB =a, and AC =b, find in terms of a and b
−→ −−→ −−→ −−→
(a) C B (b) T C (c) C M (d) T M
Solution
−→
(a) C B =-b + a = a - b
−−→ 1
(b) T C = 3b
−−→ −→
(c) C M = 12 C B = 12 (a - b)
−−→ −−→ −−→
(d) T M = T C + C M = 12 (a - b) + 13 b = 1
2
a- 1
6
b or 1
6
(3a - b)

CHAPTER 13. VECTORS
Example 3
The diagram shows a parallelogram O AC B , whose diagonals intersect at X .

−−→ −−→
The vector O A = a and the vector OB = b.
B C
b
X
O a A
Show that
−−→ −−→ 1
(i) OC = a + b, and (ii) OX = 2 (a + b).
Solution
(i) This follows very quickly from the triangle law of vector addition:
−−→ −−→ −→
OC = O A + AC = a + b.
We call this the parallelogram law of vector addition.
(ii) The diagonals of a parallel bisect each other, so
−−→
OX = 12 (a + b).
This last result tells us that
the position vector of the midpoint of a line segment is the

mean of the position vectors of its end points.

CHAPTER 13. VECTORS
The triangle law and the parallelogram law are effectively the same law
expressed in two different ways. If the two vectors we are adding begin at the
same point (as in the example above) then their sum is the diagonal of the
parallelogram formed by those two vectors.
The triangle law is the more natural one to use when the one vector begins
where the other one ends. Here are some examples. The sum of the two
vectors (sometimes called the resultant) is shown in blue.
b
a
a+b
a
a+b b
a
a+b
a a+b
b
b

CHAPTER 13. VECTORS
13.2 Column vectors

So far we have concentrated on vector principles and how we combine
vectors. Turning our attention to specific vectors, we generally define them
by giving the components in the x and y directions. For this we use two
alternative notations.
y
(i)
¡4¢
(ii) 2
¡−2¢
2 x
¡0¢
4
(iii)
Written with one number above the other, enclosed in a bracket, we call it a
column vector. The top number always corresponds to the (positive) x
direction and the bottom number to the (positive) y direction.
Write down the column vectors for (i), (ii) and (iii). The answers are in the
footnote3 .
Combining column vectors is very simple.

y
¡5 ¢
1 C
A
¡−2¢ ¡7 ¢
−6 7 x
−→ −→ −→ −2 7 5
¡ ¢ ¡¢ ¡¢
AB + BC = AC becomes + =
−6 7 1
−−→ −→ −→ −5 −2 −7
¡ ¢ ¡ ¢ ¡ ¢
Similarly C A + AB = C B becomes + =
−1 −6 −7
¡−2¢ ¡4¢
3
(iii) −4
¡ ¢
(i) 0 (ii) −2 −3 .

CHAPTER 13. VECTORS
Scalar multiples work in just the same way, but we must remember that it
applies to both components.
¡3¢ ¡−1¢
For example if a = and b =
−2 4
¡3¢ ¡3¢ ¡6¢ ¡3¢
2a = + = = 2
−2 −2 −4 −2
¡−1¢ ¡−3¢
3b = 3 =
4 12
¡ 12 ¢ ¡−3¢ ¡ 15 ¢ ¡3¢
4a - 3b = − = = 5
−8 12 −20 −4
QQ
¡4¢ ¡2¢ ¡5¢

if p = ,q= and r = , find as a single vector
−1 3 −2
(a) 3p (b) 2q + r (c) 2q - r (d) 21 (p + q) (e) 4q - 2r + p
¡ 12 ¢ ¡9¢ ¡−1¢ ¡3¢ ¡2¢

4
(a) (b) (c) (d) (e)
−3 4 8 1 15

CHAPTER 13. VECTORS
13.3 i, j notation
As an alternative to the column notation¡we ¢ can think of each vector as¡the
1 0
¢
sum of two vectors, one a multiple of i = , the other a multiple of j = .
0 1
y
(i)
j
¡4¢
=4 i+2j
(ii) i 2
j ¡−2¢
i =-2 i+2j
2 x
¡0¢
=4j
4
(iii)
Write down the vectors for (i), (ii) and (iii) in i and j form. The answers are in
the footnote5 .
Combining vectors and scalar multiplication mirrors exactly the treatment

of column vectors. We just need to remember that an i component can
never be combined with a j component; one represents what is happening
in the x direction, the other what is happening in the y direction.
For example if a = 2 i + 3 j and b = i - 5 j
a + b = (2 i + 3 j) + ( i - 5 j) = 3 i - 2 j
1 1
2 a = 2 (2 i + 3 j) = i + 32 j
a - 2b = (2 i + 3 j) - 2( i - 5 j) = 13 j
QQ
If a = 3 i + j, b = 5 i - 2 j and c = 21 ( i + 3 j), find in terms of i and j

(i) a + b (ii) a - b (iii) 2a - 3b (iv) a - c (v) b + 34 c (vi) 2(a + c)
5
(i) -2 i (ii) 4 i - 2 j (iii) -4 i - 3 j
6 5 1 17
(i) 8 i - j (ii) -2 i - 3 j (iii) -9 i + 8 j (iv) 2 i- 2 j (v) 3 i (vi) 7 i + 5 j

CHAPTER 13. VECTORS
Independence of components
The independence of the x and y components is very important: it means
that
¡a ¢ ¡ c ¢
= ⇒ a = c and b = d .
b d
a i + b j = c i + d j ⇒ a = c and b = d .
This looks odd because in each case it appears that we have a single
equation, and somehow we are turning it into two equations. It makes sense
though because the single equation we started with is a vector equation,
which contains information about the x direction and the y direction. It is
really two relationships combined into a single statement.
For example, suppose
¡6¢ ¡ µ ¢
= ⇒ µ = 6 and λ = −1.
λ −1
Or in i, j notation,
8 i + (λ + µ) j = 2µ i + 5 j
yields 2 equations
8 = 2µ
λ+µ = 5
⇒ µ = 4 and λ = 1.
We make further use of this principle of independence in the following

example.
Example 4
(i) a = 2 i + r j, b = 6 i + 12 j and c = s i + t j
(a) Find r if a is parallel to b (b) Find s and t if b = 2c
(ii) (α − 2) i + (α + 1) j = β(- i + 2 j). Find the possible values for α and β.
2
Solution
(i) (a) From the x component we can see that the scalar multiplier is 3,
i.e. b = 3a, so r = 4
(b) 6 i + 12 j = 2(s i + t j), so s = 3 and t = 6

(ii) α − 2 = −β and α2 + 1 = 2β
Substituting β = 2 − α gives α2 + 2α − 3 = 0, whose roots are α = 1 and α = −3.
So α = 1 and β = 1 or α = −3 and β = 5

CHAPTER 13. VECTORS
13.4 Magnitude of a vector
x
a
3 r
y
θ φ
O O
The magnitude or size of a vector we can easily find using Pythagoras.

p
The notation is: magnitude of a is written |a| = 42 + 32 = 5
It is sometimes useful to know what the unit vector is in a particular
direction (i.e. the vector with magnitude = 1)
¡4¢
The vector a is 3 and this has magnitude 5, so the unit vector parallel to a is
1 4
¡ ¢ ¡0.8¢ 4 3
5 3
= 0.6 , or 5 i + 5 j
p
In general, if r = x i + y j, then |r| = x 2 + y 2 and the unit vector parallel to r is
p 12 2 xy .
¡ ¢
x +y
Simple trigonometry gives us the angle the vector makes with the axes. We
normally state the angle it makes with the x axis. In the case of the vector a
3 y
tan θ = 4 = 36.87◦ and in general tan φ = x
Of course the angle won’t always be acute. It could be in any quadrant.
Some examples.
p
|6 i - 8 j| = 100 = 10
p p
|4 i + 6 j| = 52 = 2 13
p p
|k i + 2k j| = 5k 2 = 5k

CHAPTER 13. VECTORS
13.5 Direction of a vector

A vector has magnitude and direction. We describe its direction by
specifying the angle it makes with a known direction, generally the (positve)
x axis. You will remember that it is conventional to measure in the
anticlockwise direction. In the diagrams below that would be the angles θ
(theta) and φ (phi).
x
a
3 r
y
θ φ
O O
3 y
tan θ = , so θ = 36.87◦ tan φ =
4 x
Some examples of finding the magnitude and the angle each of these vectors
makes with the (positive) x direction.
Vector Magnitude Quadrant Direction

p
2i + j 5 1 arctan 0.5 = 26.57◦ above
p
2i - j 5 4 arctan 0.5 = 26.57◦ below
p
7i + 7j 7 2 1 arctan 1 = 45◦ above
p
-7 i + 7 j 7 2 2 180◦ − 45◦ = 135◦ above
p p
- i - 3j 2 3 180◦ − arctan 3 = 180◦ − 60◦ below
In the last example notice that the vector is in the 3rd quadrant. So we regard
it as below the positive
p x direction. The angle it makes with the x-axis is 60◦ ,
◦
because tan 60 = 3
Conversely, if we know the magnitude and direction of a vector we can find

its x and y components.
25◦
8 5
4.5
60◦
105◦
x = 8 cos 60 = 4 x = 5 cos 25 = 4.53 x = −4.5 cos 75 = −1.16

y = 8 sin 60 = 6.93 y = −5 sin 25 = −2.11 y = 4.5 sin 75 = 4.35

CHAPTER 13. VECTORS
QQ
1. a = 4 i + j, b = i - 2 j, c = - i + 4 j and d = k( i - j).
Find the magnitude of each fo these vectors:
(a) a| (ii) |b| (iii) |23b| (iv) |4a - c| (v) |d|
(b) Find the direction of each of these vectors:
(i) a (ii) b (iii) 23b (iv) - 3a (v) d
2. Find the x and y components of these vectors.
p
Where appropriate leave
◦ 3
p
your answer in surd form (e.g. 4 cos 30 = 4 × 2 = 2 3). “Above" and “below"
refer to the positive x direction.
Magnitude Direction
(i) 4 30◦ above
(ii) 8 30◦ below
(iii) 6 70◦ below
(iv) 10 120◦ above
(v) 10 120◦ below
In the following example, and further on in this chapter, we will be using the
Sine Rule and the Cosine Rule. If you need reminders about either of these you
will find them in sections 7 and 8 of Chapter 12.
7
p p p p
1. (a) (i) 17 (ii) 5 (iii) 23 5 (iv) 17 (v) k 2
(b) (i) 14.0◦ above (ii) 63.43◦ below (iii) 63.43◦ below (iv) 166.0◦ below (v) 45◦ below
p p p p
2. (i) 2 3, 2 (ii) 4 3, −4 (iii) 2.05, −5.64 (iv) −5, 5 3 (v) −5, −5 3

CHAPTER 13. VECTORS
Example 5
−−→ −−→
In the triangle OAB the vector O A = i + j and OB = 2 i - 3 j.
(a) Find the angle each of these vectors makes with x direction.
(b) Hence find the angle AOBb and area of the triangle.
(c) Find also the length of AB and the angle O AB
b .
Solution
(a) OB makes and angle of 45◦ with
B the x direction (above the dotted
line).
i+ j
OA makes and angle of
O θ arctan 1.5 = 56.31◦
with the x direction (below).
(b) Hence θ = 101.31◦ = 101.3◦ (3 SF)
2i - 3j −−→ p −−→ p
|O A| = 13, |OB | = 2
So theparea of
p the triangle
= 21 × 13 × 2 × sin 101.31◦
= 2.50 (3 SF)
(c) By the Cosine Rule

p p
AB 2 = 13 + 2 − 2( 13)( 2) cos 101.31◦ = 17.00
So AB = 4.1231 = 4.12 (3 SF)
p p ◦
2 4.1231 b = 2×sin 101.31 = 0.3363
By Sine Rule = ◦ ⇒ sin A
sin Ab sin 101.31 4.1231
◦ ◦
A = 19.654 = 19.7 (3 SF)
b

CHAPTER 13. VECTORS
13.6 Position Vectors

We shall use the column vector notation in the section. We could equally
well have used the i, j notation.
A vector has a magnitude and a direction; it does not have a specific position
or location. However we can use a vector to describe a position, for example
of the point P, by assuming that its starting point is the origin of coordinates.
y ¡3¢
2
¡3¢
¡3¢ 2
P
2
x
O
The position vector of point is therefore just another way of writing its
coordinates: P is the point (3, 2).
¡We
3
¢have three ways to define the position of the point P :
, 3 i + 2 j and (3, 2).
2
When choosing between the two vector forms, there is something to be said
for preferring the column vector, because the independence of the
components is more obvious than in the i , j notation. There is a rumour
that printers of text books prefer the i , j notation because it take up less
paper. You must judge for yourself.

CHAPTER 13. VECTORS
Example 6
−→ −2
¡ ¢
(a) The point A is at (4, 1) and AB = . Find the coordinates of B and
3
hence, or otherwise, find the angle O AB
b .
−−→ 4 −−→ −10
¡¢ ¡ ¢
(b) The point P is at (3, −3), PQ = and QR = .
5 2
Find the distance between R and the origin of coordinates.
Solution
(a) B −−→ −−→ −→
OB = O A + AB
¡4¢ ¡−2¢ ¡2¢
= + =
¡−2¢ 1 3 4
3 So the coordinates of B are (2, 4).
¡4¢ OB 2 = 20, so by Cosine Rule
p p
1 A 20 = 17 + 13 − 2 20 13 cos Ab
10
⇒ cos A = p p
O 2 20 13
Ab = 71.936◦ = 71.9◦ (3 SF)
Alternatively we could determine Ab by finding the angle that O A and AB

make with the x axis.
(b) Although it is usually a good idea to draw one, we don’t always need a
diagram for a vector question.
−−→ −−→ −−→ −−→
¡ 3 ¢ ¡4¢ ¡−10¢ ¡−3¢
OR = OP + PQ + QR = + + =
−3 5 2 4
Therefore OR = 5.

CHAPTER 13. VECTORS
Example 7
A is the point (4, 3) and B is the point (7, 2). Use vectors to find where the line
AB meets the coordinate axes. Find also where it meets the line 2y = x − 8.
Solution
−−→
We can specify every point on the line AB using a position vector OP . We
−−→ −−→ −→
can write OP = O A + k AB , where a different value of k gives¡a ¢different
¡4¢ point.
−→ 7
¡3¢
For example if k = 0 we have A, if k = 1 we have B etc. AB = − =
2 3 −1
−−→ 4 3 4+3k
¡ ¢ ¡ ¢ ¡ ¢
so OP = +k =
3 −1 3−k
Another way of thinking about this is to say that every point on the line has
the form8 x = 4 + 3k and y = 3 − k for some value of k. Taking an extreme
example, if k = 100 the point (304, −97) lies on the line.
•
k = −1 A
•
k =0 B
•
k =1 P
•
k =2
O
−−→ 4+3(3)
¡ ¢
The line AB will meets the x-axis where 3 − k = 0, i.e. where OP = .
0
It crosses the x-axis at (13, 0).
−−→
¡0¢
It meets the y-axis where k = − 3 so OP = 13 i.e. at (0, 13
4
3 ).
3
Now to find where the line crosses the line 2y = x − 8. Since every point on
the line through AB is of the form x = 4 + 3k, y = 3 − k we are looking for the
point with coordinates (4 + 3k, 3 − k) which satisfies the equation 2y = x − 8.
Substituting these coordinates in we have 2(3 − k) = 4 + 3k − 8 ⇒ k = 2.
The lines meet at (10, 1).
8
We call this a parametric equation of the line.

CHAPTER 13. VECTORS
13.7 Solving geometry problems using vectors

Vectors can sometimes simplify greatly a geometry problem. We shall be
making use of the principle that
¡a ¢ ¡ c ¢
= ⇒ a = c and b = d
b d
Or in i, j notation
a i + b j = c i + d j ⇒ a = c and b = d .
In the problems we will be learning about, the approach will be to find the
position vector of a point by arriving at it in two different ways. Each one will
require the introduction of a new variable. The two vectors must be equal
(they give the position of the same point). Equating them will give us 2
equations, which we can solve for the 2 variables.
Example 8
−−→ −→
PQRS is a parallelogram in which PQ = a and P S = b. M is the midpoint of
P S and MQ meets P R at X . Find the ratio in which X divides MQ and P R.
S R
M
X
P Q
Solution
We can reach X from P by going along P R, or by going via Q.
−−→ −→ −−→
We can therefore write P X = λP R, i.e. P X = λ(a + b).
−−→ −−→ −−→ −−→ −−→ µ
Also P X = PQ + Q X = PQ + µQM = a +µ(-a + 21 b) = (1 − µ) a + 2
b
−−→
Equating the two versions of P X we have
µ
λa + λb = (1 − µ) a + 2 b
µ
So λ = 1 − µ and λ =⇒ µ = 23 and λ = 13 .
2
−−→ −−→ −−→
Now we can answer the question. Q X = µQM = 32 QM .
M X : XQ = 1 : 2.
−−→ −→ −→
Similarly P X = λP R = 31 P R.
P X : X R = 1 : 2.

CHAPTER 13. VECTORS
Example 9
−−→ −−→
O AB is a triangle in which O A = a and OB = b. S is the point on O A such that
OS = 25 O A, and T is the midpoint of AB . The line OP meets AB at T .
Find the ratio in which P divides OT and B S.
T
P
O S A
Solution
We can reach P from O by going directly along OT , or by going via B or S.
−−→ −−→
We can therefore write OP = λOT = λ [b + 12 (-b + a)]= 12 λ(a + b).
−−→ −−→ −→ 2µ
Also OP = OB + B P = b + µ( - b + 25 a) = (1 − µ) b + 5 a.
−−→
Equating the two versions of OP we have
1 2µ
2
λa + 12 λb = 5
a + (1 − µ) b
So 5λ = 4µ and λ = 2 − 2µ ⇒ µ = 57 and λ = 47 .
−−→ −−→ −−→
Since OP = λOT = 74 OT .
OP : P T = 4 : 3.
−→ −→ −→
Similarly B P = µB S = 57 B S.
B P : P S = 5 : 2.

CHAPTER 13. VECTORS
13.8 Modelling with vectors

At the beginning of the chapter we saw an example of adding two
displacements, using the triangle law of vector addition. We use the same
principle to combine any kind of vectors. Some other examples are forces,
velocities and accelerations. That a force is a vector is fairly self-evident: the
direction in which a force is applied is very relevant to understanding the
effect it has.
The direction of a velocity is also very relevant of course, and when we use
the term “velocity" we mean to include the direction. When we just mean its
magnitude (how fast) we use the term “speed".
Example 10
Starting at a point O a girl walks 2 km due north to a point A, and then walks
1.5 km due East to point B . At the end of her journey how far is she from O?
What is the bearing of B from O?
What is the bearing of O from B ?
Solution
−−→ −−→ −→
OB = O A + AB
p p
Distance OB = 22 + 1.52 = 6.25 = 2.5
A 1.5 B
φ A bearing is always measured clockwise
starting from due north.
So the bearing of B from O is θ.
2 tan θ = 0.75, so θ = 36.87◦
The bearing of O from B will be
θ φ = 180◦ + 36.87◦ = 216.87◦ .
Example 11
When a force of F Newtons is applied to an object of mass m kg it causes it to
accelerate at a m s −2 , where F =ma.
(i) Find the magnitude of the force required on an object of mass 5 kg to
produce an acceleration of (3 i - 8 j) m s −2
(ii) Find the magnitude and direction of the acceleration of the same object
resulting from a force of (20 i + 80 j) Newtons?
Solution
(i) F = 5(3 i - 8 j)
p
|F | = 1825 = 42.72 Newtons.
(ii) (20 i + 80 j) = 5a , so a = 4 i + 16 j.
p p p
|a| = 42 + 162 = 4 1 + 16 = 4 17 m s −2 .
The acceleration is in the same direction as the force, i.e. at an angle of
arctan 4 = 76.0◦ with the direction of i (to 3 SF)9 .
9
The direction is arctan 16 80
4 = arctan 20 = arctan 4.

CHAPTER 13. VECTORS
Example 12
In a desert rally, a car travels from its starting point P to the first checkpoint
which is at Q, 25 km from P on a bearing of 040◦ . From Q it travels, on a
bearing of 105◦ for a further 15 km to R, where it breaks down. There is a
recovery team based at P .
(a) When the car breaks down, how far north and east of P is it?
(b) How far does the recovery team need to travel to reach the car?
(c) What bearing does the recovery truck need to follow?
Solution
N
10
Q
5
◦
15 k
75◦ m
R
km
25
40◦
α
P X
(a) The bearing of R from Q is 105◦ so the acute angle that QR makes with
due north is 75◦ .
−→ −−→ −−→
P R = PQ + QR = (25 sin 40◦ i + 25 cos 40◦ j) + (15 sin 75◦ i - 15 cos 75◦ j)
= 30.56 i + 15.27 j, i.e. 15.3 km north of P and 30.6 km east of P .
−→ p
(b) |P R| = 30.562 + 15.272 = 34.16 = 34.2 km (3 SF)
15.27
(c) tan α = ⇒ α = 26.55◦
30.56
So the bearing of R from P is 90◦ − 26.55◦ = 63.45◦ = 63.5◦ . (3 SF)

CHAPTER 13. VECTORS
QQ
1. In each case combine the two vectors and find the magnitude of the
resulting vector.
(i) Displacements 5 i + 2 j and i - 10 j, where the unions are metres
(ii) Velocities 4( i - j) and i + 2 j, where the units are m s−1
(iii) Forces 20 i + 7 j and -12 i + 8 j, where the units are Newtons
2. Find the magnitude and direction of the force required to give a particle of
mass 5 kg an acceleration of 0.4 i - 0.8 j ms −2 .
3. The relationship connecting a particle’s initial velocity u, its acceleration a
and its final velocity v after time t is v = u + at .
If the acceleration of a particle is 1.5 i + 3.5 j ms −2 and its final velocity after
travelling for 4 seconds is 14 i + 11 j ms −2 , what was its initial velocity.
10
p
1. (i) 6 i - 8 j, 10m (ii) 5 i - 2 j, 29 ms −1 (iii) 8 i + 15 j, 17 Newtons.
2. 4.47 Newtons at 63.4◦ below the direction of i.
3. 8 i - 3 j ms −2

CHAPTER 13. VECTORS
The centroid of a triangle

In a triangle a median is a line line joining a vertex to the midpoint of the
opposite side. The three medians meet at a single point, known as the
centroid of the triangle.
We know how to use vectors to find where lines meet and in what ratios they
divide each other. The centroid divides the medians in the ratio 2 : 1 (starting
at the vertex). We are going to show that this is the case in a beautifully
simple and elegant way.
With problems like this we often take one vertex to be at the origin to
simplify the working. Rather surprising we will find the exact opposite here:
we take the origin O to be some arbitrary point outside the triangle ABC
(shown in red). With respect to this origin the position vectors a, b and ct
define where the vertices are.
We want to show that the three medians meet at a point two-thirds of the
way down any one of the medians (starting at the vertex). We have labelled
that point X for the median that passes through the vertex A.
A
a
C
•X
c
•
O M
b
B
In Example 3 we saw that the position vector of a line segment is the mean of
the position vectors of the two ends of the segment.
−−→ −−→
Therefore OM = 12 (b + c), and AM = − a + 12 b + 12 c
−−→
It follows that the position vector of X , i.e. OX = a + 32 ( - a + 12 b + 12 c)
1
= 3
(a + b + c)
But this is symmetrical in a, b and c, so whichever vertex we had started with
we would have arrived at the same result. This point is therefore two thirds
of the way “down" each of the medians. In other words they all meet at this
point.
Because we can think of coordinates as position vectors, if we know the

coordinates of the the vertices of a triangle this allows us to write down the
coordinates of its centroid. For example in the triangle whose vertices are
(5, 4), (0, −2) and (7, 1) the position vector of the centroid is
"Ã ! Ã ! Ã !# Ã !
1 5 0 7 4
+ + =
3 4 −2 1 1
The centroid of the triangle is at (4, 1).

CHAPTER 13. VECTORS
Exercises 46
1. R In the diagram M is the

S
midpoint of SQ.
Express in terms of a, b and c
c
M the vectors
−−→ −→ −→
QM , SP and SR
b
Q
a
P
In the diagram M and M are Z
2.
the midpoints of XY and YZ
respectively.
b M
Show that MN = 12 ZX and
MN is parallel to ZX
N
X 4a Y
−−→ −−→ −−→
3. In the quadrilateral OABC, O A = p, OB = p + 2q and OC = 3q.
−→
Find C B in terms of a and b.
Show that AB is parallel to OC.
Write down the value of AB:OC.
−−→ −−→
4. In the triangle OAB O A = a and OB = b. The points P, Q and R lie on OA,
AB, and BO respectively, such that OP= 23 OA, AQ= 23 AB and BQ = 23 BO.
−−→ −−→ −→
Find in terms of a and b the vectors PQ, QR and RP
−−→ −−→ −→
Verify that PQ + QR + RP = 0, where 0 is the zero vector.
5. a = 3 i + 2 j, b = 4 i - j
(a) Find the unit vectors parallel to a and b.
(b) Find α and β if a + αb is parallel to i and βa + b is parallel to j.
6. c = 2 i + j, d = i + 3 j
If c + pd is parallel to i + j and c + qd is parallel to i - 3 j, find p and q.
Verify that these values are correct.
−→ −−→
7. In the parallelogram ABCD, AB = 16
¡ ¢ ¡6¢
0
and AD = 10
. The point T lies
on BD and BT:TD=3:2.
−→ −→
Find the vectors BT and AT .
Find the area of the triangle ATD.
D C
A B
¡p¢ ¡−q ¢ ¡−5¢
8. a = 2q , b = p and a + b = 4 . Find p and q.

CHAPTER 13. VECTORS
¡α+β¢ ¡ 3
¢ ¡7¢
9. a = 4
, b = 2α−β . Find α and β if 2a + b = 18
10. a = i - j, b = λ( i + 3 j). The resultant of a and b is parallel to i + 2 j.

FInd the value of λ.
¡α¢ ¡α+2¢
11. a = 2α , b = α2 and a + b is parallel to i + j.
Determine the possible values of α.
12. a = 2 i - j, b = λ i + µ j.
a + b is parallel to i - j, and a - b is parallel to i + j.
Find the values of λ and µ.
13. If |2 i + t j| find the possible values of t.
14. If c = λ2 and |c| = 4, find the possible values for λ.

¡ ¢
¡k¢
15. If a = k2 and b = −2 , find |a + b|2 in terms of k.
¡ ¢
p
Show that whatever the value of k, |a + b|2 ≥ 2 2.
¡ 2p ¢ ¡3q ¢ p
16. a = −q and b = p . Show that |a + b| = 5(p + q).
Find |a - b|2 in terms of p and q.
17. The diagram shows a regular

T S
hexagon PQRST.
Explain why the hexagon consists
O of 6 equilateral triangles.
U R
Find, in terms of a and b, the
−→ −−→ −→
vectors P S, PU and U S.
b
P a Q
−−→ 6 −→ 3
¡ ¢ ¡¢
18. The triangle PQR is defined by the vectors PQ = and P R = .
−2 5
(a) Find the angle Pb and the area of the triangle.
(b) Calculate the length of QR and find the angle R.b
19. |p i + q j| = 10 and the vector makes an angle θ with the x-axis, where
3
sin θ = 5 . Find the possible values of p i + q j.
20. In the
−−→
¡ 3quadrilateral
¢ −−→ 2
¡ OPQR,
¢ O is the origin and P is at (−1, 4).
PQ = and QR = . Find the coordinates of R.
−7 1
21. A(2, 1), B (1, 4) and C (7, 5) are three are vertices of a parallelogram. Find
the two possible positions of the fourth vertex.
22. In the parallelogram PQRS, M is the midpoint of PQ and N is the

−−→
midpoint of P S. The lines QN and R M meet at X . Let PQ =a and
−→ −−→
P S =b. By writing the vector P X in two different ways, find the ratios
in which X divides the lines QN and R M .
S R
N
X
P M Q

CHAPTER 13. VECTORS
23. The diagram shows three sides of a regular polygon. How many sides
does it have? Each side is of length 2.
A4
30◦
A3
α 30◦
A1 A2
−−−→ −−−→
Find the vectors A 2 A 3 and A 3 A 4 , leaving the values in surd form where
appropriate.
Hence find the distance A 1 A 4 and the exact angle α.
Write down the coordinates of A 5 and deduce the angle A 2 A c1 A 5 .
24. In the parallelogram ABC D, M is the midpoint of BC and D N = 31 DC .

−→ −−→
By letting AB = 3a and AD =2b, find the ratios in which P divides the
lines AN and D M .
D N C
P
M
A B
R
25. In the triangle PQR the
midpoint of P R is M and
M
PQ is extended such that
PQ = 2QS. M S meets QR at X .
−−→ −→ X
Let PQ = 2 a and P R = 2 b. P
Find the ratio that X divides
the line segments QR and M S. Q
S
26. After flying for 60 km, an aircraft alters its bearing by an angle α and flies
for a further 100 km. It is then 120 km from where it started.
Find the angle α.
27. In an orienteering exercise a soldier leaves the starting point O and

walks 12 km on a bearing of 060◦ to reach the first checkpoint at A.
From A she walks 10 km on a bearing of 200◦ to read the second
checkpoint at B . Find
(a) the position vector of A relative to O
−−→
(b) |OB |
(c) the bearing of B from O

CHAPTER 13. VECTORS
28. A ship starting at a port P travels to a point X which is 10 km due North

of P .
It then sails on a bearing 130◦ until it reaches a point Y , the bearing of
which from P is 070◦ .
Draw and label a diagram showing the movement of the ship.
(i) Find, correct to 2 DP, the distances X Y and P Y .
(ii) How far East of the port is the point Y ?
(iii) When the ship is at Y , what is the bearing of the port?
29. At 0900 hours a woman sets out from A, walking at a speed of 2km/h,
on a bearing of 120◦ . One hour later, a man set out from B , a point 5 km
due North of A. He walks on a bearing of 210◦ at a speed of 4km/h. At
1200 hours the woman is at C and the man is at D.
(a) Draw a diagram to show the positions of A, B , C and D, marking in
the lengths of AB , AC and B D.
(b) Find how far the man is from B at 1200 hours.
−−→ −→
(c) Find the vectors AD and AC .
(d) Hence, or otherwise, calculate the bearing of C from D.
30. If the velocity of a particle is given by v, find its speed in each case.
p
(a) v = 5 i + 12 j (b) v = 3 i - 3 j (c) v = 21 i + 23 j (d) v = 3k i + k j
31. (a) A particle travels with velocity 7 i + 24 j ms−1 for 8 seconds. How far
does it travel in that time?
(b) A second particle travelling with with a velocity 15 i + λ j travels the
same distance in 10 seconds. Find the possible values of λ.
32. How far does a particle travels in 3 seconds if its velocity is 6 i + 8 j ms−1 ?
If it then travels for a further 4 seconds with a velocity of 3 i - 2 j ms−1 .
How far is its final position from its original starting point?
v −u
33. The acceleration aof a particle is given by a = , where utravelling
t
is its initial velocity, and vuis its velocity at time t . Find the magnitude
of its acceleration if t = 3, v =13 i + j and u=10 i + 5 j.
34. Find the magnitude and direction of the resultant of adding the two
forces F 1 =4 i + 5 j Newtons and F 2 =-2 i + j Newtons.
If F 1 +kF 2 is parallel to i + j, find the value of k.
35. If a force F Newtons is applied to particle of mass m kg, its acceleration

is a ms −2 , where F = ma.
(a) Find the acceleration of particle of mass 5 kg when it is acted upon
by a force of 20 i - 10 j Newtons.
(b) If the particle stated at rest find its speed after 4 seconds.
36. Three forces (measured in Newtons) F 1 =12 i - 24 j, F 2 =36 i + 18 j and

F 3 =42 j act on a particle of mass 15 kg. Find the magnitude of the
resulting acceleration.
37. The three forces F 1 =3p i - 8 j, F 2 = -2q j and F 3 =2q i + p j are

equilibrium.
Find the values of p and q.

CHAPTER 13. VECTORS
Solutions 46
−−→ −→ −−→ −−→
1. QM = b - a, SP = a - 2b [SM + M P = (a - b) - b]
−→ −→ −−→ −−→
SR = 2(a - b) + d [SP + PQ + QR]
−−→
2. M Y = 12 (- b + 4a) = 2a - 12 b.
−−→ −−→
Y N = - 2a, so M N = - 12 b. Hence MN is parallel to XZ and half its length.
−→
3. From a diagram, C B = - 3q + p + 2q = p - q.
−→
AB = - p + p + 2q = 2q Hence AB is parallel to OC and AB:OC=2:3.
4.
−−→
B PQ = 13 a + 23 (b - a) = - 13 a + 32 b
Q −−→
QR = 13 (b - a) - 32 b = - 13 a - 13 b
−→
RP = − 31 b + 32 a
R A
−−→ −−→ −→
P Hence PQ + QR + RP = 0.
5. (a) p3 i+ p2 j, p4 i+ p1 j
13 13 17 17
3
(b) α = 2 [The y component = 2 − α = 0] β = − 4
6. p = 21 . c + pd = 52 i + 52 j
[The x and y components of c + pd must be equal, so 2 + p = 1 + 3p]
q = - 76 . c + qd = 65 i - 15
6
j
[The x and y components of c + qd must satisfy 1 + 3q = -3(2 + q)]
−→ ¡ ¡16¢ ¡ 6 ¢¢ ¡ ¢ −→
− 0 + 10 = −6 . AT = 16 + 6 = 10
3
¡ ¢ ¡−6¢ ¡ ¢
7. BT = 5 6 0 6
Area = 32 [Area 4 ABD = 12 16 × 10 = 80, 4 ABT = 12 16 × 6 = 48]
8. p = −2, q = 3 [p - q = -5, 2q + p = 4]
9. α = 4, β = − 2
2(α + β) + 3 = 7 and 4 + 2α − β = 18 ⇒ α + β = 2 and 2α − β = 10
10. λ = 3 [−1 + 3λ = 2(1 + λ)]
11. α = −1 or 2 [α + 2 = 2α − α2 ⇒ α2 + α − 2 = 0]
12. λ = −2, µ = −2. Because 2 + λ = 1 − µ and 2 − λ = −1 − µ
13. t = ±4 [4 + t 2 = 20]
p
14. λ = ±2 3 [λ2 + 4 = 16]
15. |a + b|2 = (2 + k)2 + (k − 2)2 = 8 + k 2 ≥ 8
16. |a + b|2 = (2p + 3q)2 + (q − p)2 = 5(p 2 + 2pq + q 2 ) = 5(p + q)2

|a - b|2 = (2p − 3q)2 + (q + p)2 = 5(p 2 − 2pq + 2q 2 )
−→ −−→ −−→ −−→ −→ −−→ −→
17. P S = 2b, PU = PO + OU = b - a, US = UT + TS = b + a

CHAPTER 13. VECTORS
18. ¡5¢ ¡1¢

R (a) φ = arctan 3 + arctan 3 =
77.47◦
¡3¢ PQ 2 = 40, P R 2 = 34
5 Area of the triangle
p p
= 12 × 40 × 34 × sin 77.47◦
φ
P = 18.0 (3 SF)
¡6¢
−2 Q
p p
(b) By the Cosine Rule RQ 2 = 40 + 34 − 2 40 34 cos 77.47◦ = 57.999
So RQ = 7.616, i.e. 7.62 (3 SF)
p p
40 7.616 40×sin 77.47◦
By Sine Rule = ⇒ sin R = = 0.8107
sin Rb sin 77.47◦
b
7.616
Rb = 54.160◦ = 54.2◦ (3 SF)
3
19. sin θ = 5 ⇒ θ = 36.9◦ or 180◦ − 36.9◦
So the vector lies in the 1st or 2nd quadrant.
In the 1st quadrant p = 10 cos θ = 10 × 54 = 8. Similarly q = 6.
Therefore p i + q j = 8 i + 6 j or - 8 i + 6 j
−−→ −1 3 2 4
¡ ¢ ¡ ¢ ¡¢ ¡ ¢
20. OR = + + = , so R is at (4, −2).
4 −7 1 −2
−−→ −→ 6
¡¢
21. The vertex could be at X where AX = BC = so X is (8, 2),
1
−−→ −→ −1
¡ ¢
Or it could be at Y where C Y = AB = so Y is (6, 8).
3
Y (6, 8)
C (7, 5)
B (1, 4)
X (8, 2)
A(2, 1)
O
22. The polygon has 12 sides.
[Each external angles is 30◦ so the no. of sides is 360 ÷ 30]
−−−→
¡p3¢ −−−→ ¡ 1 ¢
A2 A3 = , A3 A4 = p
1 3
p
[sides of right angle triangle with 30◦ and 60◦ are 2, 1 and 3]
−−−→ 2
¡ ¢ ¡p3¢ ¡ 1 ¢ ¡3+p3¢ ¡p3(1+p3)¢
So A 1 A 4 = + + p = p = p
0 1 3 1+ 3 1+ 3
p ¡p3¢
p ¡ 3¢ p
= (1 + 3) . Magnitude of is 2 so length A 1 A 4 = 2(1 + 3)
1 1
1
tan α = p , so α = 30◦
3
p p
c1 A 5 = 45◦
A 5 is (3 + 3, 3 + 3) so A 2 A

CHAPTER 13. VECTORS
−−→ −−→ −−→

23. Let P X = P M + λM R = 21 a + λ( 12 a + b) = 12 (1 + λ) a + λb
−−→ −−→ −−→
And P X = P N + µNQ = 12 b + µ(- 12 b + a) = µ a + 21 (1 − µ) b
So 21 (1 + λ) = µ and λ = 21 (1 − µ) ⇒ λ = 15 and µ = 35
M X : X R = 1 : 4 and N X : XQ = 3 : 2.
−→ −→
24. AP = λ AP = λ (2b + a).
−→
And AP = 3a + b + µ(b - 3a) = 3(1 − µ)a + (1 + µ)b
So λ = 3(1 − µ) and 2λ = (1 + µ) ⇒ λ = 67 and µ = 75 .
AP : P N = 6 : 1 and M P : P D = 5 : 2.
−−→ −−→ −−→
25. Let P X = PQ + λQR =2a + λ(-2a + 2b) = 2(1 − λ) a + 2λb
−−→ −→ −−→
And P X = P S + µSM = 3a + µ(-3a + b) = 3(1 − µ) a + µ b
So 2(1 − λ) = 3(1 − µ) and 2λ = µ ⇒ λ = 41 and µ = 12
Q X : X R = 1 : 3 and S X : X M = 1 : 1, i.e X is the midpoint of SM .
26. 62.7◦ , because the actual directions are not relevant. Use Cosine Rule
to find 180◦ − α
27. p
−−→
(a) O A = 6 3 i + 6 j
−−→
A 200◦ (b) OB
p
N = 6 3 i + 6 j - 10 sin 20 i - 10 cos 20 j
km
12 = 6.97 i - 3.40 j
m
−−→
Therefore |OB | = 7.76 km (3 SF)
10 k
60◦
O 3.40
(c) tan θ = 6.97 ⇒ θ = 26.0◦
θ
The bearing of B from O is 116.0◦
(d)
B
28. In the example there is little to be gained by using vectors. Sine Rule
and geometry is sufficient.
N
(i) X Y = 10.85 km, P Y = 8.85 km [P Yb X = 60◦ ]
(ii) 8.31 km (or 8.32) [X Y sin 50◦ or P Y sin 70◦ ]
X (iii) 250◦ [70◦ + 180◦ ]
50◦
10 km
Y
70◦
P

CHAPTER 13. VECTORS
29. N
(a) When drawing your diagram make it big,
to allow plenty of room for measurements,
angles etc.
B 210◦
(b) B D = 8 km.
−−→
(c) AD = −8 sin 30◦ i - (8 cos 30◦ − 5) j
30◦ p
= −4 i - (4 3 − 5) j
5 km −→ p
AC = 6 sin 60◦ i - 6 cos 60◦ j = 3 3 i - 3 j
30◦ A 120◦
60◦ 6k
m
D

α
C
−−→ −−→ −→ p p
(d) DC = −D A + AC = (4 + 3 3) i−(8 − 4 3) j
p
8−4p3
tan α = ⇒ α = 6.648◦ , so bearing of C from D is 096.7◦ . (3 SF)
4+3 3
p
30. (a) 13 (b) 2 3 (c) 56 (d) 2k
p p
31. 200 m. λ = ±5 7. [speed = 20 ms−1 = 152 + λ2 ]
32. 30 m. 34 m [Final position is 30 i + 16 j]
33. 15 ms−2 [|3(3 i - 4 j)|]

p
34. 2 10 Newtons. 63.4◦ above i [F 1 + F 2 =2 i + 4 j]
1
k =− [4 − 2k = 5 + 5]
3
p
35. 8 5 ms −1 [a = 4 i - 2 j = increase in its velocity each second]
16 12
36. 4 ms −2 [Resultant force is 48 i + 36 j, so acceleration = 5 i+ 5 j]
37. p = 2, q = -3 [3p + 2q = 0 and -8 - 2q + p=0]

CHAPTER 13. VECTORS

• Vector quantity compared to a scalar quantity.
• Column vectors For example

¡4¢ ¡ 5 ¢ ¡ 9 ¢ ¡ 2 ¢ ¡ 6 ¢ ¡−10¢
+ = and 4 −3 =
1 −3 −2 −3 −5 3
Scalar multiple of a vector. Parallel vectors. Triangle law of vector
addition.
• i, j notation
(4 i + j) + (5 i - 3 j) = 9 i - 2 j. Find 2a - 3b if a = 2 i + 5 j and b = i - 3 j
¡a ¢ ¡ c ¢
Independence of components = ⇒ a = c and b = d .
b d
p
• Magnitude of a vector |x i + y j| = x 2 + y 2
p p
Examples |3 i - 4 j| = 5, |7 i + 7 j| = 7 2, |λ i + 2λ j| = 5λ
• Direction of a vector Examples

2 i + 3 j makes an angle θ above the direction of i, where tan θ = 23
-4 i - 4 j makes an angle 135◦ below the direction of i [135◦ = 180◦ − 45◦ ]
Components of a vector given its magnitude and direction.
• Position vectors Fixing the starting point of a¡vector

¢ as the original
2
coordinates to specify a position, in which case ≡ 2 i - 3 j ≡ (2, −3)
−3
Using vectors to solve coordinate geometry problems.
• Solving geometry problems using vectors For example

B Given the ratios BT : T A and
AS : S A find the ratio in which P
T divides OT and B S.
P
O S A
• Modelling with vectors Applied to displacements (bearing problems),

forces, velocities and accelerations.
Bearing problems also involve the use of Sine Rule and Cosine Rule.
Combining forces to obtain a resultant force. The formula F = ma .
Formula connecting velocity, acceleration and time: v = u + at .

Chapter 14
GRANDE FINALE
14.1 Series
What we cover here is definitely not on your syllabus but it draws together
many of the themes in the book. The chapter consists of a number of short
section, most of which are not at all difficult, giving you a glimpse of some of
the fascinating topics that lie ahead if you continue to study Maths.
In the process of exploring we will see how a series containing an infinite

number of terms can still have a finite sum, unearth some unexpected
connections between apparently unrelated branches of mathematics,
discover something astonishing about the properties of functions and
discover that, viewed in a certain way, the square root of −1 makes perfect
sense. (Oh, and we’ll also prove the result that we assumed earlier that the
gradient function of e ax is ae ax .)
Don’t be put off by the fact that the whole chapter, because it is optional, is in
blue.
We will be heavily dependent on what are known as Power Series, so let’s
start by explaining what we mean by a series.
A series is a sequence of terms added together, where each term is defined in

a specified way. For example in the five term series 1 + 2 + 4 + 8 + 16, we are
adding a sequence of increasing powers of 2, starting at 20 . If a series
contains a known number of terms we can do the addition and find the sum;
in this case the answer is 31. However sometimes this would be extremely
tedious–if there were 100 terms for example. So the first thing we are going
to do is derive a formula for the sum of n terms of a series defined in this
way. Notice that the nt h term is 2n−1 because we started at 20 .
Let S n = 20 + 21 + 22 + 23 + 24 + . . . + 2n−1
then 2S n = 21 + 22 + 23 + 24 + . . . + 2n−1 + 2n
so S n = 2n − 1
We obtain our formula by subtracting the first series from the second series.
(Note that you can simply cross out the like terms across the first two rows
above.)
When n = 5 this gives 31, as expected. When n = 100 it gives 2100 − 1 ≈ 1.27 ×
1030 .
Needless to say, the more terms we add the bigger the sum. Not only that,
there is no limit to how big the sum can be. We say such a series diverges.
14-0
CHAPTER 14. GRANDE FINALE
A series that works like this one, where each term is a fixed multiple of the
previous term, is known as a geometric series. In our case that multiple is 2,
generally referred to as the common ratio, normally denoted by r .
What if the series is geometric but common ratio is < 1? The terms get
smaller and smaller rather than bigger and bigger.
1
Suppose r = 2 for example. Repeating what we did above with this common
ratio, 1 1 1 1 1 1
we let S n = 0
+ 1 + 2 + 3 + 4 + . . . + n−1
2 2 2 2 2 2
1 1 1 1 1 1 1
then Sn = 1
+ 2 + 3 + 4 + . . . + n−1 + n
2 2 2 2 2 2 2
1 1
So S n = 1 − n (taking the 2nd row from the 1st)
2 µ 2 ¶
1
Sn = 2 1 − n
2
The extraordinary thing we notice about this is that the S n is always less that
2 regardless of the value of n, and the larger n becomes the closer the sum is
1
to 2, because 2n → 0. We say the series converges to the value 2 and write
S n → 2 as n → ∞.
Of course we can never have an infinite number of terms, nevertheless to
capture the idea that the sum gets closer and closer to 2 as the number of
terms increases, we write
1 1 1
S ∞ = 1 + 1 + 2 + 3 + . . . = 2.
2 2 2
To see that this makes sense imagine an elderly frog jumping from one end
of a 2m long table to the other. Suppose her first jump is 1m, and with every
jump she gets more tired and at each jump can only manage half the
1
remaining distance. The second jump will therefore be one of the 2 m, the
1
third 4 m and so on.
1 1 1
The total distance jumped will be 1 + 21 + 22 + 23 + . . ., exactly the series we
found to have a sum of 2m (the total length of the table).

It is a small step to generalise this for any value of the common ratio < 1. For
a reason that will soon become clear, here we are going to use x rather the r
for the common ratio.
Let S∞ = 1 + x1 + x2 + x3 + x4 + . . .
then xS ∞ = x(1 + x 1 + x 2 + x 3 + x 4 + . . .)
= x1 + x2 + x3 + x4 + . . .
So (1 − x)S ∞ = 1
1
S∞ =
1−x
1 1
A quick check reassures us we haven’t made a mistake. If x = 2 , S ∞ = 1 = 2.
2
Another example of an infinite number of terms having a finite sum. If

x = 0.1 we are saying that
1
1 + 0.1 + 0.01 + 0.001 + 0.0001 . . . =
0.9
Since the series never ends the left hand side is the recurring decimal
1.11111 . . ., written 1.1̇.
10 1 1
The right hand side is = 1 9 . This is correct because 9 = 0.1̇, something
9
1
we can deduce by dividing 3 = 0.3̇ = 0.3333 . . . by 3.
As brief aside, if we multiply both sides by 3 we get 1 = 0.9999 . . . = 0.9̇.
How can this be? Are they not different numbers? If we embrace the idea of
limits, they are in effect the same number but written differently.
One way to feel more comfortable about this is to ask yourself the
question:“what is the difference between them?" The answer is a 1 preceded
by an infinite number of 0’s.

14.2 Binomial expansion: turbo-charged version

1
Writing 1−x as (1 − x)−1 makes us think of the Binomial expansion. Is it
possible to use the Binomial expansion with something like (1 − x)−1 ? The
answer is yes.
We know how to expand an expression like (1 + x)n :
n n(n − 1) 2 n(n − 1)(n − 2) 3
(1 + x)n = 1 +
x+ x + x + . . . + xn
1 2! 3!
where previously n has always been a positive integer and the series
naturally ended after n + 1 terms.
In fact n doesn’t have to be a positive integer, in fact it need not be an integer
2
at all. We can expand1 expressions like (1 + x)−3 and (1 + x) 3 using exactly
the same method. The difference is this: if n is not a positive integer the
series never ends. It is an infinite series, often referred to as a power series.
But as we have seen, it can still have a finite sum.
So let’s see what the Binomial expansion for (1 − x)−1 might look like. First
we replace +x with −x in the above expansion: the result is that even powers
of x remain positive, odd powers become negative:
n n(n − 1) 2 n(n − 1)(n − 2) 3
(1 + x)n = 1 + x + x + x +...
1 2! 3!
n n(n − 1) 2 n(n − 1)(n − 2) 3
so (1 − x)n = 1− x + x − x +...
1 2! 3!
(−1) (−1)(−2) 2 (−1)(−2)(−3) 3
therefore (1 − x)−1 = 1− x+ x − x +...
1 2! 3!
(−1) (−1)(−2) 2 (−1)(−2)(−3) 3
= 1− x+ x − x +...
1 (1)(2) (1)(2)(3)
= 1 + x + x2 + x3 . . .
Notice that the sign of every term is positive.
We have arrived at the same result as the one on the previous page by a
completely different method.
We are now going to tie this up with what we know about differentiation.
(Remember that if y = f (x), its gradient function or derived function we
dy
write as either d x or f 0 (x).)
1
How might we differentiate an expressions like ? Its power series gives
1−x
us a way in.
1
f (x) = = 1 + x + x2 + x3 . . .
1−x
f 0 (x) = 1 + 2x + 3x 2 + 4x 3 . . .
(−2) (−2)(−3) (−2)(−3)(−4)
But (1 − x)−2 = 1 + (−x) + (−x)2 + (−x)3 + . . .
1 1.2 1.2.3
= 1 + 2x + 3x 2 + 4x 3 + . . .
1 1
So if f (x) = , f 0 (x) = .
1−x (1−x)2
Of course we cheated because to find the answer we had to know the power
series expansion for (1 − x)−2 .
1
The expansion may not be valid for every value of x. In these examples we need |x| < 1.

To overcome this objection let’s see if we can differentiate y = (1 + ax)n for

any n and any (constant) a. The example we just looked would then
correspond to the case a = −1 and n = −1. We ask you to take on trust that it
is OK to differentiate infinite series like this.
n n(n − 1) 2 2 n(n − 1)(n − 2) 3 3
y = 1+ (ax) + (a x ) + (a x ) + . . .
1 2! 3!
dy n n(n − 1) n(n − 1)(n − 2)
= 0 + (a) + (2a 2 x) + (3a 3 x 2 ) + . . .
dx 1 2! 3!
n n(n − 1) 2 n(n − 1)(n − 2) 3 2
= (a) + (a x) + (a x ) + . . .
1 1! 2!
n −1
µ ¶
(n − 1)(n − 2) 2
= na 1 + (ax) + (ax) + . . .
1! 2!
= na(1 + ax)n−1
Satisfy yourself that

n −1 (n − 1)(n − 2)
(1 + ax)n−1 = 1 + (ax) + (ax)2 + . . .
1! 2!
Putting a = n = −1 we have (1 − x)−2 , the result we found above.

14.3 Power series of e x

We said we would show why e ax differentiates to ae ax . Along the way we will
also see why the derivative e x is e x itself.
The expansion (or power series) for e x is given by2
x2 x3 x4
y = ex = 1 + x +
+ + +...
2! 3! 4!
Later on we will see why this series describes e x , but for now take it on trust.
Differentiating:
dy 2x 3x 2 4x 3
= 0+1+ + + +...
dx 2! 3! 4!
x2 x3 x4
= 0+1+x + + + +...
2! 3! 4!
= ex
Isn’t that amazing!
So let’s see why the derived function for e ax is ae ax (where a is a constant).
a2x2 a3x3 a4x4

Let y = e ax = 1 + ax + + + +...
2! 3! 4!
dy (a 2 )(2x) (a 3 )(3x 2 ) (a 4 )(4x 3 )
= a+ + + +...
dx 2! 3! 4!
(a 2 )(x) (a 3 )(x 2 ) (a 4 )(x 3 )
= a+ + + +...
1! 2! 3!
(ax)2 (ax)3
µ ¶
= a 1 + ax + + +...
2! 3!
dy
i.e. = ae ax
dx
If the third line is not entirely clear to you, look at the last term
(a 4 )(4x 3 ) (a 4 )(4 x 3 ) a(ax)3

= =
4! 4 .3.2.1 3!
2
Use this approach to find the derived function e x . The answer is in the
footnote3 .
2 x0 x1
The first two terms follow the same pattern as the others because 1 = and x =
2
0! 1!
3
2xe x

14.4 Power series and logs

You may have noticed that, in our discussion about logs, we did not mention
differentiating logs, or integrating them. This is not in the syllabus, but we
can use power series to have a sneak preview. The power series expansion
for f (x) = ln(1 − x) is given by:
x2 x3 x4
f (x) = ln(1 − x) = −x − − − ...
2 3 4
Differentiating the right hand side we have
f 0 (x) = −1 − x − x 2 − x 3 . . .
= −(1 + x + x 2 + x 3 . . .)
= −(1 − x)−1
1
= −
1−x
1 1
So the derivative of the function is and its integral − ln(1 − x).
1−x (1−x)2
Can power series shed any light on the the unintuitive behaviour of the log
function? For example
log A + log B = log AB.
If we let A = 1 − x and B = 1 + x we should find that
ln(1 − x) + ln(1 + x) = ln(1 − x)(1 + x) = ln(1 − x 2 )

Let’s see:
x2 x3 x4 x5
ln(1 − x) = −x − − − − ...
2 3 4 5
x2 x3 x4 x5
so ln(1 + x) = +x − + − + . . . (replacing x with −x)
2 3 4 5
µ 2
x x4 x6 x8
¶
ln(1 − x) + ln(1 + x) = −2 + + + ...
2 4 6 8
x2 x4 x6 x8
= − − − − ...
1 2 3 4
= ln(1 − x 2 )

14.5 Power series and the trig functions

We will now tell you what the power series expansions are for the two
functions sin x and cos x. Later we’ll talk about how they, along with many
other power series, are arrived at.
x2 x4 x6
cos x = 1 − + − ...
2! 4! 6!
x3 x5 x7
sin x = x − + − ...
3! 5! 7!
What we immediately notice is the resemblance to the expansion for e x : each

xn
consists of terms of the form . This is not a coincidence, as we shall see.
n!
Nevertheless there are some important differences. For example with these
two functions the signs alternate between plus and minus, as they do with
some of the binomial expansion we looked at. The cos x expansion is also
similar to the expansion for ln(1 − x 2 ), in that it contains even powers of x
only.
Before we talk about the features of there series, the units of x are not degrees.
The notion dividing a circle into 360 divisions was chosen for purely practical
reasons4 , there is no inherent logic to it. There is a more “natural" way of
measuring angles: it is a radian, which is defined as the angle for which the
arc length of a sector of a circle is the same as its radius. 1 radian ≈ 57.3◦ .
From that definition we can see that 180◦ = π radians (because the arc length
in this case is π × 1.
π
With this in mind, if 90◦ = 2 radians, sin 90◦ = 1 and we have
π ³ π ´3 1 ³ π ´5 1 ³ π ´7 1
1= − + − ...
2 2 3! 2 5! 2 7!
π
¡ π ¢3 1 ¡ π ¢5 1 ¡ π ¢7 1
So the sum of the series − + − . . . is 1.
2 2 3! 2 5! 2 7!
We can be comfortable that it is correct by summing the first few terms:
1.5708 − 0.6460 + 0.0797 − 0.0047 = 0.9998 (to 4 DP).
However this is more curious than it is useful. Let’s look at some of the other
properties. To start with we know that sin 0 = 0, which is what we get if we
put x = 0 into its power series expansion. We also know that cos 0 = 1, which
also agrees with the power series.
You may also recall that cos x is an even function and sin x is an odd
function.
Remember that
a function f (x) is even if f (−x) = f (x) for all values of x, in other words the
curve has reflective symmetry about the y-axis;
a function is odd if f (−x) = − f (x), in other words it has rotational symmetry
about the origin.
Not only can we see that cos x is an even function and sin x is an odd
function, but we can see the reasoning behind the terms even and odd: it
corresponds to the parity5 of the terms in the corresponding power series;
cos x only has even powers of x and sin x only has odd powers.
4
It has many divisors e.g. 3, 4, 5, 6et c.
5
The term parity means whether it is even odd.

Turning again to calculus, sin x and cos x are two more functions we have not
learned how to differentiate or integrate. The power series expansions give us
the answers.
x3 x5 x7
if f (x) = sin x = x − + − ...
3! 5! 7!
3x 2 5x 4 7x 6
f 0 (x) = 1 − + − ...
3! 5! 7!
x2 x4 x6
= 1− + − ...
2! 4! 6!
= cos x
The derived function of sin x = cos x.

Similarly x2 x4 x6
if f (x) = cos x = 1 − + − ...
2! 4! 6!
2x 4x 3 6x 5
f 0 (x) = − + − ...
2! 4! 6!
x3 x5 x7
= −x + − + ...
3! 5! 7!
= − sin x
The derived function of cos x is − sin x.

Differentiating one of the functions gives the other, with a possible sign
change.
f (x) f 0 (x) f 00 (x) f i i i (x) f i v (x)
sin x → cos x → − sin x → − cos x → sin x
Such beautiful simplicity.
One of the many uses of power series is they allow for polynomial
approximations of functions when x is known to be small.
cos x+sin x
Take a function like , and suppose we know that x is so
(1+x)2
small we can ignore powers of x higher than 2.
1
The expansion of is 1 + 2x + 3x 2 + 4x 3 . . ., so if x is small, we could use
(1−x)2
the quadratic 1 + 2x + 3x 2 as an approximation for it, and derive an simple
approximation for the whole function:
(1 − x)−2 = 1 + 2x + 3x 2 + . . .
x2
cos x = 1 − +...
2
sin x = x − . . .
cos x + sin x x2
so 2
= (1 + x − )(1 + 2x + 3x 2 ) . . .
(1 − x) 2
x2
= (1 + 2x + 3x 2 ) + x(1 + 2x) − (1) . . .
2
2
x
= 1 + 2x + 3x 2 + x + 2x 2 − ...
2
9
≈ 1 + 3x + x 2
2

14.6 Maclaurin’s incredible theorem

If you tell me the value of a function and every one of its derivatives at x = 0,
I can tell you the the value of the function for every value of x e.g. f (1, 000).
In other words, provided I know everything about how the function behaves
at the origin, it’s possible to deduce how the function behaves everywhere.
Even to my ears it sounds ridiculous. Actually it only works for “smooth"
functions, it doesn’t work for functions that have infinite values or
1
discontinuities, such as y = x , or functions like y = |x|. Even so it is quite
extraordinary. The theorem that proves it was the work the 17th Century
mathematician Colin Maclaurin6 .
The theorem is surprisingly easy to prove but we shall just demonstrate why
and how it works using a simple polynomial function.
Take the function f (x) = 2x 4 − x 3 + 31 x + 7.
f (x) = 2x 4 − x 3 + 13 x + 7 f (0) = 7
f 0 (x) = 8x 3 − 3x 2 + 31 f 0 (0) = 13
f 00 (x) = 24x 2 − 6x f 00 (0) = 0
f 000 (x) = 48x − 6 f 000 (0) = −6
f i v (x) = 48 f i v (0) = 48
f v (x) = 0 f v (0) = 0
and all subsequent derivatives are also zero.
Maclaurins’ Theorem states that the Power Series of f(x) is:
x 2 00 x 3 000 x4 i v x5 v
f (x) = f (0) + x f 0 (0) + f (0) + f (0) + f (0) + f (0) . . .
2! 3! 4! 5!
Substituting in the values of the values of the derivatives for the function
above we have
x 2 00 x 3 000 x4 i v x5 v
f (x) = f (0) + x f 0 (0) + f (0) + f (0) + f (0) + f (0) . . .
2! 3! 4! 5!
1 x2 x3 x4 x5
= 7 + x + (0) + (−6) + 48 + (0) (no more terms)
3 2! 3! 4! 5!
1
= 7+ x − x 3 + 2x 4
3
To keep things simple the power series we chose was a finite one (i.e. a
polynomial). Let’s use Maclaurin’s Theorem to derive a one involving an
infinite number of terms, such as f (x) = sin x.
This is quite simple because of the circular nature of its derivatives:
f (x) f 0 (x) f 00 (x) f i v (x) f v (x)
sin x → cos x → − sin x → − cos x → sin x
so f (0) = 0 f 0 (0) = 1 f 000 (0) = 0 f i v (0) = −1 f v (0) = 0 etc
which gives us all we need to know to construct the power series for sin x
6
There is a more general theorem that uses the derivatives at any point x = a, developed
by Brook Taylor in the 18th Century.

x2 x3 x4 x5
sin x = 0 + x(1) + (0) + (−1) + (0) + (1) . . .
2! 3! 4! 5!
x3 x5 x5
= x − + − ...
3! 5! 5!
Knowing the very special properties of the function e x , you may wonder how
Macclarin sheds light on those.
Repeating he above process with e x instead of sin x, we have an extremely
simple piece of analysis:
f (x) f 0 (x) f 00 (x) f i v (x) f v (x)
ex → ex → ex → ex → ex
so f (0) = 1 f 0 (0) = 1 f 000 (0) = 1 f i v (0) = 1 f v (0) = 1 etc
Therefore
x 0 x 2 00 x 3 000 x4 i v x5 v
e = f (0) + x f (0) + f (0) + f (0) + f (0) + f (0) . . .
2! 3! 4! 5!
x2 x3 x4 x5
= 1+x + + + + ...
2! 3! 4! 5!

p
14.7 −1
We are now going to venture into an area that p is often regarded a hard to
understand, because it involves dealing with −1, a number that you have
been told does not exist. That is certainly true, if we confine ourselves to
numbers the lie on the number line: no number has a square equal to −1.
In case you think this is no more than an esoteric invention cooked up by
mathematicians with nothing better to do, you would be very wrong.
Engineers and physicists use imaginary numbers extensively, even though in
one sense they do not exist!
p
The key to understanding it is to think of −1 as lying off the number line.
i.e. above it or below it, in other words allowing for a whole plane of
numbers. We call this the complex plane, and numbers on it Complex
numbers. A diagram representing complex numbers in this way is called an
Argand7 diagram. Numbers on the number line we call Real numbers. They
constitute only a subset of the complex numbers.
The square root of −1 is denoted by i , so i 2 = −1. (Some books use j .)
A Complex number like 3 + 2i would be represented in the complex plane at
the point (3, 2) in coordinate terms. We say its real part is 3 and its imaginary
part is 2. You have met something similar in Chapter 13, where a vector has
an i -component and a j -component. (The use of i in both contexts is no
coincidence, although they mean something very different.)
Imaginary
3i
3 + 2i
2i •
−2 + i
• i
0 Real
−4 −3 −2 −1 1 2 3 4
i •
4−i
−2i•
−3i
A number that has no real part, such as 5i or −3i , is called an Imaginary

number.
By thinking of multiplication (and division) is a different way8 we are going
to see how the square roots of a negative number can make sense. We are
going to think of every number, whether Real our Complex, as having two
parts to it: a magnitude and a direction (as a vector in fact). The magnitude
will be its modulus, the direction will be the angle it makes with the positive
number line (oddly named the argument of the number). So for example
the Real number 3 is [3, 0◦ ] and −2 is [2, 180◦ ].
7
Named after Jean-Robert Argand a bookshop manager and amateur mathematician who
lived in Paris in the early 19th century.
8
In much the same way as we defined the trig functions in a new way to allow for angles
> 90◦

Now we define the product of two numbers as follows: the magnitude of

their product is the product of the two moduli; the direction is the sum of the
two arguments. Here are couple of examples, starting off with these two Real
numbers:
+3 −2
0 0
[3, 0◦ ] [2, 180◦ ]
According to this new definition:

+4
2 × 2 = [2 × 2, 0◦ + 0◦ ] = [4, 0◦ ]= 4
−6 −4 −2 0 2 4 6
−6
2 × −3 = [2 × 3, 0◦ + 180◦ ] = [6, 180◦ ]= −6
−6 −4 −2 0 2 4 6
4
−2 × −2 = [2 × 2, 180◦ + 180◦ ] = [4, 360◦ ]= +4
−6 −4 −2 0 2 4 6
These results are entirely consistent with our understanding of

multiplication.
Let’s see what happens if we apply this approach to multiplying the
Imaginary number i by itself9 . Since i is [1, 90◦ ]
i × i = [1 × 1, 90◦ + 90◦ ] = [1, 180◦ ] = −1
p
Now we have established that −1 can be given a meaning that is consistent
with the behaviour of Real numbers, let’s look at apmore interesting example:
squaring the number −1 + i . This has magnitude 2 and argument 135◦ , so
p p
(−1 + i ) × (−1 + i ) = [ 2 × 2, 135◦ + 135◦ ] = [2, 270◦ ] = −2i
Imaginary
3i
2i
−1 + i
• i
0 Real
−4 −3 −2 −1 1 2 3 4
i
−2i•
−3i
We can arrive at the same answer by multiplying the brackets, bearing in

mind that i 2 = −1:
(−1 + i )(−1 + i ) = (−1)2 − 2i + i 2
= 1 − 2i + (−1)
= −2i
9
Higher powers of i are easily calculated e.g. i 3 = −1 × i = −i , i 4 = (−1)2 = 1, i 5 = i 4 × i = i .

In effect what we have done is solve the equation

x 2 + 1 = 0.
This suggests that if we allow Complex numbers, maybe every quadratic has
two solutions . This turns out to be the case. In fact every cubic equation has
three roots, every quartic equation has four roots etc10 . (Some roots may be
repeated of course.). Take the equation
x 2 − 2x + 10 = 0.
Completing the square we have
(x − 1)2 + 9 = 0
(x − 1)2 = −9
p
x − 1 = ± −9
= ±3i
x = 1 ± 3i
The curve of the function y = f (x) = x 2 − 2x + 10 does not to meet the x-axis,
hence there are no real roots. The closest the curve gets to the x-axis (i.e. its
minimum value) is where x = 1. We can think of the two complex roots 1 + 3i
and 1 − 3i as sitting 3 units “above" and “below" x = 1 on the x-axis. They lie
in the plane passing through the vertex of the parabola perpendicular to the
x-axis.
We can check this by substitution (we will check for 1 + 3i );
(1 + 3i )2 − 2(1 + 3i ) + 10 = 1 + 6i − 9 − 2 − 6i + 10 = 0
We see a similar phenomenon with higher order polynomial equations. For

example take the cubic equation
x(x 2 + 4) = 0.
We expect to find three roots in total. One is x = 0, the other two are x = 2i
and x = −2i .
What we find is that whenever there are complex roots they come in pairs,
each with the same Real component but the opposite Imaginary
component, a so-called complex conjugate pair. That means there are only
certain combinations of real and complex roots possible:
Polynomial Roots
Quadratic 2 real or 2 complex
Cubic 3 real or 1 real, 2 complex
Quartic 4 real or 2 real, 2 complex
Quintic 5 real or 3 real, 2 complex or 1 real, 4 complex
and so on ...
10
In fact it was trying to understand the roots of equations that led mathematicians to
develop the ideas.

In the same way that with vectors, the components are independent, the Real
and Imaginery parts of a Complex number are independent too. So if we have
two Complex numbers a +i b and c +i d , they are only equal if a = c and b = d .
This must be true because we can think of each one as a point in the Argand
diagram. To be equal means they correspond to the same point.
Now consider the power series expansion of e i x .
x2 x3 x4 x5 x6
ex = 1+x + + + + + ...
2! 3! 4! 5! 6!
(i x)2 x 3 (i x)4 (i x)5 (i x)6
so e i x = 1 + (i x) + + + + + ...
2! 3! 4! 5! 6!
x2 x3 x4 x5 x6
= 1+ix − −i + +i − ...
2! 3! 4! 5! 6!
x2 x4 x6 x3 x5
µ ¶ µ ¶
= 1− + − ... +i x − + ...
2! 4! 6! 3! 5!
= cos x + i sin x
We have managed to capture in the simple

p term e i x the sine of x and the
cosine of x, but to do so we have used −1. Can we treat i just like any other
number?
We showed that the derivative of e ax is ae ax , where a is a Real number. Will
this rule apply if a is an Imaginery number? Differentiating both sides, we
would have
i e i x = − sin x + i cos x
Multiplying both sides by i gives
−e i x = −i sin x − cos x or e i x = i sin x + cos x.
It seems to obey the same rules.
We are going to use this relationship to prove a result which by another
method is very complicated. In Chapter 12 we saw that the sin and cos
functions have to be treated with care. For example sin 2θ 6= 2 sin θ. If we
think about it they cannot be equal: the curve y = sin 2θ is the curve y = sin θ
1
transformed by a stretch of scale factor 2 parallel to the x-axis; 2 sin θ is the
result of a stretch scale factor 2 parallel to the y-axis.
The result above provides a beautifully simple way to find expressions for
sin 2θ and cos 2θ in terms of sin x and cos x. We start by seeing11 that
e 2i θ = (e i θ )2
Expanding each side using the fact that e i x = cos x + i sin x:
cos 2θ + sin 2θ = (cos θ + i sin θ)2
= (cos θ + i sin θ)(cos θ + i sin θ)
= cos2 θ + 2i cos θ sin θ − sin2 θ
= (cos2 θ − sin2 θ) + i (2 cos θ sin θ)
so cos 2θ = cos2 θ − sin2 θ
and sin 2θ = 2 cos θ sin θ
11 2i θ
e = e i (2θ) .

14.8 Euler’s equation: e i π + 1 = 0

We end with an equation (or strictly speaking an identity) which combines e,
π, two numbers that a woven into the way the universe works, i (the square
root of −1) and the numbers 1 and 0, the building blocks upon which all
other numbers are formed.
Despite it being a truly astonishing and beautiful result, extremely easy to
prove using what we learnt in the previous sections. (Remember that
π = 180◦ .)
e i π = cos π + i sin π
= −1 (since cos 180◦ = −1 and sin 180◦ = 0)
so e i π + 1 = 0
It is sometimes said that the test of a true mathematician is whether he/she

is thrilled by this result.

Appendix A
Notes and points of interest

A.1 An infinity of primes
We can all list lots of prime numbers: 2, 3, 5, 7, 11, 13, and so on, but how
many primes are there altogether? Euclid showed (around 300 B.C.) that
there are infinitely many. That he was able to so by purely abstract reasoning
marked him out as a truly great thinker. His proof1 goes like this.
Let’s suppose there is a finite number of primes, say N of them. We don’t

need to know what they all are, or even what N is, only that a complete list is
possible. Call them p 1 , p 2 , p 3 . . . p N . The key idea is this: if we can produce a
new prime number that isn’t in our list, we have proved that writing a
complete list is impossible, ergo there is an infinite number of primes.
Consider the number we get when we multiply all the primes in our list
together and add one to the answer. Call this number M, i.e.
M = p 1 p 2 p 3 . . . ...p N + 1
Now M is either prime or it is not. If it is prime, then M is a prime that was
not in our list and the proof is complete. If M is not prime, then it is divisible
by some prime, call it P . Notice P can not be any of p 1 , p 2 , p 3 . . . p N , because
if we divide by any of these we get a remainder of 1. So P is a new prime and
that completes the proof.
We illustrate the reasoning with two simple examples.

(i) With the list consisting of the first three prime numbers.
N = 2 × 3 × 5 + 1 = 11.
This is a prime not in the list 2, 3, 5.
(ii) With the list ending at 17.
N = 2 × 3 × 5 . . . × 17 + 1 = 510511 = 19 × 97 × 277.
We have three new prime numbers not in the list.
1
The method he used we now call proof by contradiction.
A-1
APPENDIX A. NOTES AND POINTS OF INTEREST
A.2 The largest known prime number

The largest known prime number (as of 4 Jan 2019) is 282,589,933 − 1, a number
with 24,862,048 digits. It was found by the Great Internet Mersenne Prime
Search (GIMPS).
A prime number that can be written in this form is known as a Mersenne
prime, after their discoverer Marin Mersenne, a 17th century French monk.
The numbers are all expressed as 2 raised to the power of a (smaller) prime
number minus 1, i.e.
2p − 1.
Notice that p can’t be any prime number. For example p = 11 doesn’t work.
211 − 1 = 2047 = 23 × 89.
There is little known mathematical value to finding new primes, but they of
interest because finding new ones gets harder all the time.
The Electronic Frontier Foundation is offering various rewards for finding
even larger prime numbers. For example a $150,000 reward for discovering a
prime number with 100, 000 digits.
PURE MATHS: AS LEVEL GUIDE A-2

A.3 A formula for the primes?

We discussed prime numbers earlier in the chapter, and you may have
wondered if they follow a nice pattern. If you have, you would not be alone;
countless people over many centuries have tried to devise formulae for
them. One such formula produces all these:
43, 47, 53, 61, 71, 83, 97, 113, 131, 151, 173, 197, 223, 251, 281, 313,
347, 383, 421, 461, 503, 547, 593, 641, 691, 743, 797, 853, 911, 971,
1033, 1097, 1163, 1231, 1301, 1373, 1447, 1523, 1601, 1681.
If you care to check, they are all prime numbers, and you could be forgiven
for concluding that the formula that produced them
M n = n 2 + n + 41, for n = 1, 2, 3, 4, 5 . . .
only gave prime numbers. That would be a mistake. The next one in the list,
the value corresponding to n = 41, is 1763, which is not prime. Can you see
why? If not, the reason is in the footnote2 .
We saw a different example of unreliable intuition when we looked at the

question “is 0.9̇ = 1?". It didn’t take a lot to establish the flaw in the thinking.
In A.15 you will see a “proof" that 1 = 2. Clearly it’s not a valid proof, but
sometimes a mistake can be very hard to spot. It’s all about training
ourselves to be meticulous in our thinking.
To get to the bottom of a mathematical problem, often what we need is not

more information but a clearer understanding of an idea, or a more careful
examination of the information we have.
Examples abound in everyday life of wrong conclusions being jumped to.
Sometimes we are “urged" into less than rigorous thinking by advertisers,
and even—dare I say it—by our esteemed political masters. The best way to
protect ourselves against this sophistry is to hone our thinking skills, and
there can be few better ways to do this than to study Advanced Maths.
2
The number 412 + 41 + 41 is divisible by 41. Can you see from this that no polynomial
formula can ever generate only prime numbers?

A.4 More on recurring decimals

A juicier example of converting a recurring decimal to a fraction is
0.094̇28571̇
The fly in the ointment is the .09 that precedes the recurring part of the
number. Remember we need to generate another number with the same
digits after the decimal point: we can do this by first multiplying the number
by 100. That moves the offending digits to the left of the decimal point.
x = 0.094̇28571̇
100x = 9.4̇28571̇
Now let y = 4̇28571̇, and by the usual method
y = 4̇28571̇
1000000y = 428571.4̇28571̇
999999y = 428571
This looks as if it’s going to give us an extremely cumbersome fraction, but

look:
y = 428571 47619
999999 = 111111
15873 1443
= 37037
= 3367
dividing by 3 then by 11
111 3
= 259 = 7 dividing by 13 then by 37
We are nearly there. 100x = 9 + y, so
9 3 63 3 66 33
x= + = + = =
100 700 700 700 700 350
The answer is
33
0.094̇28571̇ =
350
You might like to try this one for yourself. The answer is in the footnote3 .
0.125̇38461̇
3 214
0.094̇28571̇ = 325

A.5 Is 0.9̇ = 1?
Is 0.999 . . . the same as 1? The obvious answer is no. How can they be the
same?
OK, well consider this: we know that 0.3̇ = 31 . Multiply both sides by 3 and we
have 0.9̇ = 1.
If you think there might be something slightly dodgy about multiplying a
recurring number, let’s look at it another way. If we add together three lots of
1
3
and three lots of 0.3̇. We still get 1 = 0.9̇.
If you are sill not convinced, let’s think about what the difference between
the numbers is. It’s 0.00000000 . . . with a 1, well, we never reach it, do we?
Is this of more than philosophical interest? Well, if 0.999 . . . = 1, dividing each
side by 9, we have 0.111 . . . = 19 . From this we can write down the decimal
version of any fractions with a denominator of 9, e.g. 49 = 0.4̇, 79 = 0.7̇. From
7
there, 18 = 0.38̇. And so on.
On a different—and rather more profound—track, since 0.111 . . . = 19
1 1 1 1 1
+ + + +... =
10 100 1000 10000 9
We have added an infinite number of fractions and got a finite sum. You will
see more examples of infinite sums in Core 2.
What all this has shown us is that we can sometimes get very surprising
results when dealing with infinity.

A.6 Complex Numbers

Since the Real numbers are all the numbers on the number line, what other
numbers could there be? Think of any number and it will correspond to
some point on the number line. Well, the 19th Century saw the addition of
Complex Numbers, denoted by C. These include numbers are allowed to
stray off the number line! We think of them as lying within a whole plane, of
which the number line is just part. For example a complex number 3 units
“north" of the number 7 on the number line we would say has a Real part
equal to 7 and an Imaginary part equal to 3. (Imaginary numbers? All we
need now is for mathematicians to start talking to them!)
They were discovered, if that is the right word, by mathematicians trying to
work out how to solve cubic equations, like x 3 + 5x 2 − 2x + 11 = 0.
We shall touch briefly on Complex numbers in the chapter on solving
equations, but to study them in depth—and find out all about these
fascinating numbers that don’t really exist—you will need to study Pure
Maths beyond the level of this book. Believe it or not, though, far from being
mere esoteric flights of fancy, no engineer or physicist would be without
them.
You will find a more detailed description of Complex numbers in Chapter 14.
A.7 Normal numbers

A normal number is a special type of irrational number. Along with every
other irrational number its decimal expansion is infinite and non-recurring,
but it has the additional property that somewhere within it there is every
possible sequence of digits. So for example take the sequence consisting of a
thousand 7’s. If we were to write down any normal number, sooner or later
we would come across one thousand consecutive 7’s.
But why stop at something as simple as a sequence of 7’s. What about this
sequence: all the mobile phone numbers of the UK population written down
in order of their birthdays. Of course we may have to go a long way before we
find it. In fact it may be so far past the decimal point that we couldn’t reach it
in our lifetime, or in the remaining lifetime of planet earth! But it’s there
somewhere.
So give me an example of a Normal number, I hear you say. Although
mathematicians have shown that there are vast numbers of them, it is
extremely difficult to establish which numbers are normal and which are
not. Apparently there are good reasons to believe that π is Normal, but it’s
not proved (you would become instantly famous if you could do so). In
Wikipedia there is an example, albeit an artificial one.

p
A.8 Proof that 2 is irrational
First we need to explain something about the method used. It’s the notion of
proving something by showing that the opposite can’t be true. It goes by the
fancy name of “proof by contradiction".
Here is a simple example.
“Today is Monday. Show that the same date next year won’t be a Monday."
We start by assuming the opposite, i.e. that it will be a Monday. This would
imply that the number of days between now and then is a multiple of 7.
However that’s impossible because there are either 365 or 366 days in a year,
neither of which is divisible by 7. Based on our starting assumption we have
deduced (using sound logic) something that’s incorrect, so our starting
assumption must have been wrong; the same date next year won’t be a
Monday.
p
So to the proof that 2 is irrational.
p
We are trying to prove that something is impossible i.e. that 2 can’t be
written as a fraction using whole numbers. This is a p perfect candidate for
proof by contradiction. We start by assuming that 2 is rational, which
would imply that it could be written as a fraction, say
p P
2=
Q
where P and Q are integers. From our understanding of equivalent fractions
we know there are infinitely many such fractions. It’s important for us to
know that the version we have is in its simplest form—you’ll see why later. If
P
is in its simplest form, fine. If not, we cancel it down until it is. Suppose
Q
that leave us with p p
2=
q
where we know that p and q have no common factors. Now we square both
sides of the equation
p2
2= 2 or 2q 2 = p 2 .
q
This means that p 2 is an even number. But that implies p itself is an even
number (the square of an odd number is an odd number). Since p is even we
can write it as p = 2r , where r is also an integer. Substituting p = 2r into our
equation, p 2 = (2r )2 , so:
2q 2 = 4r 2 or q 2 = 2r 2 .
Here’s where we get to the nub of the proof. Now we can apply the same
reasoning as before: q 2 is even and therefore q is even. In other words p and
q are both even.
p
Why is this significant? It means q is not a fraction its simplest form. Can
p
you see what’s happening? Starting with the assumption the 2 is rational we
have concluded (with correctp logic) something that cannot be correct.
p That
starting assumption that 2 is rational must have been false: 2 must be
irrational. We can say for certain that any attempt to express it as the quotient
of two integers would be futile.
In fact it’s possible to show that the square roots of any positive integer (that
isn’t a square number) is irrational
.

A.9 Solution to the chess board puzzle

The proof goes like this. It uses the idea of proof by contradiction described
in the previous section. The squares of a chessboard are coloured alternately
black, white, black and so on. Since we have cut out two opposite corners,
the missing squares will both be the same colour (satisfy yourself that this is
so). Let’s assume the missing squares are both white4 . That means the 62
remaining squares are made up of 32 black and 30 white.
Now each of our 31 dominos covers one black square and one white square.
In terms of the total area we have sufficient dominoes to cover what remains
of the board. But every domino covers one white square and one black
square, so our 31 dominoes cannot cover the 62 remaining squares. We have
a contradiction, so the assumption upon which we built our argument—that
the task is possible—must have been wrong: it is impossible.
4
The argument works just as well if they are both black.

A.10 A cautionary tale

On page A-5 we saw how it is sometimes possible to add an infinite series of
numbers together and get a finite answer. Doing this is not without its pitfalls.
An obvious way to explore the sum would be to group the numbers together.
To help with doing this we’ll first write it as a series of directed number added
together:
S = 1 +− 2 + 3 +− 4 + 5 +− 6 + 7 +− 8 + 9 . . .
Grouping the series like this
S = 1 + (− 2 + 3) + (− 4 + 5) + (− 6 + 7) + (− 8 + 9) . . .
= 1+1+1+1+...
The further we go the bigger S becomes—we say the series diverges to ∞.

But now suppose we group the series like this:
S = (1 +− 2) + (3 +− 4) + (5 +− 6) + (7 +− 8) + (9 . . .
−
= 1 +− 1 +− 1 +− 1 . . .
which diverges to −∞. There is obviously something very strange going on. It
may not be entirely surprising that we get an infinite answer, but two different
answers—one infinitely positive and the other infinitely negative?!
However, if you think what we have seen so far is odd, here is another way of
summing the series. We are going to start by doubling it. You will agree that
we could write
2S = 1 + 1 +− 2 +− 2 + 3 + 3 +− 4 +− 4 + 5 + 5 +− 6 +− 6 + 7 + 7 +− 8 +− 8 . . .
Now we group the terms like this:
2S = 1 + (1 +− 2 +− 2 + 3) + (3 +− 4 +− 4 + 5) + (5 +− 6 +− 6 + 7) . . .
Can you see that we could continue this indefinitely, and that the numbers in
each bracket will always add up to zero? This being the case we get the rather
spooky result
2S = 1
1
S=
2
which is surprising to say the least!

I think the point is clear: summing infinite series can be a minefield.
Surprisingly enough it wasn’t until the 19th Century—after mathematicians
had been using infinite series for more than 100 years—that the theory of
infinite series was put on a rigorous footing. Now the minefield has been
cleared we know which infinite series can be summed and which can’t.
Obviously this one can’t.

A.11 Irrational indices

What could a number like
p
3
7 or 10−π mean?
p
We’ll demonstrate what it means with 7 3 , but we could justp as easily use
10−π . We start with a decimal
p
approximation of the index: 3 ≈ 1.7321 (to
4 DP). This suggests that 7 has a value somewhere between 71 and 72 . It
3
also suggests that a reasonable approximation might be

17321
p
10000
71.7321 = 7 10000 = ( 7)17321 = 29.0934 (to 4 DP).
This is consistent with its value lying between 71 and 72 , which is reassuring.
Where this is different fromphaving a rational index is that we can never get
an exact answer; because 3 is irrational it doesn’t have an exact decimal
value. Nevertheless we can get it as precise as p we like, by using more and
more decimal places in our approximation for 3.
The same applies to
1 1
10−π ≈ = = 7.22 × 10−4 (to 3 SF).
103.14159265 10
314159265
100000000
All this is very interesting but do we need worry about indices that are not
rational? The answer is we might have an equation like
2x = 50
It’s hard to imagine raising 2 to some power, finding its root and—lo and
behold—getting the exact answer 50. It’s not going to happen. So we shall
have to allow for an index to be any Real number. To find x we will need the
concept of logarithms, which is covered in Chapter 11.

A.12 Combining terms

Strictly speaking, to see whether two terms can be combined it’s not the
algebraic parts that need to be the same, it’s the variable parts. Look at this
expression:
(4x 2 + 7x − 1) + (2x 2 + kx + 4)
In this case it is clear that k is a constant coefficient, so we can combine the
two x terms, to give:
6x 2 + (7 + k)x + 3.
You might be wondering whether this can lead to some confusion.
For example in an expression like
8y − x y
how do we know whether this is one y term and one x y term (with coefficients
8 and −1), or two y terms (with coefficient 8 and −x)? The answer is we will
only know from the context, although the letter convention can give us a clue.
In this case, the use of x suggests the second term is a variable not a constant.
In practice it will always be clear.

A.13 Factorising quadratics: a shortcut

The reasoning goes like this. Look back at the original expression
6x 2 − 11x − 10.
The terms have no common factor—we checked that at the beginning. Now
look at the first one in the list:
(x + 1)(6x − 10).
The right hand bracket has a factor 2, so the whole expression could be
written
(. . . . . . )(. . . . . .).
If this were the right answer it would imply that all the terms were multiples
of 2, but we know they don’t have a common factor so this can’t be the right
answer.
Six of the options can be eliminated in this way, leaving a shorter list of
options to check:
a + − − + c
(( ((
1×6 (x(
( +(1)(6x
((( − 10) (x(
( −(1)(6x
((( + 10) 1 × 10
(x + 10)(6x − 1) (x − 10)(6x + 1) 10 × 1
(x + 2)(6x − 5) (x − 2)(6x + 5) 2×5
(x + 5)(6x − 2) (x − 5)(6x + 2) 5×2
2×3 (2x + 1)(3x − 10) (2x − 1)(3x + 10) 1 × 10
(( ( (( (
(2x
((+ (10)(3x
((( − 1) (2x
( (−(10)(3x
((( + 1) 10 × 1
( (
(2x
(+ (((−(5)
(2)(3x (2x
(− (((+(5)
(2)(3x 2×5
( (
(2x + 5)(3x − 2) (2x − 5)(3x + 2) 5×2

A.14 Difference of two squares: harder examples

The expression
(3x + y)2 − (3x − y)2
fits the difference of two squares pattern if we write
X = 3x + y and Y = 3x − y.
Bearing his in mind, we have
(3x + y)2 − (3x − y)2 = [(3x + y) + (3x − y)][(3x + y) − (3x − y)]

= [3x + y + 3x − y][3x + y − 3x + y]
= (6x)(2y) = 12x y
There is nothing wrong with simplifying (3x + y)2 − (3x − y)2 by multiplying
each bracket out. It’s just a little more work
(3x + y)2 − (3x − y)2 = (9x 2 + 6x y + y 2 )

= (9x 2 + 6x y + y 2 ) − (9x 2 − 6x y + y 2 )
= 9x 2 + 6x y + y 2 − 9x 2 + 6x y − y 2
= 12x y
Here is an even more extreme example (think how tedious it would be to

multiply out the brackets):
(2x + 6y + a)2 − (2x + 6y − a)2

= [(2x + 6y + a) + (2x + 6y − a)][(2x + 6y + a) − (2x + 6y − a)]
= (4x + 12y)(2a)
= 8a(x + 3y)

A.15 “Proof" that 2 = 1

Your job, of course, is to spot the flaw in the proof.
We start by assuming that a = b, where the number can be any value except
zero.
Since ab = b 2 , we can write a 2 − b 2 = a 2 − ab.
After factorising, this becomes (a + b)(a − b) = a(a − b).
Dividing both sides of this equation by (a − b) we have
(a + b)(a − b) a(a − b)
= .
(a − b) (a − b)
Now we can cancel by a − b in each fraction
(a + b)
(a b)
−

a(a−b)

=
(a
−b)

(a−b)

leaving a + b = a.
But a = b, so 2a = a. However this seems to prove that 2 = 1.
Can you see the error in the proof? If not, the answer is in the footnote5 .
5
It’s impossible to divide by 0, which is what we did when we divided by (a − b).

A.16 Harder algebraic fractions

Here is a similar, but somewhat more complicated, example
a 2 − 74 a − 21
2a + 12
It’s harder than you’ll be expected to deal with in C1, but it doesn’t use
anything you haven’t met. We can turn numerator and denominator each
into a single fraction using the hammer and boot rule:
7 1 4a 2 − 7a − 8 1 4a + 1
a2 − a − = and 2a + =
4 2 4 2 2
4a 2 −7a−2
4 2(4a 2 − 7a − 2) (4a 2 − 7a − 2)
∴ 4a+1
= =
2
4(4a + 1) 2(4a + 1)
X − 2) a − 2
(4a + 1)(a
XXX
= =
2(4a + 1) 2
XXX
X
a
= −1
2
This is some simplification!
Another way we could tackle it would be to use the notion of equivalent
fractions. If we multiply top and bottom by 4 (the lowest common multiple
of all the fractions we want to get rid of )
4 × (a 2 − 74 a − 21 ) 4a 2 − 7a − 2
=
4 × (2a + 12 ) 8a + 2
which can be simplified as before. Either approach is fine.
In case you are thinking these are all very contrived, it’s surprising how easily
we can find ourselves with fractions within fractions. For example
substituting
1 x +2
x= into y = .
2 3x − 1
We have
1 5
+2
y = 23 = 21 = 5.
2
−1 2
For a bit of fun let’s simplify this juicy looking expression
1
x−1 +1
1 − x.
x− x1
Breaking the problem down, we start with the numerator of the big fraction:
1 1 + (x − 1) x
+1 = = .
x −1 x −1 x −1
For its denominator we’ll use the equivalent fraction approach (multiplying
top and bottom by x)6 :
1 x
1
= 2
x−x x −1
x 2 −1
6
We could also use the hammer and boot method, after writing x − x1 as x .

and we have
x
x−1 x (x 2 − 1) (x + 1)(xX−X1)
X
X
−x = −x = −x = 1
@
x
x 2 −1
x (x − 1)
@ x −1
XXX
For those of you who like a challenge, this problem was lifted from a C4 exam
paper. The candidate had to show that the expression
1 x2
is equivalent to .
1 − ( x−1
x )
2 1 − 2x
We start by tackling the squared term in the denominator:

x − 1 2 (x − 1)2 x 2 − 2x + 1
µ ¶
= =
x x2 x2
and therefore
x −1 2 x 2 − 2x + 1 x 2 − (x 2 − 2x + 1)
µ ¶
2x − 1
1− = 1− 2
= 2
= .
x x x x2
With the denominator a single fraction, the simplification is now easy:
1 1 x2
= =
1 − ( x−1
x
)2 2x−1
x2
2x − 1
If there is a lesson here, it’s not to be afraid of breaking a problem down into
bite sized chunks.

A.17 Precision
What level of precision (e.g. how many decimal points) do we choose when
giving the answer to a problem, and what level should we use in the course
of the calculation? If the exact answer is an integer obviously that’s the
answer we give. If it’s a fraction we generally leave it as a fraction, rather than
convert it to a decimal. If the fraction is top-heavy (improper) there’s usually
nothing to be gained by converting it to a mixed fraction—either version is
fine. The exception to this is if it represent some physical quantity: for
example we wouldn’t leave an answer as $ 52 .
If an exact answer isn’t possible the question will normally state the level of
precision required e.g. “give your answer correct to 2 SF". If the question
doesn’t specify a level of precision it’s safe to give the answer correct to 3
significant figures. Here is a reminder of the terminology around precision.
Decimal Place (DP) precision.
This is the number of digits—including any zeros—that must follow the
decimal point. For example:
16.0472 = 16.047 (3DP) = 16.05 (2DP) = 16.0 (1DP)
There are two things to point out about the 1DP answer. First it is 16.0 not
16: we must write in the zero because 1DP was specified. Had we written 16
the reader would be entitled to wonder if we had rounded to the nearest
whole number. Putting in the 0 makes it crystal clear that there is nothing in
the first decimal place. Secondly when we round we do so starting with all
the digits we have. In this example had we incorrectly started from the 2DP
version we’d have got 16.1 (we generally round a 5 up).
Significant figure (SF) precision.

This is the number of digits there must be starting from the first non-zero
digit. We may need to insert additional zeros to maintain the correct
position of the decimal point.
The big advantage of SF precision is that it is independent of the units of
measurement. For example:
12.68 = 12.7 (2SF), 0.007619 = 0.00762 (3SF), 9361 = 9400 (2SF)
Working precision
Whatever the precision required for the answer, we will need to do
the calculations to a greater level of precision; if we don’t, we risk A
introducing rounding errors. There is no simple rule for deciding
!
A
A
how much extra precision to work to, it depends on the context, A
but 2 or 3 extra figures is usually sufficient.

A.18 Formulating linear equations: a harder

example
A train takes 6 hours to complete a journey, travelling at an average speed of
40km h−1 . The delivery of a new locomotive increases the average speed by
v km h−1 . By how many minutes does this reduce the journey time?
Solution
We start by defining t = the reduction in journey time (in hours). We make
hours the units because all the information is given in hours ;we’ll convert
the answer to minutes at the end.
The distance is the same with both locomotives—6 × 40 km, so
(40 + v)(6 − t ) = 240
6v − 40t − v t = 0
Rearrange to make t the subject (check this for yourself )
6v
t= hours.
40 + v
Converting the answer to minutes, the journey using the new locomotive is
shorter by
360v
minutes.
40 + v
Does this answer look right? One way we can check it is to pick a very simple
case—one we can easily work out the answer for—and compare it with what
our formula tells us. If they agree it won’t prove the formula is correct, but it
will be reassuring; if they don’t it will tell us we’ve slipped up somewhere.
We could put v = 0, i.e. no increase in the speed. The formula tells us t = 0,
as expected but any formula whose numerator is a multiple of v would have
told us the same thing. It does little to reassure. A better check would come
from, say, doubling the speed, i.e. v = 40.
6×40
This gives t = = 3 hours, which is what we’d expect: doubling the speed
80
halves the journey time.

A.19 Simultaneous equations in more than two

variables
Take this equation.
x +y +z =5
It describes a relationship, a connection between three numbers x, y and z.
We don’t know what their values are; individually they could be anything, but
the equation stipulates that they must add up to 5. Some examples would
be: x = 3, y = 1, z = 1, or x = 0, y = 105, z = −100. We say these are two
possible solutions of the equation x + y + z = 5.
It’s sometimes useful to think of as a solution as a point. If the equation has
one variable it’s a point on a line; if it has two variables it is point in a plane
(a 2-dimensional space); if it has three variables it’s a point in 3-dimensional
space. So for our equation x + y + z = 5, the two possible solutions we
identified we can think of as the points (3, 1, 1) and (0, 105, −100). For any
equation in more than one dimension here is an infinite number of possible
solutions.
An equation can have any number of variables: people who use
mathematics in their work (Scientists, Engineers, Economists etc) make use
of equations having many variables. One variable might be a stress forces
within a structure, or the speed of airflow over a section of a plane’s wing, or
the production cost of a car component. The list is endless. In some
modelling contexts, hundreds of variables are involved, for example the
scheduling of aircraft and crew by an airline. Each equation represents a
restriction on some numbers, such as the total number of hours worked by a
member of the flight crew, or the maximum number of miles a plane can fly
without re-fueling. Making decisions, for example about which staff to send
where and with which aircraft, becomes a matter of solving lots of equations.
To be a solution it must satisfy all the equations at the same time; we say
they must be solved simultaneously. At A Level we never go beyond two
variables.
An example of simultaneous equations in four variables
We said that for a unique solution to a set of simultaneous linear equations,

we need the same number of equations as there are variables. In this
example there are four equations in four variables x, y, z and t .
5x + y + z + t = 4
x + 5y + z + t = 4
x + y + 5z + t = 4
x + y + z + 5t = 4
We want to solve these simultaneously, i.e. find the values of x, y, z and t

that satisfy all four equations.
There are various ways we could do this. The key thing to notice is that the
coefficients follow a nice simple pattern; there is a symmetry about them.
This is important because we haven’t learnt hope to solve simultaneous
equations in more than two variables. Exploiting the symmetry allows us to
solve this one.
Let’s play for a minute. If we add all four equations we get

8x + 8y + 8z + 8t = 16
from which we can deduce
x + y + z + t = 2.
This is a beautifully simple relationship between the four variables, one

which is worth recognising, but it doesn’t tell us what any of the individual
variables are.
Let’s go back to the symmetry idea. Suppose we swap the variables x and y.
The first two of the four equations change places and the last two remain the
same. We end up with exactly the same set of equations. By symmetry that
same applies if swap any pair of variables. The only way this can work is if all
the variables are equal, i.e. x = y = z = t .
1
Since theirs total is 2, it follows that x = y = z = t = 2 .
In fact it is not difficult to solve simultaneous equations with any number of
variable, but the lesson to take away is: always look for any symmetries or
special features that make the process simpler than it otherwise would be.

A.20 Derivation of the quadratic formula

The formula by solving the equation ax 2 + bx + c = 0 using the formula is
derived by the method of completing the square.
We begin by making the coefficient of x 2 equal to 1:
ax 2 + bx + c = 0
b c
x2 + x + = 0.
a a
Then we form a perfect square using half7 the coefficient of x:

¶2
b b 2 c
µ µ ¶
x+ − + = 0
2a 2a a
b 2 b2 b 2 − 4ac
µ ¶
4ac
x+ = − =
2a 4a 2 4a 2 4a 2
s
b b 2 − 4ac
x+ = ±
2a 4a 2
p
b 2 − 4ac
= ±
2a
p
b b 2 − 4ac
x = − ±
2a 2a
p
−b ± b 2 − 4ac
x =
2a
b b
7
We can halve a fraction by doubling its denominator:
a
→ 2a . Or we can use the
b
a b
hammer and boot rule: = .
2 2a

A.21 Quadratic formula with algebraic coefficients

To solve this equation:
x 2 + k(k + 1)x + k 3 = 0
using the quadratic formula.
Solution
a = 1, b = k(k + 1), c = k 3 .
We will do it in stages. The discriminant is
b 2 − 4ac = k 2 (k + 1)2 − 4k 3
= k 2 (k 2 + 2k + 1) − 4k 3
= k 4 + 2k 3 + k 2 − 4k 3
= k 4 − 2k 3 + k 2
= k 2 (k 2 − 2k + 1)
= k 2 (k − 1)2
= [k(k − 1)]2
We are going to need the square roots of this expression, i.e. ±k(k − 1).
Now substituting into the quadratic formula
p
−b ± b 2 − 4ac
x=
2a
we have
−k(k + 1) ± k(k − 1)
x =
2
(−k 2 − k) ± (k 2 − k)
=
2
(−k 2 − k) + (k 2 − k) (−k 2 − k) − (k 2 − k)
= or
2 2
−2k −2k 2
= or
2 2
= −k or − k2
In an example like this involving a great may steps it is always worth checking
the answer. In this case the solutions are correct if the quadratic factorises
into (x + k)(x + k 2 ). So let’s see:
(x + k)(x + k 2 ) = x(x + k 2 ) + k(x + k 2 )

= x 2 + k 2 x + kx + k 3
= x 2 + k(k + 1)x + k 3
At the end of a proof like this you will sometimes see the abbreviation QED,
which stands for the latin quod erat demonstrandum (“thus it has been
proved").

A.22 Inverse functions

Finding the inverse of a function can sometimes be very simple. For example
if the function f defines the process "add 5", its inverse function, written f −1 ,
is the process "subtract 5".
Or in mathematical shorthand:
f −1 (t ) = t − 5 or f −1 : t → t − 5
The choice of variable (in this case t ) is arbitrary—we can pick any variable:
it’s the process that we are communicating. We can say that t is merely a place
keeper (or dummy variable). So we could, for example, choose to use y, and
write
f −1 (y) = y − 5 or f −1 : y → y − 5
It’s best to think of this not as a formula, but as a process, a description of
what we have to do to a number.
In fact the convention, perhaps rather confusingly, is to use x in the
description of both f −1 and f :
f (x) = x + 5 or f : x → x +5
and
f −1 (x) = x − 5 or f −1 : x → x − 5.
The inverse of the function g : x → 2x − 5 is less obvious. One way to find it
is to break it down into into steps, and then retrace them, as it were.
If we think of g as being the process from left to right, i.e.
x - ×2 - 2x - −5 - 2x − 5
Then we can find g −1 by “unwinding" this process. We can do this by starting

on the right hand side and inverting each step as we make our way back to
the left hand side. We could do this starting with x again, but to make it clear
this is not the same x as we used for g , we use another variable, say y:
y+5
2
÷2 y +5 +5 y
We could therefore describe g −1 like this
y +5 y +5
g −1 (y) = or g −1 : y →
2 2
Or, remembering that the variable is merely a place keeper—a means of
describing the process, as
x +5 x +5
g −1 (x) = or g −1 : x →
2 2
If the process g means “multiply by 2 and then subtract 5", the process g −1
means “add 5 and then divide by 2".

We can easily check that this works. For example g (10) = 15, and going back
the other way
15 + 5
g −1 (15) = = 10.
2
It is sometimes useful to have picture in your mind when thinking about a
function. The function f operates on each number in the left hand “circle"
to a give a corresponding number in the right hand “circle". The function
f −1 does the same thing in reverse.
f
H

2 7
1 −3 6 2
0 5

H
−1
f
The functions f and g are defined in this way:
f (x) = 2x − 1 and g (x) = x 2 + 3.
(a) Find the values of

(i) f (6) (i) f (0) (ii) f ( 21 ) (iii) f (a 2 ) (iv) g (4) (v) g (−1) (vi) g (y).
(b) Find the roots of these equations

(i) f (x) = 9 (ii) f (x) = 0 (iii) f (x) = −1 (iv) g (x) = 12.
8
(a) (i) 11 (ii) −1 (iii) 0 (iv) 19 (v) 4 (vi) y 2 + 3.
1
(b) (i) 2x − 1 = 9, so x = 5 (ii) 2x = 1, so x = 2 (iii) x = 0
2
(iv) x = 9, so x = 3 or −3.

A.23 Hard problems involving surds

Here are two lovely little problems involving surds, that seem harder than
they really are.
We are required to show that

s p
3+2 2 p
p = 3+2 2
3−2 2
It is not immediately obvious how to establish this but the presence of a surd
in the denominator suggests we rationalise the fraction. So let’s do this first
p
(3 + 2 2)2 p
p = (3 + 2 2)2
32 − (2 2)2
p
because (2 2)2 = 8. The result follows.
We end with a question that is not about rationalising.
Show that
p p p
(a) 7−4 3 = 2− 3
p p p p
(b) 2 3+ 5 = 5+1
Just because we naturally read the statement from left to right doesn’t mean
we have to start on the left when proving it. In this case it is a lot easier to start
on the right.
p
(a) Squaring (2 − 3) we have
p p p
(2 − 3)2 = 4 + 3 − 4 3 = 7 − 4 3.
Now we can take the square root of both sides.

(b) Again, squaring the right hand side
p p p p
( 5 + 1)2 = 5 + 2 5 + 1 = 6 + 2 5 = 2(3 + 5).
The result follows.

A.24 Sets: HCF and LCM

Set intersection and the HCF
Take the two numbers 72 and 420 and for each one we consider the set of its
prime factors; let’s call them A and B . So
A = {2, 2, 2, 3, 3} and B = {2, 2, 3, 5, 7}
If we draw a Venn diagram showing the relationship between these two sets,
it looks like this
A
B
2 2 5
2
3 3 7
and the HCF of 72 and 420 corresponds to the intersection of the the two sets:
A ∩ B.
We can think of algebraic factorisation in the same way. For example if we
define P and Q to be the set of factors of 20b 3 and 15ab 2 respectively, i.e.
P = {2, 2, 5, b, b, b} and Q = {3, 5, a, b, b}
the Venn diagram looks like this:
P
2 Q
2 5 3
b
b b a
and the HCF of 20b 3 and 15ab 2 corresponds to the intersection P ∩Q.

Set union and the LCM

We can use the concept of the union of sets to find the LCM of two or more
numbers (typically the denominators of fractions). We’ll illustrate how it
works with this example:
1 2
3
+ .
20b 15ab 2
Let’s look at the Venn diagram showing the factors of 20b 3 and 15ab 2 (we’ll
call them P and Q):
P = {2, 2, 5, b, b, b} and Q = {3, 5, a, b, b}.
P 2 Q
2 5 3
b
b b a
If we construct the number equal to the product of everything inside P and

Q (the union of the two sets), the numbers 20b 3 and 15ab 2 will both divide
into it. But the clever thing is that because the union only includes the
numbers 5, b and b once, it will be the smallest number they both divide
into, i.e. the LCM.
P ∪Q = {2, 2, 5, 3, b, b, b, a}
LCM = 22 .5.3.b 3 .a = 60ab 3
Now we have the lowest common denominator we can add the fractions
1 2 3a 8b 3a + 8b
3
+ 2
= 3
+ 2
=
20b 15ab 60ab 15ab 60ab 3
Here are some problems for you to test your understanding of the union and
intersection of sets.

Exercise
1. A set containing 3 elements has 8 subsets, if we include the entire set

and the empty set.
(a) How many subsets does a 5-element set have?

(b) What about an N -element set?
2. (i) By reference to the Venn diagrams on the previous pages, explain

carefully why, for any two sets S and T,
(a) S ∩ T ⊆ S
(b) S ⊆ S ∪ T .
(ii) What can we say about the sets S and T if S ∩ T = S ∪ T ?

In Questions 4 and 5 the sets L, M and K are defined by
L = {1, 3, 5, 7}, M = {0, 1, 2, 3, 4} and K = {odd integers > 1}
The symbols N , Z (including Z + and Z − ), Q and R have their usual

meaning. (See Section 1.1)
3. Find
(a) L ∩ M (b) L ∪ M (c) M ∩ K (d ) L ∪ K (e) L ∩ K
4. In each case determine whether the statement is true or false. If it is

false, say why.
(a) M ⊆ Q (b) L ∩ R ⊆ L (c) L ⊆ Q (d) L ∪ R = Q

(e) If W = {x ∈ Z : x > 0}, then M ⊆ W
(f) M ∩ Z− = φ (g) M ⊆ K (h) K ⊆ L
5. Use set language to describe each of these sets:
(a) all integers greater than 20,

(b) all positive real numbers,
p p
(c) all rational numbers between 2 and 3.
6. The sets G,H and I are defines as follows:
G = {x ∈ R : x ≤ 3}
H = {x ∈ R : x > 1}
I = {x ∈ R : 0 ≤ x ≤ 4}
(i) Draw a number line diagram and use it to mark off the regions
representing these sets.
(ii) Describe in terms of G, H and I , each of these sets:
(a) {x ∈ R : 0 ≤ x ≤ 3} (b) {x ∈ R : 1 < x ≤ 4}

(c) {x ∈ R : x ≤ 4} (d) {x ∈ R : x ≥ 0}
(e) {x ∈ R : 1 < x ≤ 3} (f) R

Solutions
1. (a) 32. [Find the number of subsets for 1 and 2 elements and look for
the pattern.]
(b) 2N
2. (i)(a) If S and T don’t overlap S ∩ T = φ, and φ ⊆ every set.

If S and T overlap then S ∩ T ⊆ S (and in this case S ∩ T ⊆ T as well).
If S ⊆ T then S ∩ T = S ⊆ S; if T ⊆ S then S ∩ T = T ⊆ S.
(b) In the first two cases (one is not a subset of the other) it is clear from
the Venn diagrams.
In the third case, if S ⊆ T , then S ∪ T = T , and S ⊆ S ∪ T is equivalent to
the statement S ⊆ T , which is true.
If T ⊆ S then S ∪ T = S and it is equivalent to S ⊆ S.
(Remember that the statement X ⊆ Y includes the possibility that X =
Y .)
(ii) S = T . The statement will not hold for any other configuration.
3. (a) {1, 3} (b) {0, 1, 2, 3, 4, 5, 7} (c) {3}

(d) {odd integers ≥ 1} (e) {3, 5, 7}
4. (a) T.
(b) T.
(c) T.
(d) F. L ∪ R = R.
(e) F. The number 0 is in M but not W .
(f) T. The number 0, 2 and 4 are in M but not K .
(g) F. M ⊆K.
(h) F. Odd numbers bigger than 7 are in K but not L.
5. (a) {x ∈ Z : x > 20}

x > 0}
(b) {x ∈ R : p p
(c) {x ∈ Q : 2 < x < 3}
6. (i) G •
H ◦
I • •
−1 0 1 2 3 4 5
(ii) (a) G ∩ H (b) H ∩ I (c) G ∪ I (d ) H ∪ I (e) G ∩ H (f ) G ∪ H

Pure Maths Guide From Love of Maths

Uploaded by

Copyright:

Pure Maths Guide From Love of Maths

Uploaded by

Document Information

Original Title

Copyright

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Pure Maths Guide From Love of Maths

Uploaded by

Copyright:

This book is available (in whole or by chapter) in pdf format via digital

download from our website loveofmaths.co.uk.

"Sometimes a theorem can just be so elegant. It can draw together.

1 FUNDAMENTAL CONCEPTS 1-1

2 INDICES AND SURDS 2-1

2.8 Surds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-51

3 BRACKETS AND FACTORISING 3-1

4 ALGEBRAIC DIVISION AND FACTOR THEOREM 4-1

5 COORDINATE GEOMETRY 5-1

PURE MATHS: AS LEVEL GUIDE 2

6 LINEAR RELATIONSHIPS 6-1

7 QUADRATIC RELATIONSHIPS 7-1

8 SKETCHING AND TRANSFORMING CURVES 8-1

PURE MATHS: AS LEVEL GUIDE 3

9 BINOMIAL EXPANSION 9-1

10 CALCULUS: DIFFERENTIATION AND INTEGRATION 10-1

11 EXPONENTIAL EQUATIONS AND LOGARITHMS 11-1

PURE MATHS: AS LEVEL GUIDE 4

14 GRANDE FINALE 14-0

A Notes and points of interest A-1

0.1 Is this book for you?

• Fractions, decimals, percentages; ratios and proportions.

• The geometry of straight lines, triangles and circles.

• Linear equations. Rearranging simple formulae. Simple quadratic

• Pythagoras’ Theorem. Basic trigonometry and SOHCAHTOA.

• Coordinate axes, but not the equation of a straight line.

0.2 Navigating your way around

Ch Title Topics covered

PURE MATHS: AS LEVEL GUIDE 0-2

Ch Title Topics covered

PURE MATHS: AS LEVEL GUIDE 0-3

To make it easy to pick out crucial material

• Footnotes, where we think a short comment could be useful or

• References, usually by page number, reminding you about something

• Harder worked examples as part of the explanation of the concepts;

• Intriguing (and sometimes amazing!) diversions that relate in some

PURE MATHS: AS LEVEL GUIDE 0-4

...between double lines like this

It is impossible to overstate the importance of Exercises. Whether the ones

We complete this introduction with some practical tips on solving Maths

PURE MATHS: AS LEVEL GUIDE 0-5

0.3 How your teacher solves Maths problems

We start with a familiar scenario. You have been attempting some

PURE MATHS: AS LEVEL GUIDE 0-6

• looking for a familiar pattern or relationship ;

• experimenting with writing things in different ways.

To summarise, success with Maths is not a matter of serendipitous

PURE MATHS: AS LEVEL GUIDE 0-7

1.1.1 Natural Numbers

PURE MATHS: AS LEVEL GUIDE 1-2

1.1.2 Prime numbers

the number 1 is not a prime number.

How many prime numbers are there? In fact there is an infinite

PURE MATHS: AS LEVEL GUIDE 1-3

Z = · · · − 4, −3, −2, −1, 0, 1, 2, 3, 4, . . .

Z+ = 1, 2, 3, 4, . . . and Z− = −1, −2, −3, −4, . . .

Notice that Z+ and Z− combined do not make up Z, because the number 0

PURE MATHS: AS LEVEL GUIDE 1-4

1.1.4 Rational Numbers

Most decimals that come to mind are rational numbers.

PURE MATHS: AS LEVEL GUIDE 1-5

Recurring5 decimals are also rational, a simple example being

If you want to have a go at a slightly trickier one, try 0.094̇28571̇.