Contents 

Maths for Science

Version 1.3.0
30th January 2005

Back I 1
Contents 

Contents
Introduction 6 1.3.4 Roots and fractional exponents . 50
1.4 Doing calculations in the right order 53
1 Starting Points 8 1.5 Learning outcomes for Chapter 1 . . 57
1.1 Numbers . . . . . . . . . . . . . . . 9
1.1.1 Different types of number . . . . 10 2 Measurement in Science 58
1.1.2 Calculating with negative numbers 15 2.1 Large quantities and small quantities 59
1.1.3 Working with negative numbers
2.2 Units of measurement . . . . . . . . 65
on a calculator . . . . . . . . . . 20
2.3 Scales of measurement . . . . . . . 74
1.1.4 The number zero . . . . . . . . 23
2.3.1 Logarithmic scales in practice . 80
1.2 Fractions . . . . . . . . . . . . . . . 24
1.2.1 Using fractions . . . . . . . . . 24 2.4 How precise are the measurements? 83
1.2.2 Adding and subtracting fractions 27 2.5 Learning outcomes for Chapter 2 . . 89
1.2.3 Manipulating fractions . . . . . 31
1.2.4 Multiplying fractions . . . . . . 33 3 Calculating in Science 90
1.2.5 Dividing fractions . . . . . . . . 35 3.1 Calculating area; thinking about
1.3 Powers, reciprocals and roots . . . . 39 units and significant figures . . . . . 91
1.3.1 Powers . . . . . . . . . . . . . . 39 3.1.1 Units in calculations . . . . . . 92
1.3.2 Multiplying and dividing with 3.1.2 Significant figures and rounding
powers . . . . . . . . . . . . . . 45 in calculations . . . . . . . . . . 93
1.3.3 Powers of powers . . . . . . . . 48 3.2 Calculating in scientific notation . . 97

Back J I 2
Contents 
3.2.1 Calculating in scientific notation 4.3 Combining equations . . . . . . . . 203
without a calculator . . . . . . . 97 4.4 Putting algebra to work . . . . . . . 212
3.2.2 Using a calculator for scientific 4.4.1 Algebra is fun! . . . . . . . . . 213
notation . . . . . . . . . . . . . 100 4.4.2 Using algebra to solve scientific
3.3 Estimating answers . . . . . . . . . 102 problems . . . . . . . . . . . . 217
3.4 Unit conversions . . . . . . . . . . . 106 4.5 Learning outcomes for Chapter 4 . . 233
3.4.1 Converting units of area . . . . . 107
3.4.2 Converting units of volume . . . 110 5 Using Graphs 234
3.4.3 Converting units of distance, 5.1 Graphical representations . . . . . . 235
time and speed . . . . . . . . . 113 5.1.1 Bar charts and histograms . . . . 235
3.4.4 Concentration and density; more
5.1.2 Graphs . . . . . . . . . . . . . . 238
unit conversions . . . . . . . . . 125
5.2 Straight-line graphs . . . . . . . . . 248
3.5 An introduction to symbols, equa-
5.2.1 The gradient of a straight-line
tions and formulae . . . . . . . . . . 131
graph . . . . . . . . . . . . . . 249
3.5.1 What do the symbols mean? . . 132
3.5.2 Which symbols are used . . . . 133 5.2.2 Dependent and independent
variables . . . . . . . . . . . . . 258
3.5.3 Reading equations . . . . . . . . 136
3.5.4 Using equations . . . . . . . . . 141 5.2.3 Interpreting straight-line graphs
3.6 Learning outcomes for Chapter 3 . . 153 and gradients . . . . . . . . . . 260
5.3 The equation of a straight line . . . . 265
4 Algebra 154 5.3.1 Proportional quantities . . . . . 265
4.1 Rearranging equations . . . . . . . . 155 5.3.2 A general equation for a straight
4.2 Simplifying equations . . . . . . . . 183 line . . . . . . . . . . . . . . . 270
4.2.1 Simplifying algebraic fractions . 183 5.4 Graphs of different shapes . . . . . . 276
4.2.2 Using brackets in algebra . . . . 192 5.5 Learning outcomes for Chapter 5 . . 293

Back J I 3
Contents 
6 Angles and trigonometry 294 8.1.3 Expressing probability . . . . . 390
6.1 Measuring angles: degrees and radians 296 8.1.4 Combining probabilities . . . . 391
6.2 A quick look at triangles . . . . . . 309 8.1.5 Probability ratios . . . . . . . . 398
6.3 Calculating with angles: trigonometry 312 8.2 Descriptive statistics . . . . . . . . . 405
6.3.1 Using a calculator for trigonom- 8.2.1 Repeated measurements . . . . . 406
etry . . . . . . . . . . . . . . . 316 8.2.2 The distribution of repeated
6.3.2 Using trigonometry in science . 322 measurements . . . . . . . . . . 407
6.4 Small angle approximations . . . . . 338 8.2.3 Mean and standard deviation for
6.5 Learning outcomes for Chapter 6 . . 345 repeated measurements . . . . . 410
8.2.4 Using a calculator for statistical
7 Logarithms 346 calculations . . . . . . . . . . . 417
7.1 Logarithms to base 10 . . . . . . . . 347 8.2.5 How likely are particular results? 419
7.2 Logarithmic scales revisited . . . . . 352 8.2.6 Different types of ‘average’ . . . 422
7.3 Rules of logarithms . . . . . . . . . 355 8.2.7 Samples and populations . . . . 427
7.4 Using logarithms to make curves 8.3 Learning outcomes for Chapter 8 . . 432
straight . . . . . . . . . . . . . . . . 364
7.4.1 Log–log graphs . . . . . . . . . 364 9 Statistical hypothesis testing 433
7.4.2 Log–linear graphs . . . . . . . . 370 9.1 The principles of hypothesis testing . 435
7.5 Logarithms to base e . . . . . . . . 373 9.2 Deciding which test to use; levels of
7.6 Learning outcomes for Chapter 7 . . 382 measurement . . . . . . . . . . . . 443
9.3 The chi-square test . . . . . . . . . 446
8 Probability and descriptive statistics 383 9.4 The Spearman rank correlation coef-
8.1 Chance and probability . . . . . . . 384 ficient . . . . . . . . . . . . . . . . 457
8.1.1 Calculating probability . . . . . 385 9.5 The t-test for unmatched samples . . 469
8.1.2 Probability and common sense . 388 9.6 Other statistical tests . . . . . . . . 476

Back J I 4
Contents 
9.7 Learning outcomes for Chapter 9 . . 477 10.2.4 Differentiating sums . . . . . . . 498
10.2.5 Second derivatives . . . . . . . 500
10 Differentiation 478 10.3 Differentiating exponential functions 505
10.1 Drawing tangents to curves . . . . . 479 10.4 Learning outcomes for Chapter 10 . 511
10.2 An introduction to differentiation . . 483
A Resolving vectors 512
10.2.1 The principles of differentiation 483
10.2.2 Differentiation by rule . . . . . . 487 Glossary 520
10.2.3 Using different symbols and dif-
ferent notation . . . . . . . . . . 493 Index 831

Back J I 5
Contents 

Introduction

Welcome to Maths for Science. There are many reasons for studying maths and
a compelling motivation for many people is that it provides a way of representing
and investigating the nature of the real world. Real world contexts could include
population statistics, or economics, or engineering. Here, the context is ‘science’ in
its broadest sense.
Much of science is couched in the language of mathematics. Nearly all courses
in science will assume some mathematical skills and techniques. It is clearly not
possible for Maths for Science to discuss all the mathematical techniques you might
need to pursue your study of science to degree level, but by the end of it you will
have acquired a good array of basic mathematical tools and confidence in using
them. Equally importantly, you will have a foundation that should make it much
easier to learn further mathematics if and when required.
Maths is in some sense a language with its own alphabet, vocabulary and ‘rules
of grammar’. With any language the only route to fluency is use and practice, but

Back J I 6
Contents 
eventually the process of constructing or understanding sentences becomes auto-
matic and one can then concentrate wholly on the message behind the words. You
should aim to develop a similar confidence and fluency in carrying out certain im-
portant mathematical operations. There are few shortcuts: the route requires prac-
tice, practice and more practice! Keep paper, a pencil and your calculator to hand
as you study, and use them constantly. You may find it helpful to write out notes
and even to rework some of the examples given in the text as you go along. You
will see that there are lots of questions seeded through the text and at the ends of
sections; you should work through each question as you reach it. Links are pro-
vided to the solutions, but don’t be tempted to look at these until you have made a
serious attempt at working out the answer for yourself. If you have solved all parts
of a question successfully on your own, then you are ready to move on.
The focus of Maths for Science is maths and not science, so you are not expected
to bring specific prior knowledge of any particular branch of science. However,
most of the examples and questions involve the application of mathematical tools
to a real scientific purpose, so you will probably discover some interesting science
along the way. Enjoy the journey!

Back J I 7
Contents 

Starting Points 1
The point to start from is always what you already know. It is assumed that you
are familiar with the everyday usage of the basic arithmetic operations of addition,
subtraction, multiplication, division and the use of a calculator to carry them out,
decimal notation (e.g. for money), the representation of an idea by a formula (such
as Einstein’s famous E = mc2 ), and the interpretation of information on a chart
or graph (of the kind that might, for instance, accompany a TV news item about
economic trends). Beyond that, you will find that many of the early chapters begin
with a little revision of ideas and skills that you will probably already have met.
This chapter, which concentrates on ideas about numbers – including fractions and
powers – and the use of your calculator, is slightly different from later ones in that
it covers concepts that are the basis for what is to follow in the rest of the course,
so more of it may constitute revision.

Back J I 8
Contents 
If the points covered in the rest of this chapter are completely familiar, you need not
spend very long on them, but they are worth checking out thoroughly as they are the
foundation of much that is to come later in Maths for Science. Even if it is only for
the sake of revision, make sure you understand all the emboldened terms and test
your own skills against the learning outcomes by doing the numbered questions.
If any of the material is new to you, time spent mastering it now will pay rich
dividends later.

1.1 Numbers
‘Numbers rule the universe’ (Pythagoras)
Numbers are the bedrock of mathematics, underlying measurement, calculation and
statistics, among other branches of maths. Everybody is familiar with the counting
numbers (1, 2, 3, etc.), but scientists also make use of other kinds of numbers, so it
is appropriate to begin this course with some revision of numbers of various sorts
and the ways in which they may be combined.

Back J I 9
Contents 
1.1.1 Different types of number

One convenient way to represent numbers is on a ‘number line’, as shown in Fig-

ure 1.1. A ‘step’ to the right is taken by adding 1 to the previous number and a step
to the left by subtracting 1. Positive and negative whole numbers, including zero,
are called integers.

negative numbers zero positive numbers

−5 −4 −3 −2 −1 0 1 2 3 4 5

Figure 1.1: A number line showing the integers from −5 to 5.

Fractions (formed by dividing one integer by another) and their equivalent decimal
numbers fit on the number line between the integers. For example, (i.e. 0.5) is
halfway between 0 and 1, and −2.5 is halfway between −2 and −3. A number in
which there is a decimal point (e.g. 0.5, 2.5, 100.35, etc.) is said to be written in
decimal notation.

Back J I 10
Contents 

Figure 1.2 shows part of a thermometer. The inset portion

covers a range from about +4.4 ◦ C to −5.6 ◦ C, which might
4
represent the variation in temperature over a 24-hour period
during the winter in the UK. 3

This illustrates how subdivision of the number line forms 2

the basis of a scale for measuring physical quantities that
1
can vary continuously. In this case, the scale between the
integral values is divided into tenths. (Note that, in order to 0
describe a physical quantity the numerical value has to be
−1
accompanied by a unit of measurement, in this case the de-
gree Celsius. This aspect of measuring is covered in Chap- −2
ters 2 and 3.)
−3
In the case of a fraction such as 213
25 , the decimal equivalent
−4
is exact to two places of decimals (i.e. two digits after the
decimal point): −5

213 (a) (b)

= 8.52
25
Figure 1.2: Part of a thermometer.
This decimal equivalent of 21325 cannot be given to more than
two places of decimals except by putting zeros on the end
(e.g. 8.520 000), so it is said to terminate at the digit 2.

Back J I 11
Contents 
However, if you work out a fraction like 13 on your calculator you will get a decimal
like 0.333 333 333 (the number of digits displayed will depend on the make of your
41
calculator). 333 will come out as 0.123 123 123, and 70 9 as 7.777 777 778. These
decimals in fact recur (i.e. repeat themselves) for ever, so they are called infinite
recurring decimals. The calculator truncates them when it runs out of digits on the
display, and in the case of the final example also ‘rounds up’ the last digit from a
7 to an 8. In scientific calculations, it is usually totally inappropriate to quote so
many digits after the decimal point and in Chapter 2 we will consider the rules for
deciding how to round off such numbers in real situations.
Fractions and decimals are grouped together as the so-called rational numbers. All
the rational numbers result in a decimal that either terminates or recurs. How-
ever, there are also numbers whose decimal equivalent neither terminates nor re-
curs. These numbers cannot be obtained by dividing one integer
√ by another, so they
are called irrational numbers. Well-known examples are 2 (the number that mul-
tiplied by itself gives 2, said as ‘the square root of 2’) and π, which is defined as
the number obtained by dividing the circumference of a circle by its diameter. This
would be an appropriate moment to check that you know how to use the π button
on your calculator. You should be able to get:

2 × π = 6.283 185 307

Back J I 12
Contents 
Note that as there are so many makes of scientific and graphics calculators on the
market, each operating differently, it is impossible to state the exact sequence of
keystrokes you will need to carry out particular calculations. Whenever you meet
a new type of mathematical operation, you should therefore check that you know
how to perform it on your own calculator and refer to the manufacturer’s instruction
book if necessary. A calculator symbol in the margin will alert you to the points at
which you particularly need to carry out this kind of check.
All the integers, rational and irrational numbers can be placed somewhere on the
number line, so they are grouped together as the real numbers. All the numbers
you will use in this course will be real. However, it may interest you to know
that there are also imaginary numbers based on the square root of minus 1, which
is usually represented by the symbol i. Numbers made up of real and imaginary
parts, such as (3 + 2i) are known as complex numbers. Complex numbers are used
quite extensively in science and have practical applications in telecommunications,
electrical engineering and the beautiful patterns of fractals.
In case hearing about all these different types of numbers leads you to think that
straightforward ‘counting numbers’ hold little interest for scientists, Box 1.1 shows
how a series of numbers, which mathematicians find interesting in their own right,
have also been found to describe intricate patterns of plant growth.

Back J I 13
Contents 

Box 1.1 Fibonacci numbers

The sequence of numbers

0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89 . . .

was first defined in 1202 by the Italian mathematician

Leonardo of Pisa, nicknamed Fibonacci. Each term in
the sequence after the first two is obtained by adding
together the previous two (0+1 = 1; 1+1 = 2; 1+2 = 3;
2 + 3 = 5, etc.)
It is intriguing to discover that the spiral patterns of plant
growth correspond to pairs of numbers in this series, as
illustrated in Figure 1.3.
Part (a) shows a pinecone with 8 parallel rows of bracts
spiralling gradually and 13 parallel rows of bracts spi-
ralling steeply.
Part (b) shows a sunflower head in which the seeds spi- Figure 1.3: Fibonacci numbers in nature.
ral out from the centre: 55 rows clockwise and 89 rows
anticlockwise.

Back J I 14
Contents 
1.1.2 Calculating with negative numbers

Many scientific quantities can take negative values. For example, chemical reac-
tions may either give out heat to the surroundings or absorb heat from the surround-
ings. Scientists adopt a convention that in the case of a heat-absorbing reaction, the
change in energy has a positive value. In the case of a heat-releasing reaction (such
as combustion), on the other hand, the energy change is negative. To be able to
handle quantities like this in scientific calculations it is essential to understand the
rules for performing the arithmetic operations (addition, subtraction, multiplication
and division) when negative numbers are involved. If I amalgamate a credit card
debt of £100 with an overdraft of £150, I owe £250 in total:

£100 debt + £150 debt = £250 debt

Just in terms of numbers, this is equivalent to writing:

(−100) + (−150) = −250

Note from this example how brackets can be used to make it clear how numbers and
signs are associated. The rules for performing arithmetic operations with negative
numbers are summarized by the examples in the box ‘Arithmetic with negative
numbers’. You should check that you are familiar with all the rules exemplified in
the box.

Back J I 15
Contents 
Arithmetic with negative numbers
The numbers used as examples here are small integers between 1 and 10, but
could of course be any number. As is normally the case, positive numbers are
not preceded by a + sign.
(−3) + 5 = 2 3 + (−4) = −1 (−3) + (−3) = −6
(−5) − 2 = −7 4 − (−3) = 7 (−5) − (−4) = −1
(−2) × 2 = −4 3 × (−2) = −6 (−2) × (−2) = 4
(−3) ÷ 3 = −1 3 ÷ (−3) = −1 (−3) ÷ (−3) = 1

Thinking about some of the examples in concrete terms may help to make sense of
them. For instance, taking money from a bank account that is already overdrawn
increases the amount of the debt (i.e. makes it ‘more negative’). Doubling an
overdraft produces an even larger debt (i.e. a bigger negative number).
Brackets are included to associate negative signs with particular numbers. For ex-
ample, 3 + (−4) means that (−4) is being added to 3; this is equivalent to subtracting
4 from 3, with the result (−1).
Before reading on, test your understanding of the rules by doing Question 1.1.

Back J I 16
Contents 
Question 1.1
Without using your calculator, work out:
(a) (−3) × 4 Answer
(b) (−10) − (−5) Answer
(c) 6 ÷ (−2) Answer
(d) (−12) ÷ (−6) Answer

The examples given so far illustrate one important feature of both addition and
multiplication: both these operations are commutative. This is just the mathemat-
ical way of saying that if one adds two numbers then the result (called the sum) is
identical whichever number is written first. For example:

5 + 3 = 8 and 3 + 5 = 8
(−2) + 3 = 1 and 3 + (−2) = 1

Similarly, in multiplying two numbers the result (called the product) is unchanged
if the order of the numbers is reversed. For instance:

5 × 4 = 20 and 4 × 5 = 20
(−3) × 4 = −12 and 4 × (−3) = −12

Back J I 17
Contents 
Subtraction and division, on the other hand, are not commutative:

5 − 3 = 2 but 3 − 5 = −2
8 ÷ 4 = 2 but 4 ÷ 8 = 1
2

The commutativity of addition and multiplication may seem rather obvious when
applied to the counting numbers, but is worthy of attention because of its impor-
tance in the algebraic manipulations that will be discussed in Chapter 4.
Worked example 1.1 and Question 1.2 are two rather more realistic examples re-
quiring the use of arithmetic with negative numbers.

Back J I 18
Contents 
Worked example 1.1
One of the hottest places on Earth is Death Valley, California, where an air
temperature of 56 ◦ C has been recorded. Probably the coldest inhabited place is
the Siberian village of Oymyakon, where the temperature has fallen to −72 ◦ C.
What is the difference in temperature between these two extremes?
Answer
The difference in temperature may be worked out in two ways. The first
method involves subtracting the lower temperature from the higher, i.e. 56 ◦ C −
(−72 ◦ C), which gives a positive difference of 128 Celsius degrees. This
is the amount by which Death Valley is hotter than Oymyakon. Alterna-
tively, it is equally valid to subtract the higher temperature from the lower, i.e.
−72 ◦ C−56 ◦ C, which gives a negative difference of −128 Celsius degrees. This
is equivalent to saying that Oymyakon is 128 Celsius degrees colder than Death
Valley.
This example shows that in scientific calculations involving negative numbers it
is important to keep the physical situation in mind.

Back J I 19
Contents 
Question 1.2 Answer
The maximum temperature range within the oceans is 31.9 Celsius degrees. This
is a much smaller variation in temperature than that achievable for the air above
a landmass, in part because the lowest ocean temperature is fixed at the tem-
perature at which seawater freezes. The highest recorded ocean temperature is
30.0 ◦ C. What is the freezing point of seawater?

1.1.3 Working with negative numbers on a calculator

The calculations in Questions 1.1 and 1.2 were easy enough to work out by hand,
but many of the calculations you will encounter in science will require the use of a
calculator. It is therefore important to check that you know how to input negative
numbers into your own calculator.
Take the following examples:

6 + (−8) = −2
4 − (−3) = 7
5 × (−3) = −15
(−8) ÷ (−2) = 4

and make sure that you can carry out each sum on your calculator, obtaining the
correct sign on the display of the answer. With some makes of calculator you will

Back J I 20
Contents 
be able to enter the expression on the left-hand side more or less as it is written,
with or without brackets. With other makes you may have to use a combination of
the arithmetic operation keys and the +/− (or on some makes ±) button.
When you are confident that you can input negative numbers in association with the
first arithmetic operations, test your skill with Question 1.3.

Question 1.3
Making sure you input all the signs, use your calculator to work out the follow-
ing:
(a) 117 − (−38) + (−286) Answer
(b) (−1624) ÷ (−29) Answer
(c) (−123) × (−24) Answer

There is, however, one case in which the calculator does not fully deal with signs,
and that case concerns square roots. The ‘square root of 9’ is defined as the number
that multiplied by itself gives 9. One such number is 3:

3×3=9
√
and if you use your calculator to work out 9 you will indeed obtain the answer 3.
However, it is also true that

−3 × −3 = 9

Back J I 21
Contents 
√ root of 9 is either +3 or −3. It is a mathematical convention that the
So the square
notation 9 means ‘the positive value of the square root of 9’, and this is what your
calculator displays. In cases in which the negative value of the square root might
is indicated by use of the sign ± (plus or minus) before the square
be relevant this √
root sign, i.e. ± 9.
√
In Section 1.1.1, the number 2 was given as an example of an irrational number.
Check that you can use the square root button on your own calculator to get
√
2 = 1.414 213 562
(You may obtain more or fewer digits depending on the make and model of your
calculator. The fact that the number is irrational means that in any case it never
ends.)

Question
√
5
What is ?
3
Answer
√
5
= 0.745 355 922
3

Be sure to check that you can obtain this value on your own calculator, by ensuring
that the calculator takes the square root of 5 before dividing by 3. Otherwise, you

Back J I 22
Contents 
will get the positive value of the square root of 53 , which is not the same at all!
r
5
= 1.290 994 449
3

1.1.4 The number zero

Zero is a number to be careful about, especially when it is used in multiplication or

division.
If you try multiplying 0 by 6 on your calculator, you will get the answer 0. This
is hardly surprising. If we start off with nothing, it doesn’t matter how often we
multiply it, we still have nothing. The commutativity of multiplication shows that
6 × 0 is therefore also equal to 0, and your calculator will confirm this.

The result of multiplying any number by 0 is 0.

In a similar way, dividing 0 by any non-zero number gives zero.

Trying to divide by zero is more problematic. If you enter 6 ÷ 0 into your calcu-
lator, you will get an error message. To understand why, imagine dividing 6 by
successively smaller and smaller numbers: the answers will get successively larger
and larger. The number by which we’re dividing approaches zero, the result of the
division becomes too large for the calculator to cope with. Dividing by zero does
not produce a meaningful number and is to be avoided!

Back J I 23
Contents 
1.2 Fractions
With the increasing decimalization of everyday units of measurement, we use frac-
tions less than people used to. Nowadays adding eighths and sixteenths of inches
is about as much as you might need to do, and that only if you still have a ruler, or
some items in a toolbox, marked in inches. However the ability to add, subtract,
multiply and divide using numerical fractions is extremely important in Maths for
Science, because it is the basis for the skill of manipulating algebraic fractions
which will be discussed in Chapter 4.

1.2.1 Using fractions

Fractions are characterized by a numerator (the number on top) and a denomina-

tor (the number on the bottom). So in the fraction 38 , the numerator is 3 and the
denominator is 8.

Back J I 24
Contents 
A pictorial representation, such as that in Figure 1.4, makes it obvious
that it is possible to have fractions which have different numerators and 1
2
denominators, but are nevertheless equal. The cake can be divided into
two and the shaded half further sub-divided into two quarters or four
eighths, but half the cake still remains shaded. So the fractions 12 , 24 and
4
8 all represent the same amount of the original cake, and can therefore
be described as equivalent fractions.
2
Figure 1.4 exemplifies the most fundamental rule associated with frac- 4
tions:

The value of a fraction is unchanged if its numerator and denomina-

tor are both multiplied by the same number, or both divided by the
same number. 4
8

In the case of the half cake, numerator and denominator have been mul-
tiplied by 2 to get the equivalent two quarters and again to get the equiv-
alent four eighths. In the following example of equivalent fractions,
other multiplying and dividing numbers have been used:
Figure 1.4: Sharing out half a
6 2 8 10
= = = cake.
9 3 12 15
2
3 is the simplest form in which this fraction may be expressed, i.e. the one in which
the numerator and denominator have the smallest possible value.

Back J I 25
Contents 
A percentage means a ‘number of parts per hundred’, so is equivalent to a fraction
50
in which the denominator is 100. For example, 50% is the same as 100 or 12

Question
Express 35% as a fraction of the simplest possible form.
Answer
35
35% is the same as 100 . The value of the faction will be unchanged if the nu-
merator and denominator are both divided by the same number, and 35 and 100
can both be divided by 5. Doing this gives
35 7
=
100 20
This is the simplest form in which the fraction can be expressed.

One way to convert a fraction to a percentage is to multiply top and bottom of the
fraction by whatever number is required to make the denominator equal to 100. For
instance:
1 1 × 25 25
= =
4 4 × 25 100

1
Hence 4 is equivalent to 25%.

Back J I 26
Contents 
In the first few sections of this course, all fractions have been written in the form 34 .
However, in most maths and science texts, you will find that the alternative form,
3/4, is also very common, so you have to become equally comfortable with both
systems and also have to be able to swap between them at will. From now on,
therefore, both notations will be used.

1.2.2 Adding and subtracting fractions

Suppose we want to add the two fractions shown below:

3 7
+
4 16

We cannot just add the 3 and the 7. The 3 represents 3 ‘quarters’ and the 7 represents
7 ‘sixteenths’, so adding the 3 to the 7 would be like trying to add 3 apples and 7
penguins!

In order to add or subtract two fractions, it is necessary for them both to have
the same denominator (bottom line).

Back J I 27
Contents 
Fractions with the same denominator are said to have a common denominator. In
numerical work, it is usually convenient to pick the smallest possible number for
this denominator (the so-called lowest common denominator). In this example, the
lowest common denominator is 16; we can multiply both top and bottom of the
fraction 43 by 4 to obtain the equivalent fraction 12
16 , so the calculation becomes

3 7 12 7 19
+ = + =
4 16 16 16 16

A top heavy fraction such 19 16 (i.e. one in which the numerator is larger than the
denominator) is sometimes referred to as an improper fraction. We could also write
3
the final answer as 1 16 . This notation is called a mixed number (i.e. a combination
of a whole number and a simple fraction). However for most purposes in this course
it is better to leave things as improper fractions.

Back J I 28
Contents 
If the lowest common denominator is not easy to spot, it is perfectly acceptable to
use any common denominator when adding and subtracting fractions. It may be
most convenient to multiply the top and bottom of the first fraction by the denom-
inator of the second fraction, and the top and bottom of the second fraction by the
denominator of the first. A return to our example may make this clearer:
3 7 3 × 16 7×4 48 28 76
+ = + = + =
4 16 4 × 16 16 × 4 64 64 64

76
However, 64 is not the simplest form in which this fraction can be expressed. We
can divide both the numerator and the denominator by four to obtain 19 16 . Reassur-
ingly, this is the same answer as we obtained before!
This process of dividing the top and bottom of a fraction by the same quantity is
often referred to as cancellation, because it is commonly shown by striking through
5
the numbers being divided. For example, 15 can be simplified by dividing the
numerator and denominator by 3, and this may be shown as

5
1
15
3

Back J I 29
Contents 
Worked example 1.2
Evaluate 32 + 32
1
, giving the answer in the form of the simplest possible improper
fraction.
Note that the instruction to ‘evaluate’ simply means ‘calculate the value of’.
Answer
Choosing 2 × 32 as the common denominator,
3 1 3 × 32 1×2
+ = +
2 32 2 × 32 32 × 2
96 2
= +
64 64
98
=
64
 49
98
=

64
 32

This cannot be simplified any further, so

3 1 49
+ =
2 32 32

Back J I 30
Contents 
Question 1.4
Without using a calculator, evaluate the following, leaving your answers in the
form of the simplest possible fractions.
2 1
(a) − Answer
3 6
1 1 2
(b) + − Answer
3 2 5
5 1
(c) − Answer
28 3

1.2.3 Manipulating fractions

It is very important to remember that multiplying both numerator and denominator

by the same non-zero number, or dividing both numerator and denominator by the
same non-zero number, are the only things you can do to a fraction that leave its
value unchanged. Adding the same number to the numerator and denominator will
alter the value of the fraction, as will any other operations. The following question
will help you to convince yourself of this, so it is particularly important that you
should work through it at this point.

Back J I 31
Contents 
Question 1.5
4
Take any fraction, say 16 , and evaluate it as a decimal, using your calculator if
necessary. Now try each of the following operations in turn, using your calcula-
tor to work out the result:
(a) choose any integer and add it to the numerator and Answer
denominator
(b) subtract the same integer from the numerator and denominator Answer
(c) square the numerator and the denominator (i.e. multiply the Answer
numerator by itself, and the denominator by itself)
(d) take the square root of the numerator and the square root of Answer
the denominator.

The results you obtained for Question 1.5 confirm that, for example, adding the
same non-zero number to the top and bottom of a fraction changes its value, as
do operations such as taking the square root of the numerator and denominator.
The experience of all calculations of this type can be generalized by saying that
excluding operations involving the integer zero,

A fraction is unchanged by either the multiplication, or the division, of its nu-

merator and denominator by the same amount. All other operations carried out
on the fraction will alter its value.

Back J I 32
Contents 
In terms of numerical fractions, this rule may seem fairly obvious. But forgetting
it once the numbers are replaced by symbols is the root cause of many errors in
algebra!

1.2.4 Multiplying fractions

The expression ‘three times two’ just means there are three lots of two (i.e. 2+2+2).
So multiplying by a whole number is just a form of repeated addition. For example,

3×2=2+2+2

This is equally true if you are multiplying a fraction by a whole number:

4 4 4 4 12
3× = + + =
5 5 5 5 5
We could write the 3 in the form of its equivalent fraction 31 and it is then clear that
the same answer is obtained by multiplying the two numerators together and the
two denominators together.
3 4 3 × 4 12
× = =
1 5 1×5 5
In fact, this procedure holds good for any two fractions.

Back J I 33
Contents 
To multiply two or more fractions, multiply the numerators (top lines) together
and also multiply the denominators (bottom lines) together.

So
3 7 3 × 7 21
× = =
4 8 4 × 8 32
Multiplying three fractions together is done by simple extension of the method used
in the previous examples:
7 7 3 7×7×3 147
× × = =
16 8 4 16 × 8 × 4 512

Back J I 34
Contents 
1.2.5 Dividing fractions

How are we to interpret 4 ÷ 12 ? The analogy with dividing by an integer may help.
The expression 4 ÷ 2 asks us to work out how may twos there are in 4 (answer 2).
In exactly the same way, the expression 4 ÷ 12 asks how many halves there are in 4.
Figure 1.5 illustrates this in terms of circles. Each circle contains two half-circles,
and 4 circles therefore contain 8 half-circles. So
1
4÷ =4×2=8
2

Figure 1.5: Each circle contains two half-circles.

Back J I 35
Contents 

Figure 1.6: Each half-circle contains two quarter-circles.

1 1
Similarly, 2 ÷ 4 asks how many quarters there are in a half. Figure 1.6 illustrates
that:
• each whole circle contains 4 quarter-circles
1
• each half-circle contains 2 × 4 quarter-circles
So
1 1 1 1 4 1×4 4
÷ = ×4= × = = =2
2 4 2 2 1 2×1 2
This may be extended into a general rule

To divide by a fraction, turn it upside down and multiply.

Back J I 36
Contents 
So
4 5 4 9
÷ = ×
3 9 3 5
 12
36
=

15
5
12
=
5
Here the cancellation has been done by dividing the numerator and the denominator
of the final answer by 3. However, cancellation could equally well have been carried
out at an earlier stage,

4 9
3 12
× =
3
1 5 5

Note that divisions involving fractions are commonly written in several

, different
4 5 4/3
ways; the example above might equally well have been expressed as or .
3 9 5/9

Back J I 37
Contents 
It is always important to remember that an integer is equivalent to a fraction in
which the numerator is equal to that integer and the denominator is equal to 1: for
example, the integer 3 is equivalent to the fraction 31 . So dividing by the integer 3
is equivalent to dividing by the fraction 13 , and that, according to the general rule
about how to divide by a fraction, is the same as multiplying by the fraction 13 .
1 1 3 1 1 1×1 1
Thus ÷3= ÷ = × = =
2 2 1 2 3 2×3 6
In this context, it may be helpful to restate the general rule in terms of a
specific example:

1
Multiplying by 2 is equivalent to dividing by 2.
1
Dividing by 2 is equivalent to multiplying by 2.

The blue box and the cartoon use the integer 2 as the example, but it
could of course be replaced by any other integer: it is equally true to
1
say that dividing by 10 is equivalent to multiplying by 10.

Back J I 38
Contents 
Question 1.6
Work out each of the following, leaving your answer as the simplest possible
fraction:
2
(a) × 3 Answer
7
5
(b) ÷ 7 Answer
9
1/6
(c) Answer
1/3
3 7 2
(d) × × Answer
4 8 7

1.3 Powers, reciprocals and roots

1.3.1 Powers

Most people are familiar with the fact that 2 × 2 can also be written as 22 (said as
‘two squared’) and 2 × 2 × 2 as 23 (said as ‘two cubed’). This shorthand notation
can be extended indefinitely, so 2 × 2 × 2 × 2 × 2 × 2 becomes 26 (said as ‘two raised
to the power of six’ or ‘two to the power of six’, or more usually just as ‘two to the

Back J I 39
Contents 
six’). In these examples, 2 is called the base number and the superscript indicates
the number of ‘2’s that have been multiplied together. The superscript number is
variously called the exponent, the index (plural indices) or the power. In the rest
of this section, the term exponent will be the one used, because that ties in most
closely with the notation on calculators.
‘Power’ is a slightly confusing term because it is commonly used to denote two
different quantities:
• the value of the superscript number (as in ‘two to the power of six’),
• the complete package of base number and exponent .
The context should make it clear what is meant in any particular example.
In the following example, the base number is 5:

Exponent 1 2 3 4

Power of 5 51 52 53 54
Value 5 25 125 625

If you read this table starting at the right and stepping to the left, each time you take
a step you are subtracting 1 from the number in the top row and dividing the number
in the bottom row by five. On the basis of this pattern, mathematicians extend this
table further to the left by continuing to apply the same ‘rule’ for each step, giving:

Back J I 40
Contents 
Exponent −3 −2 −1 0 1 2 3 4

Power of 5 5−3 5−2 5−1 50 51 52 53 54

1 1 1
Value 125 25 5 1 5 25 125 625

Firstly, note the extremely important result that 50 = 1.

Any base number raised to the power of zero is equal to 1.

1
Next, notice that 5−2 = 1
25 . But since 25 = 52 , 1
25 is also . So we have developed
52
a new form of shorthand such that

1 1 1
5−1 = 5−2 = 5−3 = and so on.
5 52 53

Another way of saying this is that 5−2 is the reciprocal of 52 . The reciprocal of any
number is 1 divided by that number. Note that this also works the other way round:
1
52 is the reciprocal of 5−2 . This means that −2 = 52 .
5

Back J I 41
Contents 
The system shown above for powers of 5 could be applied to any base number,
and is especially useful when applied to powers of ten, because then it ties in with
our normal system for writing decimal numbers. In the example below, the table is
constructed the other way round to emphasise this:

thousands hundreds tens units point tenths hundredths thousandths

Value 1000 100 10 1 . 0.1 0.01 0.001

Power of 10 103 102 101 100 10−1 10−2 10−3
Exponent 3 2 1 0 −1 −2 −3

In the next chapter, you will see how useful this powers of ten notation can be in
scientific work.

Back J I 42
Contents 
Question 1.7
Without using a calculator, evaluate
(a) 2−2 Answer
1
(b) Answer
3−3
1
(c) 0 Answer
4
1
(d) Answer
104

Your calculator probably has an x2 button, and either an x−1 or a 1/x button, but
to evaluate other powers you will have to use a special ‘powers’ button. On some
calculators this is marked xy , on others it has the symbol ∧. To input a negative
exponent, you may have to combine the powers button with the +/− button. Make
sure at this point that you can operate your own calculator to obtain correctly:

54 = 625
5−1 = 0.2 (i.e. 1/5)
5−2 = 0.04 (i.e. 1/25)

Back J I 43
Contents 
Question 1.8
Use your calculator to evaluate:
(a) 29 Answer
(b) 3−3 Answer
1
(c) Answer
42

Box 1.2 An intimate knowledge of powers!

Srinivasa Ramanujan (1887–1920), an Indian mathematician of immense talent,
came to England in 1913 at the invitation of the distinguished British mathe-
matician, G. H. Hardy. In his biography of Ramanujan, Hardy wrote:
I remember once going to see him when he was lying ill at Putney. I had ridden
in taxi cab number 1729 and remarked that the number seemed to me rather a
dull one, and that I hoped it was not an unfavorable omen. “No,” he replied, “it
is a very interesting number; it is the smallest number expressible as the sum of
two cubes in two different ways.”
Indeed: 1729 = 13 + 123 = 93 + 103

Back J I 44
Contents 
1.3.2 Multiplying and dividing with powers

In scientific calculations, it is very common to have to multiply and divide by

powers, especially powers of ten. It is therefore extremely important to become
confident in manipulating powers in this way, both with and without a calculator.
However, the rules for doing so are quite easy to work out.
Suppose we wanted to multiply 103 by 102 . We could write this out more fully as

103 × 102 = (10 × 10 × 10) × (10 × 10) = 105

The exponent of the result (5) is the same as the sum of the two original exponents
(3 + 2).
The process is of course not limited to powers of ten. It works for any base number.
For example:

22 × 24 = (2 × 2) × (2 × 2 × 2 × 2) = 26

Again, the exponent of the result (6) is the same as the sum of the two original
exponents (2 + 4).
1
The process also works for negative exponents. For example, since 5−2 =
52
1
53 × 5−2 = (5 × 5 × 5) × = 5 = 51
5×5
Back J I 45
Contents 
Adding the exponents here again gives the exponent of the answer:
3 + (−2) = 1

In science and maths, general rules are often stated in terms of symbols. We could
express the rule we have discovered through the above examples in the much more
general form

N a × N b = N a+b (1.1)

where N represents any base number and a and b represent any exponents

Quantities such as those represented by the symbols N, a and b, which can take any
value we choose, are called variables.
The example involving a negative exponent we looked at previously shows immedi-
ately how to extend the rules to cover situations in which we want to divide powers.
We had:
53 × 5−2 = 53+(−2) = 51 = 5
But as you will remember from Section 1.2.5, multiplying by a fraction is the same
as dividing by that fraction turned upside down (i.e. its reciprocal). So multiplying
by 5−2 is the same as dividing by its reciprocal (52 ), and we can write
53 ÷ 52 = 53−2 = 51 = 5

Back J I 46
Contents 
This time, instead of adding the exponents, we have subtracted the second from the
first. More generally,

N a ÷ N b = N a−b (1.2)

where N represents any base number and a and b represent any exponents

Question 1.9
Without using a calculator, simplify the following to the greatest possible extent
(leaving your answer expressed as a power).
(a) 230 × 22 Answer
(b) 325 × 3−9 Answer
(c) 102 /103 Answer
(d) 102 /10−3 Answer
(e) 10−4 ÷ 102 Answer
105 × 10−2
(f) Answer
103

Back J I 47
Contents 
1.3.3 Powers of powers

Consider now what happens when a number which is already raised to a power, for
example 32 , is again raised to a power. Suppose for example 32 is itself cubed, so
 3
that we have 32 . Writing this out in full shows that
 3
32 = (32 ) × (32 ) × (32 ) = (3 × 3) × (3 × 3) × (3 × 3) = 36
This time the exponents have been multiplied together to obtain the exponent of the
answer: 3 × 2 = 6.
More generally,

n
Nm = Nm×n (1.3)

where N represents any base number and m and n represent any exponents

Equation 1.3 applies for all values of N, m and n whether positive or negative. So
for example:
!3 
1 −20 3
 1
20
= 10 = 10(−20)×3 = 10−60 = 60
10 10
This is equivalent to saying that
!3
1 13 1 1
= = =
1020 1020

3 1020×3 1060

Back J I 48
Contents 
Question 1.10
Without using a calculator, simplify the following to the greatest possible extent,
leaving your answer expressed as a power.
 2
(a) 416 Answer
 2
(b) 5−3 Answer
 −1
(c) 1025 Answer
!6
1
(d) 3 Answer
3

Back J I 49
Contents 
1.3.4 Roots and fractional exponents

Finally, how are we to interpret a power with a fractional exponent, such as 21/2 ?
The rule for multiplying powers gives a clue. Suppose we were to multiply 21/2 by
itself. Applying Equation 1.1 suggests that:
1+1
 
21/2
×2 1/2
=2 2 2 = 21 = 2

But√the positive number that multiplied

√ by itself gives 2 is more commonly written
1/2
as 2. The two shorthands, 2 and 2 are often used interchangeably.
Similarly, the
√ number that multiplied by itself three times gives 125 is sometimes
3
written as 125 (said as ‘the cube root of 125’), but more commonly written in
science as (125)1/3 . This number is clearly 5, and you should notice the correspon-
dence:

53 = 125 and conversely (125)1/3 = 5

More generally,

√
n
The positive nth root of a number N can be written as either N or as N 1/n
In practice, the first type of notation is only used when n = 2 or n = 3.

Back J I 50
Contents 
Worked example 1.3
 7
21/2
Without using a calculator, evaluate
1/2
23
Answer
From Equation 1.3
 7 1  1/2 1
21/2 = 2 2 ×7 = 27/2 and 23 = 23× 2 = 23/2
so
 7
21/2 27/2

1/2 = 23/2
23
From Equation 1.2

27/2
3/2
= 27/2 − 23/2
2
= 24/2
= 22
=4

Back J I 51
Contents 
Equation 1.3 can now be used to bring meaning to a number like 272/3 .
Since 32 = 13 × 2, applying Equation 1.3 shows that 272/3 = (271/3 )2 i.e. the square
of the cube root of 27. The cube root of 27 is 3, so 272/3 is equal to 32 or 9.

Question 1.11
Without using a calculator, simplify the following to the greatest possible extent,
expressing your answer as an integer or a decimal.
 1/2
(a) 24 Answer
√
(b) 104 Answer
(c) 1003/2 Answer
(d) (125)−1/3 Answer

Back J I 52
Contents 
1.4 Doing calculations in the right order
In Section 1.1.2, brackets were used to make it clear that the minus signs were
tied to particular numbers. Brackets can also be used to show the order in which
calculations are to be performed.
If a calculation were written as

3+2×5=

should one do the addition first or the multiplication first? Try entering this expres-
sion into your calculator exactly as it is written. Do you get the answer 13? If so,
your calculator knows the convention adopted by mathematicians everywhere that
multiplication takes precedence over addition. The calculator has ‘remembered’ the
3 until it has worked out the result of multiplying 2 by 5 and has then added the 3
to the 10. According to the rules all mathematicians follow, if you wanted to add
the 3 and the 2 first and then multiply that result by 5 you would have to write

(3 + 2) × 5 = 25

Again, check that you can use the bracket function on your calculator to enter this
expression exactly as written on the left-hand side of this equation and that you
obtain the correct answer.
There are similar rules that govern the order of precedence of other arithmetic op-
erations, which are neatly encapsulated in the mnemonic BEDMAS.

Back J I 53
Contents 
Order of arithmetic operations
Brackets take precedence over
Exponents. Then. . .
Division and
Multiplication must be done before. . .
Addition and
Subtraction.

So if we write −3 − 12 ÷ 6, the BEDMAS rules tell us we must do the division

(12 ÷ 6 = 2) before carrying out the subtraction (−3 − 2 = −5). Try this on your
calculator too; you may have to use the +/− button to input the −3.
Many people, including scientists, find it hard to visualize the rules in a string of
numbers. They often opt to use brackets to make things clear, even when those
brackets simply reinforce the BEDMAS rules. So one could choose to write

(12 ÷ 3) + 2 = 6

There is nothing wrong with adding such ‘redundant’ brackets — they are simply
there for clarity and can even be entered into your calculator (try it). Far better to
have a few additional brackets than to be confused about the order in which the
calculation must be carried out!

Back J I 54
Contents 
There is one final quirk associated with the use of brackets. In mathematics, the
multiplication sign is often left out (though its presence is implied) between num-
bers and brackets, and between brackets and brackets. So
2(3 + 1) = 2 × (3 + 1) = 8
and
(1 + 1)(4 + 3) = 2 × 7 = 14
Some calculators ‘understand’ this convention and some do not. Check your own
calculator carefully using the two examples above.
The next operation in precedence after brackets involves exponents. If there are
powers in the expression you are evaluating, deal with any brackets first, then work
out the powers before carrying out any other arithmetical operations.

Question
Evaluate 2 × 32 and (2 × 3)2
Answer
In the first case, there are no brackets so the exponent takes precedence:
2 × 32 = 2 × 9 = 18
In the second case, the bracket takes precedence:
(2 × 3)2 = 62 = 36

Back J I 55
Contents 
Question 1.12
Evaluate (preferably without using your calculator):
(a) 35 − 5 × 2 Answer
(b) (35 − 5) × 2 Answer
(c) 5(2 − 3) Answer
(d) 3 × 22 Answer
(e) 23 + 3 Answer
(f) (2 + 6)(1 + 2) Answer

Back J I 56
Contents 
1.5 Learning outcomes for Chapter 1
After completing your work on this chapter you should be able to:
1.1 carry out addition, subtraction, multiplication and division operations
involving negative numbers;
1.2 add two or more fractions;
1.3 subtract one fraction from another;
1.4 multiply a fraction by an integer or by another fraction;
1.5 divide a fraction by a non-zero integer or by another fraction;
1.6 evaluate powers involving any base and positive, negative or fractional
exponents;
1.7 multiply or divide two powers involving the same base;
1.8 evaluate any given power of a number already raised to a power.

Back J I 57
Contents 

Measurement in Science 2
Observation, measurement and the recording of data are central activities in science.
Speculation and the development of new theories are crucial as well, but ultimately
the predictions resulting from those theories have to be tested against what actually
happens and this can only be done by making further measurements. Whether
measurements are made using simple instruments such as rulers and thermometers,
or involve sophisticated devices such as electron microscopes or lasers, there are
decisions to be made about how the results are to be represented, what units of
measurements will be used and the precision to which the measurements will be
made. In this chapter we will consider these points in turn. Then in Chapter 3
we will go on to think about how measurements of different quantities may be
combined, and what significance should be attached to the results.

Back J I 58
Contents 
2.1 Large quantities and small quantities
Scientists frequently deal with enormous quantities — and with tiny ones. For
example it is estimated that the Earth came into being about four and a half thousand
million years ago. It took another six hundred million years for the first living
things — bacteria — to appear. Bacteria are so small that they bear roughly the
same proportion to the size of a pinhead as the size that pinhead bears to the height
of a four-year old child!
In the previous chapter, we saw how convenient powers of ten could be as a way of
writing down very large or very small numbers. For example,

106 = 1 000 000 (a million) and 10−3 = 1/1000 = 0.001 (a thousandth)

This shorthand can be extended to any quantity, simply by multiplying the power
of ten by a small number. For instance,

2 × 106 = 2 × 1 000 000 = 2 000 000 (two million)

(The quantity on the left-hand side would be said as ‘two times ten to the six’.)
Similarly,

3.5 × 106 = 3 500 000 (three and a half million)

7 × 10−3 = 7/1000 = 0.007 (seven-thousandths)

Back J I 59
Contents 
Scientists make so much use of this particular shorthand that it has come to be
known as scientific notation (although in maths texts you may also find it referred
to as standard index form or standard form.)

A quantity is said to be expressed in scientific notation if its value is written as

a number multiplied by a power of ten. The number can be a single digit or a
decimal number, but must be greater than or equal to 1 and less than 10.

Note the restriction: 75 × 102 is not in scientific notation and nor is 0.75 × 104 ,
though these are both equivalent to 7.5 × 103 which is in scientific notation.
Scientific notation can be defined more succinctly by making use of some of the
mathematical symbols denoting the relative sizes of quantities. These symbols are:
> greater than (e.g. 3 > 2);
≥ greater than or equal to (e.g. a ≥ 4 means that the quantity a may take the
exact value 4 or any value larger than 4);
< less than;
≤ less than or equal to.
Note that ‘a ≥ 4’ and ‘4 ≤ a’ convey exactly the same information about the
quantity a.

Back J I 60
Contents 
Using these symbols, scientific notation may be defined as a notation in which
the value of a quantity is written in the form a × 10 n , where n is an integer and
1 ≤ a < 10.
To move from scientific notation to integers or to decimal notation, first deal with
the power of ten, then carry out the multiplication or division.

Worked example 2.1

Express the following numbers as integers or in decimal notation:
(a) 4.53 × 103
(b) 8.371 × 102
(c) 6.4 × 10−3
Answer

(a) 4.53 × 103 = 4.53 × 1000 = 4530

(b) 8.371 × 102 = 8.371 × 100 = 837.1
1 6.4
(c) 6.4 × 10−3 = 6.4 × = = 0.0064
1000 1000

Note that, as in Worked example 2.1, a requirement to express a quantity in a dif-

ferent form simply involves taking the quantity and writing down its equivalent in

Back J I 61
Contents 
the new form. You may do this in one step, or write down intermediate steps as was
done in the worked example.

Question 2.1
Without using your calculator, express the following numbers as integers or in
decimal notation. Note that (a) and (b) are in scientific notation, while (c) is not.
(a) 5.4 × 104 Answer
(b) 2.1 × 10−2 Answer
(c) 0.6 × 10−1 Answer

Moving from an integer or decimal notation to scientific notation is equivalent to

deciding what power of ten you need to multiply or divide by in order to convert
the number you are starting with to a number that lies between 1 and 10.

Back J I 62
Contents 
Worked example 2.2
Express the following numbers in scientific notation:
(a) 356 000
(b) 49.7 × 104
(c) 0.831
Answer

(a) 356 000 = 3.56 × 100 000 = 3.56 × 105

(b) 49.7 × 104 = 4.97 × 10 × 104 = 4.97 × 10(1+4) = 4.97 × 105
8.31
(c) 0.831 = = 8.31 × 10−1
10

In this worked example, all the steps have been written out in full. You may be able
to manage with fewer steps in your own calculations — just use as many or as few
as you feel comfortable with in order to get the right answer!

Back J I 63
Contents 
Question 2.2
Without using your calculator, express the following numbers in scientific nota-
tion:
(a) 215 Answer
(b) 46.7 Answer
(c) 152 × 103 Answer
(d) 0.000 0876 Answer

It is only too easy to lose track of the sizes of things when using scientific notation,
so you should make a habit of thinking carefully about what the numbers mean,
bearing in mind that numbers may be positive or negative. For example:

−1 × 1010 is a very large negative number;

−1 × 10−10 is a very small negative number;
1 × 10−10 is a very small positive number.

Figure 2.1 places on the number line some numbers in scientific notation. You may
find this helps you to visualize things.

Back J I 64
Contents 
We started this section thinking about the early Earth and the first appearance of life.
Using scientific notation, the age of the Earth can be neatly expressed as 4.6 × 109
years and the size of one type of those early bacteria as 1.2 × 10−6 metres. Of
course the value we come up with for such sizes will depend on the units in which
we choose to make the measurements. If we were measuring the diameter of the
Moon, we could elect to express it in metres or in kilometres, or even in miles.

2.2 Units of measurement

In the UK, two systems of units are in common use. We still use old imperial mea-
sures for some things: milk is sold in pints and signposts indicate distances in miles.
But for many other everyday measurements metric units have been adopted: we buy
petrol in litres and sugar in kilogram bags. A great advantage of metric units is that
we no longer have to convert laboriously from imperial units, such as gallons, feet
and inches, in order to trade with continental Europe. Also, calculations are easier
in a metric (i.e. decimal) system! Similar advantages were the main consideration
when in 1960 an international conference formally approved a standard set of sci-
entific units, thus replacing at a stroke the many different systems of measurement
that had been used up until then by scientists of different nationalities. This ‘univer-
sal’ system for scientific measurement is referred to as SI units (short for Système
International d’Unités).

Back J I 65
Contents 
In SI, there are seven ‘base units’, which are listed in Box 2.1. Surprising as it
may initially seem, every unit for every other kind of quantity (speed, acceleration,
pressure, energy, voltage, heat, magnetic field, properties of radioactive materials,
indeed whatever you care to name) can be made up from combinations of just these
seven base units. For instance, speed is measured in metres per second. You will
find some other combinations of base units described in Chapter 3. In this course
we shall work mainly with the familiar base units of length, mass, time and temper-
ature, and some of their combinations, but it is worth knowing that the other base
units exist as you may meet them in other courses.

Box 2.1 The SI base units

Physical quantity Name of unit Symbol for unit

length metre m
time second s
mass kilogram kg
temperature kelvin K
amount of substance mole mol
electric current ampere A
luminous intensity candela cd

Back J I 66
Contents 
Most of these base units relate to physical descriptions that apply universally.
The SI base unit of time, the second, is defined as the period over which
the waves emitted by caesium atoms under specific conditions cycle exactly
9192 631 770 times. Then the SI base unit of length, the metre, is defined by
stating that the speed of light in a vacuum, which is a constant throughout the
Universe, is exactly 299 792 458 metres per second.

The SI base unit of mass, the kilogram, is the only fundamental unit that is
defined in terms of a specific object. The metal cylinder which constitutes the
world’s ‘standard kilogram’ is kept in France. Note that the kilogram is actually
the standard unit of mass, not of weight. In scientific language, the weight of
an object is the downward pull on that object due to gravity, whereas its mass
is determined by the amount of matter in it. When astronauts go to the Moon,
where the pull of gravity is only about one-sixth of that on Earth, their mass
remains the same but their weight drops dramatically! And in zero gravity, they
experience a condition known as ‘weightlessness’.

The SI base unit of temperature is the kelvin, which is related to the everyday
unit of temperature, the degree Celsius:

(temperature in kelvin) = (temperature in degrees Celsius) + 273.15

(You will find some of the rationale for the kelvin scale of temperature in Chap-
ter 5.)

Back J I 67
Contents 
The amount of a pure substance is expressed in the SI base unit of the mole.
Whatever the smallest particle of a given substance is, one mole of that sub-
stance will contain 6.02211367 × 1023 (known as Avogadro’s number) of those
particles. A mole of graphite contains Avogadro’s number of carbon atoms.
Carbon dioxide is made up of molecules in which one carbon atom is joined to
two oxygen atoms, and a mole of carbon dioxide contains Avogadro’s number
of these molecules.

You will have noticed that while the base unit of length is the metre, not the kilo-
metre, the base unit of mass is the kilogram, not the gram.
It is important to realize that, although in everyday usage it is common to say that
you ‘weigh so many kilos’, there are two things wrong with this usage from the
scientific point of view. First, as noted in Box 2.1, the kilogram is not a unit of
weight, but a unit of mass. (The SI unit of weight, the newton, will be discussed in
Chapter 3.) Secondly, in scientific language, ‘kilo’ is never used as an abbreviation
for kilogram, in the sense of the everyday phrase ‘he weighs so many kilos’. In
science, kilo is always used as a prefix, denoting a thousand: one kilometre is a
thousand metres, one kilogram is a thousand grams.
Another prefix with which everybody is familiar is ‘milli’, denoting a thousandth.
One millimetre, as marked on ordinary rulers, is one-thousandth of a metre; or put
the other way round, a thousand millimetres make up a metre. There are many other
prefixes in use with SI units, all of which may be applied to any quantity. Like kilo
and milli, the standard prefixes are based on multiples of 1000 (i.e. 103 ). The most

Back J I 68
Contents 
commonly used prefixes are listed in Box 2.2.
It is important to write the symbols for units and their prefixes in the correct case.
So k (lower case) is the symbol for the prefix ‘kilo’ whilst K (upper case) is the
symbol for the Kelvin; m (lower case) is the symbol for the metre or the prefix
‘milli’ whilst M (upper case) is the symbol for the prefix ‘mega’.

Box 2.2 Prefixes used with SI units

prefix symbol multiplying factor

tera T 1012 = 1000 000 000 000
giga G 109 = 1000 000 000
mega M 106 = 1000 000
kilo k 103 = 1000
– – 100 = 1
milli m 10−3 = 0.001
micro µ∗ 10−6 = 0.000 001
nano n 10−9 = 0.000 000 001
pico p 10−12 = 0.000 000 000 001
femto f 10−15 = 0.000 000 000 000 001
* The Greek letter µ is pronounced ‘mew’.

The following data may help to illustrate the size implications of some of the

Back J I 69
Contents 
prefixes:
• the distance between Pluto (the furthest planet in the Solar System) and
the Sun is about 6 Tm,
• a century is about 3 Gs,
• eleven and a half days contain about 1 Ms,
• the length of a typical virus is about 10 nm,
• the mass of a typical bacterial cell is about 1 pg.
Astronomers have long been making measurements involving very large quanti-
ties, but scientists are increasingly probing very small quantities. ‘Femtochem-
istry’ is a rapidly developing area, which involves the use of advanced laser
techniques to investigate the act of chemical transformation as molecules collide
with one another, chemical bonds are broken and new ones are formed. In this
work, measurements have to be made on the femtosecond timescale. Ahmed H.
Zewail (whose laboratory at the California Institute of Technology in Pasadena
is often referred to as ‘femtoland’) received the 1999 Nobel Prize in Chemistry
for his development of this new area.

Although scientific notation, SI units and the prefixes in Box 2.2 are universal short-
hand for all scientists, there are a few instances in which other conventions and units
are adopted by particular groups of scientists for reasons of convenience. For ex-
ample, we have seen that the age of the Earth is about 4.6 × 109 years. One way

Back J I 70
Contents 
to write this would be 4.6 ‘giga years’ but geologists find millions of years a much
more convenient standard measure. They even have a special symbol for a million
years: Ma (where the ‘a’ stands for‘annum’, the Latin word for year). So in Earth
science texts you will commonly find the age of the Earth written as 4600 Ma. It
won’t have escaped your notice that the year is not the SI base unit of time — but
then perhaps it would be a little odd to think about geological timescales in terms
of seconds!
A few metric units from the pre-SI era also remain in use. In chemistry courses,
you may come across the ångström (symbol Å), equal to 10−10 metres. This was
commonly used for the measurement of distances between atoms in chemical struc-
tures, although these distances are now often expressed in either nanometres or pi-
cometres. Other metric but non-SI units with which we are all familiar are the litre
(symbol l) and the degree Celsius (symbol ◦ C).
There are also some prefixes in common use, which don’t appear in Box 2.2 be-
cause they don’t conform to the ‘multiples of 1000’ rule, but that when applied to
particular units happen to produce a very convenient measure. One you will cer-
tainly have used yourself is centi (hundredth): rulers show centimetres (hundredths
of a metre) as well as millimetres, and standard wine bottles are marked as holding
75 cl. One less commonly seen is deci (tenth) but that is routinely used by chemists
in measuring concentrations of chemicals dissolved in water, or other solvents, as
you will see in Chapter 3. In the next section you will also come across the decibel,
which is used to measure the loudness of sounds.

Back J I 71
Contents 
Worked example 2.3
Diamond is a crystalline form of carbon in which the distance between adja-
cent carbon atoms is 0.154 nm. What is this interatomic distance expressed in
picometres?
Answer
1 pm = 10−12 m so
1
1m= pm = 1012 pm
10−12

1 nm = 10−9 m so

1 nm = 10−9 × 1012 pm
= 10−9+12 pm
= 103 pm

0.154 nm = 0.154 × 103 pm

= 154 pm

Back J I 72
Contents 
Question 2.3
Using scientific notation, express:
(a) 3476 km (the radius of the Moon) in metres. Answer
(b) 8.0 µm (the diameter of a capillary carrying blood in the body) Answer
in nm,
(c) 0.8 s (a typical time between human heartbeats) in ms. Answer

Back J I 73
Contents 
2.3 Scales of measurement
In thinking about the sizes of things, it is sometimes useful to do so in quite rough
terms, just to the nearest power of ten. For example, 200 is nearer to 100 than it is
to 1000, but 850 is nearer to 1000 than it is to 100. So if we were approximating to
the nearest power of ten we could say 200 was roughly 102 , but 850 was roughly
103 . This process is called reducing the numbers to the nearest order of magnitude.

The approximate value of a quantity expressed as the nearest power of ten to

that value is called the order of magnitude of the quantity.

The easiest way to work out the order of magnitude of a quantity is to express it
first in scientific notation in the form a × 10n . Then if a is less than 5, the order of
magnitude is 10n . But if a is equal to or greater than 5, the power of ten is rounded
up by one, so the order of magnitude is 10n+1 . For example, the diameter of Mars
is 6762 km. This can be written as 6.762 × 103 km, and because 6.762 is greater
than 5, the diameter of Mars is said to be ‘of order 104 km’.
This is normally written as:

diameter of Mars ∼ 104 km

where the symbol ∼ denotes ‘is of order’.

Back J I 74
Contents 
Question
What is the order of magnitude of the mass of the Earth, 6.0 × 1024 kg?
Answer
Mass of the Earth ∼ 1025 kg (since 6.0 is greater than 5, the power of ten has
been rounded up).

Question
What is the order of magnitude of the mass of Jupiter, 1.9 × 1027 kg?
Answer
Mass of Jupiter ∼ 1027 kg (since 1.9 is less than 5, the power of ten remains
unchanged).

Back J I 75
Contents 
Question
What is the order of magnitude of the average lifetime of unstable ‘sigma plus’
particles, 0.7 × 10−10 s?
Answer
Particle lifetime = 0.7 × 10−10 s
= 7 × 10−11 s
Since 7 is greater than
∼ 10(−11+1) s 5, the power of ten
must be rounded up
∼ 10−10 s

Back J I 76
Contents 
The phrase ‘order of magnitude’ is also quite commonly used to compare the sizes
of things, e.g. a millimetre is three orders of magnitude smaller than a metre.

Worked example 2.4

To the nearest order of magnitude, how many times more massive is Jupiter than
the Earth?
Answer
We had:
mass of Jupiter ∼ 1027 kg
and
mass of Earth ∼ 1025 kg
so
mass of Jupiter 1027
∼ 25 ∼ 10(27−25) ∼ 102
mass of Earth 10
Jupiter is two orders of magnitude (i.e. roughly 100 times) more massive than
the Earth.

Back J I 77
Contents 
Question 2.4
What is the order of magnitude of the following measurements?
(a) The distance between Pluto (the furthest planet in the Solar Answer
System) and the Sun: five thousand nine hundred million kilo-
metres.
(b) The diameter of the Sun, given that its radius is 6.97 × 107 m. Answer
(c) 2π. Answer
(d) The mass of a carbon dioxide molecule: 7.31 × 10−26 kg. Answer

Sophisticated instrumentation now allows scientists to measure across 40 orders of

magnitude, as shown in Figure 2.2. If you turn back to Figure 1.2, you will see
that the scale there is quite different to that in Figure 2.2. On the thermometer,
the interval between marked points was always the same, with marked points at
−0.1, 0, 0.1, 0.2, etc. In other words, each step from one division to the next on the
scale represented the addition or subtraction of a fixed amount (0.1 in that case).
This kind of scale is called linear. In Figure 2.2, on the other hand, each step
involves multiplication or division by a fixed power of ten (102 in this particular
case). As a result, the intervals between divisions are all different. This kind of
scale is called logarithmic. The next question allows you to investigate some of the
properties of this type of scale.

Back J I 78
Contents 
Question 2.5
Use information from Figure 2.2 to answer the following questions.
(a) What is the difference in value between: Answer
(i) the tick marks at 10−2 m and 100 m;
(ii) the tick marks at 100 m and 102 m, and
(iii) the tick marks at 102 m and 104 m?
(b) Calculate to the nearest order of magnitude, how many times Answer
taller than a child is Mount Everest.
(c) Calculate to the nearest order of magnitude, how many typical Answer
viruses laid end to end would cover the thickness of a piece of
paper. (Hint: you may find it helpful to look back at Worked
example 2.4.)

Back J I 79
Contents 
2.3.1 Logarithmic scales in practice

In Figure 2.2, a logarithmic scale was used for the purposes of display, and the
power of ten for the multiplying factor (102 ) was chosen because it was the one that
best fitted the page. In drawing diagrams and graphs we are always free to choose
the scale divisions. However, logarithmic scales are used in a number of fields to
measure quantities that can vary over a very wide range. In such cases, an increase
or decrease of one ‘unit’ always represents a ten-fold increase or decrease in the
quantity measured. The following sections give two examples.

Sound waves
The decibel (symbol dB) is the unit used to measure the relative loudness of sounds.
The ‘intensity’ of a sound is related to the square of the variation in pressure as the
sound wave passes through the air, and the range of intensities that people can detect
is enormous. The sound that just causes pain is 1012 times more intense than the
sound that is just audible! To deal with this huge range, a logarithmic scale for
loudness was devised, according to which every 10 dB (or ‘1 B’) increase in sound
level is equivalent to a 10-fold increase in intensity. The decibel is also a convenient
measure because a sound level of 1 dB is just within the limit of human hearing, and
a change of 1 dB is about the smallest difference in sound that the ear can detect.
(See Figure 2.3.)

Back J I 80
Contents 
Earthquakes
The Richter scale describes the magnitude of earthquakes. An instrument called
a seismometer is used to measure the maximum ground movement caused by the
earthquake, and a correction factor is applied to this reading to allow for the distance
of the seismometer from the site of the earthquake. Seismometers are very sensi-
tive and can detect minute amounts of ground movement (they have to be shielded
from the effects caused just by people walking near them), but some earthquakes
can produce ground movements millions of times greater than the minimum de-
tectable limit. To cope with this huge variation, the Richter scale is logarithmic: an
increase of one unit on the scale implies a ten-fold increase in the maximum ground
movement. A magnitude 2 earthquake can just be felt as a tremor. A magnitude 3
earthquake produces 10 times more ground motion than a magnitude 2 earthquake.
Damage to buildings occurs at magnitudes in excess of 6. The three largest earth-
quakes ever recorded (in Portugal in 1775, in Columbia in 1905 and in Japan in
1933) each had a Richter magnitudes of 8.9.

Back J I 81
Contents 
Worked example 2.5
A whisper corresponds to a sound level of about 20 dB, and a shout to a level of
about 80 dB. How much greater is the intensity of a shout compared to that of a
whisper?
Answer
The increase in sound level is

80 dB − 20 dB = 60 dB

This may be expressed as (10 dB + 10 dB + 10 dB + 10 dB + 10 dB + 10 dB),

and each 10 dB increase corresponds to multiplying the intensity by 10.
So the intensity of a shout is (10 × 10 × 10 × 10 × 10 × 10) = 106 times greater
than a whisper!

Question 2.6 Answer

How much more ground movement is there in an earthquake measuring 7 on the
Richter scale compared to one measuring 3?

The basis of logarithmic scales will be discussed in Chapter 7.

Back J I 82
Contents 
2.4 How precise are the measurements?
Scientists are always trying to get better and more reliable data. One way of getting
a more precise measurement might be to switch to an instrument with a more finely
divided scale. Figure 2.4 shows parts of two thermometers placed side by side to
record the air temperature in a room.

19 20 21 22 23 24 25 26 27 °C
A

17 18 19 20 21 22 23 24 25 °C
B

Figure 2.4: Parts of two thermometers A and B, measuring the air temperature in
the same place.

The scale on thermometer A is quite coarse. The marked divisions represent integer
numbers of degrees. On this scale we can see that the temperature is between 21 ◦ C
and 22 ◦ C. I might estimate it as 21.7 ◦ C, but somebody else could easily record it
as 21.6 ◦ C or 21.8 ◦ C. So there is some uncertainty in the first decimal place, and
certainly there is no way we could attempt to guess the temperature to two decimal
places using this particular thermometer.

Back J I 83
Contents 
Thermometer B has a finer scale, with divisions marked every 0.1 ◦ C. Now we can
clearly see that the temperature is between 21.6 ◦ C and 21.7 ◦ C. I might read it as
21.63 ◦ C, but a second person could plausibly read it as 21.61 ◦ C or 21.65 ◦ C. With
this scale we are sure of the first decimal place but uncertain of the second.
When quoting the result of a measurement, you should never quote more digits
than you can justify in terms of the uncertainty in the measurement. The number of
significant figures in the value of a measured quantity is defined as the number of
digits known with certainty plus one uncertain digit. With thermometer A we could
be sure of the 21 (two digits), but were uncertain about the digit in the first decimal
place, so we can quote a reading to three significant figures, as 21.7 ◦ C (or 21.6 ◦ C
or 21.8 ◦ C). With thermometer B it was the fourth digit that was uncertain, so we
can quote our reading to four significant figures, as, for example, 21.64 ◦ C.

Question 2.7 Answer

How many significant figures are quoted in each of the following quantities:
1221 m; 223.4 km; 1.487 km?

Question 2.7 emphasizes that significant figures mustn’t be confused with the num-
ber of decimal places. After all, if you had measured the length of something
as 13 mm, you wouldn’t want the precision of your result to be changed just be-
cause you converted the measurement to centimetres. Whether you write 13 mm or
1.3 cm you are expressing the result of your measurement to two significant figures.
Now suppose you convert to metres: 0.013 m. The uncertainty in your result still

Back J I 84
Contents 
hasn’t changed, so this shows that leading zeroes in decimal numbers do not count
as significant figures. Scientific notation is helpful in this regard. Expressing the
result as 1.3 × 10−2 m makes it very obvious that there are two significant figures.
Another circumstance in which one has to be careful about not using unjustified
precision occurs when the results of measurements are used as the basis for calcu-
lations. Suppose we had measured the diameter of a circular pattern to two signifi-
cant figures and obtained the result 3.3 cm. If we then needed to calculate the radius
of the circle, it might be tempting simply to divide the diameter by 2 and say ‘the
radius of the pattern is 1.65 cm’. But 1.65 cm implies that the value is known to
three significant figures! So we need to round off the figure in some way, to express
the fact that the last significant digit in this particular case is the first digit after the
decimal point. The usual rule for doing this is to leave the last significant digit un-
changed if it would have been followed by a digit from 0 to 4, and to increase it by
one if it would have been followed by a digit from 5 to 9. To two significant figures
our circular pattern therefore has a radius of 1.7 cm. The issues involved in dealing
with significant figures in more complex calculations are discussed in Chapter 3.
Scientific notation also shows up the need for care in dealing with very large num-
bers. The speed of light in a vacuum (the constant c in Einstein’s equation E = mc2
is, to six significant figures, 299 792 kilometres per second. Remembering the
rounding rule, this can quite properly be written as 3 × 105 kilometres per sec-
ond (one significant figure), or 3.00 × 105 kilometres per second (three significant
figures). But it would be misleading to write it as 300 000 kilometres per second,
because that could imply that all six digits are significant.

Back J I 85
Contents 
One of the advantages of using scientific notation is that it removes any ambiguity
about whether zeroes at the end of a number are significant or are simply place
markers. For example, if a length is measured to just one significant figure as
8 m, how should the equivalent value in centimetres be expressed? It would be
ambiguous to write 800 cm, since that could imply the value is known to three
significant figures. The only way out of this difficulty is to use scientific notation:
writing 8 × 102 cm makes it clear that the quantity is known only to one significant
figure, in line with the precision of the original measurement.

Question
If the speed of light through glass is quoted as 2.0 × 108 metres per second, how
many significant figures are being given?
Answer
Final zeroes are significant, so the speed is being given to two significant figures.

Back J I 86
Contents 
Question
Neon gas makes up 0.0018% by volume of the air around us. How many signif-
icant figures are being given in this percentage?
Answer
Leading zeroes are not significant, so this value is also being given to two sig-
nificant figures.

Worked example 2.6

The average diameter of Mars is 6762 km. What is this distance in metres,
expressed to three significant figures?
Answer
The only way to express this quantity unambiguously to fewer than the four
significant figures originally given is to use scientific notation.

6762 km = 6.762 × 103 km

= 6.762 × 103 × 103 m
= 6.762 × 10(3+3) m
= 6.762 × 106 m

The final digit is a 2, so no rounding up is required and the average diameter of

Mars is 6.76 × 106 m to three significant figures.

Back J I 87
Contents 
Question 2.8
Express the following temperatures to two significant figures:
(a) −38.87 ◦ C (the melting point of mercury, which has the un- Answer
usual property for a metal of being liquid at room temperature);
(b) −195.8 ◦ C (the boiling point of nitrogen, i.e. the temperature Answer
above which it is a gas);
(c) 1083.4 ◦ C (the melting point of copper). Answer

Back J I 88
Contents 
In the following chapter and in your future studies of science generally, you will
be doing lots of calculations with numbers in scientific notation, and will also be
expected to quote your results to appropriate numbers of significant figures. Chapter
3 will discuss the efficient way to input scientific notation into your calculator, and
how to interpret the results.

2.5 Learning outcomes for Chapter 2

After completing your work on this chapter you should be able to:
2.1 convert quantities expressed as integers or in decimal notation to scientific
notation and vice versa;
2.2 use prefixes in association with the SI base units and convert between prefixes;
2.3 express a given quantity as an order of magnitude;
2.4 state the number of significant figures in any given quantity;
2.5 express a given quantity to any stipulated number of significant figures.

Back J I 89
Contents 

Calculating in Science 3
There comes a point in science when simply measuring is
not enough and we need to calculate the value of a quantity
from values for other quantities that have been measured
previously. Take, for example, the piece of granite shown in
Figure 3.1. We can measure the lengths of its sides and its
mass. With a little calculation we can also find its volume,
its density, and the speed at which seismic waves will pass
through a rock of this type following an earthquake.
This chapter looks at several scientific calculations, and in
the process considers the role of significant figures, scien-
tific notation and estimating when calculating in science. In
Figure 3.1: A specimen of granite.
addition, it introduces unit conversions and the use of for-
mulae and equations.

Back J I 90
Contents 
3.1 Calculating area; thinking about units and significant
figures

Suppose we want to find the area of the top of the granite spec-
imen shown in Figure 3.1. The lengths of its sides, measured
in centimetres, are shown in Figure 3.2, and the area of a rect-
angle is given by

area of rectangle = length × width

Thus the area of the top of the granite is

area = 8.4 cm × 5.7 cm

Multiplying the two numbers together gives 47.88. However, Figure 3.2: The lengths of the sides of
if given as a value for the area, this would be incomplete and the specimen of granite.
incorrectly stated for two reasons.
1 No units have been given.
2 The values for length and width which we’ve used are each
given to two significant figures, but 47.88 is to four signifi-
cant figures. This is too many.

Back J I 91
Contents 
3.1.1 Units in calculations

The length and the width of the specimen of granite aren’t just
numbers, but physical quantities, with units. The area — the
result of multiplying the length by the width — is a physical
quantity too and it should also have units. The units which have
been multiplied together are cm × cm, which can be written as
(cm)2 , or more commonly as cm2 . In fact any unit of length
squared will be a unit of area. Conversely, a value given for
area should always have units of (length)2 .

All measurements should be given with appropriate units,

and when performing calculations the units of the answer
must always be consistent with the units of the quantities
you input.

Care needs to be taken when multiplying together two lengths which have been
measured in different units. Suppose, for instance, that we needed to find the area
of a 1 cm by 4 m rectangle. Units of cm × m are meaningless; we need to convert
the units to the same form before proceeding, and if in doubt it is best to convert to
SI base units. Since 1 cm = 0.01 m, this gives an area of 0.01 m × 4 m = 0.04 m2 .

Back J I 92
Contents 
Question 3.1 Answer
Which of the following are units of area:
(inch)2 ; s2 ; m−2 ; cm2 ; km3 ; square miles?
Note: the symbols used for SI units are as given in Box 2.1.

3.1.2 Significant figures and rounding in calculations

It is not appropriate to quote answers to calculations to an unlimited number of

significant figures. Suppose that, as part of a calculation, you were asked to divide
3.4 (known to two significant figures) by 2.34 (known to three significant figures).
Entering 3.4÷2.34 on most scientific calculators gives 1.452 991 453, but to quote a
result to this number of significant figures would imply that you know the answer far
more precisely than is really the case. The fact that 3.4 is quoted to two significant
figures implies that the first digit is precisely known, but there is some uncertainty
in the second digit; similarly the fact that 2.34 is quoted to three significant figures
implies that there is some uncertainty in the third digit. Yet in giving the result as
1.452 991 453 we are claiming to be absolutely confident of the answer as far as
1.452 991 45, with just some uncertainty in the final digit. This is clearly nonsense!

Back J I 93
Contents 
The sensible number of significant figures to quote in any answer depends on a
number of factors. However, in the absence of other considerations, a simple rule
of thumb is useful:

When multiplying and dividing numbers, the number of significant figures in

the result should be the same as in the measurement with the fewest significant
figures.

Applying this rule of thumb, the answer to the calculation 3.4÷2.34 should be given
to two significant figures, i.e. as 1.5.
Similarly, the result of the multiplication 8.4 cm × 5.7 cm (used in finding the area
of the top of the granite specimen) should be given as 48 cm2 , again to two signifi-
cant figures.
There are two points of caution to bear in mind when thinking about the appropriate
number of significant figures in calculations.

Avoiding rounding errors

You should round your answer to an appropriate number of significant figures at
the end of a calculation. However, be careful not to round too soon, as this may
introduce unnecessary errors, known as rounding errors. As an example of the
dangers of rounding errors, let’s return to our previous example. We found that:

3.4 ÷ 2.34 = 1.452 991 453

Back J I 94
Contents 
Or, giving the answer to two significant figures:

3.4 ÷ 2.34 = 1.5

Suppose that we now need to multiply the answer by 5.9:

1.452 991 453 × 5.9 = 8.572 649 573 = 8.6 to two significant figures

However, using the intermediate answer as quoted to two significant figures gives

1.5 × 5.9 = 8.85 = 8.9 to two significant figures

Rounding too soon has resulted in an incorrect answer.

The use of scientific calculators enables us to work to a large number of significant
figures and so to avoid rounding errors. If this is not possible, you should follow
the following advice:

Work to at least one more significant figure than is required in the final answer,
and just round at the end of the whole calculation.

In our example, the final answer should be given to two significant figures, which
means that we should work using the result of the first calculation to at least three
significant figures (1.45).

1.45 × 5.9 = 8.555 = 8.6 to two significant figures.

Back J I 95
Contents 
Applying common sense!
Always bear in mind the real problem that you are solving, and apply common
sense in deciding how to quote the answer. Particular care needs to be taken when
the calculation involves numbers which are exactly known. A light-hearted example
should illustrate this point.

Question
Suppose you have 7 apples to share between 4 children. How many apples does
each child get?
Answer
Dividing the number of apples by the number of children gives
7
= 1.75
4
If we were to assume that the number of apples and number of children were
each quoted to one significant figure, we would round the answer to one signifi-
cant figure too, i.e. to 2 apples. But we would then need eight apples, which is
more than we’ve got. In reality there are exactly 4 children and 7 apples, so the
number of significant figures need not bother us. Provided we have a knife, it is
perfectly possible to give each child 1.75 (1 43 ) apples.

Back J I 96
Contents 
Question 3.2
Do the following calculations and express your answers to an appropriate num-
ber of significant figures.
6.732
(a) Answer
1.51
(b) 2.0 × 2.5 Answer
!2
4.2
(c) Answer
3.1
(d) What is the total mass of three 1.5 kg bags of flour? Answer

3.2 Calculating in scientific notation

In science it is very often necessary to do calculations using very large and very
small numbers, and scientific notation can be a tremendous help in this.

3.2.1 Calculating in scientific notation without a calculator

Suppose we need to multiply 2.50 × 104 and 2.00 × 105 . The commutative nature of
multiplication is completely general, so it applies when multiplying two numbers

Back J I 97
Contents 
written in scientific notation too. This means that (2.50 × 104 ) × (2.00 × 105 ) can
be written as (2.50 × 2.00) × (104 × 105 ), i.e.

(2.50 × 104 ) × (2.00 × 105 ) = (2.50 × 2.00) × (104 × 105 )

= 5.00 × 104+5
= 5.00 × 109

All of the rules for the manipulation of powers discussed in Chapter 1 can be applied
to numbers written in scientific notation, but care needs to be taken to treat the
decimal parts of the numbers (such as the 2.50 in 2.50 × 105 ) and the powers of ten
separately. So, for example

2.50 × 104 2.50 104 2.50

= × = × 104−5 = 1.25 × 10−1
2.00 × 105 2.00 105 2.00

and
 2  2
2.50 × 105 = 2.502 × 105 = 6.25 × 1010

Back J I 98
Contents 
Question 3.3
Evaluate the following without using a calculator, giving your answers in scien-
tific notation.
(a) (3.0 × 106 ) × (7.0 × 10−2 ) Answer
8 × 104
(b) Answer
4 × 10−1
104 × (4 × 104 )
(c) Answer
1 × 10−5
 2
(d) 3.00 × 108 Answer

Back J I 99
Contents 
3.2.2 Using a calculator for scientific notation

In the rest of this chapter, and in your future studies of science generally, you will be
doing many calculations with numbers in scientific notation, so it is very important
that you know how to input them into your calculator efficiently and how to interpret
the results.
First of all make sure that you can input numbers in scientific notation into your cal-
culator.You can do this using the button you used to input powers in Section 1.3.1,
but it is more straightforward to use the special button provided for entering scien-
tific notation. This might be labelled as EXP, EE, E or EX, but there is considerable
variation between calculators. Make sure that you can find the appropriate button
on your calculator. Using a button of this sort is equivalent to typing the whole of
‘×10 to the power’. So, on a particular calculator, keying 2.5 EXP 12 enters the
whole of 2.5 × 1012 .

Back J I 100
Contents 
In addition to being able to enter numbers in scientific notation into
your calculator, it is important that you can understand your calculator
display when it gives an answer in scientific notation.
Enter the number 2.5 × 1012 into your calculator and look at the display. (a)

Again there is considerable variation from calculator to calculator, but it

is likely that the display will be similar to one of those shown in Figure
3.3. The 12 at the right of the display is the power of ten, but notice (b)
that the ten itself is frequently not displayed. If your calculator is one of
those which displays 2.5 × 1012 as shown in Figure 3.3e, then you will 2.5 e 12
need to take particular care; this does not mean 2.512 on this occasion.
(c)
You should be careful not to copy down a number displayed in this way
on your calculator as an answer to a question; this could cause confusion
at a later stage.
(d)

No matter how scientific notation is entered and displayed on your

calculator or computer, when writing it on paper you should always
use the form exemplified by 2.5 × 1012 . (e)

To enter a number such as 5 × 10−16 into your calculator, you may need Figure 3.3: Examples of how
to use the button labelled something like +/− (as used in Section 1.1.3) various calculators would dis-
in order to enter the negative exponent. play the number 2.5 × 1012
To enter a number such as 108 into your calculator using the scientific notation
button, it can be helpful to remember that 108 is written as 1 × 108 in scientific

Back J I 101
Contents 
notation, so you will need to key something like 1 EXP 8.
If you are at all unsure about using your calculator for calculations involving sci-
entific notation, you should repeat Question 3.3, this time using your calculator.

Question 3.4 Answer

A square integrated circuit, used as the processor in a computer, has sides of
length 9.78 × 10−3 m. Give its area in m2 in scientific notation and to an appro-
priate number of significant figures.

3.3 Estimating answers

The first time I attempted Question 3.4, my calculator gave me the answer 95.6 m2 .
This is incorrect (I’d forgotten to enter the power of ten). It is sensible to get into the
habit of checking that the answer your calculator gives is reasonable, by estimating
the likely answer. In the case of Question 3.4, the answer should be approximately
 2
1 × 10−2 m which you can see (without using a calculator!) is 1 × 10−4 m2 . So
a calculator answer of 95.6 m2 is clearly wrong.

Back J I 102
Contents 
In addition to being useful as a way of checking calculator answers, estimated an-
swers are, in their own right, quite frequently all that is needed. Chapter 2 began
with a comparison between the size of a bacterium and the size of a pinhead. We
could use precise measuring instruments to find that the diameter of a particular
bacterium is 1.69 µm (i.e. 1.69 × 10−6 m) and that the diameter of the head of a
particular pin is 9.86 × 10−4 m. The diameter of the pinhead would then be

9.86 × 10−4 m
= 5.83 × 102 times bigger than that of the bacterium.
1.69 × 10−6 m
However, to get a feel for the relative sizes, we only really need to estimate the
answer. If an estimate is all that is required, it is perfectly acceptable to work to one
significant figure throughout (indeed, working to the nearest order of magnitude is
sometimes sufficient) and since the final answer is only approximately known, the
symbol ‘≈’ (meaning ‘approximately equal to’) is used in place of an equals sign.

Back J I 103
Contents 
Worked example 3.1
Working to one significant figure throughout, estimate how many times big-
ger a pinhead of diameter 9.86 × 10−4 m is than a bacterium of diameter
1.69 × 10−6 m.
Answer
Diameter of pinhead ≈ 1 × 10−3 m.
Diameter of bacterium ≈ 2 × 10−6 m.
diameter of pinhead 1 × 10−3 m
≈
diameter of bacterium 2 × 10−6 m
1 10−3
≈ × −6
2 10
≈ 0.5 × 10−3−(−6)
≈ 0.5 × 103
≈ 5 × 102

So the diameter of the pinhead is approximately 500 times that of the bacterium.

Back J I 104
Contents 
It is important that you write out your mathematical calculations carefully, and
one of the functions of the worked examples scattered throughout the course is to
illustrate how to do this. There are three particular points to note from Worked
example 3.1.

Taking care when writing maths

1 Note that the symbols = and ≈ mean ‘equals’ and ‘approximately equals’ and
should never be used to mean ‘thus’ or ‘therefore’. It is acceptable to use
the symbol ∴ for ‘therefore’; alternatively don’t be afraid to write words of
explanation in your calculations.
2 It can make a calculation clearer if you align the = or ≈ symbols vertically,
to indicate that the quantity on the left-hand side is equal to or approximately
equal to each of the quantities on the right-hand side.
3 Note that the diameter of the bacterium and the pinhead each have metres (m)
as their units, so when one diameter is divided by the other, the units cancel
to leave a number with no units.

The handling of units in calculations is discussed further in Section 3.5.4.

Back J I 105
Contents 
Question 3.5 Answer
The average distance of the Earth from the Sun is 1.50 × 1011 m and the dis-
tance to the nearest star other than the Sun (Proxima Centauri) is 3.99 × 1016 m.
Working to one significant figure throughout, estimate how many times further
it is to Proxima Centauri than to the Sun.

3.4 Unit conversions

In calculating the area of the top of the granite specimen earlier in this chapter, we
measured the length of the sides in centimetres and hence calculated the area in
cm2 . If we had wanted the area in the SI units of m2 we could have converted the
lengths from centimetres to metres before starting the calculation. We would then
have had

area = (8.4 × 10−2 m) × (5.7 × 10−2 m) = 4.8 × 10−3 m2

It is best, whenever possible, to convert all units to SI units before starting on a

calculation.

Unfortunately it is not always possible to convert units before commencing a calcu-

lation; sometimes you will be given an area in, say, cm2 , without knowing how the

Back J I 106
Contents 
area was calculated, and you will need to convert this to an area in m2 . This section
discusses this, as well as some more complex unit conversions.

3.4.1 Converting units of area

Let’s start with an example which is relatively easy to visual-

ize. Suppose we want to know how many mm2 there are in a 10 mm

cm2 . There are 10 millimetres in a centimetre, so each side of the

square centimetre in Figure 3.4 measures either 1 cm or 10 mm.

10 mm
1 cm
To find the area, we need to multiply the length by the width.
Working in centimetres gives
1 cm
area = 1 cm × 1 cm = (1 cm)2 = 12 cm2 = 1 cm2

Working in millimetres gives

Figure 3.4: A square centimetre
area = 10 mm × 10 mm = (10 mm) = 10 mm = 100 mm
2 2 2 2
(not to scale)
1
Thus 1 cm2 = 100 mm2 and 1 mm2 = cm2 .
100
If we want to convert from cm2 to mm2 we need to multiply by 100; if we want to
convert from mm2 to cm2 we need to divide by 100.

Back J I 107
Contents 
Figure 3.5 illustrates another example which is a little harder to
visualize. Each side of the square measures either 1 km or 1000 m
10 3 m
(103 m). Working in kilometres gives

area = 1 km × 1 km = (1 km)2 = 12 km2 = 1 km2

10 3 m
1 km
Working in metres gives
 2  2
area = 103 m × 103 m = 103 m = 103 m2 = 106 m2 1 km

1
Thus 1 km2 = 106 m2 and 1 m2 = km2 . Figure 3.5: A square kilometre
106
To convert from km2 to m2 we need to multiply by 106 ; to convert
from m2 to km2 we need to divide by 106 .
The number by which we need to divide or multiply to convert from one unit to
another is known as the ‘conversion factor’. In general, to convert between units
of area we need to square the conversion factor which we would use to convert
corresponding lengths.

Back J I 108
Contents 
As a final example consider a conversion between km2 and mm2 .
There are 103 millimetres in a metre and 103 metres in a kilome- 10 6 mm
tre, so there are 106 millimetres in a kilometre as illustrated in
Figure 3.6.

10 6 mm
1 km
To convert from kilometres to millimetres we need to multiply by
106 ; however to convert from km2 to mm2 we need to multiply by
 2
106 , i.e. 1012 .
1 km
Similarly, to convert from mm2 to km2 we need to divide by
 2
106 , i.e. 1012 . Figure 3.6: A square kilometre

Question 3.6
A desk has an area of 1.04 m2 . Express this area in:
(a) cm2 Answer
(b) µm2 Answer
(c) km2 Answer

Back J I 109
Contents 
3.4.2 Converting units of volume

The volume of the piece of granite shown in Figure 3.2 is given by

volume = length × width × height

The lengths of the sides are 8.4 cm, 5.7 cm and 4.8 cm, so

volume = 8.4 cm × 5.7 cm × 4.8 cm

= 2.3 × 102 cm3 to two significant figures.

Note that the units which have been multiplied together are cm × cm × cm, so in
this case the units of volume are cm3 . A value given for volume should always have
units equivalent to those used for (length)3 , and if we had converted the lengths of
the sides to metres before doing the calculation, we would have obtained a value
for volume in m3 :

volume = (8.4 × 10−2 m) × (5.7 × 10−2 m) × (4.8 × 10−2 m)

= 2.3 × 10−4 m3 to two significant figures.

The method for converting between different units of volume is a direct extension
of the method for converting between different units of area. Suppose we want to
know how many mm3 there are in a cm3 .

Back J I 110
Contents 
There are 10 mm in 1 cm, so each side of the cubic centimetre
in Figure 3.7 measures either 1 cm or 10 mm. The volume can 1 cm
be written as either 1 cm3 or 103 mm3 . Thus 1 cm3 = 103 mm3 1c
m
1
and 1 mm3 = 3 cm3 . To convert from cm3 to mm3 we need to

10 mm
10
multiply by 103 ; to convert from mm3 to cm3 we need to divide

1 cm
by 103 .
In general, to convert between units of volume we need to cube 10 mm
the conversion factor that we would use to convert corresponding 10 mm
lengths.
We can convert a volume of 2.3 × 102 cm3 into m3 simply by say- Figure 3.7: A cubic centimetre (not
 3
ing that there are 102 cm in 1 m; hence there are 102 cm3 in to scale).
1 m3 , so
1
1 cm3 =
3 m
3
10 2

and

2.3 × 102 3
2.3 × 102 cm3 =
3 m
102
= 2.3 × 10−4 m3

This value is, of course, the same as the one we obtained from first principles!

Back J I 111
Contents 
The prefix ‘deci’ meaning one tenth was introduced in Section 2.2, thus 1 decimetre
(dm) is one tenth of a metre. The cubic decimetre (dm3 ) is sometimes used as a unit
of volume. The litre (l) (also introduced in Chapter 2) was defined in 1901 as the
volume of a kilogram of water at 4 ◦ C, under standard atmospheric pressure. This
volume turns out to be 1.000 28 dm3 , and since 1969 a litre has been defined to be
1 dm3 .

Worked example 3.2

Convert a volume of 1 dm3 to: (a) cm3 (b) m3
Answer

(a) 1 m = 10 dm and 1 m = 100 cm so 1 dm = 10 cm.

Thus 1 dm3 = 103 cm3 .
(b) 1 m = 10 dm
Thus 1 m3 = 103 dm3
1
and 1 dm3 = 3 m3 = 10−3 m3 .
10
{Thus 1 dm3 (i.e. 1 litre) is a thousand times bigger than a cubic centimetre and
a thousand times smaller than a cubic metre. You may already have been aware
that 1 litre = 1000 cm3 . Thus 1 ml = 1 cm3 . }

Back J I 112
Contents 
Figure 3.8 is a summary of unit conversions for length, area and volume, but you
should try to remember the general principles involved rather than memorizing in-
dividual conversion factors.

Question 3.7
Express each of the following volumes in scientific notation in m3 :
(a) the volume of the planet Mars, which is 1.64 × 1011 km3 ; Answer
(b) the volume of a ball bearing, which is 16 mm3 . Answer

3.4.3 Converting units of distance, time and speed

You were introduced in Box 2.1 to the metre as the base unit of distance or length
and to the second as the base unit of time. The average speed with which an object
moves is the total distance travelled divided by the total time taken, so when Marion
Jones won the women’s 100-metre final at the 2000 Sydney Olympics in 10.75 s,
her average speed was
100.0 m
average speed = = 9.302 m s−1
10.75 s
Similarly, if a girl grows a total of 116 cm in 12.5 years, her average rate of growth
is
116 cm
growth rate = = 9.28 cm year−1
12.5 years
Back J I 113
Contents 
Note that it is appropriate to give the answer to the first example to four significant
figures (assuming that the length of the running track was known to at least four
significant figures). Also note the way in which the units have been written in both
examples.

The notation of negative exponents, which we have used to represent numbers

like 1/23 as 2−3 and 1/108 as 10−8 , can also be used for units. So 1/s can
be written as s−1 , m/s can be written as m s−1 and cm/year can be written as
cm year−1 .

The SI unit of speed is m s−1 and this is usually said as ‘metres seconds to the
minus one’. Although m s−1 is the correct scientific way of writing the unit, it is
sometimes written as m/s, and quite frequently said as ‘metres per second’, even
when written as m s−1 . The ‘/’ for per is quite commonly used in other units too.

Back J I 114
Contents 
Many things move and/or grow in the world around us, and it is useful to compare
different values for speed or rate of growth. Different speeds are frequently mea-
sured in different units, so in order to be able to compare like with like it is necessary
to convert between different units for distance, time and speed. Box 3.1 considers
various examples of speed and growth, and the text immediately following the box
looks at ways of converting one unit to another.

Box 3.1 How fast?

Light (and other forms of radiation such as X-rays and radio waves) travels in a
vacuum with a constant speed of 3.00 × 108 m s−1 . It is currently believed that
nothing can travel faster than this.
Towards the opposite extreme are stalactites and stalagmites, which grow just
fractions of a millimetre each year. A typical growth rate is 0.1 mm year−1 . Sta-
lactites form when water drips from the roof of an underground cave, depositing
calcite (frequently from the limestone in the rock above the cave) in an icicle
shaped formation as it does so. Stalagmites form as the water drips onto the
floor of the cave, depositing further calcite.

Back J I 115
Contents 

Figure 3.9: The Saskatchewan Glacier, Banff National Park, Canada.

It is not normally possible to detect the motion of a glacier by eye, but there
is considerable variation in the speed with which they move. The Franz Josef
Glacier in New Zealand is particularly fast moving, with an average speed of
about 1.5 m day−1 . The speed of the Saskatchewan Glacier in Canada (Figure
3.9) is rather more typical, at about 12 cm day−1 .
In addition to geological processes such as glacier flow and stalactite formation,
the theory of plate tectonics tells us that the surface of the Earth is itself moving.
The Earth’s surface is thought to comprise seven major tectonic plates and nu-
merous smaller ones, each only about 100 km thick but mostly thousands of
kilometres in width. Evidence, including evidence from sea-floor spreading (to
be discussed in Chapter 5) indicates that plates move relative to one another with
speeds between about 10 km Ma−1 and 100 km Ma−1 (where Ma is the abbrevi-
ation for a million years, as discussed in Section 2.2).

Back J I 116
Contents 
P waves S waves Love waves Rayleigh waves

5 minutes

Figure 3.10: A seismogram (the printout from a seismometer) showing the ar-
rival of P waves, S waves, Love waves and Rayleigh waves from a distant earth-
quake. Elapsed time increases from left to right.

Earthquakes and volcanoes occur all over the Earth, but they are more common
close to the boundaries of tectonic plates than elsewhere. Following an earth-
quake, seismic waves (the word ‘seismic’ is from the Greek for ‘shaking’) travel
out from the centre of the quake and are recorded by seismometers at various lo-
cations. There are several different types of seismic waves, including P waves, S
waves, Love waves and Rayleigh waves, each travelling at different speeds (and
sometimes also by different routes), so reaching a given seismometer at different
times (see Figure 3.10). P waves travel fastest, with an average speed of about
5.6 km s−1 in rocks close to the Earth’s surface, so reach the seismometer first
(the name P wave was originally an abbreviation for primary wave). S waves (S
for secondary) travel with an average speed of about 3.4 km s−1 in rocks close
to the Earth’s surface.
Perhaps the most dangerous sort of volcanic eruption is one that leads to a high-

Back J I 117
Contents 
speed pyroclastic flow (a mixture of rock fragments and gases, moving as a
fluid) away from the volcano. Pyroclastic flows are particularly destructive both
because of their high temperatures (typically between 200 ◦ C and 700 ◦ C) and
the high speed at which they travel (up to about 100 km hour−1 ).
The speeds given so far have related to processes on the Earth, but remember
that the Earth itself is moving too! The rotation of the Earth on its axis leads to
a movement of up to 0.5 km s−1 at the surface. In addition, the Earth is orbiting
the Sun at about 30 km s−1 and the entire Solar System is moving around the
centre of the galaxy at about 250 km s−1 .

To convert from one unit of speed to another, we may need to convert both the
unit of distance and the unit of time. To start with, let’s consider the rather more
straightforward case when we only have to convert the unit of distance, for example
in converting from mm s−1 to m s−1 .
We know that 1 m = 103 mm
1
so 1 mm = 3 m = 1 × 10−3 m
10
We can therefore say straight away that 1 mm s−1 = 1 × 10−3 m s−1
We have simply applied the same conversion factor as in converting from mm to
m. Note that the answer makes sense: it is reasonable to expect that the numerical
value of a speed in m s−1 will be smaller than the same speed when given in mm s−1 .

Back J I 118
Contents 
Worked example 3.3
Convert the speed of the Earth as it orbits the Sun (given above as 30 km s−1 )
into a value in m s−1 .
Answer
1 km = 1 × 103 m
So
1 km s−1 = 1 × 103 m s−1

30 km s−1 = 30 × 103 m s−1

= 3.0 × 104 m s−1 in scientific notation.

The Earth orbits the Sun with a speed of about 3.0 × 104 m s−1 . Again the an-
swer makes sense: it is reasonable to expect that the numerical value of a speed
in m s−1 will be larger than the same speed when given in km s−1 .

Next let’s consider what happens when we need to convert only the time part of
units of speed, for instance in converting from km hour−1 to km s−1 .
We know that there are 60 minutes in an hour and 60 seconds in a minute, so
1 hour = 60 × 60 s = 3600 s
However, in this case we don’t want to convert from hours to seconds, but rather
from kilometres per hour to kilometres per second. The way forward comes in

Back J I 119
Contents 
recognizing that the word ‘per’ and the use of negative exponents in hour−1 and s−1
indicate division. So to convert from hour−1 to s−1 (or from km hour−1 to km s−1 )
we need to find the conversion factor from hours to seconds and then divide by it.

1 hour = 3600 s
1
so 1 km hour−1 = km s−1
3600

In deciding whether to divide or multiply by a particular conversion factor, common

sense can also come to our aid. It is reasonable to expect that a speed quoted in
km s−1 will be smaller than the same speed when quoted in km hour−1 , so it is
reasonable to divide by the 3600 on this occasion.

Back J I 120
Contents 
Worked example 3.4
Two tectonic plates are moving apart at an average rate of 35 km Ma−1 . Convert
this to a value in km year−1 .
Answer
We know that

1 Ma = 106 years

so
1
1 km Ma−1 = km year−1
106
and therefore
35
35 km Ma−1 = 6
km year−1
10
= 3.5 × 10−5 km year−1 in scientific notation.

The plates are moving apart at an average rate of 3.5 × 10−5 km year−1 .
This answer is reasonable: you would expect the rate of separation quoted in
km year−1 to be smaller than the same rate quoted in km Ma−1 .

Back J I 121
Contents 
Question 3.8
Convert the average speed of the Saskatchewan Glacier (12 cm day−1 ) to a value
in:
(a) m day−1 Answer
(b) cm s−1 Answer

Finally we need to consider conversions for speed in which both the units of dis-
tance and the units of time have to be converted. This is simply a combination of
the techniques illustrated in Worked examples 3.3 and 3.4. Suppose we want to
convert from km hour−1 to m s−1 .

1 km = 103 m
1 hour = 3600 s

To convert from km hour−1 to m s−1 , we need to multiply by 103 (to convert the km
to m) and divide by 3600 (to convert the hour−1 to s−1 ):

103
1 km hour−1 = m s−1 = 0.278 m s−1 to three significant figures.
3600

Back J I 122
Contents 
Worked example 3.5
Convert the average speed of separation of the tectonic plates discussed in
Worked example 3.4 (35 km Ma−1 ) to a value in mm year−1 .
Answer
1 km = 103 m and 1 m = 103 mm, so 1 km = 106 mm
1 Ma = 106 year
To convert from km Ma−1 to mm year−1 , we need to multiply by 106 (to convert
the km to mm) and divide by 106 (to convert the Ma−1 to year−1 .

106
1 km Ma−1 = mm year−1 = 1 mm year−1
106
Thus a speed given in km Ma−1 is numerically equal to one given in mm year−1 .
The plates are moving apart at a 35 mm year−1 . This is similar to the rate at
which human fingernails grow and is easier to imagine than is 35 km Ma−1 .

Back J I 123
Contents 
Question 3.9
Convert each of the following to values in m s−1 and then compare them.
(a) A stalactite growth rate of 0.1 mm year−1 . Answer
(b) The average speed of the Saskatchewan Glacier (12 cm day−1 ). Answer
(c) The speed of separation of the tectonic plates discussed in Answer
Worked examples 3.4 and 3.5 (35 km Ma−1 ).
(Note: for the purposes of this question, consider 1 year to be 365 days long.)

Back J I 124
Contents 
3.4.4 Concentration and density; more unit conversions

Methods for converting units for physical quantities, such as concentration and den-
sity, follow directly from the discussion in the previous sections.

Box 3.2 Concentration

The concentration of a solution is a term used as a measure of how much of
a certain substance the solution contains, relative to the solution’s total volume.
For example, we may want to know how much sugar has been dissolved in water
to give one litre of syrup.
The amount of the substance can be measured in moles, in which case the con-
centration will have units of mol l−1 or mol dm−3 . Alternatively, the amount can
be measured by mass, in kg, g, mg, etc., leading to units for concentration of
kg dm−3 , g m−3 , or mg l−1 , and so on.
The World Health Organization (WHO) sets limits for safe concentrations of
various impurities in water, for example, the limit for the concentration of ni-
trates in water is currently 50 mg l−1 . This means that there should be no more
than 50 mg of nitrate in each litre (dm3 ) of water.

Back J I 125
Contents 
To convert a concentration from, say, mg l−1 to µg ml−1 you need to follow a very It is very easy to
similar procedure to the one introduced in Section 3.4.3, as the following worked confuse the letter ‘l’,
example shows. used as the symbol for
litres, with the number
Worked example 3.6 1. Take care!
Convert 50 mg l−1 (the World Health Organization’s limit for the concentration
of nitrates in water) to a value in µg ml−1 .
Answer
We can easily write down the conversion factors for mg to µg and from litres to
ml.

1 mg = 103 µg
1 litre = 1 l = 103 ml

So to convert from mg l−1 to µg ml−1 , we need to multiply by 103 (to convert the
mg to µg) and divide by 103 (to convert the l−1 to ml−1 ).

103
1 mg l−1 = 3
µg ml−1 = 1 µg ml−1
10
Thus a concentration given in mg l−1 is numerically equal to one given in
µg ml−1 , in particular 50 mg l−1 = 50 µg ml−1 .

Back J I 126
Contents 
Box 3.3 Density
The density of a piece of material is found by dividing its mass by its volume.
In other words
mass
density =
volume
If mass is measured in kg and volume is in m3 , then it follows that the unit of
density will be kg/m3 (said as ‘kilograms per metre cubed’) or, written in the
form favoured in this course, kg m−3 (said as ‘kilograms metres to the minus
three’).
The density of pure water is 1 × 103 kg m−3 ; materials with a density greater
than this (such as steel of density 7.8 × 103 kg m−3 ) will sink in water
whereas materials of lower density (such as wood from an oak tree, density
6.5 × 102 kg m−3 ) will float.
If mass is measured in g and the volume is in cm3 , then the unit of density will be
g cm−3 . Note that g cm−3 is not an SI unit, but it is nevertheless quite frequently
used.

Back J I 127
Contents 
Question
The specimen of granite shown in Figure 3.2 has a mass of 6.20 × 102 g. Cal-
culate the density of the granite in g cm−3 .
Answer
The volume of the specimen = 8.4 cm × 5.7 cm × 4.8 cm, so
mass
density =
volume
6.20 × 102 g
=
8.4 cm × 5.7 cm × 4.8 cm
= 2.6977 g cm−3
= 2.7 g cm−3 to two significant figures.

Note that it was not necessary actually to calculate a value for volume before
completing the calculation of density. If you had used the value for volume
calculated at the beginning of Section 3.4.2, you would have obtained

mass 6.20 × 102 g

density = = = 2.7 g cm−3
volume 2.3 × 102 cm3
but you would have risked introducing rounding errors.

Back J I 128
Contents 
The final worked example in this section converts the units of the density of the
granite specimen from g cm−3 to kg m−3 , using a method which is a combination of
the techniques taught throughout Section 3.4. You can convert units of concentra-
tion such as mg dm−3 to g m−3 in a similar way.

Worked example 3.7

Convert 2.7 g cm−3 (the density of the specimen of granite shown in Figures 3.1
and 3.2) to a value in the SI units of kg m−3 .
Answer
1 kg = 103 g, so 1 g = kg = 10−3 kg
1
103
 3
1 m = 102 cm, so 1 m3 = 102 cm3 = 106 cm3 (from Section 3.4.2)

so 1 cm3 = 1
106
m3 = 10−6 m3

To convert from g cm−3 to kg m−3 we need to multiply by 10−3 (to convert the g
to kg) and divide by 10−6 (to convert the cm−3 to m−3 ).
10−3
1 g cm−3 = −6
kg m−3 = 10−3−(−6) kg m−3 = 103 kg m−3
10
Thus 2.7 g cm−3 = 2.7 × 103 kg m−3 .
The specimen of granite has a density of 2.7 × 103 kg m−3 .

Back J I 129
Contents 
You may have already known that you need to multiply by 1000 in order to convert
from units of g cm−3 to units of kg m−3 , but as was the case with the unit conversions
for area and volume, it is better to consider general principles rather than trying to
memorize conversion factors.

Question 3.10
The World Health Organization reduced its maximum recommended concentra-
tion for arsenic in drinking water from 50 µg l−1 to 10 µg l−1 in 1999. Convert
10 µg l−1 to a value in:
(a) µg ml−1 Answer
(b) mg dm−3 Answer
(c) g m−3 Answer

Back J I 130
Contents 
3.5 An introduction to symbols, equations and formulae
To progress further in our exploration of ways of calculating in science, we need
to enter the world of symbols, equations and formulae. The word ‘algebra’ is used
to describe the process of using symbols, usually letters, to represent quantities
and the relationships between them. Algebra is a powerful shorthand that enables
us to describe the relationships between physical quantities briefly and precisely,
without having to know their numerical values. Some people consider algebra to
be a beautiful thing: others are filled with terror by the very word. This course may
not convince you of algebra’s beauty, but it should at least illustrate its usefulness
and give you an opportunity to learn and practise new techniques or revise old ones.
Chapter 4 is devoted to algebraic techniques such as simplifying, rearranging, and
combining equations. The remainder of Chapter 3 simply introduces the language
of algebra by looking at a few equations very carefully, and substituting values into
them.
The word equation is used for an expression containing an equals sign. The quanti-
ties under consideration may be described in words, for example
mass
density =
volume
in which case the equation is known as a ‘word equation’, or represented by sym-
bols, for example
m
ρ=
V
Back J I 131
Contents 
but the important thing to remember is that what is written on the left-hand side of
the ‘=’ sign must always be equal to what is written on the right-hand side. Thus,
as explained in Taking care when writing maths in Section 3.3, you should never
use ‘=’ as a shorthand for anything other than ‘equals’.
The word formula is used in mathematics to mean a rule expressed in algebraic
m
symbols. Thus ρ = is a formula which tells you that the density ρ of a substance
V
can be obtained by dividing the mass, m, of a sample of the substance by the volume,
V, of the sample. Strictly speaking, not all equations are formulae, but the words
tend to be used interchangeably.

3.5.1 What do the symbols mean?

Mathematics textbooks teaching algebra frequently contain page after page of equa-
tions of the form:

x+3=8 (3.1)

and

y= x+5 (3.2)

In Equation 3.1, x can only have one value, i.e. it is a constant. In this case x has
the value 5. In Equation 3.2, x and y are variables which can each take an infinite
number of values, but y will always be 5 greater than x. The values (of x and y, etc.)

Back J I 132
Contents 
which satisfy a particular equation are known as solutions and if you are asked to
solve an equation you need to look for solutions.
In both Equation 3.1 and Equation 3.2, x and y represent pure numbers. Equations
in science are often rather different. Rather than representing pure numbers, the
symbols usually represent physical quantities and will therefore have units attached.

3.5.2 Which symbols are used

Box 3.4 contains a range of scientific formulae in common use, along with a brief
explanation of the meaning of each symbol used. Have a quick at these equations
now, but don’t worry about their details; you are not expected to learn them or to
understand the meanings of the scientific terms introduced. The equations in the
boxes will be used as examples throughout the rest of this chapter, and have been
numbered for ease of reference.
The symbol chosen to represent something is often the first letter of the quantity in
question, e.g. m for mass, t for time and l for length, but it isn’t always so simple.
Greek letters are also frequently used as symbols e.g. λ (lambda) for wavelength
in Equation 3.13 and ρ (rho) for density in Equations 3.9, 3.10 and 3.11. A list of
Greek letters and their pronunciation is given in the Table 3.1 and you will soon
become familiar with those that are commonly used. In a sense it doesn’t matter
which symbol you use to represent a quantity, since the symbol is only an arbitrar-
ily chosen label. For instance, Einstein’s famous equation (Equation 3.7) is usually
written as E = mc2 , but the equation could equally well be written using any sym-

Back J I 133
Contents 
bols you wanted to use, e.g. p = qr2 , provided you also made it clear that p was
used to represent energy, q was used to represent mass and r was used to represent
the speed of light. However, the use of conventional symbols, such as E for energy,
saves scientists a lot of time in explaining their shorthand. Maths for Science fol-
lows convention as far as possible in its use of symbols. Sometimes the reason for
the choice of symbol will be obvious but unfortunately this is not always the case.
Sometimes a subscript is used alongside a symbol in order to make its meaning
more specific, as in vi , vf and vav used in Equation 3.15 to mean initial, final, and
average speed, and a x in Equations 3.16 and 3.17 used to mean acceleration along
the x-axis. Note that although a x , for example, uses two letters, it represents a single
physical entity; note also that a x is not the same as ax. The symbol ∆ (the Greek
upper case delta) is frequently used to represent the change in a quantity, so ∆T in
Equation 3.14 means a change in temperature T; again a single physical entity is
represented by two letters.
A few letters have more than one conventional meaning, for example c in Equation
3.7 represents the speed of light, but in Equation 3.14 the same letter represents
specific heat capacity. Other letters have two meanings but lower case is conven-
tionally used for one meaning and upper case for the other, for example v for speed
and V for volume or t for time and T for temperature. Care needs to be taken, but
the intended meaning should be clear from the context.

Back J I 134
Contents 
Unfortunately some Greek letters look rather like everyday English ones; for exam-
ple ρ (rho), used for density, can look rather like the English lower case p. Some
textbooks use lower case p for pressure (this course uses capital P) and Equation
3.11 (P = ρgh) can then appear to have the same quantity on both the left- and
right-hand sides of the equals sign, especially when written out by hand. In reality,
this formula has pressure on the left-hand side and density (and other things) on the
right-hand side. A similar confusion can arise because the letter l can look like the
number 1.
A final possible source of confusion stems from the fact that the same letter may
sometimes be used to represent both a physical quantity and a unit of measurement.
For example, an object with a mass of 6 kilograms and a length of 2 metres might be
described by the relationships m = 6 kg, l = 2 m, where the letter m is used to rep-
resent both mass and the units of length, metres. In all material for this course, and
in most other printed text, letters used to represent physical quantities are printed in
italics, whereas those used for units are not.

Back J I 135
Contents 
3.5.3 Reading equations

To understand, and thus use, the equations in Box 3.4 you need to be aware of a
few rules and conventions. Most of these are extensions of things you have learnt
earlier in this course. First:

When using symbols instead of words or numbers, it is conventional to drop the

‘×’ sign for multiplication.

So in Equation 3.6, ma means mass times magnitude of acceleration and in Equation

3.11, ρgh means density times acceleration due to gravity times depth.

Rules of arithmetic, such as the fact that addition and multiplication are com-
mutative, and the BEDMAS order of operations, apply when using symbols too.

The fact that multiplication is commutative means that equations involving sev-
eral multiplications can be written in any order. So Equation 3.14 could be (and
sometimes is) written as q = cm ∆T instead of q = mc ∆T . Addition is also com-
mutative, so Equation 3.16 could be written as v x = a x t +u x instead of v x = u x +a x t.

Back J I 136
Contents 
Although the order in which multiplications are written doesn’t matter, various con-
ventions are generally applied. Note that in Equation 3.3 (C = 2π r), the number 2
is written first, then the constant π, then the variable r. This order (numbers, then
constants, then variables) is the one that is generally applied. Similarly, E = mc2
(Equation 3.7) could be written as E = c2 m, but it generally isn’t! Variables that
are raised to a power tend to appear at the end of equations.
BEDMAS tells us that operations within brackets take precedence, i.e. operations
inside brackets should be evaluated before those outside the brackets. When work-
ing with symbols, this means that an operation applied to a bracket
 applies to every-
2GM
thing within the bracket. So in Equation 3.19, the whole of R is raised to the
power 12 . Equation 3.20 uses two sets of brackets (different styles of brackets have
been used to avoid confusion). The inner, round brackets ( ) are used to indicate
that L should be divided by the whole of (4π F) and the outer, square brackets [ ]
are used to indicate that the whole of L/ (4π F) should be raised to the power 21 .
There are two further points to note that are linked to the use of brackets.
1 A square root sign and a horizontal line used to indicate division can both be
thought of as containing invisible brackets, i.e. the square root sign is taken to
apply to everything within the sign and the division applies to everything!above
µ
the line. So, in Equation 3.10, the square root applies to the whole of , (this
√ ρ
µ µ
r
means that could be written as √ ), and in Equation 3.15 the whole of
ρ ρ
(vi + vf ) should be divided by two.

Back J I 137
Contents 
2 Throughout this course, brackets are sometimes used for added clarity even when
this is not strictly necessary. In addition, you are encouraged to add your own
brackets whenever you think doing so would make the meaning of an equation
clearer.
The ‘E’ in BEDMAS (see Section 1.4) tells us that exponents take precedence over
divisions and multiplications, so in Equation 3.7 (E = mc2 ) the c must be squared
before being multiplied by m. This means that it is only
 the c that is squared, not
the m. For clarity you could write this as E = m c , but it is very important to
2

remember that mc2 , (mc)2 , i.e. that mc2 , m2 c2 , where the symbol , means ‘is
not equal to’.
BEDMAS also reminds us that multiplications should be carried out before addi-
tions and subtractions, so in Equation 3.16, a x and t should be multiplied together
before u x is added.
Finally, note that all of the rules discussed in Chapter 1 for the writing and manipu-
lation of fractions and powers apply when using symbols, in exactly the same way
as they do when using numbers. So, Equation 3.17 could be written as s x = u x t +
a x t2 Gm1 m2
instead of s x = u x t + 21 a x t2 ; Equation 3.18 could be written as Fg =
2 r2
m1 m2
instead of Fg = G ; and the following two representations of Equation 3.20,
r2
although they look very different, are actually identical in meaning:
r
L
d= d = [L/(4π F)]1/2
4π F

Back J I 138
Contents 
Question 3.11 Answer
Which two pairs of equations for a of those given below are equivalent? You
should be able to answer this question by just looking at the equations, but you
might like to check your answer by substituting values such as x = 3, y = 4,
z = 5.
(i) a = x(y + z)
(ii) a = xy + z
(iii) a = (y + z)x
(iv) a = x + yz
(v) a = z + yx

Back J I 139
Contents 
Question 3.12 Answer
Two of the equations given below for m are equivalent. Which two? Again, you
should attempt this question initially by simply looking at the equations.
bac2
(i) m =
d
b2 c2
(ii) m = a
d
bc2
(iii) m = a
d
abc2
(iv) m =
ad
b2 a2 c2
(v) m =
d

Back J I 140
Contents 
3.5.4 Using equations

Substituting values into equations provides a way of checking your understanding

of many of the techniques introduced in this chapter, especially the correct reading
of equations, the use of scientific notation, and the need to quote answers to an
appropriate number of significant figures. It also provides an opportunity for you to
extend your understanding of units in calculations and to begin to think about how
to choose an appropriate equation to use in answering a particular question. Don’t
worry about the science in the worked examples in this section; they are given as
illustrations of good practice for substituting values into equations.

Worked example 3.8

Use v x = u x + a x t (Equation 3.16) to find the speed reached after 0.45 s by a
stone thrown downwards from a cliff with initial speed 1.5 m s−1 . This situation
is illustrated in Figure 3.11. You can assume that the magnitude (size) of the
acceleration is 9.81 m s−2 , where m s−2 are the SI units of acceleration.
Answer
Equation 3.16 states that v x = u x +a x t, and we are trying to find v x . The question
tells us that
u x = 1.5 m s−1 a x = 9.81 m s−2 t = 0.45 s

Back J I 141
Contents 
Thus
   
v x = 1.5 m s−1 + 9.81 m s−2 × 0.45 s

where the units of a x are m s−2 and the units of t are s, so the units of a x t are
m s−2 × s. Simplifying this gives
m m ×
s m
m s−2 × s = 2
× s = = = m s−1
s s × s

s

So

v x = 1.5 m s−1 + 4.4145 m s−1

= 5.9 m s−1 to two significant figures,

i.e. the speed after 0.45 seconds is 5.9 m s−1 .

Back J I 142
Contents 
Note, from Worked example 3.8, the following points about the handling of units:

1 Calculations have been done in SI units.

2 Units have been included next to values at all times, and the units in the final
answers are both consistent with the working and what we would expect the
units of the final answer to be.

The second point follows from what was said about units in Section 3.1.1; we have
input values with units of m s−1 for initial speed, units of s for time, and units of
m s−2 for acceleration, and the units for final speed have worked out to be m s−1 .
We have not simply assumed the units for final speed to be m s−1 , but rather have
calculated the units for v x at the same time as calculating the numerical value. Han-
dling units in this way ensures that the answers are expressed as physical quantities
(with units), not just numbers. It also gives an easy way of checking a calculation.
If the final units in Worked example 3.8 had come out as m2 s−1 you might have
realized that, since these are not units of speed, you must have made a mistake.
It is good practice to work out the units in this way in all your scientific calculations.
To enable you to do this, Box 3.5 explains a little more about some of the derived
units that you will encounter in this course.

Back J I 143
Contents 
Box 3.5 Derived SI units
Box 2.1 introduced the SI base units, and since then you have encountered the
SI units of m s−1 for speed, kg m−3 for density and m s−2 for acceleration. These
units are combinations of the base units m, kg and s; other physical quantities
have units involving other base units too. Some physical quantities are so com-
monly used that their units have names and symbols of their own, even though
they could be stated as a combination of base units. Several of these derived
units are listed in Table 3.2. Note that if you become a sufficiently famous sci-
entist you are likely to end up with a unit named after you! The units in Table
3.2 are named after Sir Isaac Newton, James Prescott Joule, James Watt, Blaise
Pascal and Heinrich Hertz respectively.

Physical quantity Name Symbol Base unit

of unit for unit equivalent
force, such as weight newton N kg m s−2
energy joule J kg m2 s−2
power watt W kg m2 s−3
pressure pascal Pa kg m−1 s−2
frequency hertz Hz s−1

Table 3.2: Some derived units

Back J I 144
Contents 
Note also that many of the derived units are interlinked:

1J=1N×1m
1J
1W=
1s
1N
1 Pa =
1 m2
The following data may help to illustrate the sizes of the units:
• An eating apple has a weight of about 1 N on Earth;
• An athlete with mass 75 kg, sprinting at 9 m s−1 , has an energy of about
3000 J;
• A domestic kettle has a power rating of about 2500 W;
• Atmospheric pressure at sea-level is about 105 Pa;
• The human heart beats with a frequency of about 1.3 Hz.

Back J I 145
Contents 
To find the units of vesc in Worked example 3.9, you need to use the fact, from
Table 3.2, that 1 N = kg m s−2 . This worked example also provides a reminder of
the importance of converting to SI base units before beginning a calculation.

Worked example 3.9

!1/2
2GM
Use vesc = (Equation 3.19) to find the escape speed, vesc , needed
R
for an object to escape from the Earth’s gravitational attraction. The mass of
the Earth, M = 5.98 × 1024 kg, the radius of the Earth, R = 6.38 × 103 km and
G = 6.673 × 10−11 N m2 kg−2 .
Answer
Converting R to SI base units gives

R = 6.38 × 103 km
= 6.38 × 103 × 103 m
= 6.38 × 106 m

M = 5.98 × 1024 kg
G = 6.673 × 10−11 N m2 kg−2

Back J I 146
Contents 
Substituting in Equation 3.19
!1/2
2GM
vesc =
R
!1/2
2 × 6.673 × 10−11 N m2 kg−2 × 5.98 × 1024 kg
=
6.38 × 106 m

Rearranging this so that the units on the top of the fraction are all together we
get
!1/2
2 × 6.673 × 10−11 × 5.98 × 1024 N m2 kg−2 kg
vesc =
6.38 × 106 m

Since 1 N = 1 kg m s−2 , this can be rewritten as

!1/2
2 × 6.673 × 10−11 × 5.98 × 1024 kg m s−2 m2 kg−2 kg
vesc =
6.38 × 106 m

This can be simplified by cancelling some of the units

−2 !1/2
2 × 6.673 × 10−11 × 5.98 × 1024 kg 
 s−2 m2 
m kg

kg
vesc =
6.38 × 106 
m


Back J I 147
Contents 
Calculating the numeric value, and reordering the units, we have
 1/2
vesc = 1.2509 × 108 m2 s−2

Taking the square root of both 1.2509 × 108 and m2 s−2 gives

vesc = 1.12 × 104 m s−1 to three significant figures.

The escape speed is 1.12 × 104 m s−1 , with units of m s−1 , as expected.

Question 3.13 Answer

In a classic experiment in the USA in 1926, Edgar Transeau calculated the
amount of energy stored in the corn plants in a one-acre field in a 100-day grow-
ing period to be 1.06 × 108 kJ. This is NPP in Equation 3.8. For the same field
and the same time period, he found the energy used by the plants in respiration
(R) to be 3.23 × 107 kJ. Use Equation 3.8 to find the corresponding value of
GPP, the total energy captured by the plants.

Back J I 148
Contents 
Question 3.14 Answer
Use Equation 3.13 to find the speed of waves (in m s−1 ) which have a frequency
of 4.83 × 1014 Hz and a wavelength of 621 nm.

The final worked example in this section returns us to the piece of granite introduced
at the beginning of the chapter. It is perhaps a somewhat more realistic example than
Worked examples 3.8 and 3.9 because the question does not tell us which formula
to use.

Back J I 149
Contents 
Worked example 3.10
The rigidity modulus of granite (a measure of the rock’s ability to resist defor-
mation) near the surface of the Earth is 3.0 × 1010 Nm−2 . Use this value, and
the value you found previously for the density of granite to find the speed of S
waves travelling through granite.

Answer
Which equation shall we use? When faced by this dilemma it is best to start
by thinking carefully about what you already know and what you want to find.
On this occasion we’re told that the rigidity modulus is 3.0 × 1010 Nm−2 and we
know (from Worked example 3.7) that the density of granite is 2.7 × 103 kg m−3
(using a value to three significant figures to avoid rounding errors). We need
to find a value for S wave speed. So we need an equation r which links density,
µ
rigidity modulus and S wave speed; Equation 3.10 (vs = ) from Box 3.4 fits
ρ
the bill.

Simply finding an equation from a list, all that is possible in this course, is
somewhat unlike the situation you are likely to encounter in the real scientific
world. Nevertheless, the principle of starting each question by thinking about
what you already know and what you want to find is a good one, and on this
occasion it makes it straightforward to find an equation to use from Box 3.4.

Back J I 150
Contents 

µ
r
vs =
ρ
µ = 3.0 × 1010 N m−2
ρ = 2.70 × 103 kg m−3

So
s
3.0 × 1010 N m−2
vs =
2.70 × 103 kg m−3

Since 1 N = 1 kg m s−2 , this can be rewritten as

s
3.0 × 1010 kg m s−2 m−2
vs =
2.70 × 103 kg m−3

This can be simplified by cancelling the kg on top and bottom of the fraction
s
3.0 × 1010 kg m s−2 m−2
vs =
2.70 × 103 kg m−3

Back J I 151
Contents 
Calculating the numeric value, and combining the m and m−2 on the top of the
fraction with the m−3 on the bottom, we have
p
vs = 1.11 × 107 m2 s−2
= 3.3 × 103 m s−1 to two significant figures

So the S waves travel with a speed of 3.3 × 103 m s−1 through granite.

Question 3.15
The Earth has an average radius of 6.38 × 103 km and a mass of 5.97 × 1024 kg.
The Moon has a mass of 7.35 × 1022 kg. The distance between the Earth and
the Moon is 3.84 × 105 km and G = 6.673 × 10−11 N m2 kg−2 . Use appropriate
equations from Box 3.4 to calculate:
(a) the Earth’s volume (in m3 ); Answer
(b) the magnitude of the gravitational force between the Earth and Answer
the Moon (in newtons).
Note: on this occasion you should be able to work out the final units of your
answer without expressing newtons in the form of base units. This is further
discussed in the answer to the question.

Back J I 152
Contents 
3.6 Learning outcomes for Chapter 3
After completing your work on this chapter you should be able to:
3.1 demonstrate understanding of the terms emboldened in the text;
3.2 perform calculations to an appropriate number of significant figures;
3.3 give answers to calculations in appropriate SI units;
3.4 carry out calculations in scientific notation, both with and without the use of a
scientific calculator;
3.5 estimate answers to one significant figure;
3.6 convert between various units for quantities such as area, volume, speed,
density and concentration;
3.7 demonstrate understanding of the rules and conventions used in scientific
formulae;
3.8 substitute values (numbers and units) into scientific formulae.

Back J I 153
Contents 

Algebra 4
µ
r
At the end of Chapter 3 we used the equation vs = to calculate the S wave
ρ
speed, vs , of seismic waves passing through a rock of density ρ and rigidity modulus
µ. But suppose that, instead of knowing ρ and µ and wanting to find vs ,rwe know vs
µ
and ρ and want to find µ. The best way to proceed is to rearrange vs = to make
ρ
µ the subject of the equation, where the word ‘subject’ is used to mean the term
written by itself, usually to the left of the equals sign. Rearranging equations is the
first topic considered in Chapter 4. The rest of the chapter introduces methods for
simplifying equations and ways of combining two or more equations together, and
it ends with a look at ways of using algebra to solve problems.

Back J I 154
Contents 
4.1 Rearranging equations
There are many different methods taught for rearranging equations, and if you are
happy with a method you have learnt previously it is probably best to stick with
this method, provided it gives correct answers to all the questions in this section.
However, if you have not found a way of rearranging equations which suits you,
you might like to try the method highlighted in the blue-toned boxes throughout
this section. This method draws on an analogy between an equation and an old-
fashioned set of kitchen scales, and considers the equation to be ‘balanced’ at the
equals sign. The scales will remain balanced if you add a 50 g mass to one side of
the scales, or halve the mass on one side, provided you do exactly the same thing to
the other side. In a similar way, you can do (almost) anything you like to one side
of an equation and, provided you do exactly the same thing to the other side, the
equation will still be valid. This point is illustrated in Figure 4.1.
The following rule summarizes the discussion above:

Whatever you do mathematically to one side of an equation you must also do to

the other side.

This rule is fundamental when rearranging equations, but it doesn’t tell you what
operation to perform to both sides of an equation in order to rearrange it in the way
you want. The highlighted points below should help with this, as will plenty of
practice.

Back J I 155
Contents 
Two things are worth noting at the outset:
1 Equations are conventionally written with the subject on the left-hand side of the
equals sign. However, when rearranging an equation it can very often be helpful
simply to reverse the order.
So if you derive or are given the equation c = a + b you can rewrite it as a + b = c;
if you derive or are given the equation ab = c you can rewrite it as c = ab.
2 Even if you choose the ‘wrong’ operation, provided you correctly perform that
operation to both sides of the equation, the equation will still be valid. Suppose
we want to rearrange the equation c = a + b to obtain an expression for a. We
could divide by two, as illustrated by Figure 4.1c; this gives
c a+b
=
2 2
This is a perfectly valid equation; it just doesn’t help much in our quest for a.
The numbered points below give some hints for more helpful ways forward, and
each guideline is followed by an illustration of its use.
In the numbered hints the words expression and term are used to describe the parts
of an equation. An equation must always include an equals sign, but an expression
or term won’t. A term may be a single variable (such as v x or u x in the equation
v x = u x + a x t, or a combination of several variables (such as a x t); an expression is
usually a combination of variables (such as a x t or u x + a x t, but the words are often
used interchangeably.

Back J I 156
Contents 
Hint 1
If you want to remove an expression that is added to the term you want, subtract
that expression from both sides of the equation.

To rearrange a + b = c to make a the subject, note that we need to remove the b

from the left-hand side of the equation. The b is currently added to a, so we need
to subtract b from both sides. This gives

a+b−b=c−b

or

a=c−b (since b − b = 0)

Hint 2
If you want to remove an expression that is subtracted from the term you want,
add that expression to both sides of the equation.

To rearrange a − b = c to make a the subject, note that we need to remove the b

from the left-hand side of the equation. The b is currently subtracted from a, so we
need to add b to both sides. This gives

a−b+b=c+b

Back J I 157
Contents 
or

a=c+b (since − b + b = 0)

Hint 3
If the term you want is multiplied by another expression, divide both sides of the
equation by that expression.

To rearrange ab = c to make a the subject, note that we need to remove the b from
the left-hand side of the equation. The a is currently multiplied by b, so we need to
divide both sides of the equation by b. This gives
ab c
=
b b
The b in the numerator of the fraction on the left-hand side cancels with the b in the
denominator to give
c
a=
b

Hint 4
If the term you want is divided by another expression, multiply both sides of the
equation by that expression.

Back J I 158
Contents 
a
To rearrange = c to make a the subject, note that we need to remove the b from
b
the left-hand side of the equation. The a is currently divided by b, so we need to
multiply both sides of the equation by b. This gives
a×b
=c×b
b
The b in the numerator of the fraction on the left-hand side cancels with the b in the
denominator to give

a = cb

Hint 5
If you are trying to make a term the subject of an equation and you currently
have an equation for the square of that term, take the square root of both sides
of the equation.

To rearrange a2 = b to make a the subject, note that the a is currently squared, and
take the square root of both sides of the equation. This gives
√
a=± b

Note the presence of the ± sign, indicating that the answer could be either positive
or negative, as discussed in Section 1.1.3. In practice, the reality of the problem we
are solving sometimes allows us to rule out one of the two values.

Back J I 159
Contents 
Hint 6
If you are trying to make a term the subject of an equation and you currently have
an equation for the square root of that term, square both sides of the equation.
√
To rearrange a = b to make a the subject, note that you currently have an equation
for the square root of a, and square both sides of the equation. This gives

a = b2

Hints 1 to 6 all follow from a general principle:

To ‘undo’ an operation (e.g. +, −, ×, ÷, square, square root) you should do the

opposite, (i.e. −, +, ÷, ×, square root, square).

The following worked examples use the principles introduced in the numbered hints
above, in the context of equations which are frequently encountered in science.
Worked example 4.1 also involves substituting numerical values and units into the
equation once it has been rearranged.

Back J I 160
Contents 
Worked example 4.1
As discussed in Box 2.1, mass and weight are not the same. However, the mag-
nitude of the weight, W, of an object at the surface of the Earth and its mass, m,
are related by the equation W = mg. The magnitude of the acceleration due to
gravity, g, can be taken as 9.81 m s−2
A teenager’s weight is 649 N. What is his mass?
Answer
We need to start by rearranging W = mg to make m the subject of the equation.
It is helpful to start by reversing the order of the equation, i.e. to write it as
mg = W

To isolate m we need to get rid of g, and m is currently multiplied by g so, from

Hint 3 we need to divide by g. Remember that we must do this to both sides of
the equation, so we have
mg W
=
g g

The g in the numerator of the fraction on the left-hand side cancels with the g in
the denominator to give
W
m=
g

Back J I 161
Contents 
Substituting values for W and g gives
649 N
m=
9.81 m s−2

Since 1 N = 1 kg m s−2 (from Table 3.2) and

N m
kg  s−2

=

m s−2 m

 s−2


we then have
649 kg m s−2
m= = 66.2 kg
9.81 m s−2

So the teenager’s mass is 66.2 kg

Back J I 162
Contents 
Worked example 4.2
The time T for one swing of a pendulum is related to its length, L, by the equa-
tion
4π2 L
T2 =
g
where g is the magnitude of the acceleration due to gravity. Write down an
equation for T .
Answer
T is currently squared, so from Hint 5, we need to take the square root of both
sides of the equation. This gives
s
4π2 L
T=
g

Since T is a period of time, its value must be positive, so we only need to write
down the positive square root.

Back J I 163
Contents 
Question 4.1
(a) Rearrange v = f λ to make f the subject. Answer
(b) Rearrange Etot = Ek + Ep so that Ek is the subject. Answer
m
(c) Rearrange ρ = to obtain an equation for m. Answer
V

Back J I 164
Contents 
When rearranging more complicated equations, it is often necessary to proceed in
several steps. Each step will use the rules already discussed, but many people are
perplexed when trying to decide which step to take first. Expertise in this area
comes largely with practice, and there are no hard and fast rules (it is often possible
to rearrange an equation by several, equally correct, routes). However, the following
guidelines may help:

Hint 7
Don’t be afraid of using several small steps to rearrange one equation.
Hint 8
Aim to get the new subject into position on the left-hand side as soon as you can.
(This will not always be possible straight away.) Simply reversing an equation
can sometimes be a helpful initial step.
Hint 9
You can treat an expression within brackets as if it was a single term. This is true
whether the brackets are shown explicitly in the original equation or whether
you have added them (or imagined them) for clarity. If the quantity required
as the subject is itself part of an expression in brackets in the original equation,
it is often best to start by making the whole bracketed term the subject of the
equation.

Back J I 165
Contents 
Let’s look at these guidelines in the context of a series of worked examples, in-
terspersed with questions for you to try for yourself. Note that although ‘real’
science equations have been used as much as possible in the worked examples and
questions, the symbols have not been explained, and you do not need to know the
meaning of them. This is to allow you to concentrate, for the time being only, on
the algebra rather than getting side-tracked into the underlying science.
You may be able to rearrange the equations in the following worked examples in
fewer steps than are shown, but if you are in any doubt at all it is best to write down
all the intermediate steps in the process.

Back J I 166
Contents 
Worked example 4.3
Rearrange PV = nRT to give an equation for T .
Answer
This example is perhaps more straightforward than it looks, but it is best to
proceed in steps.
The first step is to reverse the equation so that the T is on the left-hand side
(from Hint 8). This gives

nRT = PV

We now need to remove the nR by which the T is multiplied. Dividing both

sides by nR gives
nRT PV
=
nR nR
The nR in the numerator of the fraction on the left-hand side cancels with the
nR in the denominator to give
PV
T=
nR

Back J I 167
Contents 
Worked example 4.4
m
Rearrange ρ = so that V is the subject.
V
Answer
The first step is to multiply both sides by V (thus getting V into the right position,
as in Hint 8). This gives
mV
ρV =
V
that is

ρV = m

Then dividing both sides by ρ gives

ρV m
=
ρ ρ
that is
m
V=
ρ

Back J I 168
Contents 
Worked example 4.5
Rearrange v x = u x + a x t to make u x the subject.
Answer
This equation can be written as

ux + axt = vx

which has u x on the left-hand side (Hint 8).

We can treat the expression a x t as a single term (by considering there to be
brackets around it, as in Hint 9) and subtract it from both sides of the equation
to give

ux + axt − axt = vx − axt

that is

ux = vx − axt

Back J I 169
Contents 
Worked example 4.6
Rearrange h = 12 gt2 to give an equation for t.
Answer
We can consider there to be brackets around (t2 ) and start by finding an expres-
sion for t2 (Hint 9). The equation can be written as
1 2
gt = h
2

which has t2 on the left-hand side (Hint 8). Multiplying both sides by 2 gives
1
2 × gt2 = 2h
2
that is
gt2 = 2h

Dividing both sides by g gives

gt2 2h
=
g g
that is
2h
t2 =
g

Back J I 170
Contents 
Now we can take the square root of both sides to give
s
2h
t=±
g

Back J I 171
Contents 
Worked example 4.7
µ
r
Rearrange vs = so that µ is the subject.
ρ
Answer
µ
!
We can consider there to be brackets around and start by finding an expres-
ρ
µ
!
sion for (Hint 9).
ρ
µ µ
r
The equation can be written as = vs , which has on the left-hand side
ρ ρ
(Hint 8).
Squaring both sides gives
µ
= v2s
ρ

Now we can multiply both sides by ρ, to give µ = v2s ρ .

Back J I 172
Contents 
Box 4.1 Interlude: why bother with algebra?
You may have recognized the equation rearranged in Worked example 4.7; it was
the one discussed at the beginning of the chapter. Thinking back to the beginning
of the chapter reminds us of the purpose of what we are doing. The ability to
rearrange equations is useful (arguably the most useful skill developed in this
course), but it’s not something that you should do just for the sake of doing so,
but rather because you want to work something out, and rearranging an equation
is the means to this end. Suppose you have been told that S waves pass through
rocks of density ρ = 3.9 × 103 kg m−3 with a speed vs = 3.0 × 103 m s−1 r, and
µ
you want to find the rigidity modulus µ. The equation in the form vs = is
ρ
not much use, but the rearranged form immediately tells us that

µ = v2s ρ
 2  
= 3.0 × 103 m s−1 × 3.9 × 103 kg m−3
= 3.5 × 1010 m2 s−2 kg m−3
= 3.5 × 1010 kg m−1 s−2

So the rigidity modulus is 3.5 × 1010 kg m−1 s−2 .

Back J I 173
Contents 
Question 4.2
(a) Rearrange b = c − d + e so that e is the subject. Answer
(b) Rearrange p = ρgh to give an equation for h. Answer
2GM
(c) Rearrange v2esc = to make R the subject. Answer
R
(d) Rearrange E = h f − φ so that φ is the subject. Answer
bc2
(e) Rearrange a = to give an equation for c. Answer
d
r
b
(f) Rearrange a = to make b the subject. Answer
c

Back J I 174
Contents 
Question 4.3
The mass, m, speed, v, and kinetic energy, Ek , of an object are linked by the
equation Ek = 21 mv2 .
(a) Rearrange this equation so that v is the subject. Answer
(b) Use your answer to part (a) to estimate (in m s−1 to one signif- Answer
icant figure) the speed needed in order for a tectonic plate of
mass 4 × 1021 kg to have a kinetic energy of 2 × 103 J.
(c) Use your answer to part (a) to estimate (in m s−1 to one signif- Answer
icant figure) the speed needed in order for an athlete of mass
70 kg to have the same kinetic energy as the tectonic plate in
part (b).

The final group of worked examples in this section involve equations which may
appear rather more complex than the previous ones, but they can all be rearranged
using the rules and guidelines already introduced. Some, like Worked example 4.8,
appear more complex partly because they use symbols that are rather unwieldy.
However, these final worked examples are genuinely more complicated too, and are
best solved by taking a logical stepwise approach (as the early Arab mathematicians
did; see Box 4.2). Rearranging complicated equations is rather like peeling away
layers of an onion, systematically removing layer by layer in order to get to the part
you want. But that doesn’t mean it should end in tears!

Back J I 175
Contents 
Box 4.2 Al-Khwarizmi and al-jabr
The techniques of algebra have developed over a period of several thousand
years, but the word ‘algebra’ comes from ‘al-jabr’ in the title of a book written
by Mohammed ibn-Musa al-Khwarizmi in about 830. The book, whose title
Hisab al-jabr w’al muqabela, can be translated as ‘Transposition and reduction’,
explained how it was possible to reduce any problem to one of six standard
forms using the two processes, al-jabr (transferring terms to eliminate negative
quantities) and muqabela (balancing the remaining positive quantities).
Arab mathematicians like al-Khwarizmi did not use symbols in their work, but
rather explained everything in words. Nevertheless, their stepwise approach
was very similar to the one advocated in this course. Al-Khwarizmi is also
remembered for his work on the solution of quadratic equations, discussed later
in this chapter.

A little less working is shown in Worked examples 4.8, 4.9 and 4.10 than previ-
ously, and hints are not explicitly referred to. This has been done so as to make the
working more akin to what you might reasonably write when working through the
questions in this course. You are encouraged to show as many steps as necessary in
your working, and to use words of explanation wherever they help you.

Back J I 176
Contents 
Worked example 4.8
Rearrange ∆G m = ∆Hm
− T ∆S so that ∆S is the subject.
m m

(Note: ∆G m , ∆Hm
and ∆S each represent a single physical entity.)
m

Answer
Adding T ∆S m
to both sides of the equation gives

∆G m + T ∆S m

= ∆Hm

Subtracting ∆G m from both sides gives

T ∆S m

= ∆Hm

− ∆G m

Dividing both sides by T gives

∆Hm
− ∆G
m
∆S m

=
T

Back J I 177
Contents 
Worked example 4.9
Rearrange s x = u x t + 12 a x t2 to make a x the subject.
Answer
The equation can be written as u x t + 12 a x t2 = s x .
Subtracting u x t from both sides gives
1
2 axt
2
= sx − uxt

Multiplying both sides by 2 gives

a x t2 = 2(s x − u x t)

Dividing both sides by t2 gives

2(s x − u x t)
ax =
t2

Back J I 178
Contents 
Worked example 4.10
m1 m2
Rearrange Fg = G 2 to give an equation for r.
r
Answer
m1 m2 Gm1 m2
Note that Fg = G 2 can be written as Fg = (see Section 3.5.3).
r r2
We can get the r2 into position on the left-hand side by multiplying both sides
by r2 . This gives

Fg r2 = Gm1 m2

Dividing both sides by Fg gives

Gm1 m2
r2 =
Fg

Taking the square root of both sides gives

s
Gm1 m2
r=±
Fg

Back J I 179
Contents 
Box 4.3 Using algebra in astronomy
The luminosity of a star (the total rate at which it radiates energy into space, in
all directions), L, is related to its radius, R, and the temperature (measured in
kelvin), T , of its outer layer (called the photosphere) by the equation

L = 4π R2 σT 4 (4.1)

where σ (the lower case Greek letter sigma) represents a constant known as
Stefan’s constant, with the value σ = 5.67 × 10−8 W m−2 K−4 .

It is impossible to take direct readings for the luminosity, radius or tempera-

ture of distant stars, but indirect measurements can lead to values for photo-
spheric temperature and luminosity. Figure 4.2 is a so-called Hertzsprung–
Russell diagram, comparing the photospheric temperatures and luminosity of
different stars. Note that different types of stars appear in distinct groupings on
the Hertzsprung–Russell diagram.

If we know a star’s luminosity and photospheric temperature we can find its

radius from Equation 4.1, but first of all we need to rearrange the equation to
make R the subject.

Equation 4.1 can be reversed to give

4π R2 σT 4 = L

Back J I 180
Contents 
Dividing both sides by 4π σT 4 gives

L
R2 =
4σT 4

(Note that the same results would have been achieved by dividing by 4, π, σ and
T 4 separately.)
Taking the square root of both sides gives
r
L
R=±
4σT 4
Since R is the radius of a star, we are only interested in the positive value.
The star Alcyone (in the Pleiades) has a photospheric temperature of 1.2 × 104 K
and a luminosity of 3.2 × 1029 W. So its radius is
s
3.2 × 1029 W
R=
4
4 × 5.67 × 10−8 W m−2 K−4 × 1.2 × 104 K
s
3.2 × 1029 
W
=


4 4
4 × 5.67 × 10−8 W m−2 K−4

1.2 × 104  K
p
= 2.17 × 1019 m2
= 4.7 × 109 m

Back J I 181
Contents 
The radius of Alcyone is 4.7 × 109 m.
Notice that in this example, the units of watts cancelled without having to be
expressed in SI base units.

Question 4.4
(a) Rearrange v x = u x + a x t so that a x is the subject. Answer
µ
r
(b) Rearrange vs = to make ρ the subject. Answer
ρ
L
(c) Rearrange F = to give an equation for d. Answer
4π d2

Back J I 182
Contents 
4.2 Simplifying equations
Sometimes it is possible (and helpful) to write an algebraic expression in a different
form from the one in which it is originally presented. Whenever possible you should
aim to write equations in their simplest form, i.e. to simplify them. For example,
a 3a a
you will see in this section that the equation c = + can be simplified to c = ;
4b 4b b
the latter form of the equation is rather more useful than the former.
In order to simplify equations it is often necessary to apply the rules for the manip-
ulation of fractions and brackets that were introduced in Chapter 1.

4.2.1 Simplifying algebraic fractions

Algebraic fractions can be multiplied and divided in exactly the same way as nu-
merical fractions, using the methods introduced in Section 1.2.4 and Section 1.2.5.
So just as
2 4 2×4 8
× = = (multiplying numerators and denominators together)
3 5 3 × 5 15

we can write
a c a × c ac
× = =
b d b × d bd

Back J I 183
Contents 
Similarly, just as
2 5 2 7 5
÷ = × (turning the upside down and multiplying)
3 7 3 5 7
2×7
=
3×5
14
=
15
we can write
a c a d c
÷ = × (turning the upside down and multiplying)
b d b c d
a×d
=
b×c
ad
=
bc

Back J I 184
Contents 
Worked example 4.11 illustrates a division in which several of the terms cancel.

Worked example 4.11

2ab 2
Simplify ÷
c c
Answer
2
Turning the upside down and multiplying gives
c
2ab 2 2ab c
÷ = ×
c c c 2

We can cancel the ‘2’s and the ‘c’s to give

2ab 2 2A ab c

÷ = × = ab
c c c
2A

Back J I 185
Contents 
The method described in Section 1.2.2 for adding and subtracting numerical frac-
tions can also be extended to algebraic fractions. We need to find a common de-
nominator in a similar way, so, much as we can write
2 4 2 × 5 4 × 3 10 12 10 + 12 22
+ = + = + = =
3 5 3 × 5 5 × 3 15 15 15 15
where the common denominator is the product of the denominators of the original
fractions, we can also write
a c ad cb ad + cb
+ = + =
b d bd db bd

Worked example 4.12

Electrical resistors can be combined together in various ways. You don’t need
to know or understand the scientific details, but when three resistors of resis-
tance R1 , R2 and R3 are combined in a particular way (‘in parallel’) the effective
resistance is given by the term Reff in the equation
1 1 1 1
= + + (4.2)
Reff R1 R2 R3
Rearrange Equation 4.2 to make Reff the subject.
Answer
We need to start by expressing the right-hand side of Equation 4.2 as a single

Back J I 186
Contents 
fraction. The product of R1 , R2 and R3 will be a common denominator, so we
can write
1 1 1 1
= + +
Reff R1 R2 R3
R2 R3 R1 R3 R1 R2
= + +
R1 R2 R3 R1 R2 R3 R1 R2 R3
R2 R3 + R1 R3 + R1 R2
=
R1 R2 R3
1
In order to make Reff the subject of the equation, rather than , we could mul-
Reff
tiply and divide both sides of the equations by a series of expressions. However,
it is more straightforward simply to turn the equation upside down, i.e. to take
the reciprocal of both sides. This gives
R1 R2 R3
Reff =
R2 R3 + R1 R3 + R1 R2

A note of caution when simplifying algebraic expressions

When you simplify an algebraic expression, especially one involving fractions, the
answer you arrive at doesn’t always look very simple! If you are asked to simplify
an expression which is the sum or product of two separate fractions, your answer

Back J I 187
Contents 
should normally be a single fraction, but an expression like
R1 R2 R3
Reff =
R2 R3 + R1 R3 + R1 R2
(the answer to Worked example 4.12) may be the simplest you can give. It can be
very tempting to ‘cancel’ terms incorrectly in an attempt to get to the sort of simple
fraction which is generally achievable when simplifying numerical fractions, but
less likely to be achievable when dealing with symbols.

Question
√
2c a (b + 2)
Express × √ as a single fraction of the simplest possible form.
(a + 2) 2c b
Answer
We can cancel the ‘2c’s to give
√ √
2c
 a (b + 2) a (b + 2)
√ =

× √
(a + 2)  2c
 b (a + 2) b
√
a (b + 2)
= √
b (a + 2)
It can be tempting to ‘cancel’ the square roots and the ‘+2’s too, but this would
be incorrect:
√
a a (b + 2) b
√ , and ,
b b (a + 2) a

Back J I 188
Contents 
As discussed in Section 1.2.3, a fraction is unchanged by the multiplication or
the division of both its numerator and denominator by the same amount. How-
ever, all other operations will alter its value.
√
a (b + 2) 2c a (b + 2)
r
So is as far as it is possible to simplify × √ .
b (a + 2) (a + 2) 2c b
√ √
a (b + 2)
r
a a
Note however that is equivalent to √ , so √ can also be written
b b b (a + 2)
a (b + 2)
r
as .
b (a + 2)

Back J I 189
Contents 
Question 4.5
Simplify the following expressions, giving each answer as a single fraction.
µ0 i 1 i 2
(a) × Answer
2π d
,
3a
(b) 2 Answer
2b
2b 3c
(c) + Answer
c b
2ab 2ac
(d) ÷ Answer
c b
1 1
(e) − Answer
f f +1
2b2 2c2
(f) ÷ Answer
(b + c) (a + c)

Back J I 190
Contents 

image
lens of object
on film
u v

object

Figure 4.3: The object and image of a simple camera.

Question 4.6 Answer

The distance, u, of an object from a lens (such as the lens in the simple camera
illustrated in Figure 4.3) is related to the distance, v, from the lens to the image
of the object (on the film) and the lens’s focal length, f , by the equation
1 1 1
+ =
u v f
Add the fractions 1/u and 1/v and hence rearrange the equation to give an ex-
pression for f .

Back J I 191
Contents 
4.2.2 Using brackets in algebra

You should be familiar by now with the notion that an operation applied to an
expression in a bracket must be applied to everything within the bracket, so
(2b)2 = 22 b2 = 4b2
(a + 2b) − (a + b) = a + 2b − a − b = b
(a + 2b) − (a − b) = a + 2b − a − (−b) = a + 2b − a + b = 3b
2(a + 2b) = (2 × a) + (2 × 2b) = 2a + 4b
and
2a(a + 2b) = (2a × a) + (2a × 2b) = 2a2 + 4ab
If we need to multiply two bracketed expressions, such as (a+b) and (c+d) together,
we need to multiply each term in the first bracket by each term in the second bracket
as indicated by the red lines shown below.

(a + b)(c + d)

Multiplying the terms in order gives

2
1 1 2 3 4
(a + b)(c + d) = ac + ad + bc + bd
3
4

Back J I 192
Contents 
Worked example 4.13
Rewrite the following expressions so that the brackets are removed:
(a) (x − 3)(x + 5)
(b) (x + y)(x − y)
(c) (x + y)2
(d) (x − y)2
Answer
(a) (x − 3)(x + 5) = x 2 + 5 x − 3x − 15
= x 2 + 2 x − 15

(b) (x + y)(x − y) = x 2 − xy + yx − y 2
= x 2 − y2 since xy = yx , so − xy + yx = 0

(c) (x + y) 2 = (x + y)( x + y)
= x 2 + xy + yx + y 2
= x 2 + 2 xy + y 2

(d) (x − y) 2 = (x − y)( x − y)
= x 2 − xy − yx + y 2
= x 2 − 2xy + y 2

Back J I 193
Contents 
An examination of the answers to parts (b), (c) and (d) of Worked example 4.13
serves as a reminder of the fact that

(x + y)2 , x2 + y2
(x − y)2 , x2 − y2

In other words, remember to watch out for brackets!

Question 4.7
Rewrite the following expressions so that the brackets are removed:
1
(a) (v x + u x ) t Answer
2
(a − b) − (a − c)
(b) Answer
2
(c) (k − 2)(k − 3) Answer
(d) (t − 2)2 Answer

Back J I 194
Contents 
So far, this section has discussed removing brackets from expressions, but it can
very often be useful to do the reverse.
The numbers 6 and 4 are described as factors of 24 and in general, when speaking
mathematically, ‘factors’ are terms which when multiplied together give the original
expression. Since, for example,

y (y + 3) = y2 + 3y

we can say that y and (y + 3) are factors of y2 + 3y

Similarly, since

(x + 3)( x − 1) = x 2 − x + 3x − 3
= x2 + 2 x − 3

we can say that (x + 3) and (x − 1) are factors of x2 + 2x − 3.

The verb ‘to factorize’ means to find the factors of an expression. If you are asked
to factorize y2 + 3y then you should write

y2 + 3y = y(y + 3)

and if you are asked to factorize x2 + 2x − 3 you should write

x2 + 2x − 3 = (x + 3)(x − 1)

Back J I 195
Contents 
Note, from Worked example 4.13b, that the factors of x2 − y2 are (x + y) and (x − y),
i.e.

x2 − y2 = (x + y)(x − y) (4.3)

The difference of two squared numbers can always be written as the product of
their sum and their difference.

Question 4.8
Factorize the following expressions:
(a) y2 − y Answer
(b) x2 − 25 (Hint: you may find it helpful to compare this ex- Answer
pression with Equation 4.3, remembering that 52 = 25.)

Factorizing can be useful when rearranging equations, as Worked example 4.14

illustrates.

Back J I 196
Contents 
Worked example 4.14
Rearrange q = mc ∆T + mL so that m is the subject.
Answer
Both terms on the right-hand side of this equation include m, so we can rewrite
the equation as

q = m (c ∆T + L)

This can be reversed to give

m (c ∆T + L) = q

Now we divide both sides by (c ∆T + L) to give

q
m=
c ∆T + L

Question 4.9 Answer

Rearrange Etot = 12 mv2 + mg ∆h to give an equation for m.

Back J I 197
Contents 
An ability to factorize expressions such as y2 + 3y and x2 + 2x − 3 can also help us
to find the solutions of equations such as y2 + 3y = 0 and x2 + 2x − 3 = 0. Equations
of this form are known as ‘quadratic equations’.
We know from above that

y2 + 3y = y (y + 3) (4.4)

So if y2 + 3y = 0, it follows that y (y + 3) = 0 too. Multiplying by zero gives zero (as

discussed in Section 1.1.4). So y (y + 3) = 0 implies that either y = 0 or y + 3 = 0.
y + 3 = 0 implies that y = −3, so the solutions of y2 + 3y = 0 are y = 0 and y = −3.
We can check that y = 0 and y = −3 really are solutions of the equation y2 + 3y = 0,
by substituting each value for y into the left-hand side of the equation and verifying
that it gives 0, as expected.
For y = 0, y2 + 3y = 0 + 0 = 0, as expected.
 
For y = −3, y2 + 3y = (−3)2 + 3 × (−3) = 9 + (−9) = 0, as expected.
It is sensible to check your answers in this way:

You should check your answers whenever possible.

Back J I 198
Contents 
Worked example 4.15
Use the fact that

x2 + 2x − 3 = (x + 3)(x − 1) (4.5)

to find the solutions of the equation x2 + 2x − 3 = 0.

Answer
If x2 + 2x − 3 = 0 then, from Equation 4.5, (x + 3)(x − 1) = 0
Thus x + 3 = 0 or x − 1 = 0, i.e. x = −3 or x = 1.
Checking for x = −3:

x2 + 2x − 3 = (−3)2 + 2(−3) − 3 = 9 − 6 − 3 = 0, as expected.

Checking for x = 1:

x2 + 2x − 3 = 12 + (2 × 1) − 3 = 1 + 2 − 3 = 0, as expected.

So the solutions of the equation x2 + 2x − 3 = 0 are x = −3 and x = 1.

Back J I 199
Contents 
Using factorization to solve quadratic equations relies on us being able to spot the
factors of an expression; this is quite easy for expressions like y2 + 3y (see Equa-
tion 4.4), but if we had not known or been told that x2 + 2x − 3 = (x + 3)(x − 1)
(Equation 4.5), finding the factors of x2 +2x −3 would have been largely a matter of
trial and error. An ability to find factors in this way can be developed with practice,
but it remains somewhat tedious and this method for solving quadratic equations
doesn’t work at all unless the solutions are whole numbers or simple fractions. For-
tunately help is at hand in the form of the ‘quadratic equation formula’, described
in Box 4.4.

Box 4.4 The quadratic equation formula

Al-Khwarizmi and other early Arab mathematicians developed general methods
for solving quadratic equations. A quadratic equation of the form
ax2 + bx + c = 0

will have solutions given by the quadratic equation formula

√
−b ± b2 − 4ac
x=
2a
If b2 > 4ac (i.e. b2 is greater than 4ac) then b2 − 4ac will be positive, and the
formula will lead to two distinct solutions.
If b2 = 4ac then b2 − 4ac = 0, so the two solutions will be identical
(x = −b/(2a)).

Back J I 200
Contents 
If b2 < 4ac (i.e. b2 is less than 4ac) then b2 − 4ac will be negative. This means
that the solutions will include the square root of a negative number. and hence
will involve ‘imaginary numbers’. Such numbers were mentioned in Chapter 1,
but will not be considered further in Maths for Science.

Worked example 4.16 demonstrates the use of the quadratic equation formula in
solving the equation that was solved by factorization in Worked example 4.15.

Worked example 4.16

Use the quadratic equation formula to find the solutions of the equation
x2 + 2x − 3 = 0.
Answer
Comparison of

x2 + 2x − 3 = 0

and

ax2 + bx + c = 0

Back J I 201
Contents 
shows that a = 1, b = 2 and c = −3 on this occasion, so the solutions are
√
−b ± b2 − 4ac
x=
q2a  
−2 ± 22 − 4 × 1 × (−3)
=
√ 2×1
−2 ± 4 − (−12)
=
√2
−2 ± 16
=
2
−2 ± 4
=
2
−2 + 4
So x = =1
2
−2 − 4
or x = = −3
2
The solutions can be checked in exactly the same way as in Worked example
4.15.
Once again, we have found that the solutions of the equation x2 + 2x − 3 = 0 are
x = −3 and x = 1.

Back J I 202
Contents 
Question 4.10
(a) Use your answer to Question 4.7 (c) to solve k2 − 5k + 6 = 0 Answer
by factorization.
(b) Use your answer to Question 4.7 (d) to solve t2 − 4t + 4 = 0 by Answer
factorization.
(c) Use the quadratic equation formula to check your answer to Answer
part (a).
(d) Use the quadratic equation formula to check your answer to Answer
part (b).

4.3 Combining equations

Consider the equation E = h f . This equation, first proposed by Einstein, links the
energy, E, of light to its frequency, f (h is a constant known as Planck’s constant).
Suppose that you know h and are trying to find E, but that you don’t know f . Instead
you know the values of c (speed of light) and λ (wavelength) in a second equation,
c = f λ. It would be possible to calculate a value for f from the second equation
and then substitute this value in the first equation so as to find E. However, this
approach runs the risk of numerical slips and rounding errors. It is more useful to
do the substitution algebraically, in the way shown in the following example.

Back J I 203
Contents 
Worked example 4.17
Combine the following two equations to find an equation for E not involving f :

E = hf (4.6)
c = fλ (4.7)

Answer
Rearranging Equation 4.7 gives
c
f =
λ

Substituting this expression for f into Equation 4.6 gives

c hc
E =h× =
λ λ

This mathematical technique, sometimes referred to as elimination (because a vari-

able, f on this occasion, is being eliminated), can be used in many situations, as
illustrated in the worked examples throughout this section.

Back J I 204
Contents 
Worked example 4.18
Mm
Combine Fg = G 2 and Fg = mg to give an equation for r not involving Fg .
r
Answer
Since both equations are already given with Fg (the variable we are trying to
eliminate) as the subject, we can simply set the two equations for Fg equal to
each other:
Mm
mg = G
r2
We now need to rearrange to give an equation for r. First note that there is an m
on both sides of the equation, so we can divide both sides of the equation by m
to give
M
g=G
r2
Multiplying both sides by r2 gives

gr2 = GM

Dividing both sides by g gives

GM
r2 =
g

Back J I 205
Contents 
Taking the square root of both sides gives
s
GM
r=±
g

Question 4.11
(a) Combine Ek = 12 mv2 and p = mv to give an equation for Ek Answer
not involving v.
(b) Combine E = 12 mv2 and E = mg ∆h to give an equation for v Answer
not involving E.
(c) Combine Ek = h f − φ and c = f λ to give an equation for φ not Answer
involving f .

Back J I 206
Contents 
Two (or more) different equations containing the same two (or more) unknown
quantities are called ‘simultaneous equations’ if the equations must be satisfied
(hold true) simultaneously. It is usually possible to solve two simultaneous equa-
tions by using one equation to eliminate one of the unknown quantities from the
second equation, in an extension of the method discussed above. This is illustrated
in Worked example 4.19.

Worked example 4.19

Find values of x and y which satisfy both of the equations given below:
x+y=7 (4.8)
2x − y = 2 (4.9)

Answer
If we rewrite Equation 4.8 to give an equation for y in terms of x, then we can
insert this result into Equation 4.9 to give an equation for x alone.
Subtracting x from both sides of Equation 4.8 gives
y=7−x (4.10)

Substituting for y in Equation 4.9 then gives

2x − (7 − x) = 2
i.e. 2x − 7 + x = 2
or 3x − 7 = 2

Back J I 207
Contents 
Adding 7 to both sides gives

3x = 9, i.e. x = 3

Substitution of x = 3 into Equation 4.10 shows that

y=7−x=7−3=4

So the solution (i.e. the values for x and y for which both of the equations hold
true) is x = 3 and y = 4. We can check this by substituting the values for x and
y into the left-hand side of Equations 4.8 and 4.9.
Equation 4.8 gives x + y = 3 + 4 = 7, as expected.
Equation 4.9 gives 2x − y = (2 × 3) − 4 = 6 − 4 = 2, as expected.
We could have arrived at the same result by using Equations 4.8 and 4.9 in a
different order, but there is only one correct answer.

Back J I 208
Contents 
Worked example 4.19 shows that in order to find two unknown quantities, two
different equations relating them are required. This is always true and by extension:

When you combine equations so as to find unknown quantities, it is always

necessary to have at least as many equations as there are unknown quantities.

Worked example 4.20 shows how four equations can be combined together in a case
where there are four unknown quantities (we are trying to find the total surface area,
S , but the mass, m, and volume, V, of a single particle and the number of particles,
n, are unknown too and so must be eliminated). This worked example concerns the
use of metal particles as catalysts in the chemical industry (see Box 4.5).

Box 4.5 Chemical catalysts

A catalyst is a substance which makes a chemical reaction proceed more rapidly.
The catalyst itself does not undergo permanent chemical change, and it can be
recovered when the chemical reaction is completed. Metal particles can be used
as catalysts. A large number of small particles will have a greater surface area
than a small number of larger particles, and the total surface area, S , of the par-
ticles is of critical importance to the speed of the reaction. In a typical industrial
chemical reactor, S can be approximately 5000 km2 ; roughly a third the area of
Yorkshire!

Back J I 209
Contents 
Worked example 4.20
The total surface area, S , of n metal particles of average radius r is given by the
equation
S = 4π nr2 (4.11)
The number of particles n is linked to the mass of one particle, m and the total
mass of metal, M by the equation
M
n= (4.12)
m
The mass m of one particle is linked to its volume V and the density of the metal
ρ by the equation
m
ρ= (4.13)
V
The volume V of a particle is given by
4
V = π r3 (4.14)
3
where r is the radius.
Find an equation for S in terms of M, ρ and r only.
Answer
Reversing Equation 4.13 gives
m
=ρ
V

Back J I 210
Contents 
Multiplying both sides by V gives

m = Vρ

Substituting for V from Equation 4.14 gives

4
m = π r3 ρ
3
Substituting this expression for m into Equation 4.12 gives
M
n=
m
M
=
3π r ρ
4 3

3M
=
4π r3 ρ
Substituting this expression for n into Equation 4.11 gives

S = 4π nr2
3M
=×
4π × r2

3 ρ
4πr

3M
=
rρ

Back J I 211
Contents 
4.4 Putting algebra to work
So far, Chapter 4 has been concerned almost exclusively with symbols. Equations
have been given to you and you have been told to manipulate them in a particular
way. In the real scientific world, you are likely to need to:
1 Choose the correct equation(s) to use or derive equation(s) for yourself.
2 Combine, rearrange and simplify the equation(s) using the skills introduced in
the earlier sections of this chapter.
3 Substitute numerical values, taking care over things like significant figures, sci-
entific notation and units, as you did in Chapter 3.
4 Check that the answer is reasonable.
The final section of this chapter considers these points, combining skills from Chap-
ters 3 and 4, but it starts with a more light-hearted look at the uses of algebra.

Back J I 212
Contents 
4.4.1 Algebra is fun!

Try this:
• Think of a number.
• Double it.
• Add four.
• Halve your answer.
• Subtract 1.
If you have arrived at an answer of 4, I can tell you that the number you first thought
of was 3; if your answer is 6, the number you first thought of was 5, if your answer is
11, the number you first thought of was 10, and so on. Magic? No, a demonstration
of the power of algebra! We could perform exactly the same operations for any
number; let’s represent the number by the symbol N. Then we have
• Think of a number. N
• Double it. 2N
• Add four. 2N + 4
• Halve your answer. 1
2 (2N + 4) = N + 2
• Subtract 1. (N + 2) − 1 = N + 1

Back J I 213
Contents 
So the final answer will always be one more than the number you first thought of.
Here’s another one for you to try:
• Think of a number.
• Add 5.
• Double the result.
• Subtract 2.
• Divide by 2.
• Take away the number you first thought of.
Whatever number you first thought of, the answer will always be four.

Question 4.12 Answer

Use a symbol of your choice to represent the number in the ‘think of a number’
example immediately before this question and thus show that the answer will be
four, whatever number you choose at the beginning.

You may wonder why a course entitled Maths for Science has suddenly started
discussing number tricks. There is a serious point to this, namely to illustrate how
you can get from an initial problem to a solution by using algebra. Worked example
4.21 illustrates another use of algebra.

Back J I 214
Contents 
Worked example 4.21
Chris and Jo share a birthday (but are different ages). On their birthday this year
Chris will be five times older than Jo. Their combined age on their birthday last
year was 58. How old was Chris when Jo was born?
Answer
Let C represent Chris’s age in years on her birthday this year and J represent
Jo’s age in years on her birthday this year.
Since Chris will be five times older than Jo we can say

C = 5J (4.15)

Last year Chris’s age was (C − 1) and Jo’s age was (J − 1), so we can say

(C − 1) + (J − 1) = 58
i.e. C + J − 2 = 58
C + J = 60 (4.16)

Substituting for C from Equation 4.15 in Equation 4.16 gives

5J + J = 60
i.e. 6J = 60
J = 10

Back J I 215
Contents 
Thus, from Equation 4.15, C = 5 × 10 = 50.
Thus Chris will be 50 this year and Jo will be 10. But this wasn’t the question
that was asked! When Jo was born, Chris was 50 − 10, i.e. 40 years old.

You may remember questions like Worked example 4.21 from your school days.
Problems like this can seem intimidating, but they are relatively easy to solve once
you have found the equations that describe the problem. Many people struggle with
this first step — they can’t find the equations to use. Look at Worked example 4.21
carefully; all that has been done in order to derive Equation 4.15 and Equation 4.16
has been to study carefully the information given in the question, and to write it
down in terms of symbols. So ‘On their birthday this year Chris will be five times
older than Jo’ has become C = 5J. In solving problems, it is almost always helpful
to start by writing down what you already know. Drawing a diagram to illustrate
the situation can help too; you may find this helpful in Question 4.13.

Question 4.13 Answer

Tracey is 15 cm taller than Helen, and when Helen stands on Tracey’s shoulder
she can just see over a fence 3 m tall. Assume that it is 25 cm from Tracey’s
shoulder to the top of her head and 10 cm from Helen’s eyes to the top of her
head. How tall is Helen?

Back J I 216
Contents 
4.4.2 Using algebra to solve scientific problems

In much the same way as people struggle when trying to derive equations for use
in problems like Worked example 4.21, they often have difficulty deciding which
formulae to use from those given in a book or on a formula sheet. Again, it can
be helpful to draw a diagram and it is always helpful to start by writing down what
you know and what you’re trying to find. This often helps you to decide how to
proceed.
Worked example 4.22 discusses the choice of appropriate formulae for use in an-
swering a particular question. It also works through the other steps you are likely
to follow when using algebra to solve scientific problems.

Worked example 4.22

A silver sphere (density 10.49 g cm−3 ) has a radius of 2.5 mm. What is its mass?
Use formulae given in Box 3.4.
Which equations shall we use?
We know density (ρ) and radius (r) and are trying to find mass (m), so we need
m
an equation to link these three variables. Equation 3.9, ρ = , links density
V
and mass, but it also includes volume which isn’t either given or required by
the question. Fortunately help is at hand in the form of Equation 3.5, V = 43 π r3
which gives the volume V of a sphere of radius r. We should be able to substitute

Back J I 217
Contents 
for V from Equation 3.5 into Equation 3.9. This will give an equation involving
only ρ, r and m, as required, and we can then rearrange it to make m the subject.
Combining and rearranging equations
Substituting for V from Equation 3.5 into Equation 3.9 gives
m
ρ= 4 3
3π r

Multiplying top and botom of the fraction by 3 gives

3m
ρ=
4π r3
Reversing this so that m is on the left-hand side gives
3m
=ρ
4π r3
Multiplying both sides by 4π r3 gives

3m = 4π r3 ρ

Dividing both sides by 3 gives

4
m = π r3 ρ
3

Back J I 218
Contents 
Substituting numerical values
Note that we have used symbols for as long as possible in this question, so as to
avoid numerical slips and rounding errors. However, we are now almost ready
to substitute the values given for r and ρ. First we need to convert the values
given into consistent (preferably SI) units:

r = 2.5 mm = 2.5 × 10−3 m

ρ = 10.49 g cm−3 = 10.49 × 103 kg kg−3 (1.049 × 104 kg kg−3 in scientific no-
tation), converting from g cm−3 to kg m−3 in the way described in Section 3.4.4.
Then
4
m = π r3 ρ
3
4  3
= π 2.5 × 10−3 m × 1.049 × 104 kg kg−3
3
= 6.9 × 10−4 
m3
kg 
m−3


= 6.9 × 10−4 kg

Is the answer reasonable?

It is always worth spending a few minutes checking whether the answer you
have arrived at is reasonable. There are three simple ways of doing this (it is not
normally necessary to use all three methods to check one answer):

Back J I 219
Contents 
1 We can check the units of the answer. We have given units next to all the
numerical values in the calculation, and the units on the right-hand side of the
equation have worked out to be kilograms, as we would expect for mass.
If we had made a mistake in transposing the formula for mass, and had written
it as m = 43 π r2 ρ by mistake, then the units on the right-hand side would have
been m2 × kg m−3 = kgm−1 . These are not units expected for mass by itself,
so we would have been alerted to the fact that something was wrong.
Checking units in this way provides a good way of checking that you have
written down or derived an equation correctly; the units on the left-hand side
of an equation should always be equal to the units on the right-hand side. You
can use this method for checking an equation even if you are not substituting
numerical values into it.
2 We can estimate the value (in the way described in Section 3.3), and compare
it with the answer found on a calculator. In this case
4  3
m≈ × 3 3 × 10−3 m × 1 × 104 kg m−3
3
4
≈ × 3
× 33 × 10−9 m3
× 1 × 104 kg 
m−3

3

≈ 4 × 27 × 10−9+4 kg
≈ 100 × 10−5 kg
≈ 1 × 10−3 kg

Back J I 220
Contents 
This is the same order of magnitude as the calculated value, so the calculated
value seems reasonable.
3 We can look at the answer and see if it is what common sense might lead us
to expect. Obviously this method only works when you are doing a calcula-
tion concerning physical objects with which you are familiar, but it gives a
sensible check for worked examples like the one we are considering. It seems
reasonable that a silver sphere with a diameter of 0.5 cm might have a mass
of something less than a gram. If you’d arrived at an answer of 1.1 × 102 kg
(by forgetting to cube the value given for r) you might have thought that this
mass (equivalent to more than 100 bags of sugar!) was rather large for such a
small sphere.
Note that checking doesn’t usually tell you that your answer is absolutely
correct — none of the methods described above would have spotted small
arithmetic slips — but it does frequently alert you if the answer is wrong.

Back J I 221
Contents 
Tips for using algebra to solve scientific problems

1 Start by writing down what you know and what you’re trying to find, and use
this information to find appropriate equations to use.
2 Combine, rearrange and simplify the equations, using symbols for as long as
possible so as to avoid numerical slips and rounding errors.
3 When you substitute numerical values, take care with units, scientific notation
and significant figures.
4 Check that your final answer is reasonable, by asking yourself the following
questions:
(a) Are the units what you would expect?
(b) Is the answer similar to the one you have obtained by estimating?
(c) Is the answer about what you would expect from common sense?

Worked example 4.23 shows the use of these tips in solving a different problem,
concerning the conservation of energy. This worked example uses formulae in-
troduced in Box 4.6; you may also find these formulae useful when answering
Question 4.14.

Back J I 222
Contents 
Box 4.6 The conservation of energy
Energy can never be destroyed, but it is frequently converted from one form to
another. As a child climbs the steps of a slide, he or she gains in gravitational
potential energy; as he or she slides down the slide this energy is converted into
kinetic (movement) energy. As a kettle boils, the electrical energy increases the
energy of the water molecules and so raises the temperature of the water. In both
cases some energy is ‘lost’ to other forms (such as heat to the surroundings and
sound) but very often you can assume that all of the energy initially in one form
is converted to just one other form, and so equate formulae (such as those given
below) for different forms of energy. All forms of energy should be quoted using
the SI unit of energy which is the joule (J), where 1 J = 1 kg m2 s−2 .
The kinetic energy (energy of motion), Ek , of an object with a mass m moving
at speed v is given by

Ek = 12 mv2 (4.17)

The gravitational potential energy, Eg , of an object of mass m at a height ∆h

above a reference level is given by

Eg = mg ∆h (4.18)

where g is the acceleration due to gravity.

Back J I 223
Contents 
The energy, q, needed to raise the temperature of a mass m of a substance of
specific heat capacity c by a temperature ∆T is given by

q = mc ∆T (4.19)

Worked example 4.23

A lump of putty is dropped from a height of 4.8 m. The putty’s gravitational
potential energy is all converted into kinetic energy as it falls. If, on impact, all
of this kinetic energy is used to raise the temperature of the putty, by how much
does the temperature of the putty rise? Assume that the specific heat capacity
of the putty is 5.0 × 102 J kg−1 K−1 and that the acceleration due to gravity is
9.81 m s−2 .
Which equations shall we use?
It is tempting to involve Equation 4.17, as the question talks about the putty’s
kinetic energy, but closer inspection of the question reveals that we can assume
that all the gravitational potential energy becomes kinetic energy as the putty
falls, and that all the kinetic energy is transferred to heat energy in the putty on
impact. So we can say that all the gravitational potential energy is transferred to
heat energy; we simply need to set Equations 4.18 and 4.19 equal to each other.
We have not been told the mass of the putty, but since the term m appears in both
Equation 4.18 and Equation 4.19 we will be able to cancel this term, which will

Back J I 224
Contents 
leave us with an equation linking g, ∆h, c and ∆T . We know g, ∆h and c and are
trying to find ∆T .
Combining and rearranging equations
Since we can assume that all the gravitational potential energy, Eg , is transferred
to heat energy, q, we can set Equation 4.18 and Equation 4.19 equal to each
other.

mc ∆T = mg ∆h

There is an m on both sides, so we can divide by m to give

c ∆T = g ∆h

Dividing both sides by c gives

g ∆h
∆T =
c

Substituting numerical values

g = 9.81 m s−2
h = 4.8 m
c = 5.0 × 102 J kg−1 K−1

Back J I 225
Contents 
so
g ∆h
∆T =
c
9.81 m s−2 × 4.8 m
=
5.0 × 102 J kg−1 K−1
9.81 × 4.8 
m s−2
× m
=
 

2 −2 −1
5.0 × 102 kg 
m s  kg K−1


= 0.094 K to two significant figures.

Is the answer reasonable?

In a real question you probably wouldn’t use all the checks described in the blue-
toned box after Worked example 4.22, but the answer seems about the size you
might expect (you wouldn’t expect a big temperature rise) and the units have
worked out to be kelvin, as expected for a change in temperature.
Alternatively we can estimate the answer to be

10 m s−2 × 5 m
∆T ≈ −1
≈ 10−1 K
5 × 102 J kg K−1
This is the same order of magnitude as the calculated value, so the calculated
value seems reasonable.

Back J I 226
Contents 
Question 4.14 Answer
A child climbs to the top of a 1.8 m slide and then slides to the ground. As-
suming that all of her gravitational potential energy is converted into kinetic
energy, find her speed as she reaches the ground. Take g = 9.81 m s−2 and use
appropriate formulae from Box 4.6.

In Worked example 4.24, the final worked example in Chapter 4, we return to a dis-
cussion of seismic waves travelling through the Earth’s crust (as introduced in Box
3.1). In this example there are three unknown quantities (the distance, d, from the
earthquake, the time, tp , taken for P waves to reach the seismometer and the time,
ts , taken for S waves to reach the seismometer) so we need to combine three equa-
tions to find any of the unknown quantities. You will not be expected to combine
more than two equations together in any questions associated with this course, but
Worked example 4.24 has been included because it summarizes much of what has
been discussed in Chapter 4, and also because it illustrates the usefulness of algebra
in science.

Box 4.7 Locating an earthquake

Figure 4.4 shows a seismogram recorded at the British Geological Survey in Ed-
inburgh on 12 September 1988. It is possible to see the points at which P waves
and S waves first reached the seismometer. We can assume that these seismic
waves originated in an earthquake somewhere. But where was the earthquake
and when did it occur? (although the recording was made at 2.23 p.m., it does

Back J I 227
Contents 
not tell us the time at which the earthquake occurred, since the waves will have
taken some time to reach the seismometer from the point of origin or focus of
the earthquake).
Figure 4.4 shows that the P waves reached the seismometer 20 seconds before
the S waves.
We assume that the P waves travelled with an average speed, vp = 5.6 km s−1
and that the S waves travelled with an average speed vs = 3.4 km s−1 (these val-
ues are typical for the rocks of the Earth’s crust, through which the waves will
have been travelling).
distance travelled
average speed =
time taken

d
so vp = (4.20)
tp
d
and vs = (4.21)
ts
where d is the distance from the earthquake, tp is the time taken for P waves to
travel to the seismometer and ts is the time taken for S waves to travel to the
seismometer.

Back J I 228
Contents 
Worked example 4.24
Use the information given in Box 4.7 to find the distance from Edinburgh to the
focus of the earthquake recorded on the seismogram shown in Figure 4.4.
Which equations shall we use?
d d
We know that vp = (Equation 4.20) and vs = (Equation 4.21), where
tp ts
vp = 5.6 km s−1 and vs = 3.4 km s−1 , but d, tp and ts are all unknown, so we
need another equation.
Although we don’t know the travel time of the two types of wave, we know that
the difference in the arrival time of the two waves is 20 seconds, so we can write

t = ts − tp (4.22)

where t = 20 s.
Equations 4.20, 4.21 and 4.22 give us three equations containing the three un-
knowns d, tp and ts and we need to combine and rearrange them to give an
expression for d.
Combining and rearranging equations
Multiplying both sides of Equation 4.20 by tp gives

tp vp = d

Back J I 229
Contents 
Dividing both sides by vp gives
d
tp =
vp
Similarly, from Equation 4.21,
d
ts =
vs

Substituting for ts and tp in Equation 4.22 gives

t = ts − tp
d d
= −
vs vp
!
1 1
=d −
vs vp

Combining the fractions by making vs vp a common denominator (Section 4.2.1)

gives
(vp − vs )
t=d
vs vp
Reversing the equation so that d is on the left-hand side gives
(vp − vs )
d =t
vs vp

Back J I 230
Contents 
Multiplying both sides by vs vp gives

d (vp − vs ) = t vs vp

Dividing both sides by (vp − vs ) gives

t vs vp
d=
vp − vs

Substituting numerical values

Substituting t = 20 s, vp = 5.6 km s−1 and vs = 3.4 km s−1 gives

20 s × 3.4 km s−1 × 5.6 km s−1

d=
5.6 km s−1 − 3.4 km s−1

20 s × 3.4 km s−1 × 5.6 km s−1

=
2.2 km s−1
= 1.7 × 10 km to two significant figures
2

s−1
s × km × km
s−1

The units work out to be kilometres since
= km
km
 s−1


Is the answer reasonable?

The units have worked out to be kilometres as expected for a distance. If we had
converted the speeds to values in ms−1 , we would have obtained a value for d in
metres (d = 1.7 × 105 m).

Back J I 231
Contents 
In this case it is easy to check that the answer is reasonable; many members of
the public reported a small earthquake on that day in Ambleside in Cumbria.
Ambleside is 170.5 km from Edinburgh!
In general, to use this method to uniquely identify the location of an earthquake
you need to repeat the exercise using data received at other seismometers else-
where on the Earth’s surface.

Back J I 232
Contents 
4.5 Learning outcomes for Chapter 4
After completing your work on this chapter you should be able to:
4.1 demonstrate understanding of the terms emboldened in the text;
4.2 rearrange an algebraic equation to make a different variable the subject;
4.3 simplify an algebraic expression;
4.4 add, subtract, multiply and divide algebraic fractions;
4.5 re-write an algebraic expression so that the brackets are removed;
4.6 factorize a simple algebraic expression;
4.7 eliminate one or more variables so as to combine equations together;
4.8 check the answer to a problem by checking units, estimating an answer, or
comparing the answer with what would be expected from common sense.

Back J I 233
Contents 

Using Graphs 5
The well-known saying that a picture is worth a thousand words reflects the fact
that human beings can derive a lot of information from pictorial representations of
a situation. When scientists want to condense data into a visual form that conveys
information at a glance, they most often turn to a graphical representation. Graphs
are essential tools for scientific work: they can illustrate clearly the nature of the
relationship between different quantities, they make it easy to see variations and
trends and sometimes they can be used to derive other interesting quantities or even
equations.
This chapter is mainly about the use and interpretation of graphs, rather than tech-
niques for plotting them (which are more the province of courses in practical sci-
ence). However, an understanding of the kind of information that can be derived
from different types of graph will be of considerable help when you do come to plot
your own data in the future.

Back J I 234
Contents 
5.1 Graphical representations
Although all graphs share certain characteristics, there are nevertheless a number
of different ways in which data may be presented graphically. Let us start by con-
sidering some specific examples and the features they illustrate.

5.1.1 Bar charts and histograms

‘Bar charts’ are commonly used to summarize data that require immediate com-
parison between various discrete categories. Examples of discrete categories are
human eye colour, blood group, countries and planets. The categories are listed
along a reference line, usually a horizontal one (the so-called horizontal axis). The
number or percentage of things or events falling into each category is represented
by a bar; the scale for these bars, most commonly expressed either as a number
or a percentage, is given on a second reference line, at right angles to the first. If
the categories are listed along the horizontal axis, the bars will therefore be scaled
along the vertical axis. Figure 5.1 in Box 5.1 is an example of how ecological data
might be presented in the form of a bar chart.

Back J I 235
Contents 
Box 5.1 Insects and trees
Figure 5.1 shows the number of species of herbiv- 500

orous insects associated with eight different types 450

number of insect species supported

of native and introduced tree. Tree species that 400

have been present in the country for a long time 350

and are widely distributed often support the largest 300

variety of insects. 250

200
Question
150
Roughly how many species of insect are as-
100
sociated with hawthorn?
50
Answer
0

(5 species)

(2 species)
About 220.

sycamore
hawthorn

holm oak
willows

beech
oaks

holly
The willows and oaks, which are among the com-

ash
monest tree species in the UK, can support over
400 insect species. Sycamore, which is just as
widely distributed but came to this country more
recently, supports only around 50 species, and Figure 5.1: Bar chart showing the number of
the evergreen holm oak, which was introduced a herbivorous insect species supported by some
mere 400 years ago, supports fewer than 10 insect native and introduced tree species in the UK.
species. However, one should not generalize too
far from these examples. There are other native trees, such as holly and yew, which also support very
few insect species, many of which are specialist feeders not found on other trees.

Back J I 236
Contents 
A histogram is similar to a bar chart in that numbers or
percentages are again commonly plotted vertically, but
30
on a histogram the horizontal axis is used to represent a
continuously variable quantity such as height or mass.

number of plants
The purpose of a histogram is to show how the data are 20
distributed into groups across a continuous range. Fig-
ure 5.2 shows a histogram which presents the results of 10
measurements taken of the height of 100 irises. In prin-
ciple, a plant selected at random could be of any height.
0
Of those measured, a few specimens are particularly tall

105.0 to 109.9
100.0 to 104.9

110.0 to 114.9
115.0 to 119.9
85.0 to 89.9
90.0 to 94.9
95.0 to 99.9
80.0 to 84.9
75.0 to 79.9
and a few are particularly short, but the majority are of
intermediate height. This is typical of the natural varia-
tion in populations, and Chapters 8 and 9 deal with the
statistical techniques that are required to analyse such height intervals in cm
variations. Comparing Figure 5.2 with Figure 5.1, you
will notice that on the bar chart the bars do not touch
(because they refer to different categories), whereas on
Figure 5.2: Histogram representing the heights
the histogram the columns do touch, because all possi-
of 100 of the same variety of iris. The hori-
ble heights are represented within the groups marked on
zontal axis is divided into intervals to represent
the horizontal axis. In Figure 5.2 the groups are of equal
different height groups.
intervals, and this is common practice (though there are
also ways of constructing histograms using unequal intervals). Note that the whole
range of possible heights is covered, whether or not any of the measured plants
actually fell into a particular group.

Back J I 237
Contents 
5.1.2 Graphs

On a histogram, the horizontal axis is

divided into intervals. On a graph, in
contrast, the horizontal axis is scaled 600
to represent a continuum. In Figure

estimated number of caterpillars per square metre

550
5.3, for example, time is plotted along 500
the horizontal axis, with the years be- 450
ing evenly spaced. This graph clearly 400
shows the large variation in caterpil- 350
lar numbers that can occur from year 300
to year, though no overall trend can 250
be discerned. It is not necessary to 200

join the data points on a graph of this 150

type; if this is done, as here, the lines 100

have no significance beyond simply 50

emphasizing the downturn or upturn 0

1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962
in the numbers between one year and year

the next.

Figure 5.3: Annual fluctuations in the population of Winter

Moth (Operophtera brumata) caterpillars feeding on oak trees in
Wytham Wood near Oxford.

Back J I 238
Contents 

average temperature/°C
25 20
average temperature measured in °C

20 15
15 10

10 5

5
J F M A M J J A S O N D
0 (b) month
J F M A M J J A S O N D
–5 month

–10 Figure 5.4: Average monthly temperatures for

–15 (a) Irgiz, Kazakhstan, and (b) Paris.
–20

(a)

Figure 5.4a illustrates that negative, as well as positive numbers can be plotted on
a graph; in this case the vertical axis covers temperatures from −20 ◦ C to +25 ◦ C.
The data points have been joined, but the lines are only indicators of rises or falls
in average temperature between one month and the next; they could not be used to
predict the temperature on any particular day.
The vertical axis of Figure 5.4a is labelled to show that the quantity plotted is the
average temperature measured in ◦ C. Whatever the variable we want to display

Back J I 239
Contents 
graphically, we always have to take account of its units in such a way as to plot
a pure number (i.e. a number without units) on the graph. The labelling on the
vertical axis of Figure 5.4a could have been written more succinctly as ‘average
temperature/◦ C’, and this form of labelling has been used on the vertical axis in
Figure 5.4b. The temperature values are divided by their unit (◦ C), to give pure
numbers that can be plotted on the graph:
◦
2.3 C
e.g. ◦
= 2.3
C

It is conventional always to use ‘quantity divided by its units’ (usually in the

form ‘quantity/units’) in labelling the axes of graphs.

Box 5.2 Atmospheric pollution

In industrialized countries, air pollution was historically associated mainly with
emissions of smoke and sulphur dioxide arising from the combustion of fossil
fuels (chiefly coal) for domestic heating and industrial purposes. The resulting
‘smogs’ that occurred in Northern European cities for several centuries were
the result of this kind of pollution. The problem became particularly acute in
London in the 1950s, leading to the UK Clean Air Act of 1956. Subsequent Eu-
ropean directives have further reduced emission limits and national emissions of
sulphur dioxide have fallen dramatically — by about 80% since 1962. Figure

Back J I 240
Contents 
5.5 shows air quality data recorded in the Tameside district of Greater Manch-
ester between 1963 and 2000.

3.0
concentration/10 2 µg m − 3

2.5

2.0

1.5

1.0

0.5

0 1965 1970 1975 1980 1985 1990 1995 2000

year

Figure 5.5: Average annual concentrations of sulphur dioxide in Tameside,

Greater Manchester, 1963–2000.

Note how the vertical axis in Figure 5.5 is labelled. Concentrations have been ex-
pressed in micrograms per metre cubed (µg m−3 ), so the quantity represented along
the vertical axis has been divided by µg m−3 , in the same way as the temperature in
Figure 5.4b was divided by ◦ C. But in Figure 5.5, the quantity has been divided not
only by its units but also by a power of ten. This can be a useful strategy in graph

Back J I 241
Contents 
plotting because it allows manageable numbers to be used in labelling the divisions
on the axis. To obtain the actual value of a quantity corresponding to a particular
tick mark on the axis, we have to multiply the value given at the mark by the power
of ten and by the units. For example, the mark labelled 1.5 represents:
1.5 × 102 µg m−3 = 150 µg m−3
Another way of looking at this is to say that a measured concentration has first been
expressed in scientific notation:
150 µg m−3 = 1.5 × 102 µg m−3
and then reduced to a pure number by dividing it by the power of ten and the units:
1.5 × 
10 −3
2 µg m

 = 1.5
10 −3
2 µg m

 
Figure 5.5 gives a clear visual image of a downward trend in sulphur dioxide con-
centration, but occasional blips such as occurred in 1990 mean that it is still not pos-
sible to use earlier data to predict future concentrations with any certainty. There
are simply too many variables that can affect the concentration of atmospheric sul-
phur dioxide. In other circumstances, for instance when the two quantities plotted
are linked by an equation, it is possible to use a graph for predictive purposes.
As an example of linked quantities, consider the data in Table 5.1 relating the mass
of a series of aluminium spheres to their diameter. The data are plotted in Fig-
ure 5.6. Notice that the columns of the table have been labelled according to the
same convention used to label the axes of the graph.

Back J I 242
Contents 

Diameter/10−2 m Mass/10−3 kg
14.0
0.4 0.1
0.5 0.2 12.0

0.7 0.5
10.0
1.0 1.4

mass/10− 3 kg
8.0
1.3 3.1
1.5 4.8 6.0
1.8 8.2
4.0
2.0 11.4

Table 5.1: Masses of aluminium spheres 2.0

of different diameters
0 0.5 1.0 1.5 2.0
diameter/10 − 2 m

Figure 5.6: Graph of the masses of aluminium

spheres of different diameters.

Back J I 243
Contents 
Question
What is the diameter in centimetres of the smallest sphere?
Answer
The diameter of the smallest sphere is obtained by multiplying 0.4 by the power
of ten and the units:
diameter = 0.4 × 10−2 m = 0.4 cm

In fact, the mass M of a sphere of diameter d, made of material of density ρ, is

given by the equation M = πρd3 /6. The data have been calculated and the graph
constructed using this formula, so all the points lie on a smooth curve. When the
axes of a graph represent quantities that are connected by an equation, the data
points should never be joined in the jagged point-to-point way used in Figure 5.3
and Figure 5.4. Instead, a smooth line should be drawn through them. As you will
see later in this chapter, a line described as ‘smooth’ may be straight, or may be
curved in any direction, or may have humps and dips. Smoothness depends on the
absence of abrupt changes in direction, not on shape.
Once the line has been drawn on Figure 5.6, we can use the graph to find interme-
diate values. This graph has been drawn on graph paper to make it easier to read
values from it. You should start by working out the scale used on each axis. On this
occasion tick marks have been drawn every 0.5 × 10−2 m on the horizontal axis, so
each feint grid line represents 0.05 × 10−2 m; on the vertical axis the tick marks are
every 2.0 × 10−3 kg so each feint grid line represents 0.2 × 10−3 kg.

Back J I 244
Contents 
Question
What would be the mass of an aluminium sphere of diameter 1.6 cm?
Answer
To find the mass corresponding to a diameter of 1.6 cm (i.e. 1.6 × 10−2 m) we
need to find the point on the horizontal axis representing this diameter and draw
a line vertically upwards from there until it meets the curve. We then draw a
line horizontally from that intersection to meet the vertical axis and read off the
corresponding mass. Print Figure 5.6 and draw these lines directly on to it using
the grid lines on the graph paper to help you. You should find that the mass
corresponding to a diameter of 1.6 cm is 5.8 × 10−3 kg (i.e. 5.8 g).

This process of reading between points plotted on a graph, in order to find corre-
sponding intermediate values of the plotted quantities, is called interpolation.
Provided we are sure that the equation connecting the two quantities is valid even
outside the plotted range, we can also extend the line on the graph to determine
corresponding values of the quantities that are larger or smaller than those plotted.

Back J I 245
Contents 
Question
What would be the mass of an aluminium sphere of diameter 2.1 cm?
Answer
To find the mass corresponding to a diameter of 2.1 cm (i.e. 2.1 × 10−2 m) we
need to find the point on the horizontal axis representing this diameter and draw
a line vertically upwards from there. Then (and this is the difficult bit!) we
have to extend the curve until it meets this vertical line. We then draw a line
horizontally from that intersection to meet the vertical axis and read off the
corresponding mass. Print Figure 5.6 and try drawing the lines. If your drawing
skills are high, you should obtain a mass of 13.1 × 10−3 kg, but most people find
it extremely difficult to draw smooth curves freehand, so if you obtain a value
between 12.8 × 10−3 kg and 13.4 × 10−3 kg you have done well.

This process of extending a graph beyond the highest or lowest data points, in order
to find corresponding values of the plotted quantities outside the original range, is
called extrapolation. Extrapolation is always particularly difficult in regions where
graphs curve, or have very steep or very shallow slopes. The latter situation applies
to Figure 5.6 in the region where the diameter becomes very small. It would be
practically impossible to determine by extrapolation the mass corresponding to a
diameter of, say, 0.2 cm. All we can legitimately say is that if the diameter is zero,
the mass will also be zero, so the curve must go through the point at which the axes
meet. On any graph the point at which both plotted quantities are equal to zero is
called the origin.

Back J I 246
Contents 
The fact that the graph in Figure 5.6 is curved makes both interpolation and ex-
trapolation more uncertain than they would be if the graph was a straight line. In
Question 5.1 you can practice these processes using a graph that is easier to deal
with.

Question 5.1 Answer

Five measurements have been made to inves-
tigate the way in which the voltage across the 5.0
terminals of a power supply varies according
4.0
to the current flowing in the circuit. The data

voltage/V
are plotted on Figure 5.7. (The SI unit of 3.0
voltage is the volt, symbol V; the SI unit of
electric current is the ampere, symbol A.) 2.0

(a) What is the value of the voltage when the 1.0

current is 1.5 A?
0 0.5 1.0 1.5 2.0 2.5
(b) What is the value of the voltage when the
current/A
current is zero?
(c) What is the value of the current when the
voltage is zero? Figure 5.7: Measurements of voltage against current
for the circuit in Question 5.1.

Back J I 247
Contents 
5.2 Straight-line graphs
As you have seen, it is possible to obtain useful insights and information from
curved graphs such as the one in Figure 5.6, and we will return to the interpretation
of curved graphs in Section 5.4. However, if data can be presented in the form of
a straight-line graph, the analysis becomes more straightforward. As you will have
discovered for yourself by doing Question 5.1, if you need to determine the values
of quantities lying between those that were actually measured, it is slightly easier
to perform the interpolation on a straight line than on a curve. And if you need to
estimate values of quantities outside the original range of measurements, it is con-
siderably easier to extrapolate a straight line than a curve. Furthermore, it is often
possible to use a straight-line graph to obtain additional quantities, other than those
measured. For example, the range of speeds at which the Earth’s tectonic plates
move was given in Box 3.1, but it is not possible to make a direct measurement
of these speeds. Scientists have to deduce them by measuring other quantities and
plotting graphs of their results.

Back J I 248
Contents 
5.2.1 The gradient of a straight-line graph

Box 5.3 gives a brief outline of the phenomenon of

Age of rock/Ma Separation distance/km
sea-floor spreading, the action of which is to split the
Earth’s surface and move sections of the crust apart. In 0.78 17
order to work out the rate at which the separation takes 0.99 18
place, Earth scientists date the rocks and measure the 1.07 21
separation of rocks of the same age. 1.79 32
Table 5.2 shows some typical data. (Remember from 1.95 39
Section 2.2 that Ma is the abbreviation for ‘million 2.60 48
years’.) 3.04 58
As discussed in Section 5.1.2, labelling the left-hand 3.11 59
column as ‘Age of rock/Ma’, and the right-hand column 3.22 62
as ‘Separation distance/km’, means that pure numbers 3.33 65
can be entered in each row of the table. 3.58 68

Table 5.2: Positions of some dated areas either

side of the mid-Atlantic ridge south-west of Ice-
land

Back J I 249
Contents 
Figure 5.9 shows a graphical representation of
the data from Table 5.2. Although it is obvious
just from the table that the separation distance 80

increases with age, the graph immediately gives

70
more information. First, it tells us about the rela-
tionship between the quantities plotted: the points 60

separation distance /km

lie pretty much on a straight line. The relationship
between the age and the distance is thus said to be 50

linear. Secondly, the graph provides a good test

40
of the reliability of the data. It is clear that there
are no ‘rogue points’ lying well off the straight 30
line. However, the points do not all lie exactly
on a single line. The black line that has been 20

drawn through them is the ‘best-fit line’ — i.e.

10
the line that is most representative of the data as
a whole. Best-fit lines usually only go through
0 1 2 3 4
some of the data points (and need not necessarily age/Ma
go through any); there should be approximately
the same number of points above and below the
line. The line has also been drawn to go through Figure 5.9: Graph of data in Table 5.2. The black
the origin, the point at which age is 0 Ma and line represents the ‘best-fit’ to the data. The red lines
distance is 0 km. This has been done because show that ocean crust of age 3.4 Ma has separated by
it is clear that newly-formed crust will not have 65 km.
moved any distance.

Back J I 250
Contents 
The aim of collecting the data for age and separation distance was to calculate
the rate of sea-floor spreading and this calculation can be made directly from the
graph. For an object moving at a steady rate, the speed v is related to the distance d
travelled in a time t by the equation:
v = d/t
The red lines on Figure 5.9 show that, according to the best-fit to the data, ocean
crust of age 3.4 Ma has separated by 65 km. So the average spreading rate is:
vav = 65 km/3.4 Ma = 19 km Ma−1 (to 2 significant figures)
Now you could of course carry out similar calculations using any of the individual
data pairs in Table 5.2. For example from the first data pair:
v1 = 17 km/0.78 Ma = 22 km Ma−1
and from the fourth pair:
v4 = 32 km/1.79 Ma = 18 km Ma−1
The first pair corresponds to a point that lies above the best-fit line and therefore
gives a value of v that is higher than that calculated from the graph, while the point
corresponding to the fourth data pair lies below the line and consequently gives
a value of v that is lower than that calculated from the graph. If we wanted to
calculate the average spreading rate directly from the tabulated data, the best we
could do would be to calculate values from each of the eleven data pairs in the table
(i.e. v1 to v11 ) and then average all these speeds. Plotting a graph therefore saves a
tedious amount of calculation: using the best-fit line allows vav to be calculated in
a single step. In other words, a graph provides a reliable way of averaging results.

Back J I 251
Contents 
Question
What can you deduce from the fact that all the data points are close to the best-fit
straight line, with some points lying above and others below the line?
Answer
The rate of spreading has remained roughly constant over time. Again the graph
provides this information at a glance, whereas it would require a lot of calcula-
tion to deduce it from the raw data in Table 5.2.

Another way of describing this process of calculating the spreading

rate from the distance–time graph is to say that we have determined
‘rise’
the ‘slope’ or gradient of the best-fit line. Figure 5.10 shows the = 100m
analogy with the gradients used to characterize steep hills, which
you may have seen on road signs. The gradient is defined in this ‘run’ = 300m
context as the ‘rise’ (the total change in vertical distance) divided
by the ‘run’ (the total change in horizontal distance). It is important
to remember that the actual distance travelled along the road is not Figure 5.10: Vertical cross-section
involved in the calculation. Note that in this particular case the through a road. The gradient of
gradient has no units, because it is calculated by dividing a length this road is given by: ‘rise’/‘run’ =
by another length. In general, however, gradients must, as with the 100 m/300 m,
example of Figure 5.9, be given their correct units. In the case of i.e. gradient is 1/3 or 33%.
a road, it is common to quote the gradient in the form of a percentage (33% in the
case of Figure 5.10). With a graph it is more usual to quote the gradient as a single
number.

Back J I 252
Contents 
The gradient of a straight line is the same
all the way along it, so any two points on y
the graph can be used to define the rise
y2
and the corresponding run. If, as is the
case in Figure 5.11a, a graph goes through
the origin, it may be convenient to use that
fact in calculating the gradient; here the rise
rise is (y2 − 0) and the run is (x2 − 0), so =y2 – 0
there are no subtractions to do. This was
effectively the technique used in calculat-
ing the sea-floor spreading rate from Fig-
ure 5.9, when just one point on the best-fit
0 run =x2 – 0 x
line was chosen from which to calculate
x2
the speed.

Figure 5.11a: This straight line goes through the origin, so

y2 − 0 y2
its gradient = =
x2 − 0 x2

Back J I 253
Contents 
However, not all graphs go through the y
origin, so the method illustrated by Fig-
ure 5.11a is not always applicable. Figure
5.11b shows the most general method of y2
determining the gradient of a straight-line
graph, which can be used whether or not rise
the line goes through the origin. =y2 – y1

y1
run =x2 – x1
For a straight-line graph in which the
value y2 on the vertical axis corre-
sponds to a value x2 on the horizon-
0 x1 x2 x
tal axis, and a value y1 on the vertical
axis corresponds to a value x1 on the
horizontal axis:
y2 − y1
rise y2 − y1 Figure 5.11b: For any straight line, the gradient =
gradient = = x2 − x1
run x2 − x1

Whatever points are chosen for determining the rise and run, it is always a good
idea to choose ones that are easy to read on at least one axis and preferably on both
axes! It is also good practice to choose points as widely separated as possible.

Back J I 254
Contents 

Worked example 5.1

When light is shone onto certain metals, electrons are emit-
ted from the metal. This phenomenon is called the photo- 10
electric effect, and will be described in more detail in Box
8

energy/10–19 J
5.4. Figure 5.12a shows a graph arising from a photoelec-
tric experiment on a particular metal, relating the energy of 6
the ejected electrons to the frequency (i.e. the colour) of
the light falling upon the metal. The energy is measured in 4
joules (symbol J) and the frequency in units of s−1 (which
2
are better known as hertz). What is the gradient of this
graph?
0 1 2
Answer frequency/1015 s–1
It is clear that even if the line were to be extrapolated to
smaller values of energy and frequency it would not go Figure 5.12a
through the origin, so the method shown in Figure 5.11b
is the appropriate one to use in calculating the gradient.

Back J I 255
Contents 

From the lines drawn on Figure 5.12b,

10
(9.2 × 10−19 J) − (2.6 × 10−19 J)
gradient =
(2.0 × 1015 s−1 ) − (1.0 × 1015 s−1 ) 8

energy/10–19 J
(9.2 − 2.6) × 10−19 J
= 6
(2.0 − 1.0) × 1015 s−1
6.6 × 10−19 J 4
=
1.0 × 1015 s−1 2
J
= 6.6 × 10(−19−15) −1
s
1 0 1 2
= 6.6 × 10−34 J s (remembering that = s) frequency/1015 s–1
s−1
Note that on this occasion the line drawn passes through, Figure 5.12b
or very close to, all the data points. If the best-fit line does
not go through all the data points, care must be taken to
calculate the gradient of the graph from the line rather than
from just two data points.

Back J I 256
Contents 
Question 5.2 Answer
The speed of seismic waves (see Box 3.1) may be calculated by measuring the
time for the waves to reach measuring instruments at different distances from
the epicentre of the earthquake. Some typical data from such a series of mea-
surements on P waves are plotted in Figure 5.13. Use the graph to calculate the
average speed of the P waves.
distance from epicentre/km

200

100

0 10 20 30 40
travel time after earthquake occurred/s

Figure 5.13: Graph showing how long it takes for P waves from a shallow-focus
earthquake to reach three detectors at different distances from the epicentre. (Note
that the focus is the point within the Earth at which the seismic event takes place,
and the epicentre is the point on the Earth’s surface vertically above the focus.)

Back J I 257
Contents 
5.2.2 Dependent and independent variables

In Figure 5.13, the time was deliberately plotted on the horizontal

axis and the distance travelled on the vertical axis, so that the gra-
dient would be equivalent to the seismic wave speed. However,
for this particular example, plotting the graph this way round is 40

travel time after earthquake occurred/s

not standard practice. The convention that scientists follow is to
plot on the horizontal axis the variable that is under their control.
Because they can choose the values of this quantity, it is called the 30
independent variable. In the case of the measurements described
in Question 5.2, there is a choice (within reason) of where the
seismic wave detectors are located; therefore distance from the 20
epicentre is the independent variable. The time taken for the P
waves to arrive depends on where the detectors have been posi-
tioned, so this is called the dependent variable. According to the 10
convention, the dependent variable is plotted on the vertical axis.
Figure 5.14 shows the same data as Figure 5.13, but replotted so
that the convention is followed. The three points correspond to
those on Figure 5.13 and a best-fit line has again been drawn. 0 100 200
distance from epicentre/km
The seismic wave speed can be calculated equally well from Fig-
ure 5.14 as from Figure 5.13. Figure 5.14

Back J I 258
Contents 
Question
What will be units of the gradient of the graph in Figure 5.14?
Answer
The gradient will have units of seconds divided by kilometres, which can be
written either as s/km or as s km−1 .

Question
In Question 5.2, you calculated the speed of the seismic wave in units of km/s
(or km s−1 ). How are the units s km−1 related to these units of speed?
Answer
1
The units s km−1 and km s−1 are reciprocals: i.e. = km s−1 .
s km−1

Therefore to calculate the speed of the P waves from the time against distance graph
of Figure 5.14, we need to determine the gradient and then take its reciprocal.

Question 5.3 Answer

Use Figure 5.14 to determine the average speed of the seismic waves. Remem-
ber to use the correct units at each stage of your calculation. Does your final
answer agree with the value you obtained in Question 5.2?

Back J I 259
Contents 
5.2.3 Interpreting straight-line graphs and gradients

The graphs we have looked at so far in this section

have all sloped up from left to right. But graphs
can slope the other way too. Figure 5.15 shows 25
the result of measuring the depth of snow in a par-
20
ticular location over a period of time, plotted on a

depth/cm
graph of depth against time. 15

10

When describing a graph, the convention is to 5

state the dependent variable first; a graph of
0 1 2 3 4 5
‘depth against time’ therefore plots depth on time/hours
the vertical axis and time on the horizontal axis.

For the line drawn in Figure 5.15, gradient is given, Figure 5.15: Depth of snow measured over a five-
as before, by: hour period.
rise y2 − y1
gradient = =
run x2 − x1
If x1 is 1 hour and x2 is 4 hours, the corresponding y values are y1 = 20 cm and
y2 = 5 cm, i.e. x2 is greater than x1 but y1 is greater than y2 . This means that:
(5 − 20) cm −15 cm
gradient = = = −5 cm hour−1
(4 − 1) hours 3 hours
In other words, the gradient is negative.

Back J I 260
Contents 
Question
What physical meaning do you attach to the gradient in this context?
Answer
The graph shows that depth is decreasing with time — in other words the snow
is melting. The negative value of the gradient conveys this same information.
The gradient is constant over the time during which the measurements have been
made, so the snow is melting at a steady rate.

Now look at Figure 5.16, which shows the variation of distance from
d
a given point with time, for four objects A to D moving in a variety
of situations. A scientific way to say this is that the graphs all show
distance ‘as a function of’ time, or d as a function of t. A

B
In general, it is the dependent variable (which by convention is plot-
ted along the vertical axis) that is described as being a function of D
C
the independent variable (which is plotted along the horizontal axis).

So in the situations shown in Figure 5.16, time is the independent vari-

able: the experimenter has chosen specific times at which to make the 0 t
measurements and has recorded the position of the object at those times.
Figure 5.16
As with Figure 5.13, the gradient of each line gives the speed with which
that particular object is moving.

Back J I 261
Contents 
d
Question
Which, if any, objects are moving with constant speed? Of these,
A
which is travelling the most quickly?
Answer B

The gradients of the distance–time graphs for objects A and C are D

C
constant, so their speed is constant. The gradient of the line for
object A is greater (i.e. the distance–time graph is steeper) than that
for object C, so A is moving at a higher speed than C.
0 t

Question Figure 5.16

What is happening to object B? What is the gradient of the line for
object B on the graph?
Answer
For object B the distance travelled is not changing with time. The
most likely explanation of this is that the object is stationary. On
the graph for this object, the rise is always zero, so the gradient of
the graph is also zero. This is simply another way of saying that its
speed is zero.

Back J I 262
Contents 
Question
What is happening to object D?
Answer
The gradient for object D gradually decreases (i.e. gets less steep). In other
words the object is slowing down.

Question 5.4 Answer

The lowest level of the Earth’s atmosphere is called the 20

temperature/°C
troposphere. Figure 5.17 shows the variation in temper-
ature of the troposphere from sea-level to an altitude of
about 2.5 km. Estimate to two significant figures the 10
gradient of this graph. (Because you are only being
asked for an estimate, you do not need to attempt great
precision in reading values off the graph, but you should
be careful over signs and units.) Describe clearly, in one 0 1 2
sentence, what your result means. altitude/km

Figure 5.17

Back J I 263
Contents 

Question 5.5 Answer

10
At higher levels in the troposphere, the tempera-
5 10
ture drops still further. Figure 5.18 shows the vari- 0
ation in temperature for altitudes between 4 km altitude/km
and 11 km above sea-level. Estimate to two sig- –10

temperature/°C
nificant figures the gradient of this graph. Does
your answer agree with that for Question 5.4? –20

–30

–40

–50

–60

Figure 5.18

Back J I 264
Contents 
5.3 The equation of a straight line
In the preceding sections, you saw how useful information can be derived from a
straight-line graph by interpolation, extrapolation or calculation of the gradient. But
this does not exhaust the potential of a graph as a tool: it becomes even more useful
when it can be matched to an equation.

5.3.1 Proportional quantities

Two quantities are said to be proportional to each other, or more precisely to be

directly proportional to each other, if multiplying (or dividing) one by a certain
amount automatically results in the value of the other being multiplied (or divided)
by the same amount. If I buy 500 litres of heating oil I pay twice as much as if I had
bought 250 litres, but one-half as much as if I had bought 1000 litres — assuming
that on such amounts there is no bulk discount. The cost is directly proportional to
the volume. We can write this succinctly in the form:

total cost ∝ volume

where the symbol ∝ stands for ‘proportional to’. To determine the total cost of
something we multiply the number of items we are buying by the price per item, so
we can turn our original proportionality relationship into an equation of the form:

total cost = (cost per litre) × (volume in litres)

Back J I 265
Contents 
We are assuming that the cost per litre is constant however big the delivery. This
constant factor, which is required to turn the proportionality into an equation, is
called the constant of proportionality.
Now consider how this relationship between cost and volume appears on a graph,
such as that plotted in Figure 5.19. If I don’t buy any oil, the cost is zero (but the
heating doesn’t work!), so the graph must go through the origin. If I buy 500 litres
it costs £100, and 1000 litres cost £200.

200
total cost/£

100

0 250 500 750 1000

volume/litres

Figure 5.19: The cost of heating oil.

Back J I 266
Contents 
Question
What is the gradient of this graph? What does that value represent?
Answer
£(200 − 0) £200
The gradient is = = £0.20/litre = 20 pence/litre
(1000 − 0) litre 1000 litre
In other words the gradient represents the cost per litre. The gradient of the
graph is the constant of proportionality between total cost and volume of oil.

Generalizing from this example:

y

if y = kx, where y and x are variables and k is a constant,

then y is said to be directly proportional to x, gradient = k

i.e. y ∝ x.
y = kx
A graph of y against x will go through the origin and have
gradient k, as illustrated in Figure 5.20.

0 x
Graphs like Figure 5.20, that by their shape show the nature of
the relationship between quantities but do not have scales marked
on the axes, are called ‘sketch graphs’. They can be very use-
Figure 5.20: A graph of y = kx
ful for illustrating ideas, without the need for accurate plotting or
drawing.

Back J I 267
Contents 

v
Question 5.6 Answer
Figure 5.21 shows the graphs corresponding to two dif-
ferent relationships between a variable v and another v = rz
variable z. The quantities r and s are constants. Which
is larger, r or s?
v = sz

0 z

Figure 5.21: Two proportional relation-

ships: v = rz and v = sz.

Back J I 268
Contents 
Question 5.7 Answer
Figure 5.22 shows three sketch graphs. Which of them represents a relationship
between directly proportional quantities?

f a u

0 z

0 g 0 b

(a) (b) (c)

Figure 5.22: Sketch graphs for use with Question 5.7.

Back J I 269
Contents 
5.3.2 A general equation for a straight line

Returning to the example of the oil delivery, suppose a different company decided
that it would sell at a lower cost per litre, but would impose a fixed delivery charge
in addition to the price of the oil. This situation is represented by an equation of the
form

total cost = (cost per litre × volume in litres) + delivery charge

and this is plotted on the graph in Figure 5.23.

200
total cost/£

100

0 250 500 750 1000

volume/litres

Figure 5.23: Graph of the cost of heating oil as a function of volume delivered.

Back J I 270
Contents 
Question
From Figure 5.23, estimate both the price per litre and the delivery charge.
Answer
The cost per litre is still given by the gradient of the graph, which is this case is
approximately
£(200 − 25) £175
= = 17.5 pence/litre
(1000 − 0) litre 1000 litre
The fixed charge can be estimated from the point at which the line crosses the
vertical axis: at this point, there is no charge for oil (since the volume is zero) so
the fixed charge represents the only contribution to the total cost. The delivery
charge is therefore £25.
Note that what this company is effectively doing is giving a discount for bulk
buying compared to the arrangement described by the graph of Figure 5.19. For
a delivery of 1000 litres, the cost is identical whichever company is used. For
less than 1000 litres, it would be cheaper to buy from the first company. For
volumes larger than 1000 litres the second company offers the better deal.

Back J I 271
Contents 
Generalizing from this example, if two quantities y and x are re-
lated by an equation of form y

y = mx + c (5.1)
y = mx + c
where m and c are constants, then a graph of y against x will be
a straight line that does not go through the origin. The graph will gradient = m
have gradient m. And when x = 0, then y = c, so the graph crosses
c
the vertical axis at c. The point at which a line on a graph crosses
an axis is called the intercept of the line with that axis. This is
illustrated in Figure 5.24. 0 x

The equation of a straight line is commonly written in the form

gradient of line Figure 5.24: A straight-line graph
with gradient m and intercept c on
intercept of line
with vertical axis the vertical axis.
y = mx + c

plotted on plotted on
vertical axis horizontal axis

Back J I 272
Contents 
Although the general equation of a straight line is most usually written in the form
y = mx + c, it is important to remember that the letters used and their order are quite
arbitrary: v = u + at is also the equation of a straight line. Also, although y = mx + c
does not contain any minus signs, both the gradient m and the constant c might have
a negative value.

Question
If v = u + at and v is plotted against t what, in terms of the symbols in the
equation, are the values of the gradient of the graph and the intercept on the
vertical axis?
Answer
v = u + at can be rearranged as v = at + u. Comparison with the standard
equation of a straight line
gradient intercept

y = m x + c

v = a t + u

shows that the gradient of a plot of v against t is a and the intercept with the
vertical axis is u.

An example of how the gradient and intercept of a straight line may be used to
derive quantities of real interest to scientists is given in Box 5.4.

Back J I 273
Contents 
Box 5.4 Einstein’s photoelectric equation
When light of particular colours is shone onto certain
metals, electrons are emitted from the metal, as shown
diagrammatically in Figure 5.25. Some of the energy
of the light is used to remove the electrons from the
metal; the amount of energy required to do this varies
from metal to metal, and is called the ‘work function’ emitted
electrons
φ of the metal. Any energy left over is given to the
escaping electrons:
metal
     
energy of  energy required to  energy of 
incident  = remove electrons  + ejected 
     
light
     
from the metal electrons 
Figure 5.25: The photoelectric effect.
This word equation can be rearranged as:
     
energy of  energy of  energy required to 
ejected  = incident  − remove electrons 
     
electrons  light  from the metal 

Back J I 274
Contents 
The colour of light is characterized by a quantity called its frequency f ,
and the energy of the incoming light is then given by h f , where h is a
constant called Planck’s constant. So the word equation above can be
rewritten in the form:

E = hf − φ

where the work function φ is a positive constant for any given metal.
You saw in Figure 5.12 a typical graph of energy, E, against frequency,
f . Comparison with the standard equation for a straight line shows how
such a graph could be used to determine both h and φ. (Notice that
the photoelectric equation contains a minus sign and therefore has to be
slightly rearranged to allow direct comparison.)

y = m x + c

E = h f + (− φ)

gradient intercept

The gradient calculated in Worked example 5.1 (i.e. 6.6 × 10−34 J s) is

therefore the value of Planck’s constant h, and extrapolation of the line
in Figure 5.12 to its intersection with the vertical axis could be used to
determine the work function of the metal.

Back J I 275
Contents 
5.4 Graphs of different shapes
The previous section showed that it is a relatively straightforward
matter to deduce the equation linking two variables when their 120
relationship can be represented by a straight-line graph. But of

A/cm2
course not all the quantities of interest in science are linearly 80
related to one another. Suppose you were to plot one variable 40
against another and obtained not a straight line but a curve. How
could you then determine the relationship between the variables?
0 2 4 6 8 10
r/cm
Imagine for example that you had taken a set of circular objects
with radii 1, 2, 3, . . . 6 cm and measured their respective areas.
Figure 5.26: Area A of circles plot-
Had you plotted the area A as a function of radius r you would
ted as a function of their radii r.
have obtained a graph like that in Figure 5.26.

Question
What is the equation relating the area A of a circle to its radius r?
Answer
A = πr2

This equation shows that A is not directly proportional to r, so you should not have
been surprised that plotting A against r did not give a straight line. In fact, the
curved shape of Figure 5.26 is characteristic of a relationship involving the square
of one of the quantities plotted. This particular shape is called a parabola.

Back J I 276
Contents 
Because in this case we know the equation relating A and r, it is
quite easy to see how the curve of Figure 5.26 can be ‘transformed’ 120
into a straight-line graph. A is equal to r2 multiplied by a constant

A/cm2
π. So although A is not directly proportional to r, it is directly 80
proportional to r2 :
40
2
A∝r
0 10 20 30 40 50
Therefore the result of plotting A against r2 is a straight line, as r 2 /cm2
illustrated in Figure 5.27.

Figure 5.27: Areas A of circles plot-

ted as a function of the squares of
Question their radii r2
Without measuring anything on the graph itself, can you state
the value of the gradient of the line in Figure 5.27?
Answer
Comparison with the standard equation for a straight line shows
that
y = m x + c
A = π r2 (+ 0)

so the gradient of the line is π.

Back J I 277
Contents 
Worked example 5.2
The relationship between the distance s travelled by an object which has been
dropped from a height and the time t for which it has been falling is

s = 12 gt2

where g is a constant (the magnitude of the acceleration due to gravity). If you

had measured the time as the object passed various points on the way down, how
would you use a graph to determine the value of g from your data?
Answer
Since s is directly proportional to t2 , these are the variables to plot. The descrip-
tion of the experiment shows that s is the independent variable, which according
to convention should be plotted on the horizontal axis.
We could rearrange the equation s = 12 gt2 to give
2s
t2 =
g

So if t2 is plotted against s, comparison with the standard equation for a straight

line shows that
y = m x + c

2
t2 = g
s (+ 0 )

Back J I 278
Contents 
2
so the gradient of the line = and
g
2
g=
gradient

g
{If you chose to plot s against t2 , then the gradient would be , in which case
2
g = 2 × gradient.}

Question 5.8 Answer

Table 5.1 showed the mass of a number of aluminium spheres as a function of
their diameters. When mass was plotted as a function of diameter, a curved
graph (Figure 5.6) was obtained. The mass M of a sphere of diameter d, made
of material of density ρ, is given by the equation M = πρd3 /6.
What quantities would you plot in order to obtain a straight-line graph from the
data in Table 5.1?
What expression would be given by the gradient of the line?

Back J I 279
Contents 
Question 5.9 Answer
If you have ever regulated a long-case (grandfather) clock, you will know that
the length of the pendulum, L, determines the period T (the time for one com-
plete swing) and hence affects the accuracy with which the clock keeps time.
For a simple pendulum, the period is given by
s
L
T = 2π
g

where, as in Worked example 5.2, g is a constant (the magnitude of the acceler-

ation due to gravity). If you had measured T for various values of L, how would
you use a graph to determine the value of g from your data?
{Hint: you may find it helpful to manipulate the equation so as to get rid of the
square root.}

The trick of plotting quantities in such a way as to obtain a straight line is very
useful when you want to discover the relationship between experimentally mea-
sured quantities. With practice, one can come to recognize curved graphs of vari-
ous shapes, and this helps considerably in deciding how to transform the original
data so as to obtain a straight-line plot. For example, if the result of plotting one
quantity against another is a parabolic curve, this is an immediate indication that
one of those quantities is proportional to the square of the other.

Back J I 280
Contents 
The rest of this section will simply introduce you to a few
curves of different shapes and the equations to which they pressure P
correspond. (An explanation of the techniques by which one
piston
can most easily take scientific data and discover what powers
of the variables should be used in order to get a linear plot
will come in Chapter 7.)
A completely different sort of curve is generated from experi-
ments using the apparatus in Figure 5.28. This piston arrange- gas
ment is designed for the study of a sample of gas. A pressure volume V
P can be applied to the piston and as the pressure increases so
the volume V of the gas in the chamber will decrease. Con- heat bath
versely, if the pressure is reduced, the gas in the chamber will temperature T
expand. If you have ever pumped up a bicycle tyre, you have
probably noticed that when a gas is compressed it heats up,
so in order to be sure that pressure and volume are the only Figure 5.28: An apparatus for measuring
variables involved in this particular experiment, it is impor- how the volume of a sample of gas varies
tant to ensure that each time the pressure is changed the gas with the pressure at constant temperature.
is allowed to return to its original temperature T before the
volume is measured. This temperature is maintained by the
heat bath.

Back J I 281
Contents 
A sketch graph showing the shape of a plot of V against P resulting
V
from such an experiment is shown in Figure 5.29. A plot of this
shape is called a hyperbola. A characteristic feature of the hyper-
bola is that as the variable on one axis approaches zero, the curve
approaches more and more closely to the other axis but never actu-
ally touches it.
A hyperbola arises from plotting two quantities that are linked by
one being directly proportional to the reciprocal of the other. In this
case, 0
P
1
V∝
P
Figure 5.29: A graph of volume as
This could also be expressed in words by saying that ‘V is directly a function of pressure for a fixed
proportional to one over P’ but it is more usual to say that V is amount of gas at constant temper-
inversely proportional to P. ature.

Back J I 282
Contents 
In order to obtain a straight-line plot, we would therefore have to plot V against
1/P, as illustrated by the sketch graph in Figure 5.30. In practice the volume of a
real gas can never fall to zero, but if the line were extrapolated it would go through
the origin.

0
1/P

Figure 5.30: At constant temperature, the volume of a fixed amount of gas is in-
versely proportional to the pressure.

Back J I 283
Contents 
Question

If you wanted to plot a graph using measured values of P P

and V, but with P as the dependent variable, what would
you have to plot on the horizontal axis in order to obtain a
straight line?
Answer
1 k
We know that V ∝ , i.e. that V = , where k is a constant.
P P
k 1 0
Rearrangement gives P = , i.e. P ∝ . 1/V
V V
So a graph of P against 1/V would also be a straight line, as Figure 5.31: At constant tempera-
illustrated in Figure 5.31. ture, the pressure of a fixed amount
Note that graphs of 1/V against P and 1/P against V would of gas is inversely proportional to
also be straight lines. the volume.

So far we have been primarily interested in the relationship between the pressure
and the volume of the sample of gas, so the sketch graphs of Figures 5.29 to 5.31
correspond to a situation in which the temperature has been held constant. How-
ever, it would be equally possible to use the apparatus illustrated in Figure 5.28 to
measure the volume of the gas sample as a function of temperature. Such measure-
ments are the basis of the SI (kelvin) scale of temperature, which is discussed in
Box 5.5.

Back J I 284
Contents 
Box 5.5 The absolute zero of temperature
Figure 5.31 shows that the pressure and volume of a fixed amount of gas at
constant temperature are related by an equation of form P = k/V where k is a
constant, i.e.

PV = k (at constant temperature).

This equation is a particular case of a more general equation which was intro-
duced in Box 3.4, namely
PV = nRT (3.12)
where n is the number of moles of gas, R is the so-called gas constant, and T
is the temperature (measured in kelvin).
Equation 3.12 can be rearranged to give
nR
V= T
P
and if P is held constant then

volume = C × temperature

where C is a constant equal to nR/P.

Back J I 285
Contents 
The value of C will depend on the value of P chosen, so if the volume of the
sample of gas is measured as a function of temperature in three separate experi-
ments, each one at a different constant pressure, three separate straight-line plots
will be obtained, each with a different gradient. The larger the value of P chosen,
the smaller the gradient will be.
Figure 5.32a shows how the volume of the
sample of gas measured at three different
pressures, varies over the temperature range

volume
0 ◦ C to 100 ◦ C (note that the temperatures
P1
P1

volume
here are given in degrees Celsius). The re- P2 P2
ally interesting aspect of the graph is that P3 P3

if the lines are extended to lower and lower

temperatures, as shown in Figure 5.32b, they 0 −273.15 0
100 100
all meet at the same point on the horizon- temperature/°C temperature/°C
tal axis, corresponding to a temperature of (a) (b)
−273.15 ◦ C, and to a volume of zero. Ex-
trapolation beyond this point would imply
a negative volume, which is impossible, so Figure 5.32: (a) At constant pressure, the volume of
−273.15 ◦ C is the lowest possible tempera- a fixed amount of gas is clearly related to the temper-
ture. It is therefore known as the absolute ature. Here P1 < P2 < P3 . (b) Extrapolation shows
zero of temperature. The SI (kelvin) scale of that when the volume is zero then the temperature is
temperature sets this lowest possible temper- −273.15 ◦ C.
ature at 0 K.

Back J I 286
Contents 
Temperatures may be converted from degrees
Celsius to kelvin and vice versa using the word V
P1
equation:
   
  temperature P2
temperature 
 = in degrees  + 273.15


in kelvin  
Celsius

P3

Figure 5.32b is a reminder that at −273.15 ◦ C

(i.e. at 0 K) the volume is zero.
When T is expressed in kelvin, V is directly
0
proportional to T , so the lines in Figure 5.33 T /K
go through the origin.

Figure 5.33: If T is measured on the kelvin scale, then

V ∝ T.

Back J I 287
Contents 

Yet another type of curved graph is obtained when

80
the activity of a radioactive sample is monitored over
time. The atoms of radioactive elements ‘decay’ by
emitting small particles from their nuclei, thereby 70
transforming themselves into atoms of quite different
elements. These other elements may themselves be 60
radioactive, or they may be stable. Radioactive de-
cay is a random process, in that, although the total 50
activity of a sample is predictable, one can never pre-

activity/kBq
dict which individual nuclei are going to decay at any 40
particular time. One form of polonium, the element
named after the Polish homeland of Marie Curie, de-
30
cays to leave stable atoms of lead. The activity of a
sample of polonium is plotted as a function of time in
Figure 5.34; the unit of activity is the bequerel (Bq), 20
equal to 1 disintegration per second. Because disinte-
gration of a polonium nucleus produces a stable lead 10
nucleus, the number of unstable nuclei in the sample
— and hence the activity — falls as time goes on.
0 100 200 300 400 500
time/days

Figure 5.34: The activity of a sample of polo-

nium as a function of time.

Back J I 288
Contents 
Question
How long does it take for the activity of the polonium sample in Figure 5.34 to
drop to
(a) 40 kBq
(b) 20 kBq
(c) 10 kBq
Answer
Reading from Figure 5.34,
(a) the activity has dropped to 40 kBq after 140 days
(b) the activity has dropped to 20 kBq after 280 days
(c) the activity has dropped to 10 kBq after 420 days

A little further analysis shows that the time taken for the activity to drop:
• from 80 kBq to 40 kBq = 140 days
• from 40 kBq to 20 kBq = (280 − 140) days = 140 days
• from 20 kBq to 10 kBq = (420 − 280) days = 140 days
This result demonstrates a very important property of the curve plotted in Figure
5.34; whatever value of the quantity plotted on the vertical axis is chosen, the time

Back J I 289
Contents 
taken for the quantity to fall to exactly one-half that value is a constant. This con-
stant interval of time is known as the half-life, and curves that display this property
are called ‘exponential decays’. To the precision to which it is possible to read
Figure 5.34, the half-life of the polonium sample is 140 days.
In radioactive decay, the activity is dependent on the number of radioactive nuclei
present, which is usually denoted by the letter N. Figure 5.35 shows that if N0
radioactive nuclei are present when timing starts (i.e. at time t = 0), then
• after one half-life N = N0 × 12
   2
• after two half-lives N = N0 × 12 × 1
2 = N0 1
2
 2  3
• after three half-lives N = N0 12 × 1
2 = N0 1
2
 n
• so after n half-lives N = N0 12
After a long time, and many half-lives, N will approach, though it will never reach,
zero.
The equation describing the exponential
√ decay shown in Figure 5.35 involves a
special number, e. Like π and 2, e is an irrational number and to four significant
figures its value is 2.718. The equation describing Figure 5.35 is

N = N0 e −λt (5.2)

where λ is a positive constant. In Chapter 7, you will discover the relationship

between λ and the half-life for the decay, t1/2 . Then in Chapter 10 you will find that

Back J I 290
Contents 
exponentials have another characteristic and defining property.

Question 5.10 Answer

Radium has a half-life of 1600 years. How long will it be before the number of
1
radioactive atoms in a sample is reduced to 16 of the number there are today?

Box 5.6 Dating meteorites and Moon rock

The age of many different natural materials can be determined from their ra-
dioactivity. Potassium is one element that is used to date rocks; potassium-40
has a half-life of 1.3×109 years, and decays to leave argon, an inert gas that does
not combine with other elements. When rocks first form, they are molten, so any
argon they might contain would simply escape into space. However, once the
rocks solidify, any argon resulting from the radioactive decay of potassium-40
remains trapped. Geochemists can analyse the composition of a rock to deter-
mine the ratio of potassium to argon, and hence estimate a rock’s age.
Dating using potassium and other radioactive elements has shown that almost
all known meteorites are 4.6 × 103 Ma old, so their formation was contempora-
neous with the formation of the Solar System. The oldest known Moon rock is
4.48 × 103 Ma old.

Back J I 291
Contents 
It is sometimes reported in the media that something is exhibiting
‘exponential growth’. In fact, true exponential growth, in which
the quantity being measured is multiplied by a constant factor over 60
a given period of time, is a rather unusual phenomenon although it
does occur. A general equation for exponential growth, analogous 50
to Equation 5.2 for exponential decay, is

number of cells
40
n = n0 e at
(5.3)
30
where n0 is the starting value of the quantity, n is its value after
time t and a is a positive constant. Exponential growth is some- 20
times used as a model by biologists interested in the populations
of organisms. Figure 5.36 illustrates the theoretical increase of 10
yeast cells according to such a model. In practice, the death of
organisms, as well as the influence of factors relating to over- 0
10 20
crowding, will also affect the population, so that the increase in time/hours
the number of organisms will not lie on a true exponential growth
curve. Figure 5.36: Model for the growth
of yeast cells. The population con-
sists of just two cells at time t = 0,
and it is assumed that once in every
four-hour period each cell divides
into two cells, (i.e. the multiplying
factor is 2).

Back J I 292
Contents 
5.5 Learning outcomes for Chapter 5
After completing your work on this chapter you should be able to:
5.1 demonstrate understanding of the terms emboldened in the text;
5.2 correctly interpret conventional labelling on graph axes or table columns, so
as to deduce the power of ten and the units associated with a plotted or
tabulated quantity;
5.3 use the processes of interpolation and extrapolation to read values from a
graph;
5.4 calculate the gradient of a straight-line graph;
5.5 deduce the gradient and intercept of a straight-line graph from the equation of
the line, and vice versa;
5.6 draw and interpret sketch graphs;
5.7 given the equation involving quantities raised to a power, decide what variable
should be plotted in order to obtain a straight-line graph.

Back J I 293
Contents 

Angles and trigonometry 6

It is relatively easy to measure the distance along
the ground from an observer to an object such as
a tree, but measuring the height of the tree itself is
rather less straightforward. Similarly, it is possi-
ble to find the distance from the Earth to the Moon
by measuring the time taken for a laser beam to
travel to the Moon and back, but this method can-
not be used to find the Moon’s diameter. Fortu-
nately help is at hand in both cases; we can mea-
sure angles and use these to calculate the values
we require. In the case of the tree, the angle used
is the angle between the ground (assumed to be
horizontal) and a straight line drawn to the top of Figure 6.1: Chapter 6 will show how to use angles to
the tree; this angle is marked θ (the Greek letter find the height of a tree
theta) in Figure 6.1.
Remember that a list of Greek letters and their pronunciation is given in Table 3.1.

Back J I 294
Contents 

Figure 6.2: Chapter 6 will show how to use angles to find the diameter of the Moon

In the case of the Moon the angle is the one subtended (i.e. swept out) as a straight
line drawn from an observer on the Earth moves from one side of the Moon to the
other; this angle is labelled φ (the Greek letter phi) in Figure 6.2.
Section 6.1 describes two different systems used for measuring angles and, after a
brief look at some of the properties of triangles, the rest of the chapter shows how
angles can be used in scientific calculation to determine things such as the height of
a tree and the diameter of the Moon.

Back J I 295
Contents 
6.1 Measuring angles: degrees and radians
You are probably familiar with the use of a protractor to measure angles shown on
diagrams; this gives a result in degrees (represented by the symbol ◦ , and sometimes
known as ‘degrees of arc’ to make it clear that the degrees used to measure angles
have nothing whatsoever to do with the degrees used when measuring temperature
on the Celsius scale). Figure 6.3 shows that angle θ from Figure 6.1 is about 36.5◦ .

Figure 6.3: Measuring an angle with a protractor.

Back J I 296
Contents 
If you stand facing in a particular
direction then turn through a com-
plete revolution, you will have gone
through 360◦ . The use of 360◦ to rep-
resent a complete turn is believed to
date back to the ancient Babylonian direction
of rotation
civilization; 360 subdivisions were 135°
90°
used because 360 is close to 365, the 180°
number of days in a year. Figure 6.4 45° starting
illustrates various angles encountered position
in turning through a circle. Note in 225°
particular that a right angle (the angle
between two directions that are per-
pendicular to each other) measures 360°
90◦ .
Box 6.1 on the next page describes
the use of angles to define lines of
longitude and latitude on the Earth’s
Figure 6.4: Angles encountered in turning through a circle.
surface, and hence to specify posi-
tions on the surface of the Earth.

Back J I 297
Contents 
Box 6.1 Lines of longitude and latitude
The surface of the Earth is conventionally marked with two sets of imaginary
lines, as shown in Figure 6.5. The blue lines running from left to right in Fig-
ure 6.5 are lines of latitude, the Equator being one such line, and the red lines
running from one pole to the other are lines of longitude.

Figure 6.5: A model of the Earth viewed from above the Equator, showing lines
of latitude and longitude.

Back J I 298
Contents 
In Figure 6.6, which is the view from above the North Pole, the circles are the
lines of latitude and the lines radiating out from the pole are lines of longitude.
It is easy to see, from Figure 6.6, how angles of longitude can be labelled using
degrees. A line running through Greenwich in east London, and known as the
Greenwich Meridian, is defined to be 0◦ longitude, and other lines are labelled
by measuring the angles to the east or west of the Greenwich Meridian.

Figure 6.6: A model of the earth viewed from above the North Pole, showing
lines of latitude and longitude.

Back J I 299
Contents 
Figure 6.7 shows how angles can be used to label lines of latitude too.

Figure 6.7: Using angles to label lines of latitude.

Figures 6.5, 6,6 and 6.7 show lines of longitude and latitude at 15◦ intervals only,
but in reality the lines can be drawn as close together as required, and so can be
used to specify a location very precisely. In order to specify a precise location, we
need to subdivide degrees of longitude and latitude in some way. Historically this
was done by dividing each degree into 60 minutes (or ‘minutes of arc’) in the same
way as each hour is divided into 60 minutes (of time). The symbol ‘ 0 ’ is used to

Back J I 300
Contents 
represent minutes of arc. The longitude of Heathrow Airport (approximately 30 km
west of Greenwich) is 0◦ 270 W and both Greenwich and Heathrow have a latitude
of about 51◦ 280 N.
Minutes of arc are rarely used in modern science; small angles are usually expressed
28
in decimal notation. Since 280 is 28 sixtieths of a degree and = 0.47 to two
◦ 0 ◦
60
significant figures, 51 28 can be written as 51.47 . However, astronomers continue
to use a further extension of the ‘degrees and minutes’ notation, simply because
the angles they are measuring are frequently very small (since the objects they are
measuring are such a long way from Earth). In order to measure such small angles,
minutes of arc are further divided into 60 seconds of arc, or arcsecs (in the same
way as minutes of time are subdivided into 60 seconds). So
1 1 1 1
1 arcsec = minute of arc = × degree = degree
60 60 60 3600
As the Earth orbits the Sun, the next nearest star, Proxima Centauri, appears to
0.772
move through an angle of 0.772 arcsecs across the sky; this is a mere of a
3600
degree, i.e. 2.14 × 10−4 degrees.

Back J I 301
Contents 
Angles in science are frequently measured in radians rather
than in degrees and subdivisions of degrees. Consider the X
circle shown in Figure 6.6. A part of the circumference, such
as that between point X and point Y, is known as an arc, and r
in this case the arc subtends an angle θ. The length of the arc
between X and Y is s and the radius of the circle is r. The θ s
radian is defined with reference to arc length and radius.
r

Y
The size in radians of the angle, θ, subtended by an arc
is defined to be arc length, s, divided by radius, r, thus
s
θ (in radians) = (6.1) Figure 6.8: An arc of length s subtended
r by the angle θ in a circle of radius r.

Back J I 302
Contents 
Question
What is the size in radians of the angle subtended by an arc of length 3.0 cm in
a circle of radius 2.0 cm?
Answer
From Equation 6.1 the angle is given by:
s
θ in radians =
r
3.0 cm
=
2.0 cm
= 1.5

So the size of the angle is 1.5 radians.

Note that since we have divided a length in centimetres by another length in cen-
timetres, it could be argued that the answer should have no units. However, this
course will adopt the common practice of writing the word ‘radians’ next to an-
gles given in this measuring system, to distinguish them from angles measured in
degrees or in any other system of angular measure.
An angle subtended by a longer arc in a circle of the same radius will be larger, as
expected. In the above example, an arc of length 5.0 cm would subtend an angle of
5.0 cm
, i.e. 2.5 radians.
2.0 cm

Back J I 303
Contents 
Note, however, that it is the ratio of arc length to radius which
is important in the definition of radian. This is illustrated in
X′
Figure 6.9, which shows two concentric circles (i.e. two cir-
cles with their centres at the same point). X

The smaller circle has radius r, and an arc of length s (sub- r

tended by angle θ) is shown. In the larger circle, of radius
r0 , the same angle θ is subtended by an arc of length s0 . The θ s s′
superscript ‘ 0 ’ is used to indicate that the lengths r0 and s0
(said as ‘r-prime and s-prime’, or ‘r-dash and s-dash’) both
r′ Y
relate to the same circle. The lengths s0 and r0 are bigger than
s
the values of s and r, as you would expect, but the ratios
r Y′
s0
and 0 are equal, and the angle subtended in radians is
r
s s0
θ= = Figure 6.9: Two concentric circles.
r r0

Back J I 304
Contents 
Let’s now consider two special cases. In the first, the arc length is
exactly equal to the radius, as shown in Figure 6.10a, i.e. s = r.
This means that
s r r
θ (in radians) = = =1 θ r
r r
r
i.e. the angle subtended is one radian.
In the second special case, illustrated in Figure 6.10b, the arc length
(a)
is a complete circumference. For all circles, the circumference, C,
is linked to the radius, r, by the formula C = 2π r (this formula,
given in Box 3.4, follows directly from the definition of π, given in
Section 1.1.1 as circumference divided by diameter). So when the
arc length, s, is equal to the whole circumference, C, s = 2π r so
2π r r θ
θ (in radians) = = 2π
s
r

(b)
Thus the angle subtended by a complete revolution is 2π radians,
i.e. 2π radians = 360◦ .

Figure 6.10: The angle subtended

when (a) arc length is equal to ra-
dius, and (b) arc length is equal to
circumference.
Back J I 305
Contents 
This gives us an easy way of converting between degrees and radians.
Since 2π radians = 360◦ ,
360◦
1 radian =
2π
180◦
=
π
≈ 57.3◦

where the symbol ‘≈’ means ‘is approximately equal to’, as in Chapter 3.
Similarly, since 360◦ = 2π radians,
2π
1◦ =
360
π
=
180
≈ 0.0175 radians

Note that the numerical conversion factors between radians and degrees are only
approximate (they have been given to three significant figures), so when converting
from radians to degrees or vice versa it is best to go back to first principles in
each case, remembering that a complete revolution can be represented by either
2π radians or 360◦ .

Back J I 306
Contents 
It is also worth remembering that angles in radians are frequently expressed as
multiples or fractions of π so, for example,
2π
45◦ = 45 × radians
360
π
= radians
4
π
An angle of exactly 45◦ is equal to exactly radians.
4

Worked example 6.1

π
Express radians in degrees.
6
Answer
2π radians = 360◦ so π radians = 180◦ .
π 180◦
radians = = 30◦ .
6 6

Back J I 307
Contents 
Worked example 6.2
The angle subtended as a straight line drawn from an observer on the Earth
moves from one side of the Moon to the other is 0.519◦ . (This is angle φ in
Figure 6.2, but remember that the figure is not drawn to scale). Express this
angle in radians.
Answer
2π
360◦ = 2π radians so 1◦ = radians
360
2π
0.519◦ = 0.519 × radians
360
= 9.06 × 10−3 radians to three significant figures.

Question 6.1
Convert the following from radians to degrees:
(a) 0.123 radians Answer
2π
(b) radians Answer
3
3π
(c) radians Answer
2

Back J I 308
Contents 
Question 6.2
Convert the following from degrees to radians:
(a) 36.5◦ (angle θ in Figure 6.1) Answer
(b) 90◦ Answer
(c) 210◦ Answer

6.2 A quick look at triangles

Note the labelling system used for angles in Figure 6.11. Angle
α could also be identified as angle BAC, ∠BAC or Â, but in this A
course angles will be labelled on the inside in the way angles α, α
β and γ have been labelled in Figure 6.11. If you measure the
size of the angles inside the triangle shown in Figure 6.11 with a
protractor, you will find that α = 80◦ , β = 60◦ and γ = 40◦ . Thus
β γ
α + β + γ = 80◦ + 60◦ + 40◦ = 180◦ B C

This result is true for all triangles, i.e.

Figure 6.11: The angles inside a tri-
For all triangles, the internal angles add up to 180◦ . angle.

Back J I 309
Contents 
If you wish, you can check that this result holds for all of the triangles shown in
Figure 6.12, irrespective of the shape of the triangle. Figure 6.12e and Figure 6.12f
illustrate two triangles of a particular type; each has one internal angle equal to
90◦ , i.e. a right angle, so they are known as right-angled triangles. Note that the
right angles have been labelled in the conventional way, with a square corner. In a
right-angled triangle, since the internal angles must total 180◦ and one of the three
angles is 90◦ , it follows that the other two angles must add up to a total of 90◦ too.
This result means that if you know that a triangle is right-angled, and you know one
of the other angles, you can find the remaining angle without needing to measure
it. In Figure 6.12e, α = 30◦ , so β = 90◦ − 30◦ = 60◦ .
Pythagoras’ Theorem, whose proof is accredited to the Greek philosopher Pythago-
ras or one of his followers about 2500 years ago, but which was probably known
even earlier, gives us a way of calculating the length of a third side of a right-angled
triangle from a knowledge of the lengths of the other two sides.
The side opposite the right angle in a right-angled triangle is known as the hy-
potenuse, and Pythagoras’ Theorem is commonly stated as

The square of the hypotenuse of a right-angled triangle is equal to the sum of

the squares of the other two sides.

Back J I 310
Contents 
In the triangle shown in Figure 6.13, the hypotenuse has a length h and
the other two sides have lengths a and b. Thus

h2 = a2 + b2 (6.2)
h
We are only interested in the positive square root, so b

p
h = a2 + b2

If a = 3 cm and b = 4 cm in a right-angledtriangle, then a

p p p
h = a2 + b2 = (3 cm)2 + (4 cm)2 = 9 cm2 + 16 cm2
p Figure 6.13: A right-angled
= 25 cm2 = 5 cm
triangle.

If h = 9.1 m and a = 5.1 m in a different right-angled triangle, then

h2 = a2 + b2 can be rearranged to give b2 = h2 − a2 so
p p p
b = h2 − a2 = (9.1 m)2 − (5.1 m)2 = 82.81 m2 − 26.01 m2
p
= 56.80 m2 = 7.5 m to 2 significant figures.

Back J I 311
Contents 
Question 6.3
The base of ladder of length 4.50 m is placed on level ground at a distance of
1.15 m from a vertical wall, and the top of the ladder leans against the wall. The
angle between the ground and the ladder is found to be 75.2◦ . Calculate
(a) the height that the ladder reaches up the wall; Answer
(b) the angle between the wall and the top of the ladder. Answer
Hint: you may find it helpful to start by drawing a diagram of the situation.

Pythagoras’ Theorem provides us with a way of finding unknown lengths from

known lengths; the fact that the internal angles in any triangle add up to 180◦ pro-
vides us with a way of finding unknown angles from known angles. Trigonometry,
discussed in Section 6.3, takes us one stage further by providing a way of finding
unknown lengths from known angles and unknown angles from known lengths.

6.3 Calculating with angles: trigonometry

Trigonometry is the branch of mathematics that deals with the relationships be-
tween the sides and angles of triangles. The Greek astronomer Hipparchus is cred-
ited with its invention, but the principles involved were almost certainly in use even
earlier by the ancient Egyptians surveying the land surrounding the Nile. Despite
its ancient origins, trigonometry plays an important part in modern science.

Back J I 312
Contents 
Look at the three right-angled triangles shown in Figure 6.14. These triangles are
similar, i.e. they have the same shape (although their sizes are different); note in
particular that the angle θ is exactly the same in each of the three triangles.
The superscript symbols ‘ 0 ’ and ‘ 00 ’ (‘prime and double-prime’ or ‘dash and
double-dash’) indicate lengths relating to the second and third triangles respectively.
As you look at Figure 6.14 from left to right, you will see that the triangles have
sides of increasing length; however the ratio of any one side to each of the other
sides remains constant, thus
b b 0 b 00
= = (6.3)
a a 0 a 00
b b 0 b 00
= = (6.4)
h h 0 h 00
a a 0 a 00
= = (6.5)
h h 0 h 00

If the angle θ and hence the shape of the triangle had been different, the ratios would
b b
have had different values. Thus each angle θ gives rise to unique values for ,
a h
a b b a
and , and conversely each value for , or in a triangle leads to a particular
h a h h
value for θ. This result is so important that the ratios are given the special names
tangent, sine and cosine, usually abbreviated to tan, sin and cos. Tan, sin and cos
are known collectively as trigonometric (or trig.) ratios.

Back J I 313
Contents 
The tangent of angle θ is defined by

opposite
tan θ = (6.6)
adjacent

b
This is the ratio we have been describing as , where b is the side opposite angle θ
a
and a is the side (other than the hypotenuse) that is adjacent (next to) angle θ.

The sine of angle θ is defined by

opposite
sin θ = (6.7)
hypotenuse

b
This is the ratio we have been describing as .
h

The cosine of angle θ is defined by

adjacent
cos θ = (6.8)
hypotenuse
a
This is the ratio we have been describing as .
h

Back J I 314
Contents 

hyp
opp

θ
adj

opp opp adj

Figure 6.15: tan θ = ; sin θ = ; cos θ =
adj hyp hyp

The sides opposite and adjacent to a particular angle in a right-angled triangle are
usually abbreviated to ‘opp’ and ‘adj’ and the hypotenuse is abbreviated to ‘hyp’,
as shown in Figure 6.15.
Note that the trigonometric ratios are defined with respect to a particular angle in
a right-angled triangle. If we had considered the other non right-angled angle in
the triangle in Figure 6.15, the ‘opposite’ and ‘adjacent’ sides would have been
different, and so the sine, cosine and tangent would have been different too.

Back J I 315
Contents 
The trigonometric ratios were tabulated many years ago and
generations of scientists have used tables and slide rules sim-
ilar to those shown in Figure 6.16 to calculate lengths from
angles and angles from lengths. Nowadays, trigonometric ra-
tios are available at the press of a calculator button.

6.3.1 Using a calculator for trigonometry

Make sure that you can use your calculator to find trigonomet-
ric ratios. The sine, cosine and tangent functions are likely to
be clearly marked as ‘sin’, ‘cos’ and ‘tan’. Remember, from
Section 6.1, that angles can be measured in either degrees or
radians. Your calculator should be able to cope with either Figure 6.16: Tables, slide rules and calcu-
of these (and possibly a third angular measure called ‘grad’ lators can all be used to find trigonometric
too) but you need to ensure that the calculator is in the correct ratios.
‘mode’. Angle θ in Figure 6.15 is 30 , alternatively written as
◦
π
radians, so the sine of angle θ could be expressed as either
6
π π
sin 30◦ or sin (where is in radians, though the word ‘ra-
6 6
π
dians’ is usually omitted when finding trigonometric ratios). Note that sin (which
6
π π
is the sine of the angle and could be written as sin for clarity) is not the same
6 6
sin π 1 (sin π)
as (which is th of the sine of the angle π and could be written as for
6 6 6

Back J I 316
Contents 
clarity).
Check that you can use your calculator to give:

sin 30◦ = 0.5; cos 30◦ = 0.8660; tan 30◦ = 0.5774

and also to give:

π π π
sin = 0.5; cos = 0.8660; tan = 0.5774
6 6 6
where the answers are either exact or given to four significant figures.
Note that when using trigonometric ratios you should always work to at least four
significant figures (although you should round your answer to an appropriate num-
ber of significant figures at the end of a calculation).

Question 6.4
Use your calculator to find:
(a) sin 49◦ Answer
π π
(b) cos (where is in radians) Answer
8 8
π π
(c) tan (where is in radians). Answer
4 4

Back J I 317
Contents 
You will also need to be able to use your calculator to find the angle which has a
particular sine, cosine or tangent. For example, if you know that tan θ = 0.75, then
what is θ in degrees? What you are looking for is known as the ‘inverse tangent’
or arctangent and you need to use a button on your calculator labelled as ‘tan−1 ’ or
‘arctan’. Check that you can use your calculator to give the correct answer, which
is that tan−1 (0.75) = 37◦ = 0.64 radians to two significant figures. Your calculator
should also be able to calculate ‘inverse sine’ (using a button labelled as ‘sin−1 ’
or arcsin and ‘inverse cosine’ (‘cos−1 ’ or arccos). Note that ‘tan−1 ’, ‘sin−1 ’ and
‘cos−1 ’ are properly referred to as the inverse functions of tan, sin and cos (as they
work in the opposite direction) but care needs to be taken to avoid confusion with
reciprocals:
1
tan−1 ,
tan
1
sin−1 ,
sin
1
cos−1 ,
cos
remembering that , means ‘is not equal to’.

Back J I 318
Contents 
Question 6.5
Use your calculator to find:
(a) Use your calculator to find the angle α (in degrees) for which Answer
cos α = 0.5253.
(b) Use your calculator to find the angle β (in radians) for which Answer
tan β = 1.5574.

Note that although we have only defined trigonometric ratios for angles in a right-
angled triangle, and most of the angles for which trigonometric ratios are used in
this course are acute (i.e. less than 90◦ ), values of sin, cos and tan can be found
for larger angles too. Use your calculator to check that sin π = 0, cos π = −1
and tan π = 0 (where π is an angle in radians, equal to 180◦ ). Box 6.2 considers
the sines and cosines of angles greater than 90◦ in slightly more detail, and it also
introduces you to negative angles and their trigonometric ratios.

Back J I 319
Contents 
Box 6.2 Using trigonometric ratios to describe waves θ in radians sin θ cos θ
It is possible to assign values for sin θ and cos θ for all an-
0 0 1
gles, however large they are. Table 6.1 gives values for sin θ
π/6 0.5 0.8660
and cos θ for selected values of θ up to the arbitrarily chosen
π/3 0.8660 0.5
value of 3π (540◦ ). The angles are like those encountered
π/2 1 0
in Figure 6.4 in turning through a complete circle, except
2π/3 0.8660 −0.5
that there is no need to stop at 360◦ , and the angles are now
5π/6 0.5 −0.8660
measured in radians.
π 0 −1
Two results have been omitted from Table 6.1. 7π/6 −0.5 −0.8660
4π/3 −0.8660 −0.5
Question
3π 3π/2 0
Use your calculator to find the sine of radians (270◦ ) 5π/3 −0.8660 0.5
2
13π 11π/6 −0.5 0.8660
and the cosine of radians (390◦ ) and add these values 2π 0 1
6
to Table 6.1. 13π/6 0.5
7π/3 0.8660 0.5
Answer
5π/2 1 0
3π 8π/3 0.8660 −0.5
sin = −1
2 17π/6 0.5 −0.8660
13π 3π 0 −1
cos = 0.8660 to four significant figures.
6
Table 6.1: Values of sin θ and cos θ for θ
from 0 to 3π

Back J I 320
Contents 
If instead of turning in an anticlockwise direction in the way θ in radians sin θ cos θ
used to define angles up to 360◦ and beyond (Figure 6.4), we
had turned in a clockwise direction, the angles would have 0 0 1
been measured in the opposite direction. Angles such as −π/6 −0.5 0.8660
π 3π −π/3 −0.8660 0.5
these are defined to be negative e.g. − , −π, − . Values −π/2 −1 0
6 2
for sin θ and cos θ can also be assigned for negative values −2π/3 −0.8660 −0.5
of θ, as shown in Table 6.2. −5π/6 −0.5 −0.8660
Inspection of Table 6.1 and Table 6.2 shows that sin θ and −π 0 −1
cos θ each vary between −1 and +1 across the whole range −7π/6 0.5 −0.8660
of values for θ. The form of the variation is made clearer by −4π/3 0.8660 −0.5
the graphs shown in Figure 6.17. −3π/2 1 0
−5π/3 0.8660 0.5
The graphs may remind you of the sort of wave pattern ob- −11π/6 0.5 0.8660
served when you take an instantaneous sideways look at −2π 0 1
waves on a pond. In fact, sine and cosine functions, of the −13π/6 −0.5 0.8660
form y = a sin θ and y = a cos θ (where a is a constant) are −7π/3 −0.8660 0.5
extensively used in describing the motion of waves of all −5π/2 −1 0
types. The detail is beyond the scope of this course, but it is −8π/3 −0.8660 −0.5
another application of maths in science! −17π/6 −0.5 −0.8660
−3π 0 −1

Table 6.2: Values of sin θ and cos θ for θ

from 0 to −3π

Back J I 321
Contents 
6.3.2 Using trigonometry in science

We are now in a position to be able to find

the height of the tree mentioned at the be-
ginning of the chapter. This is shown as H
in Figure 6.18. We know that θ = 36.5◦ and
suppose we have measured D, the distance
to the tree from the point at which the angle
θ was measured, to be 28.6 m. How tall is
the tree?
We can say that
opp
tan θ =
adj
Figure 6.18: Using trigonometry to find the height of a tree.
H
= in this case. θ = 36.5◦ and D = 28.6 m.
D
We need to rearrange this equation to make H the subject; we can do this in exactly
the same way as we did in Chapter 4, by reversing the equation and then multiplying
both sides of the equation by D. This gives

H = D tan θ

So H = 28.6 m × tan 36.5◦ = 21.2 m to 3 significant figures.

Back J I 322
Contents 
It was clearly stated at the beginning of the
chapter that θ was the angle between the
ground and a straight line drawn to the top
of the tree, but in reality you’re more likely
to have taken readings at eye level, per-
haps using an instrument such as a ‘gun cli-
nometer’, whose use is illustrated in Figure
6.19. The gun clinometer measures the an-
gle shown as α in Figure 6.19b, and Worked
example 6.3 shows how this can be used to
find the height of a tree.

Figure 6.19: (a) Using a gun clinometer to find the height

of a tree; (b) the gun clinometer gives angle α.
Back J I 323
Contents 
Question
When used by a man of height 1.8 m and in the way illustrated in Figure 6.19, a
gun clinometer records an angle α of 39◦ at a distance, D, of 18 m from a tree.
What is the height of the tree?
Answer
H
tan α = where α = 39◦ and D = 18 m, so
D
H = D tan α
= 18 m × tan 39◦
= 14.6 m

On this occasion, however, the reading was taken at eye level, so H is not the
height of the tree. Assuming that it is 1.7 m from the ground to the man’s eyes
and that the ground is horizontal, the height of the tree is 1.7 m more than H,
i.e. the height of the tree is 16 m to two significant figures.

Question 6.7 asks you to use trigonometry in solving another simulated ‘real world’
problem, but Question 6.6 is given first to enable you to practise the underlying
trigonometric and algebraic skills.

Back J I 324
Contents 
Question 6.6
(a) Find length h in Figure 6.20a. Answer
(b) Find length a in Figure 6.20b. Answer
(c) Find angle θ in Figure 6.20c, giving your answer in degrees. Answer

Figure 6.20: Right-angled triangles for use in Question 6.6 (not drawn to scale).

Back J I 325
Contents 
Question 6.7 Answer
A theodolite of height 1.5 m is positioned with its base at sea-level somewhere
in the Cambridgeshire Fens, and indicates that the top of Ely Cathedral’s West
Tower is at an inclination of 2.27◦ (see Figure 6.21). The base of Ely Cathedral
is 15 m above sea-level and the West Tower is 66 m tall. Approximately how
far is the theodolite from Ely Cathedral?
Hint: start by finding H, the vertical distance between the top of the theodolite
and the top of the West Tower.

66m
H
theodolite

1.5 m
θ 15m

Figure 6.21: Using trigonometry to find distance (not to scale). θ = 2.27◦ .

Back J I 326
Contents 
In addition to providing a way of finding unknown lengths and angles, trigonomet-
ric ratios appear from time to time in scientific equations. You are not expected to
remember these equations or to understand the background science; brief explana-
tions are provided in Boxes 6.3–6.6 for interest only.

Box 6.3 Angle of dip and true thickness of strata

Folding and tilting of layers of rocks, caused by pressures within the Earth,
have resulted in many layers lying at an angle to the Earth’s surface. This angle
is called the angle of dip and is illustrated in Figure 6.22. The angle of dip can
usually be measured, as can the apparent width of a stratum (layer) at the Earth’s
surface — its outcrop, but it is the true thickness of the stratum which is of real
interest to geologists.
The vertical thickness of the stratum (in Figure 6.23) may also be of interest,
especially when exploring for underground resources (such as oil) by drilling.

Back J I 327
Contents 
Question
Express sin θ in Figure 6.22 in terms of T and W. Hence find an equation for
the true thickness, T , of a stratum in terms of the width, W, of the outcrop at the
Earth’s surface, and the angle of dip, θ.
Answer
T
sin θ =
W
so

T = W sin θ (6.9)

ground
W surface
θ
angle
of dip T

Figure 6.22: The relationship between the angle of dip, θ, width of outcrop, W,
and true thickness, T , for a tilting stratum of rock (shown in darker brown).

Back J I 328
Contents 
Question
Express tan θ in Figure 6.23 in terms of V and W. Hence find an equation for
the vertical thickness, V, of a stratum in terms of the width, W, of the outcrop at
the Earth’s surface, and the angle of dip, θ.
Answer
V
tan θ =
W
so

V = W tan θ (6.10)

ground
W surface
θ

Figure 6.23: The relationship between θ, W, and the vertical thickness, V, of the
stratum.

Back J I 329
Contents 
Worked example 6.4
Suppose a stratum of rock, lying at an angle of dip of 28◦ , has an outcrop of
width of 71 m at the Earth’s surface. What is its true thickness?
Answer
From Equation 6.9, T = W sin θ where W = 71 m and θ = 28◦ , so

T = 71 m × sin 28◦ = 33 m to two significant figures.

The true thickness of the layer is 33 metres.

Question 6.8 Answer

What is the vertical thickness of a stratum of rock which has outcrop of width
65 m at the Earth’s surface, and an angle of dip of 36◦ ?
Hint: you should use an appropriate equation from Box 6.3.

Back J I 330
Contents 
Box 6.4 Using trigonometry to determine the radius of ions

The crystal structure of lithium iodide consists of lithium

and iodide ions (ions are atoms with electric charge due
to the loss or gain of electrons), as shown in Figure 6.24. r

Both types of ions can be represented by spheres and, in 45°

one model, the spheres can be considered just to touch each h
other. This enables us to use trigonometry to find the radius
of the ions.
If the distance between the centre of a lithium ion and the
centre of an iodide ion is known (this is the so-called inter-
nuclear distance, and is labelled as h on Figure 6.24) then
adj r
cos 45◦ = =
hyp h
Figure 6.24: Using trigonometry
where r is the radius of a lithium ion. to find the radius of lithium ions
(shown in green). The purple
Multiplying both sides by h gives sphere represents an iodide ion.
r = h cos 45◦ (6.11)

Equation 6.11 can be used to find the radius of a lithium ion.

Back J I 331
Contents 
Question 6.9 Answer
The internuclear distance, h, between the ions shown in Figure 6.24 is measured
to be 302 pm (where 1 pm = 10−12 m, as defined in Box 2.2).
Use Equation 6.11 to find the radius of a lithium ion.

Back J I 332
Contents 
Box 6.5 Snell’s law for seismic waves and light
µ
r
v1
You may have realized, from the equation for S wave speed, vs =
ρ
(much used in Chapters 3 and 4), that waves travel at different speeds i
in different substances. When a wave moves from one substance into
another in which it travels at a different speed, the change in speed boundary v2
will cause the wave to change direction. This behaviour is known as r
refraction and it is illustrated in Figure 6.25.
Snell’s law of refraction states that
sin i v1
= (6.12)
sin r v2 Figure 6.25: A wave undergoing
refraction on passing through a
where v1 is the speed of the wave in the first substance, v2 is the
boundary between two media in
speed of the wave in the second substance, and i and r are the angles
which the speeds of propagation,
of incidence and refraction respectively, as illustrated in Figure 6.25.
v1 and v2 , are different (in this
Refraction occurs for all types of waves, for example, seismic waves case v1 > v2 ); i is called the an-
passing from one rock type to another in the Earth, or a beam of light gle of incidence and r is called
passing from air to glass, and Snell’s Law is true whatever type of the angle of refraction.
wave motion is being considered. The law is named after the Dutch
scientist Willebrord Snel (1596–1650) but the law was stated very
much earlier, by the mathematician Abu Said al-Ala Ibn Sahl in his
book On the Burning Instruments, written in about 984.

Back J I 333
Contents 
Worked example 6.5
Calculate the angle of refraction of a seismic wave which has an angle of inci-
dence of 35◦ on crossing a boundary from rock 1 (with v1 = 6.3 × 103 m s−1 )
into rock 2 (with v2 = 8.2 × 103 m s−1 ).
Answer
We know v1 = 6.3 × 103 m s−1 , v2 = 8.2 × 103 m s−1 and i = 35◦ , and we want
r. Snell’s law states that
sin i v1
=
sin r v2

Multiplying both sides by sin r gives

v1
sin i = × sin r
v2

Reversing the equation and multiplying both sides by v2 gives

v1 sin r = v2 sin i

Dividing both sides by v1 gives

v2 sin i
sin r =
v1

Back J I 334
Contents 
Substituting gives

8.2 × 103 m s−1 × sin 35◦

sin r =
6.3 × 103 m s−1
= 0.7466

So

r = sin−1 (0.7446)
= 48◦ to 2 significant figures.

Note that in this case the angle of refraction is greater than the angle of inci-
dence. This is because v2 is greater than v1 .

Question 6.10 Answer

A beam of light strikes an air–glass interface with an angle of incidence of 45.0◦
and the angle of refraction (in the glass) is found to be 26.3◦ . The speed of light
in air is 3.00 × 108 m s−1 . Use Snell’s law to find the speed of light in glass.

Back J I 335
Contents 
Box 6.6 Using a diffraction grating
A diffraction grating is simply a series of ex- diffraction screen
tremely narrow, evenly spaced slits through grating second order

which light can pass. When a light beam of a

single colour (i.e. a single wavelength) hits the
diffraction grating at an angle of 90◦ , as shown first order
in Figure 6.26, the grating acts in such a way as θ2
to split up the incoming beam, forming what is light beam θ1
zero order
called a diffraction pattern. Some light passes θ1
straight through the grating; this is called the θ2
zero-order beam. Other beams are produced at first order
angles θ1 , θ2 , etc. from the straight-through po-
sition and are known as the 1st, 2nd, etc. order
diffracted beams.
second order
The angle θn of the nth order beam is given by
the equation
nλ Figure 6.26: The pattern formed by a diffraction grat-
sin θn = (6.13)
d ing.
where λ is the wavelength of the light and d is
the grating spacing (i.e. the distance between
two adjacent slits in the grating).

Back J I 336
Contents 
Worked example 6.6
A beam of light of wavelength 5.89 × 10−7 m passes through a diffraction grat-
ing and the second-order diffracted beam is at θ2 = 45.9◦ . Find the grating
spacing d.
Answer
In this case λ = 5.89 × 10−7 m, n = 2, θn = 45.9◦ .
Multiplying both sides of Equation 6.13 by d gives

d sin θn = nλ

Dividing both sides by sin θn gives

nλ
d=
sin θn
2 × 5.89 × 10−7 m
=
sin 45.9◦
= 1.64 × 10−6 m

So the grating spacing is 1.64 × 10−6 m.

Back J I 337
Contents 
Question 6.11 Answer
Light of a different colour (i.e. a different wavelength) passes through the same
diffraction grating as in Worked example 6.6 (so d = 1.64 × 10−6 m). The first-
order diffracted beam is at θn = 24.1◦ . Find the wavelength, λ, of this light.

Appendix A, at the back of this book, considers a further application of trigonom-

etry in science; its use when dealing with vector quantities such as velocity and
force. You may find the material useful if you intend to study physics courses in the
future.

6.4 Small angle approximations

When the angle under consideration is small, some useful approximations can be
employed.

Question
Use your calculator to find sin θ, tan θ and cos θ (each to five significant figures)
for θ = 0.5◦ .
Answer
sin 0.5◦ = 8.7265 × 10−3 , tan 0.5◦ = 8.7269 × 10−3 and cos 0.5◦ = 0.99996.

Back J I 338
Contents 
Question
Convert 0.5◦ to radians, again giving your answer to five significant figures.
Answer

360◦ = 2π
2π
so 1◦ =
360
2π
0.5◦ = 0.5 ×
360
= 8.7266 × 10−3 radians

Comparing the answers to the above questions shows that sin θ ≈ θ and tan θ ≈ θ,
when θ is measured in radians, and also that cos θ ≈ 1. These results are true for all
small angles, in other words

For all small angles

cos θ ≈ 1

For small angles stated in radians,

sin θ ≈ θ and tan θ ≈ θ

Back J I 339
Contents 
These ‘small angle approximations’ hold within 0.5% accuracy for angles less that
about 0.1 radians (6◦ ). Remember though that the final two approximations are only
valid for angles measured in radians.
Small angle approximations arise from the fact that, when θ is small in a triangle
like the one shown in Figure 6.27, h ≈ a and also the length, b, of the straight side
opposite to θ approximates to the length of an arc subtended by θ in a circle with its
centre at point P and radius h or a. In other words (on Figure 6.27)

b ≈ sh (6.14)
b ≈ sa (6.15)

arc radius h
arc radius a
h
b
θ
P a

sa sh

Figure 6.27: A right-angled triangle with a small angle θ

Back J I 340
Contents 
From trigonometry
opp b opp b
sin θ = = and tan θ = =
hyp h adj a
For small angle θ we can substitute from Equations 6.14 and 6.15 to give
sh sa
sin θ ≈ and tan θ ≈
h a
From the definitions of a radian (Equation 6.1)
sh sa
= =θ
h a
So

sin θ ≈ θ and tan θ ≈ θ

Also,
adj a
cos θ = =
hyp h
so when h ≈ a (i.e. for small angles)

cos θ ≈ 1

Back J I 341
Contents 

Figure 6.28: Calculating the Moon’s diameter (not to scale).

Small angle approximations are useful in astronomy, because objects at a great

distance subtend a very small angle when observed from the Earth.
An arc drawn from the Earth and encompassing a distant object such as the Moon
(see Figure 6.28) has a very similar curvature to a line drawn from one side of the
Moon to the other (which is the Moon’s diameter) and the distance to the centre
of the Moon is approximately equal to the radius of this arc. This gives us a way
of calculating the Moon’s diameter, the second of the problems raised at the begin-
ning of the chapter. Methodology of the sort illustrated in Worked example 6.7 is
frequently used when the size or distance to a distant object is required.

Back J I 342
Contents 
Worked example 6.7
The Moon subtends an angle φ of 9.06×10−3 radians (from Worked example 6.2)
and the distance to the Moon, L, is 3.84 × 108 mm. Find the Moon’s diameter.
Answer
From the definition of the radian (Equation 6.1) and with angles and lengths as
shown in Figure 6.28
s
φ=
r
In this case s ≈ D and r ≈ L so
D
φ≈
L
Reversing this equation and multiplying both sides by L gives

D≈ Lφ
≈ 3.84 × 108 m × 9.06 × 10−3

(remembering from Section 6.1 that strictly speaking, an angle measured in ra-
dians can be written without units).
This gives D ≈ 3.48 × 106 m, i.e. the Moon’s diameter is 3.48 × 106 m.

Back J I 343
Contents 
Question 6.12 Answer
A man standing on a beach observes that a passing car ferry subtends an angle
of 3.5◦ . The ferry is 86 m long. How far is it from the ferry to the man? Assume
that the ferry is perpendicular to the direction in which it is being observed, as
shown in Figure 6.29.

3.5° car ferry 86m

observer

Figure 6.29: A car ferry observed from a beach (not to scale).

Back J I 344
Contents 
6.5 Learning outcomes for Chapter 6
After completing your work on this chapter you should be able to:
6.1 demonstrate understanding of the terms emboldened in the text;
6.2 use degrees or radians to measure angles, and convert between these two
systems of angular measure;
6.3 find an internal angle in a triangle if you have been told the other two internal
angles;
6.4 calculate the length of any side of a right-angled triangle if you have been told
the lengths of the other two sides;
6.5 use a scientific calculator to find angles from trigonometric ratios (sin, cos and
tan only), and vice versa;
6.6 use trigonometry to find unknown angles and sides in right-angled triangles;
6.7 apply small angle approximations when appropriate;
6.8 apply knowledge gained in this chapter and earlier in the course to scientific
examples involving angles and trigonometry.

Back J I 345
Contents 

Logarithms
‘Seeing there is nothing (right well-beloved Students of the Mathematics)
7
that is so troublesome to mathematical practice, nor that doth more molest
and hinder calculators, than the multiplications, divisions, square and cubi-
cal extractions of great numbers, which besides the tedious expense of time
are for the most part subject to many slippery errors, I began therefore to
consider in my mind by what certain and ready art I might remove those
hindrances.’
Thus wrote John Napier in the preface to his book Mirifici logarithmorum canonis de-
scripio in 1614 (the quote is from the English translation of 1616). Napier (1550–1617)
was a wealthy Scottish landowner and theologian, who claimed to study mathematics
only as a hobby. Despite this, he invented both logarithms (or ‘logs’ for short) and
‘Napier’s bones’ with the express purpose of making it easier to do multiplications and
divisions. Logarithms were in regular use for this purpose well into the second half of
the twentieth century.
Nowadays we have electronic calculators and computers to help with long multiplica-
tions and divisions, so you may be wondering why this course, written in the twenty-first
century, still includes a chapter on logarithms. Over the years, in addition to being an in-
valuable aid to arithmetic, logarithms have proved themselves to have many applications

Back J I 346
Contents 
and they remain widely used in these applications. For example, the pH-scale (used to
describe acidity) is based on logarithms, and the curved graph representing the variation
of activity with time for a radioactive source (see Chapter 5 Figure 5.34) can be turned
into a straight line by plotting the logarithm of activity against time. This chapter will
explain what logarithms are, and demonstrate some of their uses in modern science.

7.1 Logarithms to base 10

Henry Briggs (1561–1630), the first professor of geometry at Gresham College, Lon-
don, visited Napier in the summer of 1615 and, with Napier’s blessing, developed the
type of logarithms known as logarithms to base 10, or ‘common logarithms’.
You know, from Section 1.3.1 that, for example,

106 = 1000 000

103 = 1000
100 = 1
1
10−1 = = 0.1
10
1
10−5 = 5 = 0.000 01
10
where 10 is known as the base number.

Back J I 347
Contents 
The process of obtaining a logarithm to base 10 (usually described as ‘taking the
log to base 10’) is the inverse of raising the base 10 to a power. In each of the
above examples the logarithm to base 10 of the number on the right-hand side of
the equation is simply the power to which the 10 on the left-hand side is raised.
The logarithm to base 10 is abbreviated log10 in this course (you may also see the
abbreviation log, without a subscript, used to describe a logarithm to base 10) so,
for example,

103 = 1000 so log10 1000 = 3

10 raised to the so the logarithm to

power 3 equals 1000 base 10 of 1000 is 3

100 = 1 so log10 1 = 0

10 raised to the so the logarithm to

power 0 equals 1 base 10 of 1 is 0

10–1 = 0.1 so log10 0.1 = –1

10 raised to the so the logarithm to

power –1 equals 0.1 base 10 of 0.1 is –1

Back J I 348
Contents 
We can say, more generally:

The logarithm to base 10 of p is the power to which 10 must be raised in order

to equal p.
i.e. if p = 10 n , then log10 p = n.

The definition of a logarithm to base 10 applies√ for fractional values of n too. For
3
example, you know, from Section 1.3.4 that 10 can be written as 101/3 . This
means that
1
101/3 = 3 10 so log10 ( 3 10 ) = 3

10 raised to the so the logarithm to

1
power 13 equals 3 10 base 10 of 3 10 is 3

In fact, n could be any number; you may like to start by using your calculator to
check the following to four significant figures (use either the ‘x y ’ or ‘^’ button or,
if your calculator has one, a button marked ‘10 x ’):

100.1235 = 1.329
103.456 = 2858
10−1.234 = 0.05834

Back J I 349
Contents 
From the last of these we can say that

10–1.234 = 0.05834 so log10 0.05834 = –1.234

10 raised to the power so the logarithm to base

–1.234 equals 0.05834 10 of 0.05834 is –1.234

Question 7.1
Without further use of a calculator, write down the values of:
(a) log10 100 Answer
(b) log10 0.001 Answer
√
(c) log10 10 Answer
(d) log10 1.329 Answer

Since taking a logarithm to base 10 is the inverse of raising 10 to a power, the ‘log10 ’
or ‘log’ button on a calculator should reverse the operation of the ‘10 x ’ button. You
can use your calculator to check this for an arbitrarily chosen number, e.g. 4.8; the
‘10 x ’ button should give 63 095.734 45 and finding the logarithm to base ten of the
latter number returns the display to 4.8.

Back J I 350
Contents 
Question 7.2
Use your calculator to find the following to 4 significant figures:
(a) log10 2 Answer
(b) log10 2000 Answer

Question 7.3
(a) Use your calculator to find 101.5 to 4 significant figures. Answer
(b) If log10 p = 1.5, what is p? Answer

It is worth noting that

• it is not possible to obtain the logarithm to base 10 of a negative number, or
of zero: if you try this on your calculator it will produce an error message.
• it is possible to obtain logarithms of pure numbers only; you cannot obtain
the logarithm of a quantity possessing units. Strictly, if a quantity possesses
units, then it should be divided by those units before taking the logarithm.
You will see how this is done in practice in Box 7.1 later in this chapter.

Back J I 351
Contents 
7.2 Logarithmic scales revisited
Logarithmic scales, such as the Richter scale for earthquake magnitude and the
decibel scale for relative loudness of sounds, were discussed in Chapter 2. The
word ‘logarithmic’ is used to describe such scales simply because they are based on
logarithms; both the decibel and the Richter scales are based on logarithms to base
10. It was stated in Chapter 2 that logarithmic scales are used when the quantities
being measured vary over a wide range (see Figure 2.2); the answer to Question 7.2
illustrates why logarithms are so useful in this context. The log to base 10 of 2 is
0.3010, but the log to base 10 of 2000, a number a thousand times bigger than 2, is
just 3.3010 and it turns out that the log to base 10 of 2000 000 is only 6.3010. Thus
taking logarithms gives us a way of coping with a scale that covers a huge range of
values.
As a more specific example of this, consider the decibel. This unit was introduced in
Chapter 2, but now it can be defined properly. The loudness of a sound in decibels,
relative to a threshold value (a sound which is just audible) is given by:
!
I
relative intensity in decibels = 10 × log10
I0

where I0 is the intensity of the threshold sound and I is the intensity of the sound in
question. So the sound of a pneumatic drill with an intensity 1012 times that of the
threshold has:
 
relative intensity in decibels = 10 × log10 1012 = 10 × 12 = 120

Back J I 352
Contents 
The pH scale, widely used as a measure of acidity, is also based on logarithms to
base 10. The pH scale is discussed further in Box 7.1.

Box 7.1 The pH scale

The pH scale was developed by the Danish biochemist Søren Sørenson in 1909.
‘pH’ is an abbreviation for ‘pondis hydrogenii’ or ‘potential of hydrogen’ and
the scale is based on a measurement of the concentration of hydrogen ions in
the solution in question. Concentration and its units, mol dm−3 , were introduced
in Box 3.2, and a hydrogen ion is a hydrogen atom which has lost an electron
and so is positively charged. The hydrogen ion concentration of pure water at
25 ◦ C is 1 × 10−7 mol dm−3 , whilst that of lemon juice (more acidic than pure
water) is about 8 × 10−3 mol dm−3 and that of household bleach (considerably
less acidic than pure water) is about 1 × 10−12 mol dm−3 . Note that the range of
values is very wide and also that all of the values are quite small, which makes
them rather tricky to deal with. The definition of pH (which handles both of
these things) is:

hydrogen ion concentration in mol dm−3

!
pH = − log10
mol dm−3

Since the hydrogen ion concentration is measured in units of mol dm−3 and we
divide by mol dm−3 before taking the logarithm, we are obtaining the logarithm
of a pure number, as required.

Back J I 353
Contents 
From the definition of pH, the pH of pure water is:

1 × 10−7 mol dm−3

!  
− log10 = − log 10 10 −7
= −(−7) = 7
mol dm−3
the pH of lemon juice is:

8 × 10−3 mol dm−3

!  
− log10 = − log10 8 × 10−3 = −(−2.1) = 2.1
mol dm−3
and the pH of household bleach is:

1 × 10−12 mol dm−3

!  
− log10 = − log 10 10 −12
= −(−12) = 12
mol dm−3
Thus we have a much more manageable scale. The entire range of values for
hydrogen ion concentration, from 1.0 mol dm−3 down to 1 × 10−14 mol dm−3 ,
is represented by pH values between 0 and 14. A pH of 7 (the value for pure
water) representing a neutral solution, with lower numbers being more acidic
and higher numbers being less acidic.

Back J I 354
Contents 
Question 7.4
Calculate, to two significant figures, the pH of the following:
(a) human blood, with a hydrogen ion concentration Answer
of 4.0 × 10−8 mol dm−3
(b) hair shampoo, with a hydrogen ion concentration Answer
of 3.2 × 10−6 mol dm−3 .

7.3 Rules of logarithms

Much of the usefulness of logarithms follows from several rules which are summa-
rized below:

log10 10 n = n (7.1)
log10 (p × q) = log10 p + log10 q (7.2)
!
p
log10 = log10 p − log10 q (7.3)
q
log10 p n = n log10 p


(7.4)

Back J I 355
Contents 
Equation 7.1 is a restatement of the definition of a logarithm to base ten. The other
rules can be derived from the rules for manipulating exponents, given in Chapter 1.
The derivation of Equation 7.2 is given in Box 7.2 for your interest (the derivations
of Equations 7.3 and 7.4 are similar).

Box 7.2 Deriving Equation 7.2

From the definition of a logarithm to base 10:

If p = 10 a then log10 p = a (7.5)

If q = 10 then log10 q = b
b
(7.6)

Multiplying p and q gives:

p × q = 10 a × 10 b = 10 a+b

from the rules for exponents given in Section 1.3.2.

Taking the logarithm to base 10 of both sides:
 
log10 (p × q) = log10 10 a+b
=a+b (from Equation 7.1)

But a = log10 p from Equation 7.5 and b = log10 q from Equation 7.6 so

log10 (p × q) = log10 p + log10 q

Back J I 356
Contents 
We can verify Equations 7.2, 7.3 and 7.4 by substituting numerical values for p, q
and n.
If p = 2 and q = 1000, then from Equation 7.2,

log10 (2 × 1000) = log10 2 + log10 1000

= 0.3010 + 3
= 3.3010

To five significant figures, this is the same value as a calculator gives for log10 2000
(as obtained in the answer to Question 7.2), so Equation 7.2 seems reasonable. Note
that log10 2000 is exactly 3 more than log10 2.
Again using p = 2 and q = 1000, now in Equation 7.3,
!
2
log10 = log10 2 − log10 1000
1000
= 0.3010 − 3
= −2.6990

To five significant figures, this is the same value as a calculator gives for log10 0.002,
so Equation 7.3 seems reasonable. Note that log10 0.002 is exactly 3 less than
log10 2.

Back J I 357
Contents 
If p = 2 and n = 3, then from Equation 7.4,
 
log10 23 = 3 log10 2
= 3 × 0.3010
= 0.9030

A calculator gives log10 8 = 0.9031 to four significant figures, almost but not ex-
actly the same as the value obtained for log10 (23 ) by using Equation 7.4. directly.
Equation 7.4 seems reasonable. If we had used exact values for log10 2 and log10 8
the answers would have been identical, but in working to a limited number of sig-
nificant figures we need to take care with rounding errors.

Worked example 7.1

Use the fact that log10 3 = 0.4771 to obtain a value for log10 3000 without using
a calculator. You should give your answer to four significant figures.
Answer
log10 3000 = log10 (3 × 1000)
= log10 3 + log10 1000 (from Equation 7.2)
= 0.4771 + log10 10 3

= 0.4771 + 3 (from Equation 7.1)

= 3.477 to four significant figures

Back J I 358
Contents 
Question 7.5
Use the fact that log10 3 = 0.4771, and Equations 7.1 to 7.4 to find the following
without using a calculator. Give your answers to four significant figures.
(a) log10 300, Answer
(b) log10 0.03, Answer
(c) log10 9. {Hint: remember that 9 = 32 .} Answer

These rules for the manipulation of logarithms explain how Napier’s invention was
used to simplify the processes of multiplication and division. Equation 7.2 gives
a way of turning multiplication into addition; Equation 7.3 gives a way of turning
division into subtraction and Equation 7.4 gives a way of calculating powers and
roots. The rules of logarithms have other uses too, as illustrated in Box 7.3 on the
next page.

Back J I 359
Contents 
Box 7.3 k-value analysis
k-value analysis provides a methodology for studying the different factors that
affect the size of a biological population. Consider, for example, a population
of 24 pairs of owls studied by H. N. Southern in Wytham Wood, near Oxford,
in 1952–1953. In order for a pair of owls to have young which themselves
will breed, various things must happen: for example, the parents must breed;
the eggs must hatch; they must produce fledglings that survive to be owlets; the
owlets must live long enough to form pairs. Things can go wrong at every stage!
The k-value (which you can think of as the ‘killing factor’) is a measure of the
killing power of each of the things that can go wrong.
At each stage:
!
NB
k = log10
NA

where NB is the number of individuals alive before this stage and NA is the
number of individuals alive afterwards.
For example, 43 eggs were laid (N2 in Table 7.1) but only 16 eggs hatched (N3
in Table 7.1) so the k-value for this stage is:
! !
N2 43
k3 = log10 = log10 = log10 (2.6875) = 0.4293
N3 16

Back J I 360
Contents 
Question
!
N4
Use the data in Table 7.1 to find k5 = log10
N5
Answer !
N4
k5 = log10
N5
!
15
= log10
9
= log10 (1.6667)
= 0.2218

k-value analysis gives an easy way of comparing the effect of different killing
factors and ktotal , the total k-value for all stages is
! !
N0 72
ktotal = log10 = log10 = 0.9031
N5 9
Question
Use the data given in Table 7.1 to find k1 + k2 + k3 + k4 + k5 .
Answer
k1 + k2 + k3 + k4 + k5 = 0.1498 + 0.0741 + 0.4293 + 0.0280 + 0.2218
= 0.9030

Back J I 361
Contents 
Note that, within rounding errors, this is the same as the value calculated for
ktotal from
!
N0
ktotal = log10
N5

The result ktotal = k1 + k2 + k3 + k4 + k5 can also be proved from the rules of

logarithms:
!
N0
k1 = log10 = log10 N0 − log10 N1 (from Equation 7.3)
N1

Similarly
!
N1
k2 = log10 = log10 N1 − log10 N2
N2

and so on, until

!
N4
k5 = log10 = log10 N4 − log10 N5
N5
So

k1 + k2 + k3 + k4 + k5 = (log10 N0 − log10 N1 ) + (log10 N1 − log10 N2 ) + . . .

+ (log10 N4 − log10 N5 )

Back J I 362
Contents 
Apart from log10 N 0 and log10 N 5 , all of the logarithms on the right-hand side
are both added and subtracted, so

k1 + k2 + k3 + k4 + k5 = log10 N0 − log10 N5
!
N0
= log10
N5
= ktotal

Back J I 363
Contents 
7.4 Using logarithms to make curves straight
You were introduced, in Chapter 5, to various graphs of dif-
ferent shapes. For example a graph of the area A of a circle
against its radius r is a parabola; the equation of this graph is A
A = π r 2 . Similarly, the graph of the number of radioactive
nuclei N against elapsed time t is an exponential; the equation
of this graph is N = N0 e−λt . Logarithms can be used to turn
these and other curved graphs into straight-line graphs, and a
knowledge of the rules of logarithms (from Section 7.3) can be
used to interpret the resulting straight-line graphs and thus to (a) 0 r
determine physical constants such as N0 and λ.
log10˚A

7.4.1 Log–log graphs

Figure 7.1a shows a graph of A against r for the equation

A = π r 2 . One method for turning this curve into a straight
line was introduced in Section 5.4. Another method is to plot 0 log10˚r
(b)
log10 A against log10 r; as shown in Figure 7.1b this also gives
a straight line. A graph of this type is known as a ‘log–log
graph’. But why should it be a straight line? Figure 7.1: Graphs of (a) A against
r, and (b) log10 A against log10 r for
the equation A = π r 2 .

Back J I 364
Contents 
Taking the log to base 10 of both sides of the equation A = π r 2 gives:
 
log10 A = log10 π r 2
= log10 π + log10 r 2 (from Equation 7.2)
= log10 π + 2 log10 r (from Equation 7.4)

We can reverse the order of the two terms on the right-hand side to give:

log10 A = 2 log10 r + log10 π

This can be compared with the general equation of a straight-line graph, y = mx + c

(Chapter 5 Equation 5.1)

log10 A = 2 log10 r + log10 π

y = m x + c

gradient intercept on
the vertical axis

This comparison implies that a graph of log10 A against log10 r should be a straight
line of gradient 2 and intercept on the vertical axis of log10 π.

Back J I 365
Contents 
Figure 7.2 is an accurately plotted graph of
log10 (A/cm2 ) against log10 (r/cm) for the data in 2.6
Table 7.2.
2.4

Question 7.6 Answer 2.2

Find the gradient and intercept on the vertical 2.0

axis of the straight line shown in Figure 7.2.
1.8

1.6

log10(A/cm )
2
1.4
 
r/cm A/cm2 log10 (r/cm) log10 A/cm2
1.2
1 π 0 0.497
2 4π 0.301 1.099 1.0
3 9π 0.477 1.451
0.8
4 16π 0.602 1.701
5 25π 0.699 1.895 0.6

0.4
Table 7.2: The radius and area of various circles,
and corresponding logarithms to base 10 0.2

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
log10(r/cm)

Figure 7.2: A graph of log10 (A/cm2 ) against log10 (r/cm).

Back J I 366
Contents 
Worked example 7.2
If a graph is plotted of log10 y against log10 x for the equation y = 3x−2 , what
will be the gradient and the intercept on the vertical axis?
Answer
Taking the log to base 10 of both sides of the equation y = 3x−2 gives
 
log10 y = log10 3x−2
= log10 3 + log10 x−2 (from Equation 7.2)
= log10 3 − 2 log10 x (from Equation 7.4)

We can reverse the order of the two terms on the right-hand side to give

log10 y = −2 log10 x + log10 3

Comparison with the general equation of a straight-line graph, y = mx + c,

reveals that m = −2 and c = log10 3, so the gradient of the graph will be −2 and
the intercept on the vertical axis will be log10 3.
Figure 7.3 shows graphs of y against x and log10 y against log10 x for the equa-
tion y = 3x−2 , but note that it is possible to answer Worked example 7.2 without
plotting either of these graphs.

Back J I 367
Contents 
Question 7.7 Answer
If a graph is plotted of log10 y against log10 x for the equation y = 2x3 , what will
be the gradient and the intercept on the vertical axis?

Plotting graphs of the logarithm of one quantity against the logarithm of another
quantity can be used to solve scientific mysteries, as is illustrated in Box 7.4.

Box 7.4 Kepler’s third law

The Danish astronomer Tycho Brahe (1546–1601) was a meticulous observer
and recorder of data. He developed accurate sighting devices and kept a de-
tailed record of the positions of the planets at regular intervals for more than 20
years. Tycho Brahe’s tables provided the data which enabled Johannes Kepler
(1571–1630) to work out mathematical relationships describing the motion of
the planets.
Plotting the time T it takes for a planet to orbit the Sun (known as its orbital
period) against its average distance, a, from the Sun gives a graph of the shape
shown in Figure 7.4.
There is clearly a relationship between T and a but what is it? It took Kepler
a long time to work this out, but we can use logarithms to help. Let’s start
by assuming that the relationship is of the form T = kan where k and n are
constants. The problem now is to find the value of n.

Back J I 368
Contents 
Taking the log to base 10 of both sides of the equation T = kan gives:
log10 T = log10 kan

= log10 k + log10 an (from Equation 7.2)

= log10 k + n log10 a (from Equation 7.4)
We can reverse the order of the two terms on the right-hand side to give:
log10 T = n log10 a + log10 k
Comparison with the general equation of a straight-line graph, y = mx+c, shows
that a graph of log10 T against log10 a will be a straight line with gradient n and
intercept log10 k.
Figure 7.5 shows the same data as Figure 7.4, but now the logarithms of the
variables have been plotted.
Question
What is the gradient of the line in Figure 7.5?
Answer
1.5 − 0.0
gradient = = 1.5
1.0 − 0.0
This means that T = ka1.5 , i.e. T = ka3/2 . Squaring both sides gives T 2 = k2 a3 ,
so the square of the time it takes for a planet to orbit the Sun is proportional to
the cube of its average distance from the Sun, i.e. T 2 ∝ a3 . This is Kepler’s
third law.

Back J I 369
Contents 
7.4.2 Log–linear graphs

We can turn graphs of equations such as N = N0 e−λt into straight-line graphs using
a similar methodology to the one employed in Section 7.4.1, but now we plot the
logarithm of one variable against the other variable itself (not its logarithm). The
resultant graph is known as a ‘log–linear graph’. Figure 7.6 shows graphs of N
against t and log10 N against t for the equation N = N0 e−λt . Note that the graph of
N against t is a curve, but that the log–linear graph of log10 N against t is a straight
line.

N log10˚N

0 t 0
(a) (b)

Figure 7.6: Graphs of (a) N against t and (b) log10 N against t for the equation
N = N0 e−λt .

Back J I 370
Contents 
To show why the graph of log10 N against t is a straight line we need to start by
taking the log to base 10 of both sides of the equation N = N0 e−λt . This gives:
 
log10 N = log10 N0 e−λt
= log10 N0 + log10 e−λt (from Equation 7.2)
= log10 N0 − λ log10 e (from Equation 7.4)

We can reverse the order of the two terms on the right-hand side to give:

log10 N = −λt log10 e + log10 N0

= −λ log10 e t + log10 N0

This can be compared with the general equation of a straight-line graph, y = mx + c

log10 N = –λlog10 e t + log10 N0

y = m x + c

gradient intercept on
the vertical axis

So a graph of log10 N against t will be a straight line with a gradient of −λ log10 e

and an intercept on the vertical axis of log10 N0 . Note that the gradient of Figure
7.6b is negative, as expected.

Back J I 371
Contents 
Question 7.8 Answer
If a graph is plotted of log10 n against t for the equation n = n0 e at (Chapter 5
Equation 5.3; note that n0 and a are positive constants), what will be the gradient
and intercept on the vertical axis?

Graphs of log10 y against log10 x and log10 y against x are plotted so frequently
(though perhaps rather less frequently now than they were in the past, because of
computer graph-plotting programs) that special graph paper is available for the pur-
pose. ‘Log–linear’ (or ‘semi-log’) graph paper has divisions corresponding to log10
on the vertical axis only, so is useful for plotting graphs of log10 y against x.
Figure 7.7 illustrates the use of log–linear graph paper in investigating the variation
of log10 N with t for real experimental data, in this case in an experiment to find the
half-life of the excited state of barium-137.
‘Log–log graph paper’ has divisions corresponding to log10 on both axes, so is
useful for plotting graphs of log10 y against log10 x.

Back J I 372
Contents 
7.5 Logarithms to base e
The previous sections of this chapter have considered logarithms based on powers
of 10. It is possible to use numbers other than 10 as the base for logarithms and the
other base which is widely used in science is ‘e’, the ‘special number’ introduced
in Chapter 5.
In much the same way as taking the logarithm to base 10 is the inverse of raising
10 to a power, so taking a logarithm to base e (abbreviated ln or loge ) is the inverse
of raising e to a power.

The logarithm to base e of p is the power to which e must be raised in order to

equal p,
i.e. if p = e q then ln p = q.

A logarithm to base e is often referred to as a ‘natural logarithm’ and the ‘n’ in the
abbreviation ‘ln’ can be thought of as a reminder of this.
Check that you can use your calculator to raise e to various powers. You are likely
to be using a button labelled ‘e x ’ in order to do this; the ‘EXP’ button has a totally
different use. There is a need to take particular care over the meaning of ‘e’, ‘EXP’
and ‘exp’ since ‘exp’ is sometimes used to mean ‘e to the power’, so N = N0 e −λt
is sometimes written as N = N0 exp(−λt) and n = n0 eat is sometimes written as
n = n0 exp(at).

Back J I 373
Contents 
Check that you can get the following results (to four significant figures):

e 3 = 20.09
e 0.6931 = 2.000
e −1 = 0.3679

Then we can say:

e3 = 20.09 so ln 20.09 = 3

e raised to the so the logarithm to

power 3 equals 20.09 base e of 20.09 is 3

e0.6931 = 2 so ln 2 = 0.6931

e raised to the so the logarithm to

power 0.6931 equals 2 base e of 2 is 0.6931

e–1 = 0.3679 so ln 0.3679 = –1

e raised to the so the logarithm to

power –1 equals 0.3679 base e of 0.3679 is –1

Back J I 374
Contents 
Since taking a logarithm to base e is the inverse of raising e to a power, the ‘ln’ or
‘loge ’ button on a calculator should reverse the operation of the ‘e x ’ button. You
can use your calculator to check this for an arbitrarily chosen number, e.g. 1.4; the
‘e x ’ button should give 4.055 199 967 and finding the logarithm to base e of the
latter number returns the display to 1.4.

Question 7.9
Use your calculator to find the following to four significant figures:
(a) ln 4, Answer
(b) the number whose natural logarithm is 4. Answer

Note that the rules of logarithms, discussed in Section 7.3, apply to logarithms in
any base, not just those to base 10. In particular, they apply to logs to base e too,
so:

ln e n = n (7.7)
ln (p × q) = ln p + ln q (7.8)
!
p
ln = ln p − ln q (7.9)
q
ln p n = n ln p


(7.10)

Back J I 375
Contents 
You may be wondering why logs to base e are useful; why don’t we always use logs
to base 10? One reason why logs to base e are important stems from the fact that
taking a logarithm to base e is the inverse of raising e to a power. This means that
equations such as N = N0 e −λt can be turned into simpler straight-line equations by
taking logarithms to base e than is possible by taking logarithms to base 10.
Taking the log to base e of both sides of the equation N = N0 e −λt gives:

ln N = N = N0 e −λt
= ln N0 + ln e −λt (from Equation 7.8)
= ln N0 − λt (from Equation 7.7)

We can reverse the order of the two terms on the right-hand side to give:

ln N = −λt + ln N0

This can be compared with the general equation of a straight-line graph, y = mx + c

ln N = −λ t + ln N0
y = m x + c

gradient intercept on
the vertical axis

Back J I 376
Contents 
So a graph of ln N against t (see Figure 7.8)
will be a straight line with a gradient of −λ
lnN
and an intercept on the vertical axis of ln N0 .

Question intercept on the

vertical axis at ln N0
A graph plotted of ln N against time t
for the decay of barium-137 is a straight
line of gradient −4.4 × 10−3 s−1 . What is
the decay constant (the constant λ in the gradient of −λ
equation N = N0 e −λt )?
Answer
0
The gradient of the graph is t
−4.4 × 10−3 s−1 so the decay constant λ
is 4.4 × 10−3 s−1 .

Box 7.5 investigates the relationship be-

tween decay constant λ and half-life, t1/2 .
The half-life of a radioactive decay process
(first introduced in Chapter 5) is the time
taken for the number of radioactive nuclei, Figure 7.8: A graph of ln N against t for the equation
and hence the activity, to fall by half. N = N0 e −λt .

Back J I 377
Contents 
Box 7.5 The relationship between decay constant and half-life
The equation N = N0 e −λt can be written as:
!
1 1
N = N0 λt since e −λt
= λt
e e
Rearranging gives:
N0
eλt =
N
1
At t = t1/2 , N = N0 × (from the definition of half-life in Section 5.4) i.e.
2
N0
=2
N
So e λ t1/2 = 2
Taking the log to base e of both sides of this equation gives:
ln e λ t1/2 = ln 2
 

i.e. λ t1/2 = ln 2 (from Equation 7.7)

ln 2 ln 2
λ= or t1/2 =
t1/2 λ
Thus a decay constant of 4.4 × 10−3 s−1 (for barium-137) corresponds to a half-
ln 2
life of = 1.6 × 102 s to two significant figures.
4.4 × 10−3 s−1

Back J I 378
Contents 
Question 7.10 Answer
If a graph is plotted of ln n against t for the equation n = n0 e at , what will be the
gradient and intercept on the vertical axis? (Note that this is the same equation
as used in Question 7.8, but now you are asked to consider a graph of ln n against
t rather than a graph of log10 n against t.)

Box 7.6 The Arrhenius equation

The Arrhenius equation, named after the Swedish chemist Svante Arrhenius
(1859–1927), is one of the most important equations in chemistry. It links the
rate of a chemical reaction to the temperature at which the reaction takes place.
The equation is given by:
kR = A e (−Ea /RT ) (7.11)
where kR is the ‘rate constant’ at a particular temperature T , A is the Arrhe-
nius A-factor (or A-factor), Ea is the Arrhenius activation energy (or activation
energy) and R is the ‘gas constant’.
Taking the log to base e of both sides of Equation 7.11 gives:
ln kR = ln A e (−Ea /RT )
 

= ln A + ln e (−Ea /RT ) (from Equation 7.8)

Ea
= ln A − (from Equation 7.7)
RT

Back J I 379
Contents 
We can reverse the order of the two terms on
the right-hand side to give:
lnkR
Ea
ln kR = − + ln A intercept on the
RT vertical axis at lnA
−Ea 1
= + ln A
R T
This can be compared with the general equa-
gradient of −Ea
tion of a straight-line graph, y = mx + c R

−Ea 1
ln kR = + ln A
R T 0
y = m x + c 1
T

gradient intercept on
the vertical axis

Figure 7.9: A graph of ln kR against 1/T for the equa-

tion kR = A e (−Ea /RT ) .
So if both A and Ea are constants, independent of temperature (a reasonable
assumption for most reactions when studied over a limited range of temper-
ature), a graph of ln kR against 1/T will be a straight line of gradient −Ea /R
and intercept on the vertical axis ln A. A graph of ln kR against 1/T (as shown
in Figure 7.9) is referred to as an Arrhenius plot.

Back J I 380
Contents 
The Arrhenius equation accounts
remarkably well for the tempera-
ture behaviour of the vast major-
5.0
ity of chemical reactions, including
those which occur in nature. For
many living organisms, the temper-
4.0
ature of their environment is hugely
important, and biological processes

ln (heart rate)
are frequently temperature depen- 3.0
dent. Biological processes gener-
ally involve complex sequences of
chemical steps, yet in common with 2.0
many other composite reactions,
they often exhibit an Arrhenius-
type behaviour. Figure 7.10 shows 1.0
an Arrhenius plot for the heartbeat
of a diamond-backed terrapin. At
lower temperatures, the plot departs 0
3.3 × 10−3 3.4 × 10−3 3.5 × 10−3
from linear behaviour, indicating a T−1/K−1
different control mechanism.

Figure 7.10: An Arrhenius plot of its heartbeat (rate) in the

temperature range 18 ◦ C to 34 ◦ C.

Back J I 381
Contents 
7.6 Learning outcomes for Chapter 7
After completing your work on this chapter you should be able to:
7.1 demonstrate understanding of the terms emboldened in the text;
7.2 use a calculator to find the logarithm (to base 10 or base e) of a positive
number;
7.3 demonstrate understanding of the relationship between powers of 10 and
logarithms to base 10;
7.4 demonstrate understanding of the relationship between powers of e and
logarithms to base e;
7.5 use the rules governing the logarithms of products, fractions and powers;
7.6 interpret a graph of log10 y against log10 x for a function of the type y = a x b ;
7.7 interpret a graph of log10 y against x for a function of the type y = C e kx ;
7.8 interpret a graph of ln y against x for a function of the type y = C e kx .

Back J I 382
Contents 

Probability and descriptive statistics 8

Statistical information is a familiar aspect of modern life, which features routinely
in, for example, news reports, sports commentaries and advertising. Scientists who
have collected large amounts of data by either counting or measuring quantities also
rely on statistical techniques to help them make sense of the data. Suppose you had
information collected from, say, three thousand patients, all with the same medical
condition but undergoing a variety of treatments. First you would need techniques
for organizing and describing the data, so that you could present a summary by
giving just a few numbers. This is the function of ‘descriptive statistics’, covered
in this chapter. Then you might want to analyse the data in some way, perhaps to
decide whether it supports the suggestion that treatment with one particular drug
is more effective than other forms of medication. Chapter 9 will look at some of
the statistical tests that may be applied to raw data in order to come to objective
conclusions about what it really shows.
Statistical techniques offer ways of dealing with variability, and natural variability
is something that scientists meet all the time. Each time an experiment or a mea-
surement is repeated, a slightly different result may be obtained; in any group of
people there will be a variation in height; the count of background radiation at any

Back J I 383
Contents 
individual location fluctuates randomly from moment to moment. It is therefore
very important to be able to decide with some measure of certainty whether a par-
ticular result could have been obtained simply by chance or whether it has some real
significance, and the mathematics of chance and probability underpin all aspects of
statistics.

8.1 Chance and probability

Probability is expectation founded upon partial knowledge. A perfect
acquaintance with all the circumstances affecting the occurrence of an
event would change expectation into certainty, and leave neither room
nor demand for a theory of probabilities.’
(George Boole, 1815–1864)
In many branches of science it is not possible to predict with any certainty what
the outcome of a particular event will be. There may be several possible outcomes
and all the scientist can offer in the way of quantitative prediction is an assessment
of the relative likelihood of each of these outcomes. For example, if a man and a
woman both carry the cystic fibrosis gene without showing symptoms of the dis-
ease, the chances are 1 in 4 that their first child will suffer from the condition. Such
assessments of probability are a routine part of genetics, nuclear physics, quantum
physics and many other scientific disciplines.
In seeking to understand the nature and rules of probability it is often best to focus

Back J I 384
Contents 
initially on everyday examples that are easily visualized. So Sections 8.1.1 to 8.1.4
feature many examples of tossed coins and rolled dice. However, you will also
get the opportunity to see how these ideas are applied to some genuine scientific
problems: for example, what is the probability that two people planning to have a
child will both turn out to be carriers of the cystic fibrosis gene?

8.1.1 Calculating probability

If a process is repeated in identical fashion a very large number of times, the prob-
ability of a given outcome is defined as the fraction of the results corresponding to
that particular outcome.

number of times that outcome occurs

probability of a given outcome = (8.1)
total number of outcomes

The nature of the fraction in Equation 8.1 shows that the probability of any given
outcome cannot be smaller than 0 or larger than 1. A probability of 0 represents
impossibility, while a probability of 1 represents inevitability. The closer the prob-
ability of a given outcome is to 1, the more likely that outcome is to occur. This is
illustrated diagrammatically in Figure 8.1.
When a coin is tossed fairly, the likelihood of it landing on heads is equal to the
likelihood of it landing on tails. If it is tossed repeatedly a great many times, it will

Back J I 385
Contents 
in theory come up heads half the time: the probability of tossing heads is 12 . The
theoretical probability of tossing tails is, of course, also 12 . The sum of these two
probabilities is 21 + 12 = 1; i.e. it is certain that when the coin is tossed it will land
either on heads or on tails. This is an example of a general rule:

The sum of the probabilities of all possible outcomes is equal to 1.

A probability of 1 represents certainty.

Dice games involve rolling six-sided dice. Each face of a dice is marked with a
different score: one, two three, four, five or six. If the dice is not loaded and the
rolling is done fairly, then all outcomes are equally likely, so the probability of any
one of the six possible outcomes (for example scoring a three) is 16 . Again, the sum
of the probabilities of all the possible outcomes is 16 + 16 + 61 + 16 + 16 + 16 = 1.
So on one roll of the dice the probability of scoring a three is 61 and the probability
of not scoring a three is 65 . Another way of expressing this is to say that on a single
roll of the dice there is only one way of scoring a three, but there are 5 ways of not
scoring a three. Clearly, it is more likely than not that a number other than three
will be scored. This is just one illustration of another general rule:

The most likely outcome is the one that can occur in the greatest number of
ways.

Back J I 386
Contents 
Provided nothing biases the result to make one outcome inherently more likely
than others, the definition given by Equation 8.1 can be rewritten to encompass the
number of ways in which a particular outcome may come about:

The probability of a given outcome =

number of ways to get that particular outcome
(8.2)
total number of possible outcomes

Question
What would be the probability of throwing an odd number on one roll of a dice?
Answer
There are three possible ways of getting an odd number (1, 3 or 5) and six pos-
sible outcomes in total, so Equation 8.2 shows that the probability of throwing
an odd number is 36 , which can be simplified to the equivalent fraction 12 .
An alternative way of arriving at this conclusion is to say that as three of the
possible outcomes are even and three are odd, the chances of one throw resulting
in an odd number are the same as of it resulting in an even number. Hence the
probability of an odd number is 12 .

Back J I 387
Contents 
Question 8.1
What is the probability of one card drawn at random from a shuffled pack of
playing cards being:
(a) a heart, Answer
(b) red, Answer
(c) an ace, Answer
(d) a picture card? Answer
Note: if you are unfamiliar with playing cards, you need the following infor-
mation. There are 52 cards in a pack, divided into four suits: hearts (red),
diamonds (red), spades (black) and clubs (black). Each suit contains 13 cards,
made up of one ace, nine ‘number’ cards (from 2 to 10 inclusive) and three
picture cards (Jack, Queen, King).

8.1.2 Probability and common sense

The concept of probability is a purely theoretical one. Strictly speaking, no ex-

periment measures a probability: all that we can measure is the fraction of times
a particular outcome occurs in a finite number of attempts. In the infinitely long
run this fraction is expected to approach the theoretical probability, but in practice
we may never attain this limit. You could easily toss a fair coin four times and get

Back J I 388
Contents 
four heads. You could even toss it 20 times and still get heads on every single toss,
though that would be fairly unlikely. But the more tosses you made the more nearly
number of heads
the fraction would approach its theoretical value of 12 .
total number of tosses
A failure to appreciate the fact that the number of attempts needs to be extremely
large before the probability of a particular outcome will reliably approach the theo-
retical value is at the root of many popular misconceptions about probabilities. One
commonly held fallacy about coin tossing is that if the first ten tosses of a coin have
produced several more heads than tails, then the eleventh toss is more likely than
not to come up tails. This is not true. Although in the extremely long run the im-
balance between heads and tails is expected to be negligible, on any one toss heads
and tails are equally likely, irrespective of previous history. Coins have no memory!

Question 8.2
(a) You toss a single coin three times. It comes down heads on the Answer
first two occasions. What is the probability that you will get
heads on the third throw?
(b) If you toss two coins simultaneously and they both come down Answer
heads, what is the probability that when you then toss a third
coin it will also come down heads?

Back J I 389
Contents 
8.1.3 Expressing probability

According to Equation 8.1, probability is defined as a fraction. However, as you

know from Chapter 1, a fraction such as 14 may also be expressed as a decimal
number or as a percentage:
1
= 0.25 = 25%
4
The following statements:
• the probability of event A is 14 ,
• the probability of A occurring is 0.25,
• there is a 25% probability of the outcome being A,
are therefore all equivalent.
In addition, particularly in spoken language, it is common to say,
• there is a 1 in 4 probability of the outcome being A
and that too is equivalent to the other three statements.
For the rest of this chapter, probabilities will usually be expressed as fractions, but
you will meet the other notations in Chapter 9.

Back J I 390
Contents 
8.1.4 Combining probabilities

The probabilities described in Sections 8.1.1 and 8.1.2 related to the

outcomes of a single process, such as repeatedly tossing one coin. Now 1 H H H
suppose you were to toss three separate coins simultaneously. What is 2 H H T
the probability that all three will show heads? One way of tackling this 3 H T H
problem is to write all the possible combinations of results. There are 4 H T T
in fact eight possible outcomes, all of which are equally likely: 5 T H H
6 T H T
Of the eight combinations, only one — shown in red — represents the 7 T T H
desired outcome of three heads. On the basis of Equation 8.2, the prob- 8 T T T
ability of all three coins coming up heads is therefore 81 .
The same result can be obtained using the ‘multiplication rule for probabilities’:
the probability that the first coin will show heads is 21 , and the same is true for the
second and third coins. The probability that all three will show heads is 21 × 12 × 21 = 18 .
Notice carefully how this situation differs from the one featured in Question 8.2:
both the scenarios described in Question 8.2 correspond to having a choice only
between outcomes 1 and 2 in the list above (because the outcome of the first two
tosses is already known as being two heads). The multiplication rule is expressed
in its most general form by saying that

If a number of outcomes occur independently of one another, the probability of

them all happening together is found by multiplying their individual probabili-
ties.

Back J I 391
Contents 
An example of how this rule applies in a common genetic disease is given in Box
8.1.

Box 8.1 Probability and cystic fibrosis

Cystic fibrosis is the most common genetic disease in white European and Amer-
ican populations. It results from one of a number of mutations (errors) in a sin-
gle gene that codes for a protein involved in the transport of salts in the cells
of the body. A person with cystic fibrosis has numerous symptoms including
sticky mucus in the lungs which makes them prone to infections, abnormally
salty sweat and problems with the digestion of food. The cystic fibrosis (CF)
gene is described as recessive, which means that individuals with only one copy
of the gene, so-called ‘carriers’, show no symptoms of the disease and may be
unaware that they carry the gene. Individuals with two copies of the faulty CF
gene will show the symptoms of the condition.
1
Among white Europeans, the probability of being a carrier is 25 .
For a child whose parents are both carriers, the probability of inheriting a copy
of the CF gene from both parents is 14 . This is therefore the probability that the
child of such parents will have symptoms of the disease.
Question
Assuming that the gene is equally likely to be carried by men and women, what
is the probability that any two people planning to have a child together would
both be carriers?

Back J I 392
Contents 
Answer
The probability of both partners being carriers is 1
25 × 1
25 = 1
625 .
Question
What is the probability of a child born to white European parents having cystic
fibrosis?
Answer
1
The probability that both parents are carriers is 625 , and the probability that
a child whose parents are both carriers will have the disease is 41 . So the
probability of a child born to white European parents having cystic fibrosis is
625 × 4 = 2500
1 1 1

(In fact the figure quoted for babies born with cystic fibrosis in the UK is about
1 in 2000, somewhat higher than this calculation would suggest.)

Question 8.3
(a) If you toss two coins at the same time, what is the probability Answer
of getting two tails?
(b) If you throw a pair of dice, what is the probability of getting a Answer
pair of sixes?

Back J I 393
Contents 
Question 8.4
Under identical conditions, a seed of each of three different species of plant A,
B, and C, has a germination probability of 12 , 31 and 14 , respectively. If we have
one of each type of seed, what is the probability that:
(a) the seed of A and the seed of B will both germinate? Answer
(b) one seed of each of the three species will germinate? Answer
(c) no seed of any of the species will germinate? (Hint: first work Answer
out the probability of non-germination for each type of seed
individually.)

Another situation in which you might need to combine probabilities occurs when
outcomes are mutually exclusive (i.e. cannot occur together). For example, what is
the probability of getting either a three or a five on a single roll of a die? One way of
working this out is to say that there are six possible outcomes altogether and two of
them correspond to the desired outcome. So from Equation 8.2, the probability of
the desired outcome is 26 = 13 . The same result can be obtained using the ‘addition
rule for probabilities’. The probability of throwing a three is 16 and the probability
of throwing a five is also 61 , so the probability of throwing either a three or a five is
6 + 6 = 6 = 3 . Again, this example illustrates a general rule:
1 1 2 1

Back J I 394
Contents 
If several possible outcomes are mutually exclusive, the probability of one or
other of these outcomes occurring is found by adding their individual probabil-
ities.

Worked example 8.1

One card is drawn from a shuffled pack of 52. What is the probability of the
card being either a heart or a diamond? (For a description of a standard pack of
cards, see the comment with Question 8.1.)
The card cannot be both a heart and a diamond: these outcomes are mutually
exclusive.
The probability of the card being a heart is 14 .
The probability of the card being a diamond is 14 .
So the probability of the card being either a heart or a diamond is:
1 1 2 1
+ = =
4 4 4 2

{Note: Since both diamonds and hearts are red suits, the question is equivalent
to asking ‘what is the probability of a single card drawn from the pack being
red?’ This was posed as Question 8.1b and answered then by a different route,
though of course the result was the same!}

Back J I 395
Contents 
Question 8.5 Answer
If you were to draw one playing card from a pack of 52, what would be the
probability of that card being either the Jack, Queen or King of diamonds?

There are also cases in which both the addition and multiplication rules operate.
For example:

Question
What is the chance that in a family of three children only one will be a boy?
Answer
Assuming that the sex of a child is independent of the sexes of its siblings, the
probability that the first child is a boy is 12 , the probability that the second is a
girl is 12 , and the probability that the third is also a girl is 12 . So the probability
of this particular combination (boy–girl–girl) is
1 1 1 1
× × =
2 2 2 8
But in a family with just one boy and two girls, the boy may be the eldest, the
middle or the youngest child, and these possibilities are mutually exclusive. So
the probability of the family consisting of a boy and two girls (born in any order)
is
1 1 1 3
+ + =
8 8 8 8

Back J I 396
Contents 
(Note that in fact the assumption that a baby is just as likely to be a boy as a girl
is not quite true. UK statistics show that for every 100 girls born, 106 boys are
born.)
As with the coin-tossing example earlier, you may find that a table of the possi-
bilities helps in visualizing the situation. Of the eight possible combinations of
three children, only three — shown in red — comprise one boy and two girls.

First child Second child Third child Boys Girls

Boy Boy Boy 3 0
Boy Boy Girl 2 1
Boy Girl Boy 2 1
Boy Girl Girl 1 2
Girl Boy Boy 2 1
Girl Boy Girl 1 2
Girl Girl Boy 1 2
Girl Girl Girl 0 3

Question 8.6 Answer

If you toss two coins simultaneously, what is the probability of getting one head
and one tail?

Back J I 397
Contents 
8.1.5 Probability ratios

Probability calculations are important in many branches of science, but nowhere more
so than in genetics. Box 8.2 describes early work in the field and provides some
illustrative data, based on plant-breeding experiments.

Box 8.2 Mendel’s peas

Gregor Mendel (1822–1884) was an Austrian monk whose experiments in breeding
the garden pea laid the foundations of the science we now call genetics. Mendel
did not know about genes in the way that they are understood today, still less about
chromosomes and DNA. The rules of inheritance he developed were based on what
he observed of the external characteristics of his plants, and the probabilities of
plants with particular characteristics arising from specific breeding crosses carried
out in the following way:
1. Mendel used pollen from a plant to fertilize the flowers of the same plant
(so called ‘self-pollination’) for several generations until he was sure he had
‘pure breeding’ plants, i.e. plants that always produced offspring identical
in appearance to themselves. He called these pure-breeding plants the P-
generation (‘P’ for parental).
2. He then took pollen from one P-generation plant and used it to fertilize an-
other P-generation plant with a different characteristic. By this process of
‘cross-pollination’ a pure-breeding purple-flowered variety could be crossed

Back J I 398
Contents 
with, for example, a pure-breeding white-flowered one. Mendel called the
offspring of this cross the F1 (first filial) generation.
3. Finally members of the F1 -generation were self-pollinated and the offspring
of this process were called the F2 (second filial) generation.
Mendel investigated seven pairs of contrasting characteristics of his pea plants.
His results relating to three of these pairs of characteristics — flower colour, seed
shape and stem length, are shown below. Mendel found these characteristics to be
independent: the fact that a particular plant had white flowers had no bearing on
whether its seeds were round or wrinkled or on what height the plant was.
Flower colour: P (purple) crossed with P (white)
F1 all purple-flowered
F2 705 purple- and 224 white-flowered
Seed shape: P (round) crossed with P (wrinkled)
F1 all seeds round
F2 651 seeds round and 207 seeds wrinkled
Stem length: P (tall) crossed with P (short)
F1 all plants tall
F2 787 tall plants and 277 short plants
Note that in the case of Mendel’s peas, the heights of the plants were not distributed
across a continuous range: there was no difficulty in deciding whether a particular
plant was ‘tall’ or ‘short’.

Back J I 399
Contents 
Before working with this data, it is important to understand how the results have
been presented. Raw data from breeding experiments come in terms of descrip-
tions and numbers, as with the examples given in Box 8.2, but results are often
reported by expressing the numbers in the form of a ratio. For example, in the
F2 -generation, Mendel obtained 705 plants with purple flowers and 224 with white
flowers. Another way of expressing this is to say that purple- and white-flowered
plants appeared in the ratio 705 : 224 (said as ‘705 to 224’).
We can think of ratios as simply another way of writing fractions. If, for instance,
we discovered from a paint chart that a green paint had been mixed from yellow
paint and blue paint in the ratio 3 : 2, we would understand that the green paint
was made up of three parts yellow paint and two parts blue paint. In other words, 53
of the mixture was yellow and 25 was blue. Adding both sides of the ratio together
has given us the denominator of the fractions. Knowing the denominator, it is then
easy to express the ratio in terms of percentages: 35 = 100
60
so 60% of the mixture is
yellow and 40% is blue. A 60 : 40 ratio is exactly the same as a 3 : 2 ratio — it is
just a matter of multiplying or dividing both sides of the ratio by 20. Sometimes it
is convenient to simplify even further, in this case by dividing both sides by two to
express the 3 : 2 ratio in the equivalent form of 1.5 : 1. Note that, like fractions,
ratios do not have units attached to them.

Back J I 400
Contents 
Ratios are quoted in many applications. For example, fertiliz- N P K Others
ers are characterized on their labelling by the ratio of two or
three major ingredients, each indicated by a letter. These let- bone meal 1 5 0 19
ters are N (for nitrogen, which is required for leaf growth), lawn tonic 4 1 0 5
P (for phosphorus, which in the form of phosphates is re- tomato food 6 5 9 80
quired for root development) and K (for potassium, which in
the form of potash is required for flowers and fruit). Typical Table 8.1: Ratios of ingredients in
ratios for three common types of fertilizer are shown in Table common fertilizers expressed as ratios
8.1. N : P : K : others

Question
What is the fraction of P in bone meal?
Answer
5 5 1
The fraction of P in the whole is = = .
1 + 5 + 19 25 5

Question
What is the percentage of N in lawn tonic?
Answer
4 4 40
The fraction of N in the whole is = = .
4 + 1 + 5 10 100
So lawn tonic contains 40% N.

Back J I 401
Contents 
Question
What is the percentage of K in tomato food?
Answer
9 9
The fraction of K in the whole is = .
6 + 5 + 9 + 80 100
So tomato food contains 9% K.

As already noted for the paint example, it is quite common for ratios to be expressed
in a form such that one of the parts is 1, even if this means that the other part is a
decimal number. Question 8.7 gives an illustration of a ratio expressed in such a
way.

Question 8.7 Answer

In the atmosphere, the ratio of the volume of oxygen to the volume of other
gases is 0.26 : 1. What percentage of the atmosphere is oxygen?

The ratio of 705 : 224 that Mendel obtained for purple- to white-flowered plants
(see Box 8.2) can be simplified by dividing both sides of the ratio by 224 to obtain
the equivalent ratio of 3.15 : 1. Notice that one side of this ratio is exact: 224
224 is
exactly equal to 1. However, the other side is not exact and a choice has to be
made about how many significant figures to quote; 2 or 3 significant figures are
usually sufficient in this context. His data relating to the other independent pairs of

Back J I 402
Contents 
characteristics involving seeds and stem lengths can be simplified in a similar way
by dividing the larger number by the smaller, to obtain:

flowers purple : white = 705 : 224 = 3.15 : 1

seeds round : wrinkled = 651 : 207 = 3.14 : 1
stems tall : short = 787 : 277 = 2.84 : 1

In each case the ratio is close to 3 : 1. In other words, the character from the P-
generation that was present in all members of the F1 -generation is present in only
about 34 of the F2 -generation. By the same token, the character that completely
vanished in the F1 -generation reappears in about 41 of the F2 -generation. In fact,
modern understanding of genetics leads to the theoretical prediction of a 3 : 1 ratio;
the slight deviations observed in experiments like Mendel’s are the same as those
observed when tossing a coin. The more tosses of the coin, the more nearly the
ratio of heads : tails approaches 1 : 1. Similarly, the more pea plants included in the
experiments, the more nearly the ratios would be expected to approach 3 : 1.
The examples of Mendel’s experiments on peas concerned the inheritance of just
a single pair of alternative characteristics: flowers were either purple or they were
white; seeds were either round or they were wrinkled; stems were either tall or they
were short. When there are more than two options for particular characteristics, the
calculations become a little more complicated, but the principles remain exactly the
same, as demonstrated by the following worked example.

Back J I 403
Contents 
Question
On a maize cob, four types of grain can be distinguished: dark smooth ones,
dark wrinkled ones, pale smooth ones and pale wrinkled ones. The aggregate
results of counting numbers of the four types on 20 cobs all from the same plant
were:
dark smooth dark wrinkled pale smooth pale wrinkled
4791 1587 1617 531

Assuming that the theoretical ratios for these characteristics are whole numbers,
what would be the theoretical probability that a single grain chosen at random
from a large number of cobs would be a pale smooth one?
Answer
Dividing through by the smallest number in the sample, which in this case is
531, gives:

dark smooth dark wrinkled pale smooth pale wrinkled

9.02 2.97 3.05 1

If it is assumed that the theoretical ratios are whole numbers, these data strongly
suggest that the ratios would be:

dark smooth dark wrinkled pale smooth pale wrinkled

9 3 3 1

Back J I 404
Contents 
The theoretical fraction of grains that are pale and smooth is therefore
3 3
=
9 + 3 + 3 + 1 16
This is also the probability of one grain selected at random
  being pale and
3
smooth. This probability could be expressed as a fraction 16 , a decimal num-
ber (0.1875) or a percentage (18.75%).

8.2 Descriptive statistics

Scientists collect many different types of information, but sets of data may be very
loosely classified into two different types. In the first type, so-called ‘repeated mea-
surement’, an individual quantity is measured a number of times. An astronomer
wanting to determine the light output of a star would take many measurements on a
number of different nights to even out the effects of the various possible fluctuations
in the atmosphere that are a cause of stars ‘twinkling’. In the second type of inves-
tigation, so-called ‘sampling’, a proportion of the members of a large group are
measured or counted. A botanist interested in the average size of Primrose plants
in a wood would try to choose representative samples of plants from different parts
of the wood and measure those.

Back J I 405
Contents 
8.2.1 Repeated measurements

Scientists are always concerned with the reliability and precision of

their data, and this is the prime reason for them to repeat measure-
ments many times. Consider the photograph in Figure 8.2, which
was produced by a diffraction grating illuminated with red light (see
Box 6.6). To determine the wavelength of the light it would be
Figure 8.2: The pattern formed by a
necessary to measure the distances between the lines. Because the
diffraction grating
lines are rather fuzzy each measurement would need to be repeated
a number of times.
Generally the process of repeating measurements of a particular quantity would
lead to a number of slightly different results being obtained. Measured values of
one quantity that are scattered over a limited range like this are said to be subject
to ‘random uncertainty’. Measurements for which the random uncertainty is small
(i.e. for which the range over which the measurements are scattered is small) are
described as precise. The ‘best’ estimate scientists could make of the distance be-
tween each line in Figure 8.2 would be some sort of average of the measured values.
The scatter inevitably associated with raw data begs various questions. For instance,
• how close to the ‘true’ value is this calculated average value?
• how close to the ‘true’ value is one typical measurement likely to be?
• conversely, how probable is it than any given measurement will be close to
the average value?

Back J I 406
Contents 
Matters will be further complicated if there is some inherent error or bias in the mea-
suring instrument, such that all the readings are, say, too large by a fixed amount.
Such measurements are said to have a ‘systematic uncertainty’. Note that unless
measuring instruments can be constantly checked against one another, it is easy for
quite large systematic uncertainties to creep unnoticed into measurements. Mea-
surements for which the systematic uncertainty is small are described as accurate.
Of course to get anywhere near to the ‘true’ value of a quantity, measurements have
to be both accurate and precise!

8.2.2 The distribution of repeated measurements Measurement Cell constant/nm

1 2.458
As noted in the previous section, if the same quantity is mea- 2 2.452
sured repeatedly, the results will generally be scattered across a 3 2.454
range of values. This is perhaps best illustrated using a real ex- 4 2.452
ample. Table 8.2 shows 10 measurements of a quantity called the 5 2.459
‘unit cell constant’ for an industrial catalyst used in the refining 6 2.455
of petrol; this is an important quantity which determines how well 7 2.464
the catalyst works, and can be measured by X-ray diffraction tech- 8 2.453
niques. Notice that the cell constant is very small and is measured 9 2.449
in nanometres. 10 2.448

Table 8.2: Repeated measurements

of the unit cell constant for a batch
of industrial catalyst

Back J I 407
Contents 

number of measurements
It is always difficult to see patterns in lists or ta- 4
bles of numbers. If the data is put into the form
of a histogram, as has been done in Figure 8.3, 3
the task becomes much easier. The histogram 2
provides a visual representation of the way in
which the measurements are distributed across 1
a range of values. In fact the pattern on Figure
8.3 is not particularly obvious, because the data 0
set is quite small, consisting of only ten mea- 2.44 2.45 2.46 2.47
cell constant/nm
surements.
Figure 8.3: Histogram of data from Table 8.2.
When the number of measurements is in-
creased, the variation in the height of the bars

number of measurements
gradually becomes smoother, as illustrated in 25
Figure 8.4. 20
15
10
5
0
2.44 2.45 2.46 2.47
cell constant/nm
Figure 8.4: Distribution of a larger number of
repeated measurements.

Back J I 408
Contents 
When substantially more measurements have

number of measurements
300
been accumulated, the size of the intervals can
be reduced while still having a reasonable num-
200
ber of measurements within each interval. This
again tends to produce a smoother distribution,
as shown in Figure 8.5. Note the changes of 100
vertical scale between Figure 8.3, 8.4 and 8.5.
0
With an extremely large number of measure- 2.44 2.45 2.46 2.47
ments and very small intervals on the horizontal cell constant/nm
axis, the ‘envelope’ of the distribution will tend Figure 8.5: The distribution becomes smoother as the
to become a smooth bell-shaped curve, like that number of measurements increases.
in Figure 8.6.
These distributions all give some impression of

number of measurements
the spread of the measurements, and the way the
results cluster at the peak of the distribution in
Figure 8.6 suggests that this peak might repre-
sent the average or ‘best estimate’ value. How-
ever, a scientist would want a more quantitative
and succinct way to describe such results and to
communicate them to other people working on
similar problems. The mean and standard de- 2.44 2.45 2.46 2.47
viation are the measures most commonly used cell constant/nm
to summarize large sets of data with just a few Figure 8.6: The distribution for an extremely large
numbers. number of measurements.

Back J I 409
Contents 
8.2.3 Mean and standard deviation for repeated measurements

In everyday terms, everybody is familiar with the word ‘average’, but in science
and statistics there are actually several different kinds of average used for different
purposes. In the kind of situation exemplified by Table 8.2, the sort to use is the
mean (or more strictly the ‘arithmetic mean’) For a set of measurements, this is
defined as the sum of all the measurements divided by the total number of measure-
ments made.

Question
What is the mean of the results in Table 8.2?
Answer
The sum of all the measurements is 24.544 nm. There are 10 results, so the
24.544 nm
mean value is , or 2.4544 nm to 5 significant figures. (The reason for
10
giving the result to this number of significant figures will be discussed shortly,
but for the moment let us proceed without worrying too much about this aspect
of the calculation.)

To turn this description of how to calculate a mean into a formula, each element
has be allocated a symbol. So let us say that we have made n measurements of a
quantity x. Then we can call the individual measurements x1 , x2 , x3 , . . . xn (where
x1 is properly said either as ‘x subscript one’ or as ‘x sub one’, but also sometimes
as ‘x one’ provided the meaning remains clear). The mean value of any quantity is

Back J I 410
Contents 
usually denoted by writing a bar over the quantity so the mean of x is written as x
(and said ‘x bar’). Possible (and correct) formulae are therefore:
x1 + x2 + x3 + . . . + xn
x=
n
or
1
x= (x1 + x2 + x3 + . . . + xn )
n
P
However, the sum is tedious to write out, so a special ‘summation’ sign, (the
Greek capital letter sigma), is used to denote the adding up process, and the mean
of n measurements can be neatly written as:

n
1 X
x= xi (8.3)
n i =1

The i = 1 below the summation sign indicates that the first value for xi in the sum is
x1 , and the n above it indicates that the last value in the sum is xn . In other words,
all integer values of i (x1 , x2 , x3 , etc.) are to be included up to xn . (The summation
sign with the information attached to it is usually said as ‘sum of x sub i from one
to n’.)
We now want a quantitative way of describing the spread of measurements, i.e.
the extent to which the measurements ‘deviate’ from the mean. There are 5 steps

Back J I 411
Contents 
required to do this, which are laid out below, and Table 8.3 shows the results of
following this ‘recipe’ for the data in Table 8.2:
Step 1
Calculate the deviation of each measurement. The deviation di of any individual
measurement is defined as the difference between that measurement and the mean
of the set of measurements:

di = xi − x (8.4)

Notice that the value of di may be positive or negative depending on whether a par-
ticular measurement is larger or smaller than the mean of the set of measurements.
At this stage the deviations have been expressed as decimal numbers.
Step 2
Calculate the squares of each of the deviations (i.e. di2 ). These will, of course, all
have positive values.
By this stage the values have become very small so the column has been headed in
such way that the numbers entered in the column represent the value of di2 divided
by 10−5 .
Step 3
n
X
Add together all the squares of the deviations (i.e. di2 ).
i =1

Back J I 412
Contents 
Step 4
Divide by the total number of measurements (i.e. n) to obtain the mean of all the Table 8.3
square deviations. This may be written as:
n
1X 2
di2 = d (8.5)
n i =1 i

Step 5
Take the square root of this mean to obtain the ‘root mean square deviation’ sn . It
is this quantity sn that is known as the standard deviation. Step 5 may be written as:
q
sn = di2

or, substituting for di2 from Equation 8.5, as:

v
n
t
1X 2
sn = d (8.6)
n i =1 i

Since di was defined in Equation 8.4 as (xi − x), one final substitution into Equation
8.6 gives sn in its most frequently used format:

Back J I 413
Contents 
The standard deviation sn for n repeated measurements of the same quantity x
is given by
v
t n
1X
sn = (xi − x)2 (8.7)
n i =1

At the end of this process, we can summarize all the data in Table 8.2 just by
saying that the ten measurements had a mean of 2.4544 nm and standard deviation
0.0046 nm. The calculation of standard deviation is given in Table 8.3.
There are several things worth noting about this result and the data in Table 8.3.
First, all the quantities have units associated with them. The values of xi were
measured in nanometres, so deviations will also be in nanometres and the squares
of the deviations in nm2 , as shown in the column headings in the table.
A second useful feature to notice is that the sum of all the deviations is equal to
zero.
n
X
di = 0
i =1

If you are interested in knowing why this is always true, there is an explanation in
Box 8.4 (though you do not need to work through the full explanation in order to
make use of the result). At the end of Step 1 it is well worth adding up all the values

Back J I 414
Contents 
you have calculated for the deviations to ensure that they do indeed total zero. If
they don’t, you have made an arithmetic slip somewhere which needs to be put right
before you proceed to Step 2.

Box 8.4 The sum of the deviations is always equal to zero

It is quite easy to work out from first principles the reason for the sum of the
deviations being zero in the special case in which the set consists of just two
measurements, x1 and x2 . The mean would then be:
x1 + x2
x=
2
x1 + x2 x1 x2
so d1 = x1 − x = x1 − = x1 − −
2 2 2
x1 + x2 x1 x2
and d2 = x2 − x = x2 − = x2 − −
2 2 2
Therefore
 x1 x2   x1 x2 
d1 + d2 = x1 − − + x2 − −
2 2 2 2
 x1 x1   x2 x2 
= x1 − − + x2 − − (rearranging the terms)
2 2 2 2
=0
This argument can be extended to any number of values of x; as an exercise in
algebra you might like to try it for three measurements. However many values
of x are chosen, it is always the case that the sum of the deviations is zero.

Back J I 415
Contents 
Looking now at the details of the calculation, the original measurements of length
were made to the nearest picometre (i.e. 0.001 nm), represented by 3 places of
decimals (i.e. 3 digits after the decimal point). More digits were carried in the
calculations to avoid rounding errors. However, what is the appropriate number of
digits to quote in the final answer? Well, when we added up all the 10 results in
Xn
Table 8.3, we obtained xi = 24.544 nm (i.e. 5 digits in total). We divided this
i =1
sum by an exact number (10) so we are entitled to retain 5 digits in the result of
this division, giving x as 2.4544 nm. It is therefore valid to retain one more decimal
place in the mean value than we had in each of the measurements individually.
After all the whole point of repeating the measurement many times and averaging
is to improve our confidence in our final result! Having quoted the mean as x =
2.4544 nm, it then makes sense to quote the standard deviation as 0.0046 nm.
The fact that here the standard deviation is quite small in comparison to the mean
shows why, in this context, it is more sensible to think in terms of places of decimals
rather than significant figures. Because leading zeroes do not count as significant,
the standard deviation is actually only given to 2 significant figures, whereas the
mean is given to 5. In such circumstances it is easier to think of the mean and
the standard deviation as being expressed to the same number of decimal places
(always assuming of course that they are given the same units).
In summary, it is often reasonable to give the mean to one more decimal place (or
one more significant figure) than was used for each of the individual measurements,
and then to quote the standard deviation to the same number of decimal places as
the mean.
Back J I 416
Contents 
8.2.4 Using a calculator for statistical calculations

Table 8.3 shows all the values for each step in the process of calculating a standard
deviation, so that you can see what the operations encapsulated by Equation 8.7
actually entail, but you will probably be relieved to hear that it is not usually neces-
sary to carry out such detailed calculations. Scientific and graphics calculators (or
computer spreadsheets) can do most of the drudgery for you.
You will need to consult the instructions for your own calculator in order to find out
how to do this, but usually the process involves the following steps.
Step 1
Put the calculator into statistical mode.
Step 2
You should then be able to input all the data; sometimes the data is stored via a
memory button, in other cases it can be entered and displayed as a list. Try this out
with the following set of numbers:
8, 6, 9, 12, 10
Step 3
Having input the data, you can then get most calculators to tell you the number of
items of data. If your calculator can do this, it should return the answer ‘5’ here. It
doesn’t matter if your calculator doesn’t have this function, but if it does it’s well
worth using this checking device. If you have to input a long string of data values,
it’s quite easy to miss one out inadvertently!
Back J I 417
Contents 
Step 4
When you know you have the data correctly stored, find out how to display the
mean; you should get the answer ‘9’ here.
Step 5
Now find out how to display the standard deviation. Many calculators use the sym-
bol σn for standard deviation, rather than sn (σ is the lower case version of the
Greek letter sigma). Do be careful with this step: your calculator may also have a
button labelled σn−1 or sn−1 . Don’t use it by mistake! You should get the answer
‘2’ here.
Once you are sure you know how to use your calculator to perform calculations
of mean and standard deviation, apply this skill to Question 8.8. To answer such
questions, you could choose to work out a full table similar to Table 8.3, but that
it is a very time-consuming process, so it is worth becoming confident in using the
statistics buttons on your calculator.

Back J I 418
Contents 
Question 8.8 Answer Measurement Diameter/mm
A sample of a particular manufacturer’s ‘coarse 1 1.09
round wire’ was measured at ten points along 2 1.00
its length. The data is given in Table 8.4. 3 1.25
Calculate the mean and standard deviation of 4 1.24
these measurements. 5 1.29
6 0.89
7 1.09
8 1.14
9 1.22
10 1.01

Table 8.4: Repeated measurements

of the diameter of a wire.

8.2.5 How likely are particular results?

In real experiments, as opposed to hypothetical ones, it is very rare that scientists

make a sufficiently large number of measurements to obtain a smooth continuous
distribution like that shown in Figure 8.6. However, it often convenient to assume a
particular mathematical form for typically distributed measurements, and the form
that is usually assumed is the normal distribution, so-called because it is very com-
mon in nature. The normal distribution corresponds to a bell-shaped curve which

Back J I 419
Contents 
is symmetric about its peak, as illustrated in Figure 8.6. Repeated independent
measurements of the same quantity (such as the breadth of an object, or its mass)
approximate to a normal distribution. The more data is collected, the closer it will
come to describing a normal distribution curve.
The peak of the normal distribution curve
corresponds to the mean value of the distri- mean value

bution, as shown in Figure 8.7. This figure

also illustrates how the standard deviation
of a set of measurements is related to the
spread. Although it is beyond the scope of sn
this course to prove this, the area under a sn

portion of a distribution curve within a cer-

±sn
tain range represents the number of mea-
surements that lie within that range, as a
proportion of the whole set. For a normal Figure 8.7: The shaded area under this normal distribution
distribution, it turns out that 68% of the curve represents the measurements that lie within one stan-
measurements lie within one standard de- dard deviation of the mean.
viation (i.e. within ±sn ) of the mean value.
Conversely, 32% of the measurements will lie outside this range. If you make a
single additional measurement, it is therefore just over twice as probable that this
one measurement will fall within one standard deviation of the mean than that it
will fall outside this range. For a normal distribution, it also turns out that 95%
of measurements are likely to fall within two standard deviations of the mean and
99.7% within three standard deviations of the mean.

Back J I 420
Contents 
Remembering that precise mea-
surements were defined in Sec-
tion 8.2.1 as those for which
the scatter was small, you will
appreciate that the more pre-
cise a repeated set of the same
number of measurements of a
particular quantity, the more
highly peaked the distribution w w y y z z
curve and the smaller the stan-
dard deviation will be. A very precision of measurement increasing
(i.e. random uncertainty decreasing);
broad distribution on the other standard deviation decreasing
hand, corresponds to measure-
ments with considerable scatter
and the standard deviation will Figure 8.8: Normal distribution curves for three independent sets of
be large. These trends are illus- measurements, with the same number of measurements in each set. The
trated in Figure 8.8. measurements of quantity w are subject to large random uncertainties,
while those of quantity y are more precise and those of z more precise
still.

Back J I 421
Contents 
8.2.6 Different types of ‘average’

Figure 8.7 showed that if the data has a normal distribution the mean value corre-
sponds to the peak of the distribution. Normal distributions of data are very com-
mon in science, but by no means universal. Figure 8.9 shows some other possible
distributions, three of which are symmetric and one of which is skewed (i.e. not
symmetric).

Figure 8.9: Types of

mean distribution. (a) is the
mean normal distribution: a
symmetric bell shaped
curve. (b) is also
symmetric, but the
(a) (b)
shape of the curve
does not approximate
mean to the normal
mean
distribution. (c) is a
skewed distribution.
(d) is symmetric, but
displays two equal
(c) (d) maxima.

Back J I 422
Contents 
In many cases, especially if the distribution is skewed, the mean is not the best way
of representing an average or typical value. Imagine for example a small company
with a single owner who pays himself £900 000 a year and 10 employees who
are each paid £10 000. The statement that the mean annual income of these 11
workers is more than £90 000 (i.e. £1000 000 divided by 11) — although true —
is somewhat misleading! In such cases, two other quantities, the mode and the
median, may represent the data more fairly.
The mode is the most frequently occurring value in the set of data. If the data is
plotted on a histogram or a bar chart, the mode will be the value corresponding to
the tallest bar.

Question
What is the mode of the earnings in the company described above?
Answer
The mode is £10 000. This is certainly more representative of the typical earn-
ings than the mean would be!

Note that in some cases there may be more than one value for the mode; for exam-
ple, that would be the case for the distribution shown in Figure 8.9d.

Back J I 423
Contents 
The median is the middle value in a series when the values are arranged in order
of size. This means that half the measurements have values that are bigger than the
median and half have values that are smaller than the median. If there are an odd
number of measurements, the median is the middle measurement; if there are an
even number of measurements it is the mean of the middle two values.
To see how this works, consider the following example. Ten plants of a particu-
lar species were chosen at random and the number of flowers on each plant were
counted. The results were:
8; 7; 4; 8; 10; 7; 9; 7; 8; 7;

Question
What is the mode for this data?
Answer
The best way of answering this is to compile a table showing the number of
plants with particular numbers of flowers:

number of flowers 4 7 8 9 10
number of plants 1 4 3 1 1
The mode is 7 flowers. There are more plants with 7 flowers than with any other
number of flowers.

Back J I 424
Contents 
Question
What is the median for this data?
Answer
To answer this, we have to order the data. In increasing numbers of flowers, the
results obtained were:
4; 7; 7; 7; 7; 8; 8; 8; 9; 10.
With a sample of 10 plants the median is the mean number of flowers on the 5th
and 6th plants (counted in either ascending or descending order). In ascending
2 = 7.5.
order, the 5th plant has 7 flowers and the 6th has 8, so the median is 7+8

Question 8.9
The heights of nine different specimens of the same type of plant were measured
in centimetres, and the results in descending order were:
8.6; 8.3; 8.2; 7.9; 7.8; 7.8; 7.4; 7.3; 7.1
(a) What is the median of this data? Answer
(b) What is the mean of this data? Answer

Box 8.5 illustrates a case in which the mode and median give a more representative
summary of the data than the mean.

Back J I 425
Contents 
Box 8.5 Seabird migration
In a study of Storm Petrels (small seabirds), sev-
eral thousand birds were marked with identifying Recovery place Distance/km Number
rings when they were at their nests on a Shet- of birds
land island. After nesting, the birds dispersed. Shetland (Lerwick) 49 8
Twenty-eight of the birds were subsequently re- Shetland (Foula) 77 5
ported as having been recovered in other areas as Fair Isle 114 5
shown in Table 8.5. Orkney 157 2
Sule Skerry 248 3
Taking all 28 observations into account, the mean
Summer Isles 382 1
distance from their nest site at which the birds
St Kilda 529 2
have been recovered is 554.5 km. However, this
Cape Clear 1114 1
is not a very useful way in which to summarize
South Africa (Durban) 10568 1
the data, because in fact 13 out of the 28 birds
(i.e. nearly half) moved less than 100 km, and Table 8.5: The recovery location of Storm Petrels
only two moved further than the mean distance. ringed at their nests on one of the Shetland islands
The median distance is 114 km and this is a more
typical value.
This example shows how the mean can be highly dependent on a small number of measurements that
are a long way from the mode. In this case, the single recovery from South Africa has an enormous
influence on the mean. The median is ‘resistant’ to extreme values. Even if the bird recovered in South
Africa had stopped in Morocco, or alternatively if it had gone to New Zealand, the median value would
have remained 114 km.

Back J I 426
Contents 
8.2.7 Samples and populations

It is no accident that the examples used in Sections 8.2.3 and 8.2.4 to illustrate
the statistics for repeated measurements of individual quantities were drawn from
chemistry and physics. Experiments involving repeated measurements of some
quantity are typical of the physical sciences. There are, however, many other types
of scientific work in which a typical procedure is to collect data by measuring or
counting the members of a sub-set of things which form part of a larger group,
and Section 8.2.6 contained several examples. In this type of work, the sub-set of
members that are measured or counted is called the sample and the larger group is
called a population. Although often employed in the context of biology to describe
a group of organisms that might breed with one another, the term ‘population’ is
used much more widely in statistics to mean a collection of things or events. Ex-
amples of statistical populations could include all the sand grains on a beach, all
the leaves on a tree, all the people in England with blood group AB, or all the visits
made to the Science Museum in March.
It is generally the case that the members of any one population display some vari-
ability; for instance, not all the leaves on an oak tree will be exactly the same size.
Furthermore, different populations often overlap with respect to whatever we might
be measuring or counting. But despite this variability and overlap, what scientists
often want to know is whether there seem to be systematic differences between the
populations. Indeed, only if there do seem to be such differences do they accept
that they really are dealing with more than one population. Failure to find evidence
of systematic differences between the leaves of oak trees growing on sandy soil and

Back J I 427
Contents 
those of oak trees growing on clay soils would suggest that the leaves (and trees)
were members of the same population, or in other words that soil conditions have
no overall effect on the leaves of oak trees. The statistical techniques used in look-
ing for systematic differences between populations are the subject of Chapter 9, but
in order to make use of these techniques it is necessary to be able to summarize
the data that has been collected. You saw in Section 8.2.3 that for repeated mea-
surements data sets could be summarized by quoting just two quantities: the mean
and the standard deviation. This is also true for samples drawn from populations,
but the mean and the standard deviation take on slightly different meanings in this
context.
It is normally the case that data cannot be collected on all members of a population.
It would indeed be impractical to attempt to measure every leaf on an oak tree! By
the same token, it is usually impossible to know the true mean of some quantity for a
whole population. This ‘true mean’ (also known as the ‘population mean’) is given
the symbol µ (the Greek letter ‘mew’), with the understanding that this quantity
is generally not only unknown but unknowable. What we can easily calculate,
however, is the mean of the quantity as measured for a sample drawn from the
population. This is given the symbol x and calculated using Equation 8.3, exactly
as we did in Section 8.2.3. Provided the sample is unbiased, x is the best estimate
of µ that we can obtain.
As with the mean, the true standard deviation of a population can usually never
be known with certainty. Again, the best estimate we can obtain must come from
the distribution of values in a sample drawn from the population. However, this
time it isn’t appropriate to use the formula for the standard deviation of repeated

Back J I 428
Contents 
measurements of one quantity which was:
v
n
t
1X
sn = (xi − x)2 (8.7)
n i =1

Instead a slightly different formula is used, namely:

v
n
t
1 X
sn = (xi − x)2 (8.8)
n − 1 i =1

sn−1 is often called the ‘sample standard deviation’ because it is calculated from
data taken for a sample of the population.

The value determined for sn−1 provides the best estimate of the standard deviation
of the population. It will not have escaped your notice that the only difference be-
tween the two formulae is that in Equation 8.8 we are dividing by (n − 1), whereas
in Equation 8.7 we were dividing by n. This means that sn−1 must always be larger
than sn (because we are dividing by a smaller number). This allows for the possi-
bility that within the whole population there may be a few extremely high or low
values of the measured quantity which will not necessarily be picked up in a sample
drawn from that population.

Back J I 429
Contents 
sn−1 is also often called the ‘estimated standard deviation of a population’ be-
cause, provided the sample is chosen without bias, it is the best estimate that can
be made of the true standard deviation of the population.

You should now check that you can use your calculator to determine the sample
standard deviation sn−1 for a set of data. For this purpose, try taking the same set
of numbers you used in Section 8.2.4 to check how to calculate sn . These numbers
were:
8, 6, 9, 12, 10.
The first four steps are the same as before, only Step 5 will be different.
Step 1
Put the calculator into statistical mode.
Step 2
Input all the data.
Step 3
If your calculator can tell you the number of items of data, check that it gives the
answer ‘5’ here.
Step 4
When you know you have the data correctly stored, display the mean; you should
get the answer ‘9’ here.

Back J I 430
Contents 
Step 5
Now find out how to display the sample standard deviation. The appropriate button
will probably be marked σn−1 or sn−1 . You should get the answer ‘2.2’ here (to one
decimal place). Don’t use the σn or sn button by mistake!
While this example is useful to familiarize yourself with the process, it doesn’t rep-
resent a realistic scenario, not least because the hypothetical data set is so small.
Because the aim is to estimate the mean and standard deviation for a whole popu-
lation by carrying out measurements just on a sample, it is important to ensure that
the sample is representative of the population as a whole and that usually requires
it not only to be chosen without bias, but also to be reasonably large. In Question
8.10, the sample consists of 20 plants.

Question 8.10 Answer

Suppose that the number of flowers were counted on 20 orchid plants in a colony,
and that the results were:
8; 8; 4; 8; 8; 7; 9; 7; 7; 5; 9; 10; 6; 9; 7; 4; 8; 5; 11; 5.
From this data, estimate to 3 significant figures the mean number, µ, of flowers
per plant in the colony and the standard deviation of the population. You may
if you wish construct a table similar to Table 8.3, but it will be much quicker
simply to use your calculator.

Back J I 431
Contents 
8.3 Learning outcomes for Chapter 8
After completing your work on this chapter you should be able to:
8.1 demonstrate understanding of the terms emboldened in the text;
8.2 calculate the probability of a particular outcome from information about
possible outcomes;
8.3 express a probability as a fraction, a decimal number or a percentage;
8.4 combine probabilities appropriately from information about possible
outcomes;
8.5 interpret data in which the relative values of quantities are expressed as ratios;
8.6 calculate the mean, mode and median for a set of data;
8.7 calculate the standard deviation sn for a set of repeated measurements of a
particular quantity;
8.8 calculate the estimated standard deviation of a population, sn−1 , from a set of
measurements made on a sample drawn from the population.

Back J I 432
Contents 

Statistical hypothesis testing 9

Samples and populations can be described in terms of their actual or estimated
means and standard deviations, as discussed in Chapter 8. However, the ultimate
aim of collecting data is usually not simply to describe, but also to answer sci-
entific questions as objectively as possible. An extensive collection of statistical
techniques has been developed over many years to provide answers to some such
questions, bearing in mind that most data are intrinsically variable. In this chapter,
you will be introduced to the general principles that underpin almost all of these
techniques and then shown how to perform three particular statistical tests com-
monly used to answer different sorts of question. In order to illustrate these ideas
and to provide some of the data with which you can practice, this chapter is partly
based on a small ecological study. The study is described in Box 9.1.

Back J I 433
Contents 
Box 9.1 Green-winged Orchids and ridge-and-furrow topography
A conspicuous feature of parts of the English Midlands is ridge-and-furrow topography. Some of this
is medieval and some is much later in age. It has been known for some time that Bulbous Buttercup
(Ranunculus bulbosus) tends to occupy the drier ridges and Creeping Buttercup (R. repens) the wetter
furrows. Also found in the same area is Green-winged Orchid (Orchis morio), a rare plant in England. A
study was undertaken to find out whether the distribution and/or performance of Green-winged Orchid
might also be influenced by ridge-and-furrow topography.
Various measurements were made
on a sample of plants growing in a
local nature reserve. Figure 9.1 il-
lustrates some of the measurements
taken. These included the horizon- position of
ridge crest
tal and vertical distances of each plant height

plant from the nearest ridge crest,

the height of the plant, the number
of leaves and the number of flowers. vertical distance
Whether a plant was growing on the
horizontal distance
north-west or the south-east slope
of the ridge was also recorded, since
the two slopes might differ with re-
spect to mean temperature, mois- Figure 9.1: Measurement of horizontal and vertical distances
ture availability, etc. of a plant from the nearest ridge crest and plant height.

Back J I 434
Contents 
9.1 The principles of hypothesis testing
Many of the questions that arise out of scientific investigations are driven by hy-
potheses, tentative explanations of observations that may be tested by experiment
or by making further observations. Taking the study briefly described in Box 9.1
as an example, it might be proposed that Green-winged Orchid (like Bulbous But-
tercup) occurs more frequently — and/or grows better — nearer the drier crests of
ridges than the wetter furrows. Alternatively, it might be that Green-winged Orchid
(like Creeping Buttercup) ‘prefers’ the wetter furrows to the drier ridges. Notice
that these tentative ideas contain the unproven assumption that ridges are indeed
drier than furrows. Statistical hypothesis testing provides a universally agreed set
of procedures for answering questions such as ‘Do Green-winged Orchids tend to
occur nearer to ridge crests than expected by chance?’, ‘Does the amount of water
in soil increase with distance from the nearest ridge crest?’, ‘Do the Green-winged
Orchids growing nearer ridge crests tend to be taller or have more leaves and/or
flowers than those growing further away?’.
There are two major branches of statistical hypothesis-testing: ‘tests of association’
(e.g. ‘Are Green-winged Orchids found in association with ridge crests significantly
more frequently than would be expected by chance?’) and ‘tests of difference’ (e.g.
‘Is there a significant difference between the mean height of plants growing on the
north-west rather than the south-east slopes of ridges?’).

Back J I 435
Contents 
Question
Would an investigation into whether there is a significant increase in the water
content of soil with increasing distance from the nearest ridge crest be a test of
association or a test of difference?
Answer
Since we would be looking to see if there is an association between soil water
content and distance from ridge crest, this would be a test of association.

Having ascertained which statistical test is appropriate in any particular circum-

stance, most scientists look up the details of that test and then use it almost as if
it were a ‘black box’ (a piece of equipment that users trust to perform a particular
task reliably without understanding how it actually works). Sometimes, however,
it is helpful to stand back from the details of any particular statistical test and to
consider those features that are common to nearly all such tests. These common
features can best be illustrated by considering in general terms a test of difference
between two means.

Back J I 436
Contents 
Suppose that a scientist collected measurements
from two samples of plants, one of which had
been exposed to a particular experimental treat-

number of measurements
ment and the other (the so-called control sample)
which had not. Almost certainly, there would be
some variation within each of these two sets of
measurements and this would be reflected in their
standard deviations. Moreover, even if the differ- sn−1 sn−1

ence between the means of the experimental and

control plants was relatively large, it would not
be surprising if there was some overlap between
x x
the two sets of measurements, as shown in Figure
9.2.
Now it might suit the scientist’s favoured theory Figure 9.2: Diagram summarizing a possible result
to convince others that the treatment did have of an experiment in which a sample exposed to ex-
a statistically significant effect on the measured perimental treatment (shown with blue shading) was
character. On the other hand, it might be in the compared to a control (pink shading). Note the over-
scientist’s interests to show that the treatment did lap between the two distributions. Each x is the mean
not have a statistically significant effect. Either of a sample (and estimates the mean of the population,
way, the scientist is required by the procedures of µ) and each sn−1 is the sample standard deviation (es-
statistical hypothesis testing to put forward a so- timated standard deviation of the population).
called null hypothesis in the first instance. As the Note: The diagram shows the two distributions over-
name suggests, a null hypothesis is one of ‘no dif- lapping at exactly sn−1 from the mean. This would
ference’. In this case the null hypothesis would be not normally be the case.

Back J I 437
Contents 
that there is no difference between the population mean of the treated plants (µ1 )
and the population mean of the control plants (µ2 ). Expressing this statement math-
ematically, the null hypothesis would be that

µ1 = µ2

or, equivalently, that µ1 − µ2 = 0.

At the same time, the scientist has to put forward an alternative hypothesis that is the
logical ‘mirror image’ of the null hypothesis. In this case the alternative hypothesis
would be that there is a difference between the means of the treated and control
plants. Expressing this statement mathematically, the alternative hypothesis would
be that

µ1 , µ2

or µ1 − µ2 , 0.

Question
Is it possible for both the null and alternative hypotheses to be false?
Answer
No. If either is false, then the other must be true.

Back J I 438
Contents 
If the null hypothesis is true, then the alternative hypothesis must be false and
vice versa.

Once statements of the null and alternative hypotheses have been made, a quantity
called the test statistic is calculated. The test statistic is a number, on the basis of
which a decision can be made to accept or reject the null hypothesis. The value
of the test statistic depends on the characteristics of the samples being compared,
and in most cases it is calculated using one or more equations. Things are often so
arranged that the value of the test statistic comes out to be zero if the null hypothesis
is true (for instance, by including the term (x1 − x2 ) in the numerator of the equation,
where x1 and x2 , the means of the two samples, are the best available estimates of
the unknowable values of µ1 and µ2 ). However, because of the vagaries of sampling,
it would be extremely unlikely for the means of two samples drawn from even the
same population to be identical (for instance, two samples of control plants are very
unlikely to have exactly the same mean). So, the question is ‘How large does the
test statistic have to be before one can be reasonably confident that the samples
were drawn from different populations (and therefore conclude, in this example,
that the experimental treatment probably did have a significant effect)?’ In fact, it
is impossible to give a definitive answer to this question; it can be answered only in
terms of probabilities.
Ideally, the precise probability that the calculated value of the test statistic could
have arisen by chance if the null hypothesis were true would be determined. In a
1
particular instance, this might turn out to be something like 1 in 63, i.e. 63 , which
is 0.015 87 to four significant figures. In practice, the value of the test statistic is

Back J I 439
Contents 
usually compared to lists of critical values calculated for a few pre-determined sig-
nificance levels expressed in terms of probabilities. In this context, the probabilities
are usually abbreviated to P and expressed in decimal notation, e.g. 0.1, 0.05 and
0.01. For any particular significance level, the critical value is the most extreme
(usually largest) value that the test statistic could be expected to have if the null
hypothesis were true. Of course, if the null hypothesis is true then any deviation
from the test statistic’s expected value (which, as noted above, is usually zero) must
have arisen purely by chance. If the significance level corresponding to the value of
the test statistic turns out to be quite low (usually because the test statistic is rather
high), then it must be accepted that the null hypothesis is unlikely to be true. If the
null hypothesis is false, then the alternative hypothesis must be true. Only at this
stage can the scientist conclude:
• either that the treatment did have a significant effect (because the null hy-
pothesis was probably false and therefore the alternative hypothesis probably
true)
• or that the treatment did not have a significant effect (because the null hy-
pothesis is likely to have been true).
It is extremely important to realize that the particular significance level at which a
null hypothesis is rejected — and hence the alternative hypothesis is accepted — is
a matter of convention. The usual convention in science is to reject a null hypothesis
if the probability P is less than the 0.05 significance level, i.e. if P < 0.05. However,
in employing this convention, it is also important to realize that you could either be
rejecting a true null hypothesis or accepting a false one. Indeed, you are explicitly

Back J I 440
Contents 
accepting that on average, if you were to carry out 100 statistical tests, you would
reach the wrong conclusion for 5 of these tests (although you would not know
which ones). If the work you are engaged in really matters, for example, medical
research in which human lives might be at stake, then you would probably employ
more exacting criteria (such as rejecting null hypotheses only if P < 0.01 or even
P < 0.001). On the other hand, insisting on the use of such rigorous criteria for even
routine scientific work would mean that many null hypotheses that really are false
would have to be accepted, and this would undoubtedly hinder scientific progress.
The important features of statistical hypothesis testing are summarized below:

1. A null hypothesis (e.g. µ1 = µ2 ) and an alternative hypothesis (e.g.

µ1 , µ2 ) are proposed.
2. The value of a test statistic is calculated.
3. If the probability of this value arising by chance if the null hypothesis
were true is low — conventionally, less than 0.05 — then the null hy-
pothesis is rejected and the alternative hypothesis accepted. (There
is always the possibility of either rejecting a true null hypothesis or
of accepting a false one.)

When null hypotheses are rejected, the results are described as being statistically
significant or sometimes just as ‘significant’. A consequence of this is often a feel-
ing that ‘non-significant’ results are of less value than ‘significant’ ones. Indeed,

Back J I 441
Contents 
there is probably a ‘reporting bias’ whereby significant differences are more likely
to be published in scientific papers than non-significant ones. This undervaluing of
non-significant differences is unfortunate because the whole point of the exercise is
to try to find out what is happening in the real world. It may be just as important to
know that an effect is not produced by one experimental treatment as to know that
another treatment does produce the effect.

Question 9.1
Should the null hypothesis be accepted or rejected if the result of a statistical
hypothesis test turned out to be:
(a) P < 0.01, Answer
(b) P > 0.05, Answer
(c) P > 0.01? Answer

Back J I 442
Contents 
9.2 Deciding which test to use; levels of measurement
The expression ‘levels of measurement’ refers to important distinctions between
different sorts of data that might be collected during the course of a scientific inves-
tigation.
An example of data collected at the categorical level is the sex of animals. In most
cases, an animal is unambiguously either ‘male’ or ‘female’. Furthermore, there is
no logical way in which the category ‘male’ can be ranked as ‘higher’ or ‘better’
than the category ‘female’ or vice versa. All that can be said is that these two
categories are different. Of course, a data set may include more than two categories.
It is possible to rank ordinal level data in a sensible way. For instance, plants may
be listed in order of their heights or grouped by the approximate number of leaves
they possess without knowing the actual heights or the actual numbers of leaves. If
the actual heights or numbers of leaves are known, then these data are at the interval
level.
Data collected at the interval level can, if necessary, be analysed at the ordinal
level. For instance, you might know that Plant A has 8 leaves and that Plant B has 5
leaves (interval level data). Nevertheless, you could choose to ignore some of this
information and simply treat Plant A as having more leaves than Plant B (ordinal
level data). Of course, if all you knew was that Plant A has more leaves than Plant
B, then you could not convert this information into interval level data for analysis.
Categorical level data cannot usually be treated as if they were at interval or ordinal

Back J I 443
Contents 
level (although you might argue that, for instance, red-flowered plants have more of
a particular pigment than pink-flowered plants of the same species). However, by
applying arbitrary criteria, interval or ordinal level data can sometimes be converted
into categorical data for analysis. For instance, one of the seven pairs of contrasting
characters used by Mendel in his pioneering research on the genetics of garden
peas (see Box 8.2) was ‘tall’ versus ‘short’. This categorical distinction made sense
only because, in this particular case, there was no overlap between ‘tall’ and ‘short’
plants.
The reason for distinguishing between the different levels of measurement is that
different statistical tests must be used to analyse categorical, ordinal and interval
level data. Sometimes, when analysis of data at the interval level fails to reveal sta-
tistically significant differences, such differences may be shown up when the data
are re-analysed at the ordinal level. However, because some information about the
samples has effectively been ‘thrown away’ in the process, any statements eventu-
ally made about the populations from which the samples were drawn are necessarily
less complete than they might have been.

Back J I 444
Contents 
Question 9.2
In each of the following cases, explain briefly whether the data should be treated
as being at the categorical, the ordinal or the interval level.
(a) A count is made of the number of parasites on each member of Answer
a sample of sheep.
(b) A sample of sheep are counted as either ‘parasitized’ (i.e. car- Answer
rying one or more parasites) or ‘unparasitized’ (i.e. carrying
no parasites).
(c) A sample of sheep are counted as ‘unparasitized’ (i.e. carry- Answer
ing no parasites), ‘lightly parasitized’ (i.e. carrying 1–5 par-
asites), ‘moderately parasitized’ (i.e. carrying 6–10 parasites)
or ‘heavily parasitized (i.e. carrying more than 10 parasites).

Back J I 445
Contents 
9.3 The χ2-test
The χ2 -test (where χ is the Greek letter ‘chi’, said to rhyme with ‘sky’) is very com-
monly employed when scientists wish to test whether data on a single categorical
variable match a particular theoretical pattern. Since ‘presence’ versus ‘absence’
is a categorical variable, the χ2 -test is often used to compare the numbers of in-
dividuals present in different areas with the numbers expected on the basis of an
appropriate null hypothesis. This is more precisely called the χ2 goodness-of-fit
test. (There are other χ2 -tests, e.g. for possible associations between two categori-
cal variables.)
In the Green-winged Orchid study, described in Box 9.1, horizontal distance from
the nearest ridge crest (as shown in Figure 9.1) was recorded for 210 plants grow-
ing on several ridges. Because the ridge crest-to-furrow distance varied slightly
between ridges, each of these distances was divided into five equal categories (cat-
egory 1 being 0.00–19.9% of the distance from the crest, category 2 being 20.0–
39.9% of the distance, category 3 being 40.0–59.9% of the distance, etc.) so that
the data from different ridges could be pooled for analysis. This procedure enables
us to treat interval level data (the horizontal distance of each plant from the nearest
ridge crest) as categorical level data (the distance category into which each plant
falls). If the 210 plants were distributed uniformly with respect to the ridge crest,
then a fifth of them (i.e. 42) would be expected to occur within each distance cate-
gory. A reasonable null hypothesis would be that, if it were possible to collect data
on the entire population of Green-winged Orchids growing in fields with ridge-
and-furrow topography, then there would be no difference between the number of

Back J I 446
Contents 
plants observed in each distance category and the number that would be expected
on the assumption that the plants were distributed uniformly. The alternative hy-
pothesis would be that the number of plants observed in each distance category was
not equal to the number of plants expected. Accepting this alternative hypothesis
implies accepting that the plants were distributed non-uniformly.
In fact, of the sample of 210 plants, 105 occurred in the first distance category, 74
in the second, 28 in the third, 3 in the fourth and none in the fifth. It certainly ap-
pears that the plants were not uniformly distributed. The χ2 -test allows a definitive
statement to be made on the probability that the population of plants from which
the sample was drawn could have been distributed uniformly despite the appar-
ently non-uniform distribution observed in the sample. Only if this probability is
sufficiently low (conventionally if P < 0.05) can the null hypothesis be rejected
and the alternative hypothesis (with its implication that the plants were distributed
non-uniformly) accepted.

Back J I 447
Contents 
The first stage in performing a χ2 -test is Distance category Observed Expected
usually to draw up a table to compare ob- number (Oi ) number (Ei )
served and expected numbers in different
categories. The table for the sample of 1 (nearest to ridge) 105 42
210 orchid plants is given in Table 9.1, and 2 74 42
compares the number of individuals, Oi , 3 28 42
that were observed in each distance cate- 4 3 42
gory, with the number Ei expected on the 5 (furthest from ridge) 0 42
basis of the null hypothesis. As a check, total 210 210
the total number in the Oi column should
equal the total number in the Ei column. Table 9.1: Table comparing the observed distribution of a
The trickiest part of most χ2 -tests is decid- sample of Green-winged Orchids across 5 categories of dis-
ing the ‘expected’ numbers. In this case, if tance from the nearest ridge crest with the distribution ex-
the null hypothesis were true, a fifth of the pected if the plants were distributed uniformly
plants (i.e. 42) would be expected to fall
into each distance category.

Back J I 448
Contents 
The test statistic is χ2 and this is found in the following way.
1. For each distance category, the ‘expected’ number is subtracted from the ‘ob-
served’ number.
This gives (Oi − Ei ).
2. Each result from step 1 is squared.
This gives (Oi − Ei )2 .
3. Each result from step 2 is divided by the appropriate ‘expected’ number.
(Oi − Ei )2
This gives .
Ei
4. The results from step 3 are totalled.
n
X (Oi − Ei )2
This gives which is the test statistic χ2 .
i =1
Ei
In summary,

n
X (Oi − Ei )2
χ =2
(9.1)
i =1
Ei

Back J I 449
Contents 
The easiest way to calculate χ2 is to (Oi − Ei )2
extend Table 9.1 to include columns Distance Oi Ei (Oi − Ei ) (Oi − Ei )2
Ei
for each of these steps. This has been category
done in Table 9.2, and χ2 is the to- 1 105 42 63 3969 94.500
tal of the values in the right-hand col- 2 74 42 32 1024 24.381
umn. Notice that, as a further check, 3 28 42 −14 196 4.667
the total of the (Oi − Ei ) column must 4 3 42 −39 1521 36.214
be zero, since the total number of in- 5 0 42 −42 1764 42.000
dividuals observed is equal to the to-
tal number of individuals expected. total 210 210 0 201.762

As an example of the way in which Table 9.2: Extension of Table 9.1 to calculate χ2 .
each value is calculated, consider the
first distance category. Oi = 105 and
Ei = 42 so

(Oi − Ei )2 (105 − 42)2 632 3969

= = = = 94.500
Ei 42 42 42
This value, and all the other values in the right-hand column of Table 9.2 have been
calculated to three places of decimals. This is normal practice when finding χ2
because this is how values are generally stated in tables of critical values for χ2 .

Back J I 450
Contents 
The next stage is to compare the value of the test statistic χ2 (which, in this case,
is 201.762) with the critical values listed in Table 9.3. The sizes of the critical
values in such a table depend on both the significance level (P = 0.1, P = 0.05 and
P = 0.01, given across the top of the table) and the number of degrees of freedom
(given down the left-hand side of the table).
The number of degrees of freedom can be found by counting the number of ‘cells’
in the table that contain observed counts (i.e. ignoring expected counts, totals etc.).

For the χ2 -test, the number of degrees of freedom is given by

number of cells − 1

In this case, Table 9.1 has five cells, so

number of degrees of freedom = 5 − 1

=4

Box 9.2 gives a brief explanation of why it is reasonable for the number of degrees
of freedom to be four in this case.

Back J I 451
Contents 
Box 9.2 Degrees of freedom
Why should the number of degrees of freedom be four for the data given in
Table 9.1? The total numbers of both ‘observed’ and ‘expected’ plants became
fixed (in this case, at 210) the moment data collection ceased. The number of
expected plants in each of the distance categories (42) is fixed by a combination
of the null hypothesis being tested (i.e. that equal numbers of plants would be
expected in each distance category) and the sample size (i.e. 210). In contrast,
the number of plants that could have been observed in each of any four of the
distance categories is completely free to vary, although the number of plants that
could have been observed in the final category is not free to vary in this way —
it must be such that the total of the numbers in the observed column equals the
sample size (i.e. 210). In this case, there are therefore four degrees of freedom.
Similar arguments to the above underpin the concept of degrees of freedom in
other statistical tests.

Back J I 452
Contents 
The parts of Table 9.3 that are relevant to our example Degrees of P = 0.1 P = 0.05 P = 0.01
are reproduced in Table 9.4. Reading across the row freedom
for 4 degrees of freedom, it can be seen that the χ2
1 2.706 3.841 6.635
value of 201.762 is greater than 7.779 (corresponding
2 4.605 5.991 9.210
to a significance level of 0.1), greater than 9.488 (cor-
3 6.251 7.815 11.341
responding to a significance level of 0.05) and greater
4 7.779 9.488 13.277
than 13.277 (corresponding to a significance level of
5 9.236 11.070 15.086
0.01). In fact, the significance level is considerably less
than 0.01 (because 201.762 is much larger than 13.277). Table 9.4: Part of Table 9.3.
Thus, the probability that the plants in the population
from which the sample was drawn were distributed uni-
formly is much less than 0.01 (i.e. P  0.01). There can be little doubt that the
plants were not distributed uniformly with respect to distance from the ridge crest.
The null hypothesis can therefore be rejected — and the alternative hypothesis ac-
cepted — with a great deal of confidence. In reporting such a result, it is often stated
that the null hypothesis is rejected at the P = 0.01 significance level or (probably
more commonly) at the 1% significance level.
Although statistics shows that the plants were distributed non-uniformly, it does
not reveal the nature of the non-uniform distribution. The data should now be
re-inspected to confirm that the plants did indeed occur closer to the ridge crests
than expected by chance — rather than nearer to the furrows or clustered halfway
between the ridges and furrows. The conclusion that can be drawn from this in-
vestigation is that Green-winged Orchids tend to occur significantly closer to ridge
crests than to furrows.

Back J I 453
Contents 
Precautions for the χ2 -test
You will not be expected, in this course, to decide which statistical test to use in any
given situation. However, in general, before performing a χ2 -test you should check
that:
• the data are at the categorical level;
• the ‘observed’ numbers are actual counts (not proportions or percentages);
• none of the ‘expected’ numbers is less than 5 (a design feature of the test).
Worked example 9.1 shows the use of a χ2 -test in investigating whether or not an
observed distribution of organisms is consistent with a particular theoretical ratio.
Questions of this type are quite common, and the first step is always to work out
the number of organisms expected in each category if the null hypothesis — that
the theoretical ratio holds — is true. The worked example also illustrates that,
while observed numbers of organisms must always be whole numbers, the numbers
expected on the basis of theory or prediction often come out to be fractions.

Worked example 9.1

A biologist makes the prediction that flies of type A, type B and type C will
occur in the ratio 0.16 : 0.48 : 0.36 in a wild population, if this population is in
so-called Hardy–Weinberg equilibrium. A representative sample drawn from a
population was found to contain 28 type A flies, 134 type B flies and 78 type C
flies. Was this population in Hardy–Weinberg equilibrium?

Back J I 454
Contents 
Answer
The total number of flies in the sample was 28 + 134 + 78 = 240. If the ratio
in the sample was 0.16 type A flies : 0.48 type B flies : 0.36 type C flies, then
there would be

0.16 × 240 = 38.4 type A flies

0.48 × 240 = 115.2 type B flies
0.36 × 240 = 86.4 type C flies

These are therefore the ‘expected’ numbers.

(Oi − Ei )2
A table, extended to give values for (Oi − Ei ), (Oi − Ei )2 and is given
Ei
in Table 9.5.

(Oi − Ei )2
Fly type Oi Ei (Oi − Ei ) (Oi − Ei )2
Ei
A 28 38.4 –10.4 108.16 2.817
B 134 115.2 18.8 353.44 3.068
C 78 86.4 –8.4 70.56 0.817
total 240 240 0 6.702

Table 9.5: An extended table for Worked example 9.1

Back J I 455
Contents 
The number of degrees of freedom is given by
!
number of cells containing
−1
observed numbers
=3−1
=2

Reading across the row for 2 degrees of freedom in Table 9.3, it can be seen that
the χ2 value of 6.702 corresponds to a significance level of less than 0.05 but
more than 0.01 (i.e. 0.05 > P > 0.01).
The probability that the ratio of different types of fly in the entire population
from which the sample of 240 was drawn is 0.16 type A : 0.48 type B : 0.36
type C is less than 0.05. This means that the null hypothesis (that the population
is in Hardy–Weinberg equilibrium) must be rejected at the 5% significance level.
On the basis of this investigation, it must be concluded that the population is not
in Hardy–Weinberg equilibrium.

Back J I 456
Contents 
Question 9.3 Answer
The prediction is made on the basis of theory that, if a particular genetic cross
were to be performed, the ratio of plants in the next generation should be 1
red-flowered : 2 pink-flowered : 1 white-flowered.
The next generation of a sample comprised 185 red-flowered plants, 305 pink-
flowered plants and 146 white-flowered plants. Is this data compatible with the
1 : 2 : 1 ratio predicted?

9.4 The Spearman rank correlation coefficient

In the study described in Box 9.1, soil samples were taken right across a ridge
at different horizontal distances from the crest, in order to test whether the water
content of the soil varies significantly from ridge crests to furrows. Both the original
mass of each sample and its mass after drying in an oven were measured using
a scientific balance. The water content of each sample was then expressed as a
percentage of its dry mass. For example, since the original mass of one sample was
22.85 g and its dry mass was 11.32 g, its water content was
(22.85 g − 11.32 g)
× 100% = 102%
11.32 g
(This percentage is greater than 100% because there was slightly more water than
soil in the sample.)

Back J I 457
Contents 
In fact, several soil samples (known as ‘replicate’ samples) were taken at each hori-
zontal distance, and their mean water content was calculated and used for the rest of
the investigation. Figure 9.3 shows how the mean water content of the soil samples
taken on the north-west slope of the ridge varied with horizontal distance from the
nearest ridge crest.

140
mean water content/% dry mass

120

100

80

60

40

20

0 50 100 150 200 250 300 350

horizontal distance from ridge crest/cm

Figure 9.3: Mean water content (as a percentage of dry mass) of soil samples plotted
against horizontal distance from ridge crest.

Back J I 458
Contents 
There certainly seems to be a trend of water content increasing with increasing horizontal

variable 2
distance from ridge crest. But is this trend, or apparent correlation between these two
variables, statistically significant? The strength of a possible correlation between two
variables is summarized in the value of a correlation coefficient (r). The value of r can
range from +1 (i.e. a perfect positive correlation, in which the two variables increase or
decrease precisely in step with one another; Figure 9.4a) to (a) −1 (i.e. avariable
perfect 1 negative

correlation, in which one variable increases as the other decreases and vice versa; Figure
9.4b). Where there is no correlation between two variables, the value of r is zero (Figure
9.4c). Figure 9.3 suggests that, for mean soil water content and horizontal distance from
variable 2

variable 2
nearest ridge crest, r lies somewhere between 0 and +1. However, we need to determine
the actual value of r and hence determine the probability that — for the population of all
possible soil water contents — the null hypothesis (that there is no correlation between
water content and horizontal distance
(a)
from ridge crest) is true.
variable 1 (b) variable 1
variable 2

variable 2

variable 2
(a) variable 1 (b) variable 1 (c) variable 1

Figure 9.4: (a) A perfect positive correlation between two variables (i.e. r = +1).
(b) A perfect negative correlation (i.e. r = −1). (c) No correlation (i.e. r = 0). A graph
variable 2
variable 2

with points scattered over it in a random way also represents zero correlation.

Back J I 459
Contents 
Several different sorts of correlation coefficient have been devised. In this case it is
appropriate to calculate the Spearman rank correlation coefficient (rs ). This, as the
term ‘rank’ suggests, is based on ordinal level data. The null hypothesis is that there
is no correlation between soil water content and horizontal distance from ridge crest
(i.e. rs = 0) and the alternative hypothesis that the two variables are correlated (i.e.
rs , 0).
The measurements of mean soil water content for the north-west slope of the ridge
are summarized in Table 9.6.

Horizontal distance/cm Mean water content/% dry mass

0 76
50 83
100 93
150 80
200 102
250 95
300 120
350 130

Table 9.6: Mean soil water content (as percentage of dry mass) for samples taken
at various horizontal distances from the nearest ridge crest on the north-west slope
of a ridge

Before the test statistic can be calculated, the following steps should be completed:

Back J I 460
Contents 
1. Work out the rank (order) of each of the 8 horizontal distances, (RA )i (which
will range between 1 and 8).
2. Work out the rank of each matching value for mean water content, (RB )i
(which will also range between 1 and 8).
3. Calculate each difference, Di = (RA )i − (RB )i .
4. Square each difference, to give D2i .
n
X
5. Total all the values for D2i from Step 4 to give D2i .
i =1
As an example of Steps 1 to 4, consider the horizontal distance 150 cm, which has
(RA )i = 4 and (RB )i = 2 (since its distance is fourth from the crest while its water
content is second lowest).
Therefore, Di = (RA )i − (RB )i = 4 − 2 = 2.
So D2i = 22 = 4.
The other values for D2i are shown in Table 9.7, and the total of the numbers in
n
X n
X
2
the right-hand column of this table gives Di . Notice that Di (the sum of the
i =1 i =1
differences of the ranks) should always be zero, which provides a check that the
ranks have been worked out correctly.

Back J I 461
Contents 
Horizontal Rank Mean water Rank Di = (RA )i − (RB )i D2i
distance/cm (RA )i content/% (RB )i
0 1 76 1 0 0
50 2 83 3 –1 1
100 3 93 4 –1 1
150 4 80 2 2 4
200 5 102 6 –1 1
250 6 95 5 1 1
300 7 120 7 0 0
350 8 130 8 0 0
n
X n
X
Di = 0 D2i = 8
i =1 i =1

Table 9.7: Extension of Table 9.6 to include ranks

In the case of the data in Table 9.7, it was possible to assign a unique rank to each
value for horizontal distance and mean water content, but sometimes quantities ‘tie’
(i.e. have the same rank). Worked example 9.2, at the end of this section, illustrates
what to do when this is the case.

Back J I 462
Contents 
The test statistic, the Spearman rank correlation coefficient, is rs and this is
calculated using Equation 9.2:
n
X
6 D2i
i =1
rs = 1 − (9.2)
n(n2 − 1)
n
X
where D2i is the sum of the squares of the differences of the ranks and n is
i =1
the number of pairs of measurements.
n
X
Substituting D2i = 8 (from Table 9.7) and n = 8 into Equation 9.2 gives
i =1

6×8
rs = 1 − = 0.905
8 × (82 − 1)
The final stage is to compare the value of the test statistic rs (0.905 in this case)
with the critical values listed in Table 9.8. The critical values are again given to
three places of decimals and the size of the critical values depends on both the
significance level (P = 0.1, P = 0.05 and P = 0.01, given across the top of the
table) and the number of pairs of measurements (given down the left-hand side of
the table). In this case the number of pairs of measurements is 8, and looking across
the appropriate row it can be seen that the calculated rs value of 0.905 is greater than

Back J I 463
Contents 
0.881, corresponding to a significance level of 0.01. Thus the probability, P, that
there is no correlation between water content and horizontal distance from the ridge
crest is less than 0.01; the null hypothesis must be rejected at the 1% significance
level, and the alternative hypothesis accepted. There is a statistically significant
(positive) correlation between mean soil water content and horizontal distance from
ridge crest.
It is extremely important to appreciate that even a statistically significant correlation
between two variables does not prove that changes in one variable cause changes
in the other variable.

Correlation does not imply causality.

A time-honoured, but probably apocryphal, example often cited to illustrate this

point is the statistically significant positive correlation reported for the late nine-
teenth century between the number of clergymen in England and the consumption
of alcoholic spirits. Both the increased number of clergymen and the increased
consumption of spirits can presumably be attributed to population growth (which is
therefore regarded as a ‘confounding variable’) rather than the increase in the num-
ber of clergymen being the cause of the increased consumption of spirits or vice
versa!

Back J I 464
Contents 
Precautions for the Spearman rank correlation test
Before calculating a Spearman rank correlation (rs ) it is necessary to check that:
• the data was collected at, or can be converted into, ordinal level (i.e. ranks);
• there are 7 to 30 pairs of measurements (though the test can be performed
with more than 30 pairs if you have access to a more extensive table of critical
values);
• these measurements are reasonably scattered.
Worked example 9.2 illustrates how to rank data when two or more measurements
are identical. They must be given the same mean rank, and then account must be
taken of all the identical measurements before the rank of the next, non-identical,
value is decided.
! So, if two measurements tie for first place, they are each given a
1+2
rank of = 1.5, and the next available rank is 3.
2

Back J I 465
Contents 
Worked example 9.2
The number of Stonefly nymphs counted in standard Water speed/m s−1 Number of nymphs
samples taken at 13 stations along a stream, together 0.8 35
with the water speed measured at these stations, is 1.1 28
presented in Table 9.9. Calculate the Spearman rank 0.5 11
correlation coefficient (rs ) for this data and use this to 0.7 12
determine whether there is a statistically significant 0.2 7
correlation between water speed and the number of 0.4 5
Stonefly nymphs present. 0.5 6
Answer 1.3 21
0.9 23
Table 9.10 is an extension of Table 9.9, to include val-
n
X 1.7 43
2
ues for (RA )i , (RB )i , Di , Di and D2i for the data in 0.2 10
i =1 0.1 6
this worked example. 0.7 19
Note, for example, that the water speed was mea-
Table 9.9: Number of Stonefly nymphs in re-
sured to be 0.2 m s−1 at two sampling stations, so
lation to the speed of water flow at 13 sam-
these stations ‘tie’ for second place in the ranking of
pling stations in a stream
water speed (after the station with a water ! speed of
2 + 3
0.1 m s−1 ). Each is given a rank of = 2.5, and
2
the next available rank (for the station with a water
speed of 0.4 m s−1 ) is 4.

Back J I 466
Contents 
n
X
Substituting D2i = 47.5 (from Table 9.10) and n = 13 into Equation 9.2:
i =1

6 × 47.5
rs = 1 − = 0.870
13 × (132 − 1)
Reading across the row for 12 pairs of measurements (in the absence of a row for
13 pairs) in Table 9.8, it can be seen that P < 0.01. The null hypothesis must
therefore be rejected at the 1% significance level and the alternative hypothesis
accepted. There is a statistically significant positive correlation between water
speed and number of Stonefly nymphs.

Back J I 467
Contents 
Question 9.4 Answer
Returning to the study described in Box 9.1, Figure 9.5 shows how the mean water content
of the soil samples taken from the north-west slope of the ridge varies with vertical distance
from ridge crest. Use the data given in Table 9.11 to determine whether there is a statis-
tically significant correlation between soil water content and vertical distance from ridge crest.

140 Vertical Mean water content/

distance/cm % dry mass
mean water content/% dry mass

120
0 76
100 4 83
80
7 93
9 80
60 7 102
40
11 95
10 120
20 13 130

0 2 4 6 8 10 12 14 Table 9.11: Vertical distances from the

vertical distance from ridge crest/cm nearest ridge crest and mean soil water
content (as a percentage of dry mass) for
samples taken at various horizontal dis-
Figure 9.5: Mean water content (as a percentage of dry tances from the nearest ridge crest on the
mass) of soil samples plotted against vertical distance north-west slope of a ridge
from ridge crest.

Back J I 468
Contents 
9.5 The t-test for unmatched samples
Several t-tests are widely used to test whether the means of two samples are suffi-
ciently different to conclude that the samples were probably drawn from different
populations. Such a conclusion might allow an experimenter to conclude further
that, for example, an experimental treatment did produce a statistically significant
effect compared to the experimental control (Section 9.1). t-tests are often referred
to as ‘Student’s t-tests’. This is not because people such as yourself use them a
lot — although this is true! ‘Student’ was the pseudonym used by W. S. Gossett
when he published the first version of the test in 1907. His employer, a well-known
brewing company based in Dublin, would not allow him to publish under his own
name.

Question
State the null and alternative hypotheses that would be appropriate for a t-test.
Answer
Since a t-test would be concerned with the difference between the means of two
populations (1 and 2), the appropriate null hypothesis would be either µ1 = µ2
or its equivalent µ1 − µ2 = 0 and the appropriate alternative hypothesis either
µ1 , µ2 or µ1 − µ2 , 0 (see Section 9.1).

As indicated by the section heading, the t-test introduced here is specifically for un-
matched samples. It is therefore necessary to discuss what is meant when samples

Back J I 469
Contents 
are said to be either ‘matched’ or ‘unmatched’.
The soil samples (and hence their mean water content) discussed in Section 9.4 were
uniquely matched to particular horizontal distances from the nearest ridge crest. If
data was collected from individual patients before and after they were given either
an experimental medicine or a placebo (i.e. a ‘dummy’ medicine), this data would
also be matched. Another example of matched samples would be the test scores
achieved by individual employees before and after a training event.
A typical situation in which a t-test for un- North-west slope South-east slope
matched samples would be used is if the heights (i.e. Sample 1) (i.e. Sample 2)
of two samples of Green-winged Orchids were
measured, one sample growing on the north-west x/cm 18.6 21.1
slope of a ridge and the other sample growing sn−1 /cm 5.5 3.9
on the south-east slope (Table 9.12). Since there n 14 16
is no logical connection between any one plant
Table 9.12: Mean plant height (x), estimated popu-
growing on the north-west slope and any one
lation standard deviation (sample standard deviation)
plant growing on the south-east slope, these two
of plant height (sn−1 ) and sample size (n) for a sample
samples are unmatched.
of plants growing on the north-west slope of a ridge
(Sample 1) and another sample growing on the south-
east slope (Sample 2).

Back J I 470
Contents 
Question 9.5
In each of the following cases, explain whether the samples are matched or
unmatched.
(a) A comparison is made between the heights of a sample of Answer
Green-winged Orchids growing in one nature reserve and those
of a sample growing in another nature reserve.
(b) The numbers of nymphs of two species of Stonefly are counted Answer
in each of 10 samples taken at different positions along a
stream.

In order to calculate the test statistic in this particular t-test it is necessary to solve
three equations one after another. The test statistic itself is t and this is calculated
using Equation 9.3, in which x1 and x2 are the means of the two samples, 1 and 2,
that may, or may not, have been drawn from different populations.
x1 − x2
t= (9.3)
SED
Notice that, if x1 = x2 (which would mean that x1 − x2 = 0), then t = 0. So,
if the null hypothesis were true, then it would be expected that t = 0. The term
SED represents the ‘standard error of the differences in the sample means’. SED is
calculated using Equation 9.4, in which n1 and n2 are the two sample sizes.
s
(S c )2 (S c )2
SED = + (9.4)
n1 n2
Back J I 471
Contents 
The term (S c )2 (which appears twice in Equation 9.4) represents the ‘common pop-
ulation variance’. (S c )2 is calculated using Equation 9.5, in which s1 and s2 are
the two estimated population standard deviations (also known as sample standard
deviations, as discussed in Section 8.2.7). In fact, each of s1 and s2 should really be
written as sn−1 , but if this were done the subscripts would be getting out of hand!

(n1 − 1)(s1 )2 + (n2 − 1)(s2 )2

(S c )2 = (9.5)
(n1 − 1) + (n2 − 1)
Inspection of Equation 9.3 shows that, other things being equal, the greater the
difference between x1 and x2 , the larger the value of t. In addition, if the sample
means are well separated, it seems reasonable to expect that there is likely to be a
statistically significant difference between the true means of the populations from
which the samples were drawn. Similar arguments can be used to link small values
of s1 and s2 and large values of n1 and n2 to both high values of t and an increased
likelihood of a statistically significant difference between the means from which
the samples were drawn. In general, high values of t are associated with greater
statistical significance.
Returning to the data summarized in Table 9.12, notice that the mean height of
the sample 2 (21.1 cm) is greater than that of sample 1 (18.6 cm). What needs
to be established is whether or not the difference observed (2.5 cm) is statistically
significant.

Back J I 472
Contents 
Substituting the relevant values into Equation 9.5:
(14 − 1)(5.5 cm)2 + (16 − 1)(3.9 cm)2
S c2 =
(14 − 1) + (16 − 1)
(13 × 30.25 cm2 ) + (15 × 15.21 cm2 )
=
13 + 15
= 22.193 cm2

Substituting the relevant values into Equation 9.4:

r
22.193 cm2 22.193 cm2
SED = +
14 16
= 1.724 cm

Substituting the relevant values into Equation 9.3:

18.6 cm − 21.1 cm
t=
1.724 cm
= −1.450

What does a value of t = −1.450 mean? Did the populations of plants growing
on the north-west and south-east slopes of this ridge really differ in mean height or
could the observed difference in mean height between the two samples (i.e. 2.5 cm)
have arisen by chance?
The fact that the test statistic t turns out to have a negative value can be ignored. If
it happened that the mean height of the sample of plants growing on the north-west

Back J I 473
Contents 
slope of the ridge had been deducted from that of the sample growing on the south-
east slope, rather than the other way around, then t would have been +1.450. Only
the absolute value of t (i.e. the number without its sign, in this case 1.450) is of any
consequence.
The critical values of t are given in Table 9.13.

For the t-test for unmatched samples, the number of degrees of freedom is given
by

(n1 − 1) + (n2 − 1)

Since in this case n1 is 14 and n2 is 16, the number of degrees of freedom is

(14 − 1) + (16 − 1) = 13 + 15 = 28

Reading across the row corresponding to 28 degrees of freedom to find the highest
critical value exceeded by the value of the test statistic (i.e. 1.450), it can be seen
that all that can said is that P > 0.1. Since P is not less than 0.05, the null hypothesis
(that µ1 = µ2 ) must be accepted and the alternative hypothesis (that µ1 , µ2 )
rejected. There is therefore no evidence that the samples were taken from different
populations of plants. The plants growing on the north-west and south-east slopes
of this ridge do not differ statistically significantly from one another in mean height.

Back J I 474
Contents 
Precautions for the t-test for unmatched samples
Before performing a t-test for unmatched samples it is necessary to check that:
• the samples are unmatched (if the samples are matched, then a different ver-
sion of the t-test must be used);
• population means are to be compared (different statistical tests must be used
if population modes or medians are to be compared);
• the data are at the interval level (again, different statistical tests must be used
if the data are at either ordinal or categorical level);
• there are fewer than about 25 items of data in each sample (if the samples are
larger than this, then a different — more straightforward! — statistical test
known as a z-test should be used);
• the assumption can be made that the population(s) from which the samples
were drawn have normal distributions and approximately equal standard de-
viations.

Back J I 475
Contents 
Question 9.6 Answer
Descriptive statistics on the number of flowers per plant for samples of plants
growing on the north-west and south-east slopes of another ridge are given in
Table 9.14. Is there a statistically significant difference between the slopes in
the mean number of flowers per plant?

North-west slope South-east slope

(i.e. Sample 1) (i.e. Sample 2)
x 7.7 7.2
sn−1 2.7 2.1
n 18 15

Table 9.14: Mean number of flowers per plant (x), estimated population standard
deviation of number of flowers per plant (sn−1 ) and sample size (n) for a sample
of plants growing on the north-west slope of a ridge (Sample 1) and another
sample growing on the south-east slope (Sample 2).

9.6 Other statistical tests

You have been introduced to three particular statistical hypothesis tests in Sections
9.3–9.5. Over the years, many tests have been devised to perform a wide range of

Back J I 476
Contents 
statistical tasks in the context of science. Some of these tests (for example the t-test
for matched samples and the χ2 -test for association) are similar to those covered
here, but most are designed to answer different sorts of scientific questions or to be
used in rather different circumstances.
Many excellent books have been written to help you select which particular sta-
tistical test is most appropriate for the task at hand and then guide you through
performing that test. Sections 9.1 and 9.2 of this chapter should enable you get
to grips quickly and relatively painlessly with unfamiliar statistical tests when the
time comes for you to branch out.

9.7 Learning outcomes for Chapter 9

After completing your work on this chapter you should be able to:
9.1 demonstrate understanding of the terms emboldened in the text;
9.2 propose null and alternative hypotheses in familiar circumstances;
9.3 perform a χ2 -test and interpret the results;
9.4 calculate a Spearman rank correlation coefficient (rS ) and then test its
statistical significance;
9.5 perform a t-test for unmatched samples and interpret the results.

Back J I 477
Contents 

Differentiation 10
In science, and in everyday life, we often want to know how one quantity varies
with respect to another. We may be interested in the actual value of one quantity for
a particular value of the other quantity, but it is often the rate at which one quantity
varies with respect to another that is of more importance. Consider, for example,
a small plant. The height of the plant as we look at it is of some interest, but we
also want to know whether the plant is growing, and if so, how fast. Also, is the
plant growing at an ever increasing rate or is its rate of growth slowing down? If
the growth is slowing down the plant may fit in the space we’ve made for it on the
windowsill; if the rate of growth is increasing we may need to think again!
Chapter 5 introduced the concept of the gradient of a graph as a way of finding rate
of change, whether that be positive (as in Figure 5.9), negative (Figure 5.15) or zero
(Figure 5.16 for Object B). However, Chapter 5 considered the gradient of straight-
line graphs only; we need to extend the concept to enable us to find the gradient of
curves.
Section 10.1 discusses a method for finding the gradient of a curve graphically,

Back J I 478
Contents 
by drawing a tangent to the curve at a particular point. Section 10.2 introduces
a method for deriving an equation for the gradient from the equation of the curve
itself; this method is known as differentiation. Differentiation is one of the branches
of calculus (integration, the other major branch of calculus, is beyond the scope of
this course), where the word calculus comes from the Latin for ‘a stone’ and relates
to the use of stones for counting, or calculating. This chapter is about calculating
rate of change.

10.1 Drawing tangents to curves

For a straight line, the gradient is the same at all points. However, the
gradient of a curve varies from point to point. If you look at Figure 10.1
from left to right, you will see that the slope of the curve is initially gentle;

height of plant
then it gets steeper; then it reduces again. If this graph represents the way
in which the height of our plant varies with time, this means that growth
is initially slow, before increasing to a more rapid rate and then slowing
again.
The straight lines drawn in red at various points on Figure 10.1 each have
a slope that exactly matches the slope of the curve at the point at which time

it is drawn. These lines are called tangents, and the gradient of a curve at
a point is defined to be the gradient of a tangent drawn at that point. The
word tangent comes from the Latin ‘tangere’ which means ‘to touch’, and Figure 10.1: A curve, repre-
a tangent is a line which touches the curve but doesn’t cross it. Note that senting the growth of a hypo-
thetical plant.
Back J I 479
Contents 
the use of the word ‘tangent’ here is different from its use in trigonometry (Chapter
6).
Figure 10.2 illustrates the fact that, at each point, there is only one line that touches
a smooth curve without crossing it, so each point on the curve has a unique tangent
and thus a unique gradient. This result is true for all points on all smooth curves.

tangent to
line 1 point P

P line 2

Figure 10.2: The tangent to a curve at a point P. Note that there is only one tangent
at P. Line 1, with a gradient slightly greater than that of the tangent, and line 2, with
a gradient slightly smaller than that of the tangent, both cross the curve.

Back J I 480
Contents 
A tangent is a straight line, so we can find its gradient using y
the method discussed in Chapter 5. Figure 10.3 is a graph
of y = x2 and tangents have been drawn at x = 1 and at 18
x = 3.
Using the triangle drawn on the graph, the gradient of the 16

tangent at x = 3 is
14
rise (15.0 − 9.0) 6.0
gradient = = = = 6.0
run (4.0 − 3.0) 1.0 12

Note that, because on this occasion x and y are variables

10
without units, the gradient also has no units. tangent
at x = 3
The gradient of the curve at a point is the same as that of the 8
tangent touching the curve at that point, so we can say that
the gradient of the curve at x = 3 is 6.0 to two significant 6
figures.
4

2 tangent
at x = 1

0 1 2 3 4 5 x

Figure 10.3: A graph of y = x2

Back J I 481
Contents 
Question 10.1
(a) Find the gradient of y = x2 at x = 1 by finding the gradient of Answer
the tangent which has been drawn to Figure 10.3 at x = 1.
(b) Find the gradient of y = x2 at x = 2 by drawing an additional Answer
tangent to the curve in Figure 10.3.

Box 10.1 Rate of change of concentration in chemical reactions

As a chemical reaction involving substances in solution proceeds, the concen-
trations of the substances (called ‘reactants’ and ‘products’) vary with time.
Figure 10.4 shows the way in which the concentration of one of the products
of a particular reaction increases with time. The product in this case is called a
hypobromite ion.
To find the rate of change of concentration of hypobromite ions with time at any
instant, we can draw a tangent to the curve and find its gradient. A tangent has
been drawn to the curve in Figure 10.4 at 1500 s.

2.40 × 10−3 mol dm−3 − 1.16 × 10−3 mol dm−3

The gradient of the tangent =
3000 s − 0 s
= 4.13 × 10 mol dm−3 s−1
−7

So the rate of change of concentration of hypobromite ions with time at 1500 s

is 4.13 × 10−7 mol dm−3 s−1 .

Back J I 482
Contents 
10.2 An introduction to differentiation
In answering Question 10.1 you probably realized that drawing tangents to curves
is not a very accurate way of finding gradients. Using this method, the gradient of
y = x2 at x = 2 could reasonably be anything between 3.5 and 4.5, although the
correct answer is exactly 4 (as you will discover in Section 10.2.1). Fortunately,
when the equation of the curve is known (as it is in this case), differentiation gives
us an exact method for finding the gradient, without even having to draw a graph.

10.2.1 The principles of differentiation

The reason why drawing a tangent to a curve is tricky is that, by definition, a tangent
only goes through one point on the curve and this makes it difficult to draw a line
with the correct gradient. Drawing a chord (a line between two points on the curve)
and finding its gradient is very much easier.
The chord shown joining point P and point Q in Figure 10.5 (next page) has gradient
∆y
, where ∆y is the difference between the y values of P and Q and ∆x is the
∆x
corresponding difference between x values (∆, the Greek upper case delta, is used
to indicate the change in a quantity, as discussed in Chapter 3).

Back J I 483
Contents 
As point Q moves along the curve towards P, passing
through Q1 , Q2 and Q3 , two things happen. y

1. The values of ∆x and ∆y get smaller and

smaller. Q

2. The gradient of the chord gets closer and

closer in value to the gradient of the tangent Q1

at P.
∆y
If we reduce ∆x all the way to zero, ∆y will also be Q2
tangent to
∆y point P
zero, making rather difficult to define, but we can
∆x Q3
make ∆x as small as we like in order to get an accu- P
rate measurement of the gradient. This situation is ∆x

described as a ‘limit’; as ∆x approaches zero, the ap-

∆y
proximation approaches ever closer to the exact
∆x x
gradient of the curve at the specified point. In this
∆y dy dy
limit, is written as where (said as ‘dee
∆x dx dx Figure 10.5: Finding the gradient of a curve at P.
y by dee x’) is called the derivative (or, strictly, the
‘first derivative’) of y with respect to x.
dy
Note that should be regarded as a single symbol. It does not mean a quantity
dx
dy divided by another quantity dx, and the ‘d’s are not separate quantities so they
dy y
cannot be cancelled: , .
dx x
Back J I 484
Contents 
Differentiation is simply the process of finding a derivative. Box 10.2 shows how
this can be done from first principles for the example we have been considering, y =
x2 . This box is included for interest only; you do not need to be able to differentiate
from first principles. All you need to be able to do is to apply some very simple
general rules (the first of which is discussed in Section 10.2.2, immediately after
the box) that enable you to find the derivative directly from the original equation.
dy
It turns out that for y = x2 we can say straight away that = 2x, so the gradient
dx
at x = 1 is (2 × 1) = 2, the gradient at x = 2 is (2 × 2) = 4, and the gradient at
x = 3 is (2 × 3) = 6; reassuringly these are the same results that we obtained earlier
by drawing tangents to the curve, but now the answers are exact and we have found
them without having to draw a graph.

Back J I 485
Contents 
Box 10.2 Differentiating y = x2 from first principles
Consider the chord drawn between points P and Q on Figure y

10.6. P could be any point on the curve, so its x and y values

are related by the equation y = x2 .
The x value at Q is (x + ∆x) and the y value is (y + ∆y). Since
Q
point Q lies on the curve too, we can say

(y + ∆y) = (x + ∆x)2

Multiplying out the bracket on the right-hand side, in the way

discussed in Chapter 4, gives ∆y

y + ∆y = x2 + 2x∆x + (∆x)2

Since y = x2 , we can subtract y from the left-hand side and x2

from the right-hand side to give P
∆x

∆y = 2x∆x + (∆x)2

Dividing both sides by ∆x gives x

∆y
= 2x + ∆x
∆x
Figure 10.6: Points P and Q on the
curve y = x2 .

Back J I 486
Contents 
In the limit as ∆x approaches zero, the second term on the right-hand
∆y dy
side will disappear, and will become equal to , so we can say
∆x dx
dy
= 2x
dx

10.2.2 Differentiation by rule

dy
It was shown, in Box 10.2, that the derivative of y = x2 with respect to x is = 2x.
dx
By similar methods, it can be shown that:
dy
• the derivative of y = 2 x2 with respect to x is = 4x;
dx
dy
• the derivative of y = 3 x2 with respect to x is = 6x;
dx
dy
• the derivative of y = 4 x2 with respect to x is = 8x.
dx
or, more generally, the derivative of y = C x2 with respect to x, where C is a con-
stant, is
dy
= C × 2x
dx
Back J I 487
Contents 
Similarly, it can be shown that:
dy
• the derivative of y = C x3 with respect to x is = C × 3x2 ;
dx
dy
• the derivative of y = C x4 with respect to x is = C × 4x3 ;
dx
dy
• the derivative of y = C x5 with respect to x is = C × 5x4 .
dx
These results can be summarized in the general rule

The derivative of y = C xn with respect to x is

dy
= C n xn−1
dx
where C and n are constants.

Back J I 488
Contents 
Worked example 10.1
dy
If y = x5 , what is and what is the gradient of a graph of y = x5 at x = 2?
dx
Answer
dy
In this case C = 1 and n = 5, so = 1 × 5x4 = 5x4 .
dx
dy
When x = 2, = 5 × 24 = 5 × 16 = 80.
dx
So at x = 2 the gradient of the graph is 80.

Worked example 10.2

dy
If y = 4x3 , what is and what is the gradient of a graph of y = 4x3 at x = 3?
dx
Answer
dy
In this case C = 4 and n = 3, so = 4 × 3x2 = 12x2 .
dx
dy
When x = 3, = 12 × 32 = 12 × 9 = 108.
dx
So at x = 3 the gradient of the graph is 108.

Worked example 10.3 considers the application of the rule for differentiation in the
special case when n = 1, and Worked example 10.4 considers what happens when

Back J I 489
Contents 
n = 0; you may like to think about what you expect the results to be.

Worked example 10.3

dy
If y = 4x, what is ?
dx
Answer
dy
In this case C = 4 and n = 1, so = 4 × 1x1−1 = 4x0 = 4, (since x0 = 1 for all
dx
values of x, as discussed in Section 1.3.1).

Note that y = 4x is a linear equation of the form y = kx, so the result of Worked
example 10.3 should not have surprised you; differentiating an equation of the form
y = kx will always result in a derivative which is a constant. This constant is equal
to the gradient, k, of a graph of y against x (as discussed in Section 5.3.1).

Worked example 10.4

dy
If y = 3, what is ?
dx
Answer
y = 3 can be written as y = 3x0 (since x0 = 1), so C = 3 and n = 0.
dy
Thus = 3 × 0 × x−1 = 0 (since multiplying anything by 0 gives 0).
dx

Back J I 490
Contents 
Differentiating a constant always gives zero. This should not surprise you either,
since the graph of y = 3 is a horizontal line and the gradient of a horizontal line is
always zero.

Question 10.2
Differentiate the following with respect to x and in each case find the gradient
of the graph of y against x at x = 4.
Note that the instruction ‘to differentiate’ simply requires you to find the deriva-
dy
tive .
dx
(a) y = x4 Answer
(b) y = 5x Answer
(c) y = 3x2 Answer
(d) y = 5 Answer

The rule for differentiation that we have been using applies for negative and frac-
tional values of n too, as illustrated in Worked examples 10.5 and 10.6.

Back J I 491
Contents 
Worked example 10.5
3
Differentiate y = with respect to x.
x
Answer
3
y = can be written as y = 3x−1 (see Section 1.3.1 for a reminder of the use of
x
negative exponents), so C = 3 and n = −1.
dy 3
Thus = 3 × (−1)x−1−1 = −3x−2 = − 2
dx x

Worked example 10.6

√
Differentiate y = x with respect to x.
Answer
√
y = x can be written as y = x1/2 (see Section 1.3.4), so C = 1 and n = 12 .
dy 1 1 −1 1 −1/2 1 1
Thus = x2 = x = 1/2 = √ .
dx 2 2 2x 2 x

Back J I 492
Contents 
Question 10.3
Differentiate the following with respect to x and in each case find the gradient
of the graph of y against x at x = 4.
1
(a) y = √ Answer
x
2
(b) y = Answer
x2

10.2.3 Using different symbols and different notation

So far we have found derivatives only of y with respect to x. For example we

dy dy
differentiated y = x2 and found that = 2x. Note that y and are both functions
dx dx
dy
of x; this means that the values of y and depend on the value of x. A derivative is
dx
sometimes called a ‘derived function’ because it is a function that has been derived
from another function.
Functions in science are often expressed in terms of variables other than x and y.
For example, we may know that as time, t, changes, the distance, s, of an object
from a certain position varies according to the equation s = 5t2 . The graph of this
function is illustrated in Figure 10.7.

Back J I 493
Contents 
The speed at which the object is moving is given by
the rate of change of distance with time, so to find s
the object’s speed we need to find the gradient of the
graph shown in Figure 10.7, i.e. to differentiate s 120
with respect to t.
ds 100
= 5 × 2t 2−1 = 10 t
dt

Similarly, we know from Chapter 5 that the volume, 80

V, of a gas at constant temperature is inversely pro-
portional to its pressure, P, 60

1 k
V∝ or V = = kP −1
P P 40

where k is a constant.
20
Differentiating V with respect to P gives
dV k
= k × (−1)P −1−1 = −kP −2 = − 2
dP P 0 1 2 3 4 5 t

This expression gives the gradient of the graph

shown in Figure 5.29.
Figure 10.7: A graph of s = 5t2

Back J I 494
Contents 
Question 10.4
(a) Differentiate x = t7 with respect to t. Answer
C dE
(b) If E = where C is constant, what is ? Answer
r dr

An entirely different notation, called function or prime notation, is sometimes used

for derivatives. This notation makes it very clear that both the expression being
differentiated and its derivative are functions, and it identifies the variable on which
the functions depend. In this notation, the function shown in Figure 10.7 would be
written as f (t) = 5t2 and its first derivative would be written as f 0 (t) = 10t. The
term f (t), usually said as ‘ f of t’, does not mean f times t, but simply implies that
f is a function of t. The term f 0 (t) (said as ‘ f prime of t’) is the first derivative of
f with respect to t.
dy
Unfortunately both f 0 (x) and notation are in common use, as is a variation
dx
d  2
of the latter which writes 5t = 10t for the derivative of 5t2 with respect to
dt
dy
time. This course uses only notation as discussed in the preceding sections, but
dx
you should be aware that other notations are also widely used. f 0 (x) notation is
dy
relatively modern but notation was invented by Gottfried Leibniz, one of the
dx
founders of calculus, and is known as Leibniz notation.

Back J I 495
Contents 
Yet another notation, less commonly used in modern times, writes ṡ for the first
derivative of s = 5t2 with respect to t. This notation was first used by Newton,
and the fact that we are left with such a plethora of notations for differentiation is
a lasting reminder of the bitter dispute between Newton and Leibniz over which of
them invented calculus (see Box 10.3).

Box 10.3 Newton and Leibniz: a story of reluctant publishers and letters
‘lost in the post’
Sir Isaac Newton (1642–1727) and the German mathematician and philoso-
pher Gottfried Wilhelm Leibniz (sometimes spelt Leibnitz) (1646–1716) both
claimed to have invented calculus. It is probable that they developed the ideas
independently; they certainly described their work in very different ways. New-
ton thought in terms of ‘fluxions’ whilst Leibniz used ‘differences’ (hence the
dy
word ‘differentiation’) and developed the notation still in use today.
dx
Leibniz published a paper about differentiation in 1684 and another about inte-
gration in 1686. Newton had problems getting his mathematical work into print;
the publisher of his colleague Isaac Barrow’s work had gone bankrupt and pub-
lishers were wary of mathematical works after this. Works written by Newton
in 1669 and 1671 were not published until 1711 and 1736 respectively.

Back J I 496
Contents 
Another source of the controversy seems to have been the length of time it took
for a letter to get from Newton in Cambridge to Leibniz in Paris. Newton’s letter
listed many of his results, and when Leibniz’s reply took a long time to arrive,
Newton assumed that Leibniz had spent six weeks refining his own work in the
light of Newton’s before replying. According to Leibniz the original letter had
spent these six weeks on its way from Cambridge to Paris, and he had replied
immediately he had received it.
It is beyond doubt that Newton accused Leibniz of plagiarism and that, despite
the fact that both men were well respected within their lifetimes and famous
afterwards, they ended their lives in acrimonious dispute with each other.

Back J I 497
Contents 
10.2.4 Differentiating sums

Suppose we need to differentiate y = x2 − 4x + 3, the function

y
shown in Figure 10.8, with respect to x.
It is possible to do this differentiation from first principles, 3
as shown in Box 10.4. Once again, this box is included for
interest only, as it turns out that it is possible to differentiate
y = x2 − 4x + 3 by the application of the rule already intro- 2

duced, and another simple rule which is stated after the box.
It would be possible to differentiate all functions from first
1
principles, but it is a lot quicker simply to apply the rules!
It is shown in Box 10.4 that if
0
1 2 3 4 x
y = x2 − 4x + 3 (10.1)

then −1

dy
= 2x − 4 (10.2)
dx
We could write y = x2 − 4x + 3 as y = u + v + w where u = x2 ,
v = −4x and w = 3. We know (from the rule introduced in Figure 10.8: A graph of y = x2 − 4x + 3.
Section 10.2.2) that if u = x2 , then
du
= 2x (10.3)
dx
Back J I 498
Contents 
if v = −4x, then
dv
= −4 (10.4)
dx
and if w = 3, then
dw
= 0. (10.5)
dx

Comparing Equation 10.2 with Equations 10.3, 10.4 and 10.5 shows that
dy du dv dw
= + +
dx dx dx dx
This rule is a general one, in other words:

The derivative of the sum of a number of functions is equal to the sum of the
derivatives of these functions. If

y=u+v+w

then
dy du dv dw
= + +
dx dx dx dx

Back J I 499
Contents 
Worked example 10.7
Differentiate x = t 5 + 6t 3 with respect to t.
Answer
Differentiating each of the terms separately gives
dx
= (1 × 5t 5−1 ) + (6 × 3t 3−1 )
dt
= 5t 4 + 18t 2

Question 10.5 Answer

Differentiate z = 4y 2 + y with respect to y.

10.2.5 Second derivatives

dy
gives the gradient of a graph of y against x. It is often also useful to know
dx
the rate of change of the gradient with respect to x, i.e. to differentiate again with
respect to x to find the derivative of the derivative. Such a quantity is referred to as
d2 y
the second derivative of y with respect to x and it is written 2 (said as ‘dee-2-y
dx
by dee-x-squared’) or f 00 (x) (said as ‘ f double prime of x’) in function notation.

Back J I 500
Contents 
Consider again the example used in Section 10.2.4.
We had

y = x2 − 4x + 3

and
dy
= 2x − 4
dx
dy
is itself a function of x and differentiating again gives
dx
d2 y
=2
dx2

dy d2 y
The graphs of y against x, against x and 2 against x for this example are
dx dx
dy
shown in Figure 10.10. The graph of against x shows how the gradient of the
dx
d2 y
graph of y against x varies with x, and the graph of 2 against x shows how the
dx
dy
gradient of the graph of against x varies with x. In this particular case, the graph
dx
of y against x (Figure 10.10a) is a parabola (as discussed in Section 5.4). Note that
this graph is horizontal at x = 2; at this point its gradient is zero. It should not

Back J I 501
Contents 
dy dy
surprise you, therefore, that the graph of against x (Figure 10.10b) has =0
dx dx
dy
at x = 2. Similarly, the graph of against x is a straight line of gradient 2, so the
dx
d2 y
fact that 2 has a constant value of 2 (see Figure 10.10c) should not surprise you.
dx

Question 10.6
Find the first and second derivatives of:
(a) x = 2t3 + 4t2 − 2t + 3 with respect to t; Answer
2
(b) z = with respect to y. Answer
y

Box 10.5 considers an application of differentiation to science, in this case the mo-
tion of an object falling because of the action of gravity. Note that the variables are
now real physical quantities, so they have units attached to them.

Box 10.5 Objects falling under gravity

Suppose that an object is dropped from the Clifton Suspension Bridge, which
crosses the River Avon as it flows through a gorge near Bristol. The bridge is
75 m above the river, as illustrated in Figure 10.11.

Back J I 502
Contents 
If we assume that the object starts from rest (i.e. it is dropped not thrown from
the bridge) then the distance, s, that it has travelled downwards from the bridge
in a time t is given by the equation
1
s = gt2 (10.7)
2
where g is the magnitude of the acceleration due to gravity, which we can take
to be 9.81 m s−2 .
We can differentiate Equation 10.7 twice in order to find out more about the way
the object’s speed changes as it falls. However, first let’s find the time taken for
the object to reach the river. Rearranging Equation 10.7 to make t2 the subject
gives
2s
t2 =
g
Taking the square root of both sides (recognizing that t is a period of time so we
are only interested in the positive square root) gives
s
2s
t=
g
Thus, when s = 75 m,
r
2 × 75 m p
t= = 15.29 s2 = 3.91 s = 3.9 s to two significant figures.
9.81 m s−2

Back J I 503
Contents 
So the object takes 3.9 seconds to hit the water.
The object starts from rest, but will be travelling quite fast when it hits the water.
How fast? To find speed we need to find the rate of change of distance, i.e. to
differentiate Equation 10.7 with respect to t.
The speed v is then
ds 1
v= = g × 2t = gt
dt 2
This implies that speed is proportional to time; the speed is zero as the object is
dropped but then it increases in a linear way as time increases.
Since it takes 3.9 s for the object to hit the water (or 3.91 s, working to three
significant figures to avoid rounding errors), its speed as it hits the water is

v = 9.81 m s−2 × 3.91 s = 38 m s−1 to two significant figures.

Differentiating Equation 10.7 for a second time tells us the rate at which the
object’s speed is changing. This is the object’s acceleration, a

dv d2 s
a= = 2 = g = 9.81 m s−2
dt dt
Thus the object is accelerating at 9.81 m s−2 (the acceleration due to gravity) as
you might have expected. The fact that the final answer is reasonable provides

Back J I 504
Contents 
a useful check of Equation 10.7 (which was assumed to be the correct equation
from which to start). Note that the acceleration is constant all the time the object
is falling, and the fact that acceleration is positive is consistent with the observed
fact that speed increases as the object falls.
Figure 10.12 shows the variation of the object’s distance s from the bridge, speed
v, and acceleration in a downwards direction a, with increasing time. Note that
the gradient of the first graph (s against t) leads to the second graph (v against t)
and that the gradient of the second graph (v against t) leads to the final graph (a
against t).

10.3 Differentiating exponential functions

Chapter 5 introduced graphs of exponential growth, such as n = n0 e at (see Figure
5.36) and graphs of exponential decay, such as N = N0 e−λt (see Figure 5.35),
where e is the number whose value is 2.718 to four significant figures. In general
a function of the type y = C e kx , where x and y are variables and C and k are
constants, is called an exponential function. Finding the gradient of exponential
functions reveals another reason why e is such a special number.
Figure 10.13 is a graph of the simplest imaginable exponential function; in this case
C = 1 and k = 1, so y = e x . Tangents have been drawn to the curve in Figure 10.13
at y = 1, y = 5 and y = 10.

Back J I 505
Contents 
Question
Use the tangents that have been drawn on Figure 10.13 to find the gradient of the
graph of y = e x at y = 1, y = 5 and y = 10. You should work to two significant
figures in each case.
Answer
The gradient of the tangent drawn at y = 1 is
(2.0 − 0.0) 2.0
gradient = = = 1.0
(1.0 − (−1.0)) 2.0
The gradient of the tangent drawn at y = 5 is
(9.2 − 4.2) 5.0
gradient = = = 5.0
(2.5 − 1.5) 1.0
The gradient of the tangent drawn at y = 10 is
(17.0 − 7.0) 10.0
gradient = = = 10
(3.0 − 2.0) 1.0
In each case, to two significant figures, the gradient of the tangent (and thus of
the graph itself) is equal to the value of y at the point where the tangent was
drawn.

Back J I 506
Contents 
Question
Predict the gradient of a tangent drawn to the curve in Figure 10.13 at y = 2.
Answer
It seems likely that a tangent drawn at y = 2 will have a gradient of 2 too. It
turns out that this is indeed the case.

The rule that has emerged from this sequence is generally true; the gradient of a
graph of y = e x at a particular point is equal to the value of y at that point, i.e. for
y = e x , the derivative of y with respect to x is equal to y itself:
dy
If y = e x then =y
dx
dy
or, put another way, if y = e x then = e x.
dx
dy
is only equal to y for this one specific exponential function. However, more
dx
generally,

dy
If y = C e kx , where C and k are constants, then = Ck e kx .
dx
dy dy
Since y = C e kx this means that = ky i.e. is proportional to y.
dx dx

Back J I 507
Contents 
This rule is the final rule for differentiation given in this course, and its use is illus-
dy
trated in Worked examples 10.8 and 10.9. The fact that is proportional to y for
dx
all functions of the form y = C e kx also explains the shape of graphs of exponential
growth and decay, and is the reason why e is such a special number.

Worked example 10.8

dy
If y = e 3x what is ? Express your answer (a) in terms of x, (b) in terms of y.
dx
Answer
dy
(a) C = 1 and k = 3 in this case, so = 1 × 3 e 3x = 3 e 3x .
dx
dy
(b) Alternatively, we could write = 3 y.
dx

Worked example 10.9

Find the first and second derivatives of z = 3 e−2t with respect to t.
Answer
dz
C = 3 and k = −2 in this case, so = 3 × −2e−2t = −6 e−2t .
dt
dz
This could also be written as = −2z.
dt

Back J I 508
Contents 
Differentiating again gives

d2 z
= −6 × −2 e−2t = 12 e−2t
dt2
d2 z
This could also be written as = 4z.
dt2

Question 10.7
(a) Differentiate y = 2 e x with respect to x. Answer
(b) Differentiate z = e t/2 with respect to t. Answer

dy
Since is proportional to y for all exponential functions we can use exponen-
dx
tial functions to describe situations in which the rate of change of some quantity
at an instant is proportional to the actual value of that quantity at the same in-
stant.

Radioactive decay, first described in Chapter 5 and further discussed in Box 10.6,
is an example of a situation like this and this is why it can be described by an
exponential function.

Back J I 509
Contents 
Box 10.6 Radioactive decay revisited
As previously discussed, radioactive decay is an inherently N
random process; we can never know when a particular nucleus N0

is going to decay. However, if the sample contains a large num-

ber of radioactive nuclei, the average rate at which the nuclei
decay will be proportional to the number of radioactive nuclei
remaining. It is this proportionality which leads to the process
being an exponential one, in this case an exponential decay.
t
As discussed in Chapter 5 and Chapter 7, we can describe the
process by the equation N = N0 e−λt where λ is a positive con- Figure 10.14: Radioactive decay
stant called the decay constant, N0 is the initial number of ra- can be described by the equation
dN
dioactive nuclei and N is the number remaining after time t. N = N0 e−λt , so = −λN. Tan-
The constant λ is related to the half-life, t1/2 by the equation dt
gents to the curve are shown in red.
ln 2
t1/2 = (see Box 7.5). Differentiating N = N0 e−λt with
λ
respect to t gives
dN
= −N0 λ e−λt = −λ N (since N = N0 e−λt ).
dt
In this case the gradient is negative as you would expect, since
the number of radioactive nuclei remaining is reducing with
increasing time. The larger the value of N, the faster the rate
of decay, as shown in Figure 10.14.

Back J I 510
Contents 
10.4 Learning outcomes for Chapter 10
After completing your work on this chapter you should be able to:
10.1 demonstrate understanding of the terms emboldened in the text;
10.2 find the gradient of a curve at a particular point by means of drawing a
tangent to the curve at that point;
10.3 demonstrate understanding of the fact that the derivative of a function gives
the gradient of the corresponding graph;
10.4 demonstrate understanding of the fact that the second derivative of a function
is obtained by differentiating twice;
10.5 differentiate functions of the form y = C x n ;
10.6 differentiate simple sums of functions;
10.7 differentiate exponential functions of the form y = C e kx ;
dy
10.8 demonstrate understanding of the fact that, for exponential functions, is
dx
proportional to y.

Back J I 511
Contents 

Resolving vectors A
The letter v has been used throughout the course to represent speed; but why v rather
than s? The letter v reminds us of the word ‘velocity’ which, in everyday speech,
is used interchangeably with speed. However, in science the two words have subtly
different meanings. Velocity is an example of a vector, a quantity that has direction
as well as magnitude (size). In contrast, speed is a scalar quantity; it has magnitude
only.

Question
In terms of a strict interpretation of vector and scalar quantities, what is wrong
with the statement ‘the car has a velocity of 50 km hour−1 ?
Answer
No direction has been given, so this is a scalar quantity, i.e. the speed of the car.
To turn it into a vector we would need to say, for example, that ‘the car has a
velocity of 50 km hour−1 due north’.

Back J I 512
Contents 
Another example of the difference between speed and velocity comes when con-
sidering an object orbiting another object at constant speed. Consider, for example,
the Earth orbiting the Sun at about 30 km s−1 (as discussed in Box 3.1). The Earth’s
speed relative to the Sun is approximately constant, but its direction of movement
is constantly changing, so its velocity is constantly changing too.
The quantities considered elsewhere in this course have been almost exclusively
scalars (mass, temperature, energy, magnitude of acceleration) but velocity is not
the only scientific quantity to be a vector, by a very long way. Other such quantities
include, force, weight and acceleration.
A vector may be represented diagrammatically by an arrow, the length of which
specifies the vector’s magnitude, and the direction of which is the same as the vec-
tor’s. By convention, vectors are printed as bold symbols, e.g. v, while the magni-
tude of the vector is written normally, e.g. v. Handwritten vector symbols should
be written with a wavy underline, as shown in Figure A.1.

(a) (b)

Figure A.1: Representing a vector: (a) in printed text; (b) by hand.

Back J I 513
Contents 
To specify a vector fully, both its magnitude (which is always positive) and its
direction must be stated, e.g. ‘F is a force of 10 N acting vertically downwards’.
The magnitude of F may be written as

F = |F| = 10 N

The vertical lines drawn either side of the F provide an alternative way of indicating
the modulus (magnitude) of the vector.
Adding vector quantities together is not as straightforward as adding scalar quanti-
ties, since both magnitude and direction need to be taken into account. Fortunately
the trigonometry from Chapter 6 comes to our aid.
Imagine an object being acted on by the two forces shown in Figure A.2.
You want to know the overall effect; what is the total force acting on
b
the object as a result of a and b? It is not immediately obvious how to
proceed since the two forces have different sizes and are acting in different
a
directions.

Figure A.2: Two forces a and

b acting on an object.

Back J I 514
Contents 
One way forward is to resolve each vector into compo-
nents; any two dimensional vector (such as one drawn y
on the page of a book, as here) can be characterized by
its components along two perpendicular axes. Figure A.3
shows the components of the vector a along two axes x and
y. Note that the components a x and ay are scalar quantities.
We can use trigonometry to find a x and ay .
ay a
adj
Since cos θ =
hyp
ax
we can say that cos θ = , thus a x = a cos θ.
a θ
opp ax
Similarly sin θ = , x
hyp
ay
so we can say that sin θ = , thus ay = a sin θ. Figure A.3: The x- and y-components of a.
a
If a has magnitude a = 6.0 N and acts at 60◦ to the x-axis,
we can say

a x = 6.0 N × cos 60◦ ay = 6.0 N × sin 60◦

= 3.0 N = 5.2 N

Back J I 515
Contents 
Similarly, if b has magnitude b = 2.8 N and acts at 25◦ to the x-axis, we can say

b x = 2.8 N × cos 25◦ by = 2.8 N × sin 25◦

= 2.5 N = 1.2 N

We can find the x-component of the resultant force c, by

adding the x-components of a and b: y

c x = a x + b x = 3.0 N + 2.5 N = 5.5 N

Similarly, the y-component of c is given by

cy = ay + by = 5.2 N + 1.2 N = 6.4 N

cy c
The resultant force c is shown in Figure A.4.
We can use Pythagoras’ Theorem to find the magnitude c, φ
so
cx x
c =
2
c2x + c2y
q
c= c2x + c2y Figure A.4: Finding c from its x- and y-
q components.
= (5.5 N)2 + (6.4 N)2
= 8.4 N

Back J I 516
Contents 
opp cy
And since tan φ = = we can find the angle between c and the x-axis, which
adj cx
gives us the direction in which the force acts:
6.4 N
tan φ = = 1.1636
5.5 N
Thus φ = tan−1 (1.1636) = 49◦ to two significant figures.
So the resultant force c has a magnitude of 8.4 N and acts at an angle of 49◦ to the
horizontal axis.

Back J I 517
Contents 
Question A.1 Answer
Find the x- and y-components of the vector v y
shown in Figure A.5. The vector has a magni-
tude of 8.6 m s−1 and acts at an angle, α, of 42◦
to the x-axis.

vy
v

α
vx x

Figure A.5

Back J I 518
Contents 
Question A.2 Answer
Find the magnitude and direction of the vector y
F shown in Figure A.6.
F x = 4.0 N and Fy = 3.0 N.

Fy
F

Fx x

Figure A.6

Back J I 519
Contents 

Glossary

absolute-value The absolute value of a number is the number given without its +
or − sign.
accurate Description of a set of measurements for which the systematic
uncertainty is small. Compare with precise.
acute-angle An angle of less than 90◦ .
addition rule for probabilities A rule stating that if several possible outcomes
are mutually exclusive, the probability of one or other of these outcomes
occurring is found by adding their individual probabilities.
adjacent (trigonometry) The side other than the hypotenuse which is next to a
particular angle in a right-angled triangle.
algebra The process of using symbols, usually letters, to represent quantities and
the relationships between them.
alternative hypothesis The logical ‘mirror image’ of the null hypothesis

Back J I 520
Contents 
proposed at the start of a statistical hypothesis test (e.g. that the means of
two populations are not identical, µ1 , µ2 ).
arc A portion of a curve, particularly a portion of the circumference of a circle.
arccosine See inverse cosine.
arcsec An abbreviation for ‘second of arc’. A 60th part of a minute of arc i.e. a
3600th part of a degree (of arc).
arcsine See inverse sine.
arctangent See inverse tangent.
arithmetic mean Measure of the average of a set of numbers. For a set of n
measurements of a quantity x, the arithmetic mean x (often abbreviated to
‘the mean’) is defined as the sum of all the measurements divided by the
total number of measurements:
n
1X
x= xi
n i=1

See also the true mean.

arithmetic operations The operations of addition, subtraction, multiplication and
division.
axis (of a graph) A horizontal or vertical reference line which carries a set of
divisions. In the case of a bar chart the divisions may be a list of categories.

Back J I 521
Contents 
In the case of a graph the divisions indicate a linear or logarithmic scale, and
are used to locate points on the graph.
bar chart A diagrammatic method of presenting data grouped into discrete
categories. The categories are listed along one axis (usually the horizontal
axis), and each category is represented by a bar (usually vertical). The bars
are separated by gaps, and their height (or length) is directly proportional to
the number or percentage of things or events in each category. Compare
with histogram
base number When using exponents, the quantity that is raised to a power, e.g. 5
is the base in the statement 5 × 5 × 5 = 53 and a is the base in the statement
a3 × a4 = a7 .
best-fit line A line (usually a straight line) drawn on a graph and chosen to be the
best representation of the data as a whole. A best-fit line need not
necessarily go through any of the data points (although it will typically go
through some of them), and should be drawn in such a way that there are
approximately the same number of data points above and below the line.
calculus The branch of mathematics which includes differentiation and
integration.
cancellation The process of dividing both the numerator and denominator of a
fraction by the same quantity. With numbers it may be quicker to use
cancellation than to work out the value of the numerator and denominator
separately, e.g.

Back J I 522
Contents 
5 × 13
 5
=
13 × 8 8

Cancellation is also useful in simplifying algebraic expressions or units, e.g.
abc 2 bc2

=
2
ad 2d

1 Nm 1 kg m s−2
= = 1 m s−2

1 kg × 1 
m
 1 kg
categorical level A level of measurement in which the data comprise distinct
non-overlapping classes that cannot logically be ranked (e.g. presence
versus absence, male versus female). See also ordinal level, interval level.
centi A prefix, used with units, to denote hundredths, and indicated by the symbol
c. Thus one centimetre, denoted 1 cm, is the hundredth part of a metre.
Centi is not one of the recognized submultiples in the system of SI units, but
is nevertheless in common use, especially in association with units of length
and volume.
χ2 test (chi-squared test) A statistical hypothesis test used to determine whether
there is a statistically significant association between two categorical level
variables.
chord A line drawn between two points on a curve.
common denominator The same number or term occurring as the denominator

Back J I 523
Contents 
5 7
of two or more fractions. For example, the numerical fractions 16 and 16
have the common denominator 16. It is often necessary to use equivalent
fractions in order to find common denominators: for example 52 ( = 15
6
= 12
30
8
) and 15 ( = 30
16
) have common denominators 15 and 30 (as well as many
other numbers). The algebraic fractions ab and dc have the common
denominator b × d.
common logarithm See logarithm to base 10.
commutative An operation for which the result is unchanged if the order of terms
is reversed is described as commutative. Only two of the arithmetic
operations are commutative: addition (a + b = b + a) and multiplication
(a × b = b × a).
complex number A number of the form √ n + mi, where n is any real number, m is
any non-zero real number, and i = −1.
component (of a vector) The component of a vector along a chosen axis is
obtained by drawing a line from the head of the arrow representing the
vector onto the axis, such that the line meets the axis in a right angle. For
example, the x-component of a vector a is a x = a cos θ where a is the
magnitude of the vector and θ is the angle between the x-axis and the
direction of the vector.
concentric Two circles are described as being concentric if they have the same
centre.

Back J I 524
Contents 
constant of proportionality The constant factor that is required to turn a
proportionality into an equation. The direct proportionality of y ∝ x can be
written as y = kx, where k is the constant of proportionality.
conversion factor The number by which one needs to divide or multiply in order
to convert from one unit to another.
correlation Two variables at ordinal level or interval level are said to be
correlated if, as the value of one variable increases, the value of the second
variable either increases (i.e. positive correlation) or decreases (i.e. negative
correlation). If the values of the two variables increase precisely in step with
one another, the positive correlation can be described as ‘perfect’. In a
‘perfect’ negative correlation, the value of one variable decreases precisely
as the other increases. Correlations may or may not be statistically
significant.
correlation coefficient The correlation coefficient (r) of a ‘perfect’ positive
correlation is +1, while that of a ‘perfect’ negative correlation is −1. When
there is complete lack of correlation between two variables, r = 0. For a
positive correlation that is less than ‘perfect’, 1 > r > 0. For a negative
correlation that is less than ‘perfect’, 0 > r > −1.
cosine The cosine of an angle θ in a right-angled triangle is defined by
adjacent
cos θ =
hypotenuse
where ‘adjacent’ is the length of the side adjacent to θ and ‘hypotenuse’ is

Back J I 525
Contents 
the length of the hypotenuse.
critical value At a particular number of degrees of freedom (in many statistical
hypothesis tests), the critical value is the most extreme (usually the largest,
but in some statistical tests the smallest) value that the test statistic is
expected to have for a particular significance level.
deci Prefix, used with units, to denote tenths, and indicated by the symbol d. Thus
one decibel, denoted 1 dB, is equal to one tenth of a bel. Deci is not one of
the recognized submultiples in the system of SI units, but is commonly used
in certain areas: for example the concentration of a chemical dissolved in a
solvent is often expressed in units of moles per decimetre cubed (mol dm−3 ).
decimal notation Method of representing numbers, according to which the
integral and fractional parts of a number are separated by a decimal point.
The decimal point is written as a full stop, with the integral part of the
number to the left of it. The first digit after the decimal point indicates the
number of tenths, the second indicates the number of hundredths, the third
the number of thousandths, etc.
decimal places See places of decimals.
degree (of arc) A 360th of a complete revolution.
degree-Celsius An everyday unit of temperature, given the symbol ◦ C. Pure
water freezes at 0 ◦ C and boils at 100 ◦ C. Temperatures may be converted
from degrees Celsius to the SI unit of temperature, kelvin, using the word
equation

Back J I 526
Contents 
(temperature in kelvin) = (temperature in degrees Celsius) + 273.15
degrees of freedom A device used in many statistical hypothesis tests to allow for
the fact that the more data that are collected, the more scope there is for the
test statistic to deviate from the value expected (generally, zero) if the null
hypothesis were true.
denominator The number or term on the bottom of a fraction. For example, in
1 mn
the fraction , the denominator is 2π; in the fraction , the denominator
2π pq
is pq. See also: numerator.
dependent variable A quantity whose value is determined by the value of one or
more other variables. On a graph, the dependent variable is, by convention,
plotted along the vertical axis. Compare with: independent variable.
derivative The derivative (or derived function) of a function f (x) with respect to x
is another function of x that is equal to the rate of change of f (x) with
∆f
respect to x. Its value at any given value of x is equal to the ratio in the
∆x
df
limit as ∆x becomes very small, and is usually written as or f 0 (x). The
dx
df
value of at each value of x is also equal to the gradient of the graph of f
dx
plotted against x at that value of x. A derivative of the type is sometimes
called the first derivative to distinguish it from the second derivative of the
function.

Back J I 527
Contents 
derived function See derivative.
differentiation A mathematical process that enables the derivative of a function
to be determined.
directly proportional (quantities) Two quantities x and y are said to be directly
proportional to each other if multiplying (or dividing) x by a certain amount
automatically results in y being multiplied (or divided) by the same amount.
Direct proportionality between x and y is indicated by writing y ∝ x. The
direct proportionality can also be written as an equation of form y = kx,
where k is a constant called the constant of proportionality. A graph in
which y is plotted against x will be a straight line with gradient equal to k.
See also inversely proportional.
elimination A method of combining two or more equations by eliminating
variables that are common to them.
equation An expression containing an equals sign. What is written on one side of
the equation must always be equal to what is written on the other side.
equivalent fractions Fractions that have the same value, e.g. 23 , 46 , 8 20
12 , 30 , etc.
estimated standard deviation of a population The best estimate that can be
made for the standard deviation of some quantity for a whole population.
This estimate is usually set equal to sn−1 , which is calculated from
measurements of the quantity made on an unbiased sample drawn from the
population. If the sample consists of n members and the quantity x is

Back J I 528
Contents 
measured once for each member, then
v
n
t
1 X
sn−1 = (xi − x)2
n − 1 i=1

where x is the arithmetic mean of the measurements. The symbol σn−1 is

also widely used (especially on calculators) as an equivalent to sn−1 .
evaluate An instruction to work out the value of an expression.
exponent When raising quantities to powers, the number to which a quantity is
raised, e.g. in the term 23 , the exponent is 3.
exponential decay Decay in which the time taken for a quantity to fall to half its
original value is always the same; this time is known as the half-life. A
quantity N with an initial value of N0 at time t = 0 decays exponentially if
N = N0 e−λt , where λ is a constant known as the decay constant.
exponential function A function of the type y = Cekx where C and k are
dy
constants. A function of this type has the property that is proportional to
dx
y.
exponential growth Growth in which the quantity being measured increases by a
constant factor in any given time interval. A quantity n with a starting value
of n0 at time t = 0 grows exponentially if n = n0 eat , where a is a positive
constant.

Back J I 529
Contents 
expression A combination of variables (such as a x t or u x + a x t). Unlike an
equation, an expression is unlikely to contain an equals sign.
extrapolation The process of extending a graph beyond the highest or lowest data
points in order to find the values of one or both of the plotted quantities
outside the original range within which data were obtained.
factor A term which when multiplied to other terms results in the original
expression, so 6 and 4 are factors of 24 and (a − 3) and (a + 5) are factors of
a2 + 2a − 15.
factorize To find the factors of an expression.
first derivative See derivative.
formula A rule expressed in algebraic symbols.
fraction A number expressed in the form of one integer divided by another, e.g.
1 3 21
; ; . One algebraic term divided by another may also be described as a
4 8 13
fraction. See also: improper fraction, mixed number, equivalent fractions,
numerator and denominator.
function If the value of a variable f depends on the value of another variable x,
then f is said to be a function of x and is written as f (x). In general, there is
only one value of f (x) for each value of x.
gradient (of a graph) The slope of a line on a graph. The gradient is a measure of
how rapidly the quantity plotted on the vertical axis changes in response to a

Back J I 530
Contents 
change in the quantity plotted on the horizontal axis. If the graph is a
straight line, then the gradient is the same at all points on the line and may
be calculated by dividing the vertical ‘rise’ between any two points on the
line by the horizontal ‘run’ between the same two points. If the graph is a
curved line, the gradient at any point on the curve is defined by the gradient
of the tangent to the curve at that point. See also: derivative.
graph A method of illustrating the relationship between two variable quantities
by plotting the measured values of one of the quantities using a linear or
logarithmic scale along a horizontal axis, and the measured values of the
other quantity using a linear or logarithmic scale along a vertical axis. See
also: dependent variable, independent variable, sketch graph.
half-life The time taken for half the nuclei in a radioactive sample to decay. See
also exponential decay.
histogram A diagrammatic method of presenting data, in which the horizontal
axis is divided into (usually equal) intervals of a continuously variable
quantity. Rectangles of width equal to the interval have a height scaled to
show the value of the quantity plotted on the vertical axis that applies at the
particular interval. For example, the intervals could be the months in the
year and the vertical axis could represent the mean (monthly) rainfall in
millimetres. Compare with bar chart.
hyperbola A curve, part of which may be obtained by plotting inversely
proportional quantities against each other on a .

Back J I 531
Contents 
hypotenuse The side opposite to the right-angle in a right-angled triangle.
hypothesis A plausible idea tentatively put forward to explain an observation.
Traditionally, a hypothesis is tested by making predictions that would follow
if the hypothesis is correct. If these predictions are borne out by experiment
or further observation, then this lends weight to the hypothesis but does not
prove it to be correct. If the predictions are not borne out, then the
hypothesis is either rejected or modified.
imaginary number A√number of the form mi, where m is any non-zero real
number and i = −1.
improper fraction A fraction in which the numerator is greater than the
12
denominator, e.g. . An improper fraction may also be written as a mixed
7
number.
independent variable The quantity in an experiment or mathematical
manipulation whose value(s) can be chosen at will within a given range. On
a graph, the independent variable, is by convention, plotted along the
horizontal axis. Compare with dependent variable.
index (plural indices) See exponent.
integer A positive or negative whole number (including zero).
integral Pertaining to an integer. For example the statement that m can take
integral values from −2 to +2 means that the possible values of m are −2,

Back J I 532
Contents 
−1, 0, 1 and 2.
intercept The value on one axis of a graph at which a plotted straight line crosses
that axis, provided that axis does pass through the zero point on the other
axis. If the plotted line has an equation of form y = mx + c, the intercept on
the y axis is equal to c.
interpolation The process of reading between data points plotted on a graph, in
order to find the value of one or both of the plotted quantities at intermediate
positions.
interval level A level of measurement in which the actual values of measurements
or counts are known and used in statistical analysis (e.g. dry mass in grams,
number of flowers per plant). See also categorical level, ordinal level.
inverse cosine x is the inverse cosine (arccosine) of y if x is the angle whose
cosine is y. i.e. x = cos−1 y (x = arccos y) if y = cos x.
inverse sine x is the inverse sine (arcsine) of y if x is the angle whose sine is y. i.e.
x = sin−1 y (x = arcsin y) if y = sinx.
inverse tangent x is the inverse tangent (arctangent) of y if x is the angle whose
tangent is y, i.e. x = tan−1 y (x = arctan y) if y = tan x.
inverse trigonometric function If y is a trigonometric ratio of the angle x, then x
is the inverse trigonometric function of y. For example, if y = sin x, the
inverse trigonometric function is x = sin−1 y (or arcsin y) where sin−1 y
(arcsin y) is the angle whose sine is y.

Back J I 533
Contents 
inversely proportional (quantities) Two quantities x and y are said to be inversely
proportional to each other if an increase in x by a certain factor
automatically results in a decrease in y by the same factor (e.g. if the value
of x doubles, then the value of y halves). Inverse proportionality between x
1
and y is indicated by writing y ∝ . A graph in which y is plotted against x
x
will be a hyperbola. See also: directly proportional.
irrational number A√number that cannot be obtained by dividing one integer by
another, e.g. π, 2 and e. See also rational number.
latitude Part of the specification of the position of a point on the Earth’s surface:
the distance north or south of the Equator measured in degrees. A line of
latitude is an imaginary circle on the surface of the Earth.
level of measurement The three levels of measurement that data may be known
or analysed at are categorical level, interval level or ordinal level.
linear scale A scale on which the steps between adjacent divisions correspond to
the addition or subtraction of a fixed quantity.
logarithm The logarithm of a number to a given base is the power to which the
base must be raised in order to produce the number.
logarithm to base 10 The logarithm to base 10 (or ‘common logarithm’, log10 ) of
p is the power to which 10 must be raised in order to equal p. i.e. if p = 10n ,
then log10 p = n.

Back J I 534
Contents 
logarithm to base e The logarithm to base e (or ‘natural logarithm’) of p is the
power to which e must be raised in order to equal p, i.e. if p = eq , then
ln p = q.
logarithmic scale Scale on which the steps between adjacent divisions correspond
to multiplication or division by a fixed amount, usually a power of ten.
log-linear graph A graph of the logarithm of one quantity against the actual
value of another quantity. For an exponential function of the type y = Cekx ,
graphs of log10 y against x and of ln y against x will both be straight lines.
log-log graph A graph of the logarithm of one quantity against the logarithm of
another quantity. For a function of the type y = axb (e.g. y = 2x3 ) graphs of
log10 y against log10 x and of ln y against ln x will both be straight lines.
longitude Part of the specification of the position of a point on the Earth’s
surface. A line of longitude is an imaginary semicircle that runs from one
pole to the other. The line of zero longitude passes through Greenwich in
London. Other lines of longitude are specified by the angle east or west of
the line of zero longitude.
lowest common denominator The smallest common denominator of two or more
fractions.
magnitude The size of a quantity, also referred to as the ‘modulus’. Vector
quantities have both magnitude and direction; scalar quantities have only
magnitude.

Back J I 535
Contents 
matched samples When data are collected from two samples such that each item
of data from one sample can be uniquely matched with just one item of data
from the other sample (e.g. blood glucose levels measured in individuals
before and after they have taken medication), the samples are described as
matched. See also unmatched samples.
mean Term commonly used as an abbreviation for arithmetic mean.
median The middle value in a series when the values are arranged in either
increasing or decreasing order. If the series contains an odd number of
items, the median is the value of the middle item; if it contains an even
number of items, the median is the arithmetic mean of the values of the
middle two items.
minute (of arc) A 60th part of an degree (of arc).
1
mixed number A number consisting of a non-zero integer and a fraction, e.g. 3 .
2
Any improper fraction may also be written as a mixed number: for example
8 2
=2 .
3 3
mode The most frequently occurring value in a set of data.
modulus See magnitude.
multiplication rule for probabilities A rule stating that if a number of outcomes
occur independently of one another, the probability of them all happening
together is found by multiplying the individual probabilities.

Back J I 536
Contents 
natural logarithm See logarithm to base e.
normal distribution Distribution of measurements or characteristics which lie on
a bell-shaped curve that is symmetric about its peak, with the peak
corresponding to the mean value. Repeated independent measurements of
the same quantity approximate to a normal distribution, as do quantitative
characters in natural populations (e.g. height in human beings).
null hypothesis A ‘no difference’ hypothesis proposed at the start of a statistical
hypothesis test (e.g. that the means of two populations are identical,
µ1 = µ2 ). Compare with alternative hypothesis.
numerator The number or term on the top of a fraction. For example, in the
3 a+b
fraction , the numerator is 3; in the fraction , the numerator is a + b.
4 c
See also denominator.
opposite (trigonometry) The side opposite to a particular angle in a right-angled
triangle.
order of magnitude The approximate value of a quantity, expressed as the nearest
power of ten. If the value of the quantity is expressed in scientific notation
as a × 10n , then the order of magnitude of the quantity is 10n if a < 5 and
10n+1 if a > 5. The phrase is also used to compare the sizes of quantities, as
in ‘a metre is three orders of magnitude longer than a millimetre’ or ‘a
picogram is twelve orders of magnitude smaller than a gram’.
ordinal level A level of measurement in which the data can be logically ranked

Back J I 537
Contents 
but in which the actual values of the measurements or counts are either not
known or not used in statistical analysis (e.g. tallest to shortest, heaviest to
lightest). See also categorical level, interval level.
origin (of a graph) The point on a graph at which the quantities plotted on the
horizontal axis and the vertical axis are both zero.
parabola A curve that may be described by an equation of the form
y = ax2 + bx + c, where x and y are variables, a is a non-zero constant, and b
and c are constants that may take any value.
percentage A way of expressing a fraction with a denominator of 100. For
example, 12 per cent (also written 12%) is equivalent to twelve parts per
12
hundred or .
100
places of decimals In decimal notation, the number of digits after the decimal
point (including zeroes). Thus 21.327 and 3.000 are both given to three
places of decimals.
population Statistical term used to describe the complete set of things or events
being studied.
power See exponent.
powers of ten notation A method of representing a number as a larger or smaller
number multiplied by ten raised to the appropriate power. For example,
2576 can be written in powers of ten notation as 25.76 × 102 or 2.576 × 103 ,

Back J I 538
Contents 
or 0.02576 × 105 or 257600 × 10−2 . See also scientific notation.
precise Description of a set of measurements for which the random uncertainty is
small. Compare with accurate.
probability If a process is repeated a very large number if times, then the
probability of a particular outcome may be defined in terms of results
obtained as the fraction of results corresponding to that particular outcome.
If the process has n equally likely outcomes and q of those outcomes
correspond to a particular event, then the probability of that event is defined
as q/n. There are, for example, 6 equally likely outcomes for the process of
rolling a fair die. Only one of those outcomes corresponds to the event
‘throwing a six’, so the probability of throwing a six is 16 . Five of the
outcomes correspond to the event ‘not throwing a six’, so the probability of
not throwing a six is 65 .
product The result of a multiplication operation. For example, the product of 3
and 5 is 15.
proportional See directly proportional, inversely proportional.
Pythagoras’ Theorem The square of the hypotenuse of a right-angled triangle is
equal to the sum of the squares of the other two sides.
quadratic equation An algebraic equation for x of the form ax2 + bx + c = 0,
where a , 0 and b and c can take any value. For example, 2x2 − x + 3 = 0 is
a quadratic equation.

Back J I 539
Contents 
quadratic equation formula The solutions of a quadratic equation of the form
ax2 + bx + c = 0 are given by the formula
√
−b ± b2 − 4ac
x=
2a

radian The angle subtended at the centre of a circle by an arc equal in length to
the radius. In general, the angle θ subtended by an arc length s in a circle of
s
radius r is given by θ (in radians) = .
r
random uncertainty Measured values of one quantity that are scattered over a
limited range about a mean value are said to be subject to random
uncertainty. The larger the random uncertainty associated with the
measurements, the larger will be the scatter. See also precise and systematic
uncertainty.
ratio The relationship between the sizes of two comparable quantities. For
example, if a group of 11 people is made up of 8 women and 3 men, the
ratio of women to men is said as 8 to 3 and written as 8 : 3. Ratios may be
8 8
fairly easily converted into fractions. In this particular example =
8 + 3 11
3
of the group are women and are men.
11
a
rational number Any number that can be written in the form , where a and b
b
7 −6 1 25
are integers and b , 0, e.g. 7 = ; −6 = ; − ; 3.125 = . Every
1 1 3 8
Back J I 540
Contents 
terminating or recurring decimal is a rational number. See also: irrational
number.
real number A number that can be placed on the number line. The set of real
numbers is made up of all the rational and irrational numbers.
2 3
reciprocal A term that is related to another as is related to . The reciprocal of
3 2
y x
is , and vice versa, for any non-zero values of x and y. The reciprocal of
x y
N m is N −m and vice versa.
recurring decimal A number in which the pattern of digits after the decimal
point repeats itself indefinitely. Every recurring decimal is a rational number
1
and can therefore be written as a fraction, e.g. 0.3333 . . . = ;
3
41 2345
0.123 123 123 . . . = ; 0.2345 2345 2345 . . . = .
333 9999
right angle The angle between two directions that are perpendicular (i.e. at 90◦ )
to each other.
right-angled triangle A triangle where the angle between two of the sides is a
right angle.
rounding error An error introduced into a calculation by working to too few
significant figures. To avoid rounding errors you should work to at least one
more significant figure than is required in the final answer, and just round at
the end of the whole calculation.

Back J I 541
Contents 
sample Statistical term used to describe an unbiased sub-set of a population.
sample standard deviation See estimated standard deviation of a population.
scalar A quantity with magnitude but no direction. Compare with vector.
scientific notation Method of writing numbers, according to which any rational
number can be written in the form a × 10n where a is either an integer or a
number written in decimal notation, 1 ≤ a < 10, and n is an integer. Thus
5 870 000 may be written in scientific notation as 5.87 × 106 , and 0.003 261
may be written in scientific notation as 3.261 × 10−3 . The terms ‘standard
form’ and ‘standard index form’ are equivalent to the term scientific
notation.
second (of arc) See arcsec.
df
second derivative A derivative of a derivative, for example the derivative of
dx
d2 f
with respect to x. A second derivative is usually written as or 2
or f 00 (x).
dx
SI units An internationally agreed system of units. In this system, there are seven
base units (which include the metre, kilogram and the second) and an
unlimited number of derived units obtained by combining the base units in
various ways. The system recognizes a number of standard abbreviations (of
which SI, standing for Système International, is one). The system also uses
certain standard multiples and submultiples, represented by standard
prefixes. See also centi and deci.

Back J I 542
Contents 
significance level The probability that the value of a test statistic could be as
extreme (usually as large, but in some statistical tests as small) as the value
obtained in a statistical hypothesis test if the null hypothesis were true.
significant figures The number of digits, excluding leading zeroes, quoted for the
value of a quantity, and defined as the number of digits known with certainty
plus one uncertain digit. Thus if a measured temperature is given as 23.7◦ C
(i.e. quoted to three significant figures) this implies that the first two digits
are certain, but there is some uncertainty in the final digit, so the real
temperature might be 23.6◦ C or 23.8◦ C. The larger the number of significant
figures quoted for a value, the smaller is the uncertainty in that value.
Leading zeroes in decimal numbers do not count as significant figures (e.g.
0.002 45 is expressed to three significant figures). Numbers equal to or
greater than 100 can be unambiguously expressed to two significant figures
only by the use of scientific notation (e.g. 450 can only be unambiguously
expressed to two significant figures by writing it in the form 4.5 × 102 ).
Similarly, scientific notation must be used to express numbers equal to or
greater than 1000 unambiguously to 3 significant figures.
similar Two triangles (or other objects) are described as being similar if they have
the same shape but different size.
simplify To write an equation or expression in its simplest form.
simultaneous equations Two or more equations which must hold true
simultaneously.

Back J I 543
Contents 
sine The sine of an angle q in a right-angled triangle is defined by
opposite
sin(θ) =
hypotenuse
where ‘opposite’ is the length of the side opposite θ and ‘hypotenuse’ is the
length of the hypotenuse.
sketch graph A graph drawn to illustrate the nature of the relationship between
quantities, but not involving accurate plotting. On a sketch graph the origin
is usually indicated, but the axes are not scaled.
skewed Description of distributions that are not symmetric about their mean
value.
small angle approximation For small angles (less than about 0.1 radian)
cos θ ≈ 1, and if the angle is stated in radians, sin θ ≈ θ, tan θ ≈ θ.
solution The answer, especially numerical value or values which satisfy an
algebraic equation.
solve To find an answer, usually to find the numerical values which satisfy an
algebraic equation.
Spearman rank correlation coefficient (r s ) A test statistic calculated in a
statistical hypothesis test used to determine whether or not there is a
statistically significant correlation between two ordinal level variables.
square root The number or expression that multiplied by itself gives N is called

Back J I 544
Contents 
the square root of N. The positive square root of N can be written as either
√ 1
N or N 2 .
standard deviation A quantitative measure of the spread of a set of
measurements. For n repeated measurements of a quantity, with arithmetic
mean x, the standard deviation sn is given by
v
t n
1X
sn = (xi − x)2
n i=1

The symbol σn is also widely used (especially on calculators) as an

equivalent to sn . See also: sample standard deviation, estimated standard
deviation of a population.
standard form See scientific notation.
standard index form See scientific notation.
statistically significant In science, the result of a statistical hypothesis test is
conventionally regarded as statistically significant if the probability of the
value of the test statistic being as large (or, in some statistical tests, as small)
as the one obtained is less than 0.05.
subject The term written by itself, usually to the left of the equals sign in a
mathematical equation.
subtend A straight line rotating about a certain point is said to subtend the angle
it passes through.

Back J I 545
Contents 
sum The result of an addition operation. For example, the sum of 3 and 2 is 5. A
summation sign may be used as shorthand for more complicated addition
operations, e.g.
n
X
xi = x1 + x2 + . . . + xn .
i=1

systematic uncertainty Measured values of one quantity that are consistently too
large or too small because of bias in the measuring instrument or the
measurement technique are said to be subject to systematic uncertainty. See
also accurate, random uncertainty.
t-test One of a number of statistical tests of a hypothesis used to determine
whether there is a statistically significant difference between the estimated
population means calculated from two samples. Different versions of the test
are available for matched samples and unmatched samples.
tangent (to a curved graph) The tangent to a curve at a given point P is the
straight line that just touches the curve at P and has the same gradient as the
curve at the point P.
tangent (trigonometry) The tangent of an angle θ in a right-angled triangle is
defined by
opposite
tan θ =
adjacent

Back J I 546
Contents 
where ‘opposite’ is the length of the side opposite and ‘adjacent’ is the
length of the side adjacent to θ.
term A single variable (such as v x or u x in the equation v x = u x + a x t) or a
combination of variables, such as a x t.
test of association A statistical hypothesis test used to determine whether there is
a statistically significant association between two categorical level variables
(e.g. χ2 test) or a statistically significant correlation between two variables at
ordinal level (e.g. Spearman rank correlation (r s )) or at interval level (other
correlation coefficients (r)).
test of difference A statistical hypothesis test used to test whether there is a
statistically significant difference between, for example, the estimated
population means (e.g. t-tests) or estimated population medians (other tests)
calculated from two samples.
test statistic In most statistical tests of a hypothesis, the value of a test statistic is
calculated using an equation. The value of the test statistic is then compared
with a table of critical values in order to determine whether the null
hypothesis ought to be accepted or rejected at a particular significance level.
trigonometric ratios The ratios of the sides of a right-angled triangle, including
tangent, sine, cosine.
trigonometry The branch of mathematics which deals with the relations between
the sides and angles of triangles, usually right-angled triangles.

Back J I 547
Contents 
true mean The arithmetic mean of some quantity for a whole population, usually
denoted by the symbol µ. For a large population, the true mean is generally
unknowable and the best estimate that can be made of it is the mean of the
quantity for an unbiased sample drawn from the population.
unmatched samples When data are collected from two samples such that there is
no logical connection between any particular item of data from one sample
and any particular item of data from the other sample (e.g. the heights of
plants randomly assigned to either an experimental or a control group), the
samples are described as unmatched. See also matched samples.
variable A quantity that can take a number of values.
vector A physical quantity that has a definite magnitude and points in a definite
direction.
word equation An equation in which the quantities under consideration are
described in words.

Back J 548
Contents 

Hidden material

This ‘chapter’ contains material which you won’t normally read through in se-
quence, but will access it through the links from the main text.

Back 549
Contents 
Question 1.1 (a)
(−3) × 4 = −12

Back 550
Contents 
Question 1.1 (b)
(−10) − (−5) = −5

Back 551
Contents 
Question 1.1 (c)
6 ÷ (−2) = −3

Back 552
Contents 
Question 1.1 (d)
(−12) ÷ (−6) = 2

Back 553
Contents 
Question 1.2
The lowest temperature in the oceans, which corresponds to the freezing point, is
31.9 Celsius degrees colder than the highest recorded temperature, which is
30.0 ◦ C.

Therefore, freezing point of seawater = 30.0 ◦ C − 31.9 ◦ C

= −1.9 ◦ C

Back 554
Contents 
Question 1.3 (a)
117 − (−38) + (−286) = −131

Back 555
Contents 
Question 1.3 (b)
(−1624) ÷ (−29) = 56

Back 556
Contents 
Question 1.3 (c)
(−123) × (−24) = 2952

Back 557
Contents 
Question 1.4 (a)
The lowest common denominator is 6, so
2 1 2×2 1 4 1 3
− = − = − =
3 6 3×2 6 6 6 6

Dividing top and bottom by 3 gives

3 1
=
6 2

Alternatively,

2 1 2 × 6 1 × 3 12 3 9
− = − = − =
3 6 3 × 6 6 × 3 18 18 18

Dividing top and bottom by 9 gives

9 1
=
18 2

as before.

Back 558
Contents 
Question 1.4 (b)
The lowest common denominator is 30, so
1 1 2 1 × 10 1 × 15 2 × 6
+ − = + −
3 2 5 30 30 30
10 15 12
= + −
30 30 30
13
=
30

Back 559
Contents 
Question 1.4 (c)
In this case, the lowest common denominator isn’t immediately obvious, but a
common denominator will certainly be given by the product of 3 and 28, so
5 1 5×3 1 × 28
− = −
28 3 28 × 3 3 × 28
15 28
= −
84 84
13
=−
84

Back 560
Contents 
Question 1.5 (a)
4 1
The original fraction, = = 0.25.
16 4
You may have chosen any number for your calculations. In this answer the number
2 is used, but the principles hold good whatever choice of (non-zero) number is
made.
Suppose we were to add 2 to the numerator and to the denominator
4+2 6
= = 0.333 to three places of decimals
16 + 2 18
This is not the same as the original fraction. (There is just one special case in
which this kind of operation would not change the value of the fraction and that is
adding 0 to top and bottom, which obviously leaves the fraction unchanged.)

Back 561
Contents 
Question 1.5 (b)
Suppose we were to subtract 2 from the numerator and from the denominator
4−2 2
= = 0.143 to three places of decimals
16 − 2 14
This is not the same as the original fraction. (Again, subtracting 0 from top and
bottom is the only case in which this operation leaves the fraction unchanged.)

Back 562
Contents 
Question 1.5 (c)
If we square the numerator and the denominator
4×4 16
= = 0.0625
16 × 16 256
This is not the same as the original fraction.

Back 563
Contents 
Question 1.5 (d)
If we take the square root of the numerator and of the denominator
√
4 2
√ = = 0.5
16 4
This is not the same as the original fraction.
Incidentally, checking a general rule by trying out a specific numerical example is
a helpful technique, which will be useful for algebra in Chapter 4.

Back 564
Contents 
Question 1.6 (a)
2 2×3 6
×3= =
7 7 7

Back 565
Contents 
Question 1.6 (b)
5 5 1 5×1 5
÷7= × = =
9 9 7 9 × 7 63

Back 566
Contents 
Question 1.6 (c)
1/6 1 1 1 3 3 1
= ÷ = × = =
1/3 6 3 6 1 6 2

Back 567
Contents 
Question 1.6 (d)
3 7 2 3×7×2 42
× × = =
4 8 7 4 × 8 × 7 224
Dividing top and bottom by 2, and then by 7
42 21 3
= =
224 112 16

Alternatively, the original could have been simplified in the same way before
carrying out any multiplication:

3 7
1 2
1 3
× × =
4
2 8 7
1 16

Back 568
Contents 
Question 1.7 (a)
1 1 1
2−2 = 2
= =
2 2×2 4
You might have gone one step further and expressed this in decimal notation as
0.25.

Back 569
Contents 
Question 1.7 (b)
1
= 33 = 3 × 3 × 3 = 27
3−3

Back 570
Contents 
Question 1.7 (c)
1 1
0
= =1
4 1

Back 571
Contents 
Question 1.7 (d)
1 1
4
= = 0.000 1
10 10 000

Back 572
Contents 
Question 1.8 (a)
29 = 512

Back 573
Contents 
Question 1.8 (b)
1
3−3 = = 0.037 to three places of decimals
33
It doesn’t matter if you quoted more digits in your answer than this. There is more
explanation in Chapter 2 about how and when to round off the values given on
your calculator display.

Back 574
Contents 
Question 1.8 (c)
1
= 4−2 = 0.0625
42

Back 575
Contents 
Question 1.9 (a)
230 × 22 = 2(30+2) = 232

Back 576
Contents 
Question 1.9 (b)
325 × 3−9 = 3(25+(−9)) = 316

Back 577
Contents 
Question 1.9 (c)
102 /103 = 102 ÷ 103 = 10(2−3) = 10−1 (or 1/10)

Back 578
Contents 
Question 1.9 (d)
102 /10−3 = 102 ÷ 10−3 = 10(2−(−3)) = 105
or alternatively
1
102 /10−3 = 102 × −3
= 102 × 103 = 105
10

Back 579
Contents 
Question 1.9 (e)
10−4 ÷ 102 = 10(−4−2) = 10−6

Back 580
Contents 
Question 1.9 (f)
105 × 10−2
3
= 10(5+(−2)−3) = 100 (or 1)
10

Back 581
Contents 
Question 1.10 (a)
 2
416 = 416×2 = 432

Back 582
Contents 
Question 1.10 (b)
 2
5−3 = 5(−3)×2 = 5−6
1
This could also be written as .
56

Back 583
Contents 
Question 1.10 (c)
 −1
1025 = 1025×(−1) = 10−25
1
This could also be written as .
1025

Back 584
Contents 
Question 1.10 (d)
!6
1 16 1 1
= = 3×6 = 18
33 33

6 3 3
or alternatively
!6
1  6 1
= 3−3 = 3−3×6 = 3−18 = 18
33 3

Back 585
Contents 
Question 1.11 (a)
From Equation 1.3
 1 
4× 21

24 2
=2 = 22 = 4

Back 586
Contents 
Question 1.11 (b)
From Equation 1.3
p  1 1
104 = 104 2 = 104× 2 = 102 = 100

Back 587
Contents 
Question 1.11 (c)
From Equation 1.3

1 3
3
 
100 2 = 100 2 = 103 = 1000

Alternatively

3  1  1
100 2 = 1003 2 = 106 2 = 106/2 = 103 = 1000

Back 588
Contents 
Question 1.11 (d)
1 1
125−1/3 = 1/3
= = 0.2
125 5
Since the cube root of 125 is 5.

Back 589
Contents 
Question 1.12 (a)
Multiplication takes precedence over subtraction, so

35 − 5 × 2 = 35 − (5 × 2)
= 35 − 10
= 25

Back 590
Contents 
Question 1.12 (b)
Here the brackets take precedence, so

(35 − 5) × 2 = 30 × 2
= 60

Back 591
Contents 
Question 1.12 (c)
Again, the brackets take precedence over the (implied) multiplication, so

5(2 − 3) = 5 × (−1)
= −5

Back 592
Contents 
Question 1.12 (d)
Here the exponent takes precedence:

3 × 22 = 3 × 4
= 12

Back 593
Contents 
Question 1.12 (e)
The exponent takes precedence again:

23 + 3 = 8 + 3
= 11

Back 594
Contents 
Question 1.12 (f)
Here both brackets take precedence over the (implied) multiplication:

(2 + 6)(1 + 2) = 8 × 3
= 24

Back 595
Contents 
Question 2.1 (a)

5.4 × 104 = 5.4 × 10 000

= 54 000

Back 596
Contents 
Question 2.1 (b)

1
2.1 × 10−2 = 2.1 ×
100
2.1
=
100
= 0.021

Back 597
Contents 
Question 2.1 (c)

1
0.6 × 10−1 = 0.6 ×
10
0.6
=
10
= 0.06

Back 598
Contents 
Question 2.2 (a)

215 = 2.15 × 100

= 2.15 × 102

Back 599
Contents 
Question 2.2 (b)

46.7 = 4.67 × 10
= 4.67 × 101

Back 600
Contents 
Question 2.2 (c)

152 × 103 = 1.52 × 100 × 103

= 1.52 × 102 × 103
= 1.52 × 10(2+3)
= 1.52 × 105

Back 601
Contents 
Question 2.2 (d)

8.76
0.000 0876 =
100 000
8.76
=
105
= 8.76 × 10−5

Back 602
Contents 
Question 2.3 (a)
A kilometre is 103 times bigger than a metre, so

3476 km = 3.476 × 103 km

= 3.476 × 103 × 103 m
= 3.476 × 106 m

Back 603
Contents 
Question 2.3 (b)
A micrometre is 103 times bigger than a nanometre, so

8.0 µm = 8.0 × 103 nm

Back 604
Contents 
Question 2.3 (c)
A second is 103 times bigger than a millisecond, so

0.8 s = 0.8 × 103 ms

To express this in scientific notation, we need to multiply and divide the right-hand
side by 10:

103
0.8 × 103 ms = (0.8 × 10) × ms
 10
= 8 × 103 × 10−1 ms
= 8 × 10(3−1) ms
= 8 × 102 ms

Back 605
Contents 
Question 2.4 (a)
One million = 106 , so the distance is

5900 × 106 km = 5.9 × 109 km

∼ 1010 km (or 1013 m)

Back 606
Contents 
Question 2.4 (b)
The diameter of a spherical object is given by twice its radius. So for the Sun,

diameter = 2 × 6.97 × 107 m

= 13.94 × 107 m
= 1.394 × 108 m
∼ 108 m

Back 607
Contents 
Question 2.4 (c)

2π = 2 × 3.14 (to two places of decimals)

= 6.28

This is greater than 5, so can be rounded up to the next power of ten to give the
order of magnitude, i.e. 2π ∼ 10 (or 101 ).

Back 608
Contents 
Question 2.4 (d)

7.31 × 10−26 kg ∼ 10 × 10−26 kg

∼ 10(−26+1) kg
∼ 10−25 kg

Back 609
Contents 
Question 2.5 (a)
(i) 100 m = 1 m and 10−2 m = 0.01 m, so the difference between them is
(1 − 0.01) m = 0.99 m.
(ii) 102 m = 100 m and 100 m = 1 m, so the difference between them is 99 m.
(iii) 104 m = 10 000 m and 102 m = 100 m, so the difference between them is
9900 m.
It is quite clear that as one goes up the scale the interval between each successive
pair of tick marks increases by 100 times.

Back 610
Contents 
Question 2.5 (b)
The height of a child is about 100 m, i.e. 1 m. The height of Mount Everest is
about 104 m (actually 8800 m, but it is not possible to read that accurately from
the scale on Figure 2.2). So Mount Everest is ∼104 times taller than a child.

Back 611
Contents 
Question 2.5 (c)
The length of a typical virus is 10−8 m and the thickness of a piece of paper is
10−4 m, so it would take ∼ 10−4 /10−8 = 10−4−(−8) = 10−4+8 = 104 viruses laid end
to end to stretch across the thickness of a piece of paper.

Back 612
Contents 
Question 2.6
Magnitude 7 on the Richter scale represents four points more than magnitude 3,
and each point increase represents a factor 10 increase in maximum ground
movement. So a magnitude 7 earthquake corresponds to 104 (i.e. 10 000) times
more ground movement than a magnitude 3 earthquake.

Back 613
Contents 
Question 2.7
Each of the quantities is quoted to four significant figures.

Back 614
Contents 
Question 2.8 (a)
The third digit is an 8, so the second digit must be rounded up:

−38.87 ◦ C = −39 ◦ C to two significant figures

Back 615
Contents 
Question 2.8 (b)
There is no way of expressing a number greater than or equal to 100
unambiguously to two significant figures except by the use of scientific notation.
The third digit is a 5, so again the second digit must be rounded up.

−195.8 ◦ C = −1.958 × 102 ◦ C

= −2.0 × 102 ◦ C to two significant figures

{Note that the final zero does count.}

Back 616
Contents 
Question 2.8 (c)
Again, this quantity cannot be expressed unambiguously to two significant figures
without the use of scientific notation. The third digit is an 8, so the second digit
must be rounded up.

1083.4 ◦ C = 1.0834 × 103 ◦ C

= 1.1 × 103 ◦ C to two significant figures

Back 617
Contents 
−5 × 10 2

−500 −400 −300 −200 −100 0 100 200 300 400 500

−5 × 10 0 5 × 101

−50 −40 −30 −20 −10 0 10 20 30 40 50

−5 × 10 0 −5 × 10 −1

−10 −8 −6 −4 −2 0 2 4 6 8 10

−5 × 10 −1 5 × 10 −2

−1 −0.5 0 0.5 1

Figure 2.1: Portions of the number line, showing the positions of a few large and
small numbers expressed in scientific notation.

Click on Back to return to text

Back 618
Contents 
Height of a four
year old child Height of Mt Everest
Size of a pinhead Radius of the Earth
Thickness of a Radius of the Sun
piece of paper Average distance to the
Size of a pollen grain furthest planet (Pluto)
Length of a typical virus Distance to the nearest star to
the Sun (Proxima Centauri)
Length of a coil of DNA
Diameter of the
Radius of a hydrogen atom Milky Way galaxy
Radius of a proton Distance to the edge
Quarks must be of the observable
smaller than 10−18 m Universe

10−18 10−14 10−10 10−6 10−2 102 106 1010 1014 1018 1022 1026
10−16 10−12 10−8 10−4 100 104 108 1012 1016 1020 1024 1028
Length in metres

Figure 2.2: The scale of the known Universe.

Click on Back to return to text

Back 619
Contents 
Jet taking off (at 30 m)
Pneumatic drill (at 2 m)
Threshold of pain
Rock group
Underground train

Food blender
Alarm clock
Ordinary conversation
Pedestrianized city street
Quiet whisper
Rustling leaves
Threshold of
human hearing

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

sound level in dB

Figure 2.3: Some common sounds on the decibel scale of sound level.

Click on Back to return to text

Back 620
Contents 
Question 3.1
(inch)2 , cm2 and square miles all have units of (length)2 , so they are all units of
area.
s2 cannot be a unit of area because the unit which has been squared, the second, is
a unit of time not of length.
m−2 cannot be a unit of area because the metre is raised to the power minus 2, not
2.
km3 cannot be a unit of area because the kilometre is cubed not squared. In fact, it
is a unit of volume.

Back 621
Contents 
Question 3.2 (a)
6.732
= 4.458 = 4.46 to three significant figures.
1.51
{6.732 is known to four significant figures, and 1.51 is known to three significant
figures. The number of significant figures in the answer is the same as in the input
value with the fewest significant figures, i.e. three.}

Back 622
Contents 
Question 3.2 (b)
2.0 × 2.5 = 5.0 to two significant figures.
{2.0 and 2.5 are both given to two significant figures, so the answer is given to two
significant figures too.}

Back 623
Contents 
Question 3.2 (c)
Working to three significant figures and rounding to two significant figures at the
end of the calculation gives:
!2
4.2
= (1.35)2 = 1.82 = 1.8 to two significant figures.
3.1

{Squaring is repeated multiplication, so it is reasonable to quote the final answer

to two significant figures. However, working to two significant figures throughout
introduces a sizeable rounding error and gives a final answer of 2.0.}

Back 624
Contents 
Question 3.2 (d)
The total mass = 3 × 1.5 kg = 4.5 kg.
{Note that you have exactly 3 bags of flour, so it would not be correct to round the
answer to one significant figure.}

Back 625
Contents 
Question 3.3 (a)

(3.0 × 106 ) × (7.0 × 10−2 ) = (3.0 × 7.0) × 106+(−2)

= 21 × 104
= 2.1 × 105

{Note that 21 × 104 is a correct numerical answer to the multiplication, but it is not
given in scientific notation.}

Back 626
Contents 
Question 3.3 (b)
8 × 104 8
−1
= × 104−(−1) = 2 × 105
4 × 10 4

Back 627
Contents 
Question 3.3 (c)
104 × (4 × 104 ) 104+4
−5
= 4 × −5
= 4 × 108−(−5) = 4 × 1013
1 × 10 10

Back 628
Contents 
Question 3.3 (d)
 2  2
3.00 × 108 = (3.00)2 × 108
= 9.00 × 108×2
= 9.00 × 1016

Back 629
Contents 
Question 3.4
 2
Area = 9.78 × 10−3 m
 2
= 9.78 × 10−3 m2
= 9.56 × 10−5 m2 to three significant figures.

Back 630
Contents 
Question 3.5
To one significant figure,

distance to Proxima Centauri ≈ 4 × 1016 m

distance to the Sun ≈ 2 × 1011 m

Thus,

distance to Proxima Centauri 4 × 1016 m

≈
distance to the Sun 2 × 1011 m
4 1016 m
≈ × 11
2 10 m
≈ 2 × 1016−11
≈ 2 × 105

Thus Proxima Centauri is approximately 2 × 105 times further away than the Sun.

Back 631
Contents 
Question 3.6 (a)
1 m = 100 cm, so 1 m2 = 1002 cm2
Thus 1.04 m2 = 1.04 × 1002 cm2 = 1.04 × 104 cm2

Back 632
Contents 
Question 3.6 (b)
 2
1 m = 106 µm, so 1 m2 = 106 µm2
 2
Thus 1.04 m2 = 1.04 × 106 µm2 = 1.04 × 1012 µm2

Back 633
Contents 
Question 3.6 (c)
 2
1 km = 103 m, so 1 km2 = 103 m2
1
Thus 1 m2 =
2 km2
103
1.04
and 1.04 m2 =
2 km = 1.04 × 10 km
2 −6 2
10 3

Back 634
Contents 
Question 3.7 (a)
 3
1 km = 103 m, so 1 km3 = 103 m3 = 109 m3

Volume of Mars = 1.64 × 1011 km3

= 1.64 × 1011 × 109 m3
= 1.64 × 1020 m3

Back 635
Contents 
Question 3.7 (b)
 3
1 m = 103 mm, so 1 m3 = 103 mm3 = 109 mm3
1
Thus 1 mm3 = m3 = 10−9 m3
109

Volume of ball bearing = 16 mm3

= 16 × 10−9 m3
= 1.6 × 10−8 m3

Back 636
Contents 
Question 3.8 (a)

1 m = 100 cm

So

1
1 cm = m
100

Thus

1
1 cm day−1 = m day−1
100
and
12
12 cm day−1 = m day−1
100
= 0.12 m day−1

Back 637
Contents 
Question 3.8 (b)

1 day = 24 × 60 × 60 s = 8.64 × 104 s

So

1
1 cm day−1 = cm s−1
8.64 × 104
and
12
12 cm day−1 = 4
cm s−1
8.64 × 10
= 1.4 × 10−4 cm s−1

Back 638
Contents 
Question 3.9 (a)

1
1 m = 103 mm, so 1 mm = m = 10−3 m
103
1 year = 365 × 24 × 60 × 60 s = 3.154 × 107 s

To convert from mm year−1 m s−1 we need to multiply by 10−3 (to convert the mm
to m) and divide by 3.154 × 107 (to convert the year−1 to s−1 ).

10−3
1 mm year−1 = m s−1
3.154 × 107
so
10−3
0.1 mm year−1 = 0.1 × m s−1
3.154 × 107
= 3 × 10−12 m s−1 to one significant figure

So the stalactite is growing at about 3 × 10−12 m s−1 .

Back 639
Contents 
Question 3.9 (b)

1
1 m = 100 cm, so 1 cm = m = 10−2 m
100
1 day = 24 × 60 × 60 s = 8.64 × 104 s

To convert from cm day−1 to m s−1 we need to multiply by 10−2 (to convert the cm
to m) and divide by 8.64 × 104 (to convert the day−1 to s−1 ).

10−2
1 cm day−1 = 4
m s−1
8.64 × 10

10−2
12 cm day−1 = 12 × m s−1
8.64 × 104
= 1.4 × 10−6 m s−1

So the glacier is moving at about 1.4 × 10−6 m s−1 .

Back 640
Contents 
Question 3.9 (c)

1 km = 103 m
1 Ma = 106 × 365 × 24 × 60 × 60 s = 3.154 × 1013 s

To convert from km Ma−1 to m s−1 , we need to multiply by 103 (to convert the km
to m) and divide by 3.154 × 1013 (to convert the Ma−1 to s−1 ).

103
1 km Ma−1 = 13
m s−1
3.154 × 10

103
35 km Ma −1
= 35 × 13
m s−1
3.154 × 10
= 1.1 × 10−9 m s−1 to two significant figures.

So the plates are moving apart at an average rate of 1.1 × 10−9 m s−1 .
Comparing the answers to parts (a), (b) and (c) shows that the tectonic plates are
moving apart approximately 300 times faster than the stalactite is growing. The
glacier under consideration moves about 1000 times faster still, but remember that
there is considerable variation in the speeds at which all of these processes take
place.

Back 641
Contents 
Question 3.10 (a)
1 l = 103 ml
To convert from µg l−1 to µg ml−1 we need to divide by 103 .

1
1 µg l−1 = 3
µg ml−1 = 10−3 µg ml−1
10

10 µg l−1 = 10 × 10−3 µg ml−1

= 1.0 × 10−2 µg ml−1 to two significant figures.

Back 642
Contents 
Question 3.10 (b)
Note that 10 µg l−1 = 10 µg dm−3 , since 1 litre is defined to be equal to 1 dm3
(Section 3.4.2).

1 mg = 103 µg

so

1
1 µg = mg = 10−3 mg
103

To convert from µg dm3 to mg dm3 we need to multiply by 10−3 .

1 µg dm3 = 10−3 mg dm3

10 µg dm3 = 10 × 10−3 mg dm3

= 1.0 × 10−2 mg dm3 to two significant figures.

So a concentration of 10 µg l−1 is equal to 1.0 × 10−2 mg dm3 .

Back 643
Contents 
Question 3.10 (c)
Note that 10 µg l−1 = 10 µg dm−3 .

1 g = 106 µg
1
so 1 µg = 6 g = 10−6 g
10

1 m = 10 dm
so 1 m3 = 103 dm3
1
and 1 dm3 = 3 m3 = 10−3 m3
10

To convert from µg dm−3 to g m−3 we need to multiply by 10−6 (to convert the µg
to g) and divide by 10−3 (to convert the dm−3 to m−3 ).

10−6
1 µg dm−3 = −3
g m−3
10

Back I 644
Contents 
10−6
10 µg dm−3 = 10 × g m−3
10−3
= 10 × 10−6−(−3) g m−3
= 10 × 10−3 g m−3
= 1.0 × 10−2 g m−3 to two significant figures.

So a concentration of 10 µg l−1 is equal to 1.0 × 10−2 g m−3 .

Back J 645
Contents 
Question 3.11
(i) and (iii) are equivalent. Multiplication is commutative, so x(y + z) = (y + z)x
(ii) and (v) are equivalent. Both multiplication and addition are commutative, so
xy + z = z + yx
Note that (i) is not equivalent to (ii) since, in (i), the whole of (y + z), not just y, is
multiplied by x.
Substituting x = 3, y = 4 and z = 5 gives
(i) a = x(y + z) = 3 × (4 + 5) = 27
(ii) a = xy + z = (3 × 4) + 5 = 17
(iii) a = (y + z)x = (4 + 5) × 3 = 27
(iv) a = x + yz = 3 + (4 × 5) = 23
(v) a = z + yx = 5 + (4 × 3) = 17

Back 646
Contents 
Question 3.12
The equivalent equations are (i) and (iii), since

bc2 abc2 bac2

a = =
d d d

b2 c2 b2 a2 c2
Note that only the c is squared, so (ii) m = a and (v) m = are different.
d d
abc2
Only the numerator of the fraction is multiplied by a, so (iv) m = is different
ad
too.

Back 647
Contents 
Question 3.13

NPP = 1.06 × 108 kJ

R = 3.23 × 107 kJ

From Equation 3.8,

GPP = NPP + R
= 1.06 × 108 kJ + 3.23 × 107 kJ
= 1.38 × 108 kJ to three significant figures.

Back 648
Contents 
Question 3.14
λ = 621 nm, f = 4.83 × 1014 Hz
Converting to SI base units gives

λ = 621 × 10−9 m = 6.21 × 10−7 m

f = 4.83 × 1014 Hz = 4.83 × 1014 s−1

From Equation 3.13,

v = fλ
= 4.83 × 1014 s−1 × 6.21 × 10−7 m
= 3.00 × 108 m s−1 to three significant figures.

{Note that this is the speed of light in a vacuum. Light of this frequency and
wavelength is in the red part of the visible spectrum.}

Back 649
Contents 
Question 3.15 (a)
From Equation 3.5
4
V = π r3
3
r = 6.38 × 103 km = 6.38 × 103 × 103 m = 6.38 × 106 m

So
4  3
V = π 6.38 × 106 m
3
= 1.09 × 1021 m3 to three significant figures.

The Earth’s volume is 1.09 × 1021 m3 .

Back 650
Contents 
Question 3.15 (b)
From Equation 3.18
m1 m2
Fg = G
r2
G = 6.673 × 10−11 N m2 kg−2
m1 = 5.97 × 1024 kg
m2 = 7.35 × 1022 kg

r = 3.84 × 105 km
= 3.84 × 105 × 103 m
= 3.84 × 108 m

Substituting values into the equation gives

5.97 × 1024 kg × 7.35 × 1022 kg

Fg = 6.673 × 10−11 N m2 kg−2 ×
2
3.84 × 108 m

Rearranging to collect the units together

6.673 × 10−11 × 5.97 × 1024 × 7.35 × 1022 N m2 kg−2 kg kg

Fg =
2
3.84 × 108 m2
Back I 651
Contents 
Many of the units can be cancelled
−2
6.673 × 10−11 × 5.97 × 1024 × 7.35 × 1022 N 
m2 kg

kg kg
Fg =

2 
3.84 × 108  m2

Calculating the numeric value gives

Fg = 1.99 × 1020 N to 3 significant figures.

{Note that there was no need to express the newtons in terms of base units on this
occasion; all the other units cancelled to leave N as the units of force, as expected.}
The magnitude of the gravitational force between the Earth and the Moon is
1.99 × 1020 N.

Back J 652
Contents 
nvert divide by 10 3 nvert divide by 10 3
to co to co

length in mm length in m length in km

to c 3 to c 3
o n v er t o n v er t
m u l t i pl y b y 1 0 m u l t i pl y b y 1 0

nvert divide by (10 3 2 nvert divide by (10 3 2

to co ) to co )

area area area

in mm 2 in m 2 in km2

to c 3 )2 to c 3 )2
o n v er t o n v er t
m u l t i pl y b y ( 1 0 m u l t i pl y b y ( 1 0

nvert divide by (10 3 3 nvert divide by (10 3 3

to co ) to co )

volume volume volume

in mm 3 in m 3 in km3

to c 3 )3 to c 3 )3
o n v er t o n v er t
m u l t i pl y b y ( 1 0 m u l t i pl y b y ( 1 0

Figure 3.8: Unit conversions for length, area and volume.

Click on Back to return to text

Back 653
Contents 

ux = 1.5ms−1

ax = 9.81ms−2

Figure 3.11: A stone being thrown from a cliff.

Click on Back to return to text

Back 654
Contents 
Box 3.4 Some scientific formulae

C = 2π r (3.3)

where C is the circumference of a circle of radius r.

A = π r2 (3.4)

where A is the area of a circle of radius r.

4
V = π r3 (3.5)
3
where V is the volume of a sphere of radius r.

F = ma (3.6)

where F is the magnitude of force on an object, m is its mass and a is the

magnitude of its acceleration.

Back I 655
Contents 

E = mc2 (3.7)

where E is energy, m is mass and c is the speed of light.

GPP = NPP + R (3.8)

where GPP is the gross primary production of energy by plants in an

ecosystem, NPP is net primary production and R is energy used in plant
respiration.

m
ρ= (3.9)
V
where ρ is the density of an object of mass m and volume V.

µ
r
vs = (3.10)
ρ

where vs is the speed of an S wave travelling through rocks of density ρ and

rigidity modulus µ.

Back J I 656
Contents 

P = ρgh (3.11)

where P is the pressure at depth h in a liquid of density ρ, and g is the

acceleration due to gravity.

PV = nRT (3.12)

where P is the pressure of n moles of a gas in a container of volume V held at

temperature T and R is a constant called the gas constant.

v = fλ (3.13)

where v is the speed of a wave, f is its frequency and λ is its wavelength.

q = mc ∆T (3.14)

where q is the heat transferred to an object, m is its mass, c is its specific heat
capacity and ∆T is the change in its temperature.

Back J I 657
Contents 

vi + vf
vav = (3.15)
2
where vav is average speed, vi is initial speed and vf is final speed.

vx = ux + axt (3.16)

where u x , v x and a x are respectively initial speed, final speed and acceleration,
all in the direction of the x-axis, and t is time.

1
s x = u x t + a x t2 (3.17)
2
where s x , u x and a x are respectively distance, initial speed and acceleration, all
in the direction of the x-axis, and t is time.

Back J I 658
Contents 

m1 m2
Fg = G (3.18)
r2
where Fg is the magnitude of the gravitational force between two objects of
masses m1 and m2 , a distance r apart. G is a constant called Newton’s universal
gravitational constant.

!1/2
2GM
vesc = (3.19)
R

where vesc is the escape speed, i.e. the speed with which an object must be fired
from the surface of a planet of mass M and radius R in order just to escape
from it. G is Newton’s universal gravitational constant.

d = [L/ (4π F)]1/2 (3.20)

where d is the distance at which light from a star of luminosity L has a flux
density of F.

Return to Section 3.5.2

Back J 659
Contents 
alpha A α nu (new) N ν
beta B β xi (csi) Ξ ξ
gamma Γ γ omicron O o
delta ∆ δ pi (pie) Π π
epsilon E  rho (roe) P ρ
zeta Z ζ sigma Σ σ
eta H η tau (taw) T τ
theta Θ θ upsilon Y υ
iota I ι phi (fie) Φ φ
kappa K κ chi (kie) X χ
lambda Λ λ psi Ψ ψ
mu (mew) M µ omega Ω ω

Table 3.1: The Greek alphabet. The pronunciation is given in parentheses where it
is not obvious.

Click on Back to return to text

Back 660
Contents 
Question 4.1 (a)
v = f λ can be reversed to give f λ = v.
To isolate f we need to remove λ, and f is currently multiplied by λ so, according
to Hint 3, we need to divide by λ. Remember that we must do this to both sides of
the equation, so we have
fλ v
=
λ λ
The λ in the numerator of the fraction on the left-hand side cancels with the λ in
the denominator to give
v
f =
λ

Back 661
Contents 
Question 4.1 (b)
Etot = can be reversed to give Ek + Ep = Etot .
To isolate Ek we need to remove Ep , and Ep is currently added to Ek so, according
to Hint 1, we need to subtract Ep . Remember that we must do this to both sides of
the equation, so we have

Ek + Ep − Ep = Etot − Ep
Ep − Ep = 0, so
Ek = Etot − Ep

Back 662
Contents 
Question 4.1 (c)
m m
ρ= can be reversed to give = ρ
V V
To isolate m we need to remove V, and m is currently divided by V so, according
to Hint 4, we need to multiply by V. Remember that we must do this to both sides
of the equation, so we have
mV
= ρV
V
The V in the numerator of the fraction on the left-hand side cancels with the V in
the denominator to give

m = ρV

Back 663
Contents 
Question 4.2 (a)
b = c − d + e can be written as c − d + e = b (with e on the left-hand side).
Adding d to both sides gives

c−d+e+d =b+d

i.e.

c+e=b+d

Subtracting c from both sides gives

c+e−c=b+d−c

i.e.

e = b + d − c.

Back 664
Contents 
Question 4.2 (b)
p = ρgh can be written as ρgh = p (with h on the left-hand side).
Dividing both sides by ρ gives
ρgh p
=
ρ ρ

i.e.
p
gh =
ρ

Dividing both sides by g gives

gh p
=
g ρg
i.e.
p
h=
ρg

Back 665
Contents 
Question 4.2 (c)
2GM
v2esc =
R
Multiplying both sides by R (to get R onto the left-hand side) gives
2GMR
v2esc R =
R
= 2GM

Dividing both sides by v2esc gives

v2esc R 2GM
=
v2esc v2esc

i.e.

2GM
R=
v2esc

Back 666
Contents 
Question 4.2 (d)

E = hf − φ

Adding φ to both sides (to get φ onto the left-hand side) gives

E + φ = hf − φ + φ

i.e.

E + φ = hf

Subtracting E from both sides gives

E + φ − E = hf − E

that is

φ = hf − E

Back 667
Contents 
Question 4.2 (e)
We need to start by finding an equation for c2 .
bc2 bc2
a= can be written as = a (with c on the left-hand side).
d d
Multiplying both sides by d gives
bc2 d
= ad
d
i.e.

bc2 = ad

Dividing both sides by b gives

bc2 ad
=
b b
i.e.
ad
c2 =
b
Taking the square root of both sides gives
r
ad
c=±
b

Back 668
Contents 
Question 4.2 (f)
r r
b b
a= can be written as = a (with b on the left-hand side)
c c
Squaring both sides gives
b
= a2
c

Multiplying both sides by c gives

bc
= a2 c
c

i.e.

b = a2 c

Back 669
Contents 
Question 4.3 (a)
We need to start by finding an equation for v2 .
Ek = 12 mv2 can be written as 12 mv2 = Ek . (with the v2 on the left-hand side).
Multiplying both sides by 2 gives

mv2 = 2Ek

Dividing both sides by m gives

2Ek
v2 =
m

Taking the square root of both sides gives

r
2Ek
v=±
m

but we are only interested in the positive value on this occasion.

Back 670
Contents 
Question 4.3 (b)
If Ek = 2 × 103 J and m = 4 × 1021 kg
r
2Ek
v=
m
s
2 × 2 × 103 J
=
4 × 1021 kg
s
kg m2 s−2
= 1 × 10−18
kg
= 1 × 10−9 m s−1

{At this speed, the plate would move 3 cm in a year.}

Back 671
Contents 
Question 4.3 (c)
If Ek = 2 × 103 J and m = 70 kg
r
2Ek
v=
m
s
2 × 2 × 103 J
=
70 kg
= 8 m s−1

{The sprinter, having a smaller mass, has to move rather faster than the tectonic
plate!}

Back 672
Contents 
Question 4.4 (a)
v x = u x + a x t can be written as

ux + axt = vx

Subtracting u x from both sides gives

axt = vx − ux

Dividing both sides by t gives

vx − ux
ax =
t

Back 673
Contents 
Question 4.4 (b)
µ
r
Squaring both sides of vs = gives
ρ
µ
v2s =
ρ

Multiplying both sides by ρ gives

ρ v2s = µ

Dividing both sides by v2s gives

µ
ρ=
v2s

Back 674
Contents 
Question 4.4 (c)
L
Multiplying both sides of F = by d2 gives
4π d2
L
Fd2 =
4π

Dividing both sides by F gives

L
d2 =
4π F

Taking the square root of both sides gives

r
L
d=±
4π F

{Note that if we consider just the positive value, we have arrived at Equation 3.20,
albeit written rather differently.}

Back 675
Contents 
Question 4.5 (a)
µ0 i1 i2 µ0 × i1 i2 µ0 i1 i2
× = =
2π d 2π × d 2π d

Back 676
Contents 
Question 4.5 (b)
,
3a 3a
Note that 2 means divided by 2.
2b 2b
,
3a 3a 1 3a
2= × =
2b 2b 2 4b

Back 677
Contents 
Question 4.5 (c)
The product c × b will be a common denominator, so we can write

2b 3c 2b × b 3c × c 2b2 + 3c2
+ = + =
c b c×b b×c cb
This is the simplest form in which this fraction can be expressed.

Back 678
Contents 
Question 4.5 (d)

2ab 2ac 2ab b

÷ = ×
c b c 2ac

Cancelling the ‘2a’s gives

2ab 2ac 2ab

 b b2
÷ = × = 2
c b c 2ac

 c

{Note that, for all parts of Question 4.5 and for many other questions involving
simplification, it is possible to check that the algebraic expression you end up with
is equivalent to the one that you started with by substituting numerical values for
the variables. For example, setting a = 2, b = 3 and c = 4 in the original
expression gives
! !
2ab 2ac 2×2×3 2×2×4
÷ = ÷
c b 4 3
12 16 16 3 9
= ÷ =3÷ =3× =
4 3 3 16 16
b2 32 9
Substituting the same values in the answer gives 2
= 2 = }
c 4 16

Back 679
Contents 
Question 4.5 (e)
The product f ( f + 1) will be a common denominator, so we can write
1 1 ( f + 1) f
− = −
f f +1 f ( f + 1) ( f + 1) f
f +1− f
=
f ( f + 1)
1
=
f ( f + 1)

Back 680
Contents 
Question 4.5 (f)

2b2 2c2 2A b2 (a + c)
÷ = ×
(b + c) (a + c) (b + c) 2A c2
b (a + c)
2
= 2
c (b + c)
!2
b (a + c)
The expression can be written as but cannot be simplified further.
c (b + c)

Back 681
Contents 
Question 4.6
The equation can be written as
1 1 1
= +
f u v
v u
= + (taking the product uv as the common denominator)
uv vu
v+u
=
uv
Taking the reciprocal of both sides of the equation gives
uv
f =
v+u

Back 682
Contents 
Question 4.7 (a)

1 1 1
(v x + u x ) t = v x t + u x t
2 2 2

or alternatively

1 vxt uxt vxt + uxt

(v x + u x ) t = + or
2 2 2 2

Back 683
Contents 
Question 4.7 (b)

(a − b) − (a − c) a − b − a + c
=
2 2
c−b
=
2

since a − a = 0, and −b + c is more tidily written as c − b.

Back 684
Contents 
Question 4.7 (c)

(k − 2)(k − 3) = k 2 − 3k − 2k + 6
= k 2 − 5k + 6

Back 685
Contents 
Question 4.7 (d)

(t − 2) 2 = (t − 2)(t − 2)
= t 2 − 2t − 2 t + 4
= t 2 − 4t + 4

Back 686
Contents 
Question 4.8 (a)
y2 − y = y (y − 1)

Back 687
Contents 
Question 4.8 (b)
x2 − 25 = (x + 5)(x − 5), by comparison with Equation 4.3.
We can check that the factorization is correct by multiplying the brackets out. This
gives

(x + 5)( x − 5) = x 2 − 5x + 5 x − 25
= x 2 − 25

Back 688
Contents 
Question 4.9
Both the terms on the right-hand side of Etot = 12 mv2 + mg ∆h include m, so we can
rewrite the equation as
 
Etot = m 21 v2 + g ∆h

Reversing the order gives

 
m 1 2
2v + g ∆h = Etot
 
Dividing both sides by 1 2
2v + g ∆h gives

Etot
m= 1 2
2v + g ∆h

This is a perfectly acceptable equation for m, but the fraction in the denominator
looks a little untidy. Multiplying the numerator and denominator by 2 gives

2Etot
m=
v2 + 2g ∆h

Back 689
Contents 
Question 4.10 (a)
From the answer to Question 4.7 (c)
k2 − 5k + 6 = (k − 2)(k − 3)
Thus, if k2 − 5k + 6 = 0, then (k − 2)(k − 3) = 0 too,
so k − 2 = 0 or k − 3 = 0.
i.e. k = 2 or k = 3
Checking for k = 2:
k2 − 5k + 6 = 22 − (5 × 2) + 6 = 4 − 10 + 6 = 0, as expected.
Checking for k = 3:
k2 − 5k + 6 = 32 − (5 × 3) + 6 = 9 − 15 + 6 = 0, as expected.
So the solutions of the equation k2 − 5k + 6 = 0 are k = 2 and k = 3.

Back 690
Contents 
Question 4.10 (b)
From the answer to Question 4.7 (d)
t2 − 4t + 4 = (t − 2)2
Thus, if t2 − 4t + 4 = 0, then (t − 2)2 = 0 too,
so t − 2 = 0,
i.e. t = 2.
Checking:
t = 2 gives t2 − 4t + 4 = 22 − (4 × 2) + 4 = 4 − 8 + 4 = 0, as expected.
So the solution of the equation t2 − 4t + 4 = 0 is t = 2.

Back 691
Contents 
Question 4.10 (c)
Comparison of k2 − 5k + 6 = 0 with ax2 + bx + c = 0 shows that a = 1, b = −5 and
c = 6 on this occasion, so the solutions are
√
−b ± b2 − 4ac
k=
2ap
−(−5) ± (−5)2 − (4 × 1 × 6)
=
√ 2×1
5 ± 25 − 24
=
2
5±1
=
2
5+1 6 5−1 4
so k = = = 3 or k = = = 2.
2 2 2 2
So the solutions of the equation k2 − 5k + 6 = 0 are k = 2 and k = 3. This is the
same answer as was obtained in part (a) and could be checked in the same way.

Back 692
Contents 
Question 4.10 (d)
Comparison of t2 − 4t + 4 = 0 with ax2 + bx + c = 0 shows that a = 1, b = −4 and
c = 4 on this occasion, so the solutions are
√
−b ± b2 − 4ac
k=
2ap
−(−4) ± (−4)2 − (4 × 1 × 4)
=
√ 2×1
4 ± 16 − 16
=
2
4±0
=
2
=2

So there is just one solution to t2 − 4t + 4 = 0; namely t = 2. This is the same

answer as was obtained in part (b) and could be checked in the same way.

Back 693
Contents 
Question 4.11 (a)
Rearranging p = mv to make v the subject gives
p
v= (dividing both sides by m)
m

Substituting in Ek = 12 mv2 gives

 p 2
Ek = 12 m
m
p2
= 12 m 2
m
p 2
=
2m

Back 694
Contents 
Question 4.11 (b)
Since both equations are already written with E (the variable we are trying to
eliminate) as the subject, we can simply set the two equations for E equal to each
other:
1
2 mv
2
= mg ∆h

There is an m on both sides of the equation; dividing both sides of the equation by
m gives

1 2
2v = g ∆h

Multiplying both sides of the equation by 2 gives

v2 = 2g ∆h

Taking the square root of both sides of the equation gives

v = ± 2g ∆h
p

Back 695
Contents 
Question 4.11 (c)
Rearranging c = f λ to make f the subject gives
c
f = (dividing both sides by λ)
λ

Substituting in Ek = h f − φ gives

hc
Ek = −φ
λ

Adding φ to both sides of the equation gives

hc
Ek + φ =
λ

Subtracting Ek from both sides gives

hc
φ= − Ek
λ

Back 696
Contents 
Question 4.12
Let the number selected be represented by x:
Adding 5 gives x+5
Doubling the result gives 2(x + 5) = 2x + 10
Subtracting 2 gives (2x + 10) − 2 = 2x + 8
2x + 8
Dividing by 2 gives = x+4
2
Taking away the number you first thought of gives (x + 4) − x = 4.

Back 697
Contents 
Question 4.13

Let H represent Helen’s height in cm and T represent Tracey’s height

Helen
in cm. Since Tracey is 15 cm taller than Helen we can write

T = H + 15 (i)

The height of the wall is equal to Tracey’s height up to her shoulders wall H H − 10
(T − 25) plus Helen’s height up to her eyes (H − 10), thus

(T − 25) + (H − 10) = 300 (ii)

Simplifying (ii) gives

T
T + H − 35 = 300 T − 25

Adding 35 to both sides gives

Tracey
T + H = 335

Substituting for T from (i) gives

(H + 15) + H = 335
2H + 15 = 335

Back I 698
Contents 
Subtracting 15 from both sides gives

2H = 320

Dividing both sides by 2 gives

H = 160

i.e. Helen is 160 cm tall.

Back J 699
Contents 
Question 4.14
The equations required are Eg = mg ∆h (Equation 4.18) and Ek = 21 mv2 (Equation
4.17).
Assuming that the child’s gravitational potential energy is converted into kinetic
energy, Ek = Eg .
1
2 mv
2
= mg ∆h

Dividing both sides by m gives

1 2
2v = g ∆h

Multiplying both sides by 2 gives

v2 = 2g ∆h

Taking the square root of both sides gives

v = ± 2g ∆h
p

On this occasion we are only interested in the positive square root, i.e. v = 2g ∆h
p

Back I 700
Contents 
Substituting ∆h = 1.8 m and g = 9.81 m s−2 gives
p
v = 2 × 9.81 m s−2 × 1.8 m
= 5.9 m s−1 to two significant figures

√
(noting that m2 s−1 = m s−1 ).
Checking
The units have worked out to be m s−1 , as expected.
An estimated value is
p
v ≈ 2 × 10 m s−2 × 2 m
p
≈ 40 m2 s−2
√ √
≈ 6 m s−1 , since 40 ≈ 36

The speed seems quite high; in reality not all of the child’s gravitational potential
energy would be converted into kinetic energy.

Back J 701
Contents 
c a+b
c + 50 = a + b + 50 =
c=a+b 2 2

(a) (b) (c)

Figure 4.1: (a) The analogy between an equation and a set of kitchen scales. The
scales remain balanced if (b) 50 g is added to both sides or if (c) the weight on both
sides is halved.

Click on Back to return to text

Back 702
Contents 

high luminosity
supergiants

luminosity in W
Alcyone red
giants
Sirius A
m
ai
n
se
qu Sun
en
ce
α Centauri B
Sirius B
low luminosity

white dwarfs

high temperature low temperature

photospheric temperature in K

Figure 4.2: A Hertzsprung–Russell diagram showing the Sun and a number of other
stars.
Click on Back to return to text

Back 703
Contents 

P−wave arrival S−wave arrival

20seconds time

Figure 4.4: Seismogram recorded at the British Geological Survey in Edinburgh on

12 September 1988 at 2.23 p.m.

Click on Back to return to text

Back 704
Contents 
Question 5.1
(a) The red lines on the graph show that, by interpolation, when current = 1.5 A
then voltage = 2.0 V.
(b) The line through the data points can be extended at each end, as shown below.
This process of extrapolation to the vertical axis shows that when the current is
zero the voltage has a value of 5.0 V.
(c) Extrapolation to the horizontal axis shows that when the voltage is zero the
current has a value of 2.5 A.

5.0

4.0
voltage/V

3.0

2.0

1.0

0 0.5 1.0 1.5 2.0 2.5

current/A

Back 705
Contents 
Question 5.2

Using the red lines on Figure 5.38,

distance from epicentre/km

rise 200
gradient =
run 170
(170 − 10) km
=
(32 − 4) s 100
160 km
=
28 s
= 5.7 km s−1 10
0 10 20 30 40
4 32
Therefore the average speed of the travel time after earthquake occurred/s
P waves is 5.7 km s−1 (to two sig-
nificant figures). Figure 5.38

{You may have chosen different points from which to calculate your gradient, but
you should still have got the same answer. Note that the scale of the graph does
not really allow points to be specified to more than two significant figures, so this
is the precision to which the answer should be given.}

Back 706
Contents 
Question 5.3

Using the red lines on Figure 5.39,

rise 40
gradient =

travel time after earthquake occurred/s

run
(32 − 2) s
=
(170 − 0) km 32
30
30 s
=
170 km
= 0.176 s km−1
20
Therefore
1
speed =
0.176 s km−1 10
= 5.7 km s−1 (to two significant figures)

To the precision to which it is possible to read the

graph, this is the same value as before.
0 100 200
170
distance from epicentre/km

Figure 5.39

Back 707
Contents 
Question 5.4

Using the red lines on Figure 5.40,

rise 20
gradient =

temperature/°C
run
(7 − 20) ◦ C
=
(2 − 0) km 10
−13 ◦ C
= 7
2 km
= −6.5 ◦ C km−1
0 1 2
This could also have been written as −6.5 ◦ C/km. The negative altitude/km
value of the gradient implies that temperature decreases with in-
creasing height above sea-level and your sentence should reflect Figure 5.40
this. For example you could write: ‘For each successive kilome-
tre of height gained above sea-level, the atmospheric temperature
falls by 6.5 ◦ C’.

Back 708
Contents 
Question 5.5
Using the red lines on Figure 5.41,
(−50 − (−20)) ◦ C
gradient = 10
(10 − 5.5) km 5 10
−30 ◦ C 0
= altitude/km
4.5 km
= −6.7 ◦ C km−1 –10

temperature/°C
–20
This agrees quite well with the value obtained in
the answer to Question 5.4. In fact temperature –30
does decrease with altitude at an almost constant
rate through the troposphere. –40

–50

–60

Figure 5.41

Back 709
Contents 
Question 5.6
The line corresponding to v = rz has the larger v
(steeper) gradient. Therefore r is larger than s.

v = rz

v = sz

0 z

Back 710
Contents 
Question 5.7
If two quantities are directly proportional to each other, a graph in which one is
plotted against the other will be a straight line through the origin. Therefore, only
(c) corresponds to a proportional relationship: u ∝ z. In this case, the gradient is
negative, i.e. the constant of proportionality is negative.

f a u

0 z

0 g 0 b

(a) (b) (c)

Back 711
Contents 
Question 5.8
Since M is directly proportional to d3 , these are the quantities to plot. The spheres
are selected and then their masses are measured, so d is the independent variable,
and so according to convention d3 should therefore be plotted on the horizontal
axis. In other words, the convention would be to plot M against d3 .
Slightly rearranging the equation and comparing with the standard equation of a
straight line

y = m x + c
πρ
M = d 3 (+ 0)
6
πρ
shows that the gradient would be .
6
{If you chose to defy convention and plot d3 against M, the gradient would have
6
the reciprocal value, i.e. .}
πρ

Back 712
Contents 
Question 5.9
There are at least two equally valid ways to plot the data here. Since
s
L
T = 2π
g

squaring both sides gives

4π2 L
T =
2
g
L is the independent variable, which according to convention should be plotted on
4π2
the horizontal axis. A graph of T 2 against L has gradient = so
g

4π2
g=
gradient
√
Alternatively, you could have chosen to plot T against L. The gradient of this
2π √ 2π
line would be √ . So g = and
g gradient

4π2
g=
(gradient)2

Back 713
Contents 
Question 5.10
 n
1
After n half-lives, the number of radioactive atoms is reduced to 2 of the
original number.
!4
1 1
Since =
16 2
1
four half-lives must elapse before the number of radioactive atoms will be 16 of
the number there are today. So 4 × 1600 years = 6400 years must elapse for this to
happen.

Back 714
Contents 
N

N0 Figure 5.35: Radioactive

decay.
N0 radioactive nuclei are
present at time t = 0.
During each half-life, the
t 1/2 = half-life number of radioactive nu-
clei is halved. The half-life
is denoted by the symbol
N1 = N0 × 1
2
t1/2 .

N2 = N1 × 1
2

N3 = N2 × 1
2

0 t 1/2 2t 1/2 3t 1/2 time t

Click on Back to return to text

Back 715
Contents 
Box 5.3 Sea-floor spreading
Plate tectonics describes how the outer layer of the Earth is made up of a series of
‘plates’ which move relative to one another. The top layer of these plates is known
as the ‘crust’. Ocean crust is about 7 km thick, continental crust up to 80 km thick.
The crust is split at mid-ocean ridges and material is ejected at the ridge crests to
form new sea-floor. This creation of new crust is balanced at the opposite end of the
plate by material being forced under an adjacent plate. As eruption at a mid-ocean
ridge continues, older sea-floor crust is moved aside to make way for younger crust
and the sea-floor ‘spreads’ symmetrically away from the ridge as shown in Figure
5.8a. Recently formed ocean crust is largely inaccessible, so scientists interested in the
speed at which this spreading occurs have to resort to indirect means of measuring it.

ocean

material comes out of the

ocean crust mid-ocean ridge to form new ocean crust
moves to the left ocean crust (sea floor). moves to the right

Figure 5.8a: Production of new sea-floor at a mid-ocean ridge.

Back I 716
Contents 
A record of their age is held in rocks by their magnetism. The orientation of the Earth’s
magnetic field has reversed at irregular intervals throughout its history, and the orienta- 0
tion of the magnetic field at the time a rock was formed is ‘locked into’ the rock. The
times at which these changes in orientation took place are known from measurements on
a great many surface rocks that can be dated by a variety of means. 1

Figure 5.8b shows the timescale for reversals in the Earth’s magnetic field over the last

age/Ma
4 Ma. Black denotes ‘normal’ polarity (i.e. what we experience today) and white de- 2

notes reversed polarity.

Marine magnetic surveys reveal patterns in the orientation of the magnetization of rocks 3

near mid-ocean ridges; an idealized pattern is shown in Figure 5.8c. Correlation of

patterns like this with the pattern in Figure 5.8b provides a method of dating the rocks at
4
various distances from a mid-ocean ridge.

magnetic
stripes mid-ocean
Figure 5.8c: Idealized symmetrical Figure 5.8b
ridge ‘magnetic stripes’ either side of a
ridge; rocks shown black are magne-
tized in the opposite direction to those
shown white and can be matched with
the timescale of Figure 5.8b

Return to Section 5.2.1

Back J 717
Contents 
Question 6.1 (a)

2π radians = 360◦
360◦
so 1 radian =
2π

360◦
0.123 radians = 0.123 ×
2π
◦
≈ 7.05 to three significant figures.

Back 718
Contents 
Question 6.1 (b)
2π radians = 360◦
2π 360◦
so radians = = 120◦
3 3

Back 719
Contents 
Question 6.1 (c)
2π radians = 360◦
so π radians = 180◦

3π 3 × 180◦ 540◦
radians = = = 270◦
2 2 2

Back 720
Contents 
Question 6.2 (a)
360◦ = 2π radians
2π
so 1◦ = radians
360
2π
36.5◦ = 36.5 × radians
360
≈ 0.637 radians to three significant figures.

Back 721
Contents 
Question 6.2 (b)
360◦ = 2π radians
2π
so 1◦ = radians
360
2π
90◦ = 90 × radians
360
π
= radians.
2

{This answer could have been written as 1.57 radians (to 3 significant figures), but
π
note that radians is an exact answer which 1.57 radians is not.}
2

Back 722
Contents 
Question 6.2 (c)
360◦ = 2π radians
2π
so 1◦ = radians
360
2π
210◦ = 210 × radians
360
7π
= radians.
6

{This answer could have been written as 3.67 radians (to 3 significant figures), but
7π
note that radians is an exact answer which 3.67 radians is not.}
6

Back 723
Contents 
Question 6.3 (a)
We are trying to find length a in the diagram. wall

From Pythagoras’ Theorem

a2 + (1.15 m)2 = (4.50 m)2 θ
so
a2 = (4.50 m)2 − (1.15 m)2
p 4.50m
a = 20.25 m2 − 1.3225 m2 a

= 4.35 m to three significant figures. ladder

75.2°

ground 1.15m

Back 724
Contents 
Question 6.3 (b)
We are trying to find angle θ in the diagram. wall

The interior angles in a triangle add up to 180◦ so

θ + 75.2◦ + 90◦ = 180◦ θ
i.e.

θ = 180◦ − 75.2◦ − 90◦

4.50m
= 14.8◦ . a

ladder

75.2°

ground 1.15m

Back 725
Contents 
Question 6.4 (a)
sin 49◦ = 0.7547

Back 726
Contents 
Question 6.4 (b)
π
cos = 0.9239
8
{Since the angle was given in radians, your calculator needs to be in ‘radians
mode’ in order to obtain the correct answer to this part.}

Back 727
Contents 
Question 6.4 (c)
π
tan =1
4
{Since the angle was given in radians, your calculator needs to be in ‘radians
mode’ in order to obtain the correct answer to this part.}

Back 728
Contents 
Question 6.5 (a)
cos−1 (0.5253) = 58.31◦
{Your calculator needs to be in ‘degrees mode’ in order to obtain the correct
answer.}

Back 729
Contents 
Question 6.5 (b)
tan−1 (1.5574) = 1.0000 radians
{Your calculator needs to be in ‘radians mode’ in order to obtain the correct
answer.}

Back 730
Contents 
Question 6.6 (a)
adj
cos θ =
hyp
so
4.3 m
cos 32◦ =
h
Multiplying both sides by h gives

h cos 32◦ = 4.3 m

Dividing both sides by cos 32◦ gives

4.3 m
h=
cos 32◦
= 5.1 m to two significant figures.

Back 731
Contents 
Question 6.6 (b)
opp
sin θ =
hyp
so
π a
sin =
3 10 m
Multiplying both sides by 10 m gives
π
a = 10 m × sin
3
= 8.7 m to two significant figures.

Back 732
Contents 
Question 6.6 (c)

opp
tan θ =
adj
5.0 m
=
1.0 m
= 5.0

So θ = tan−1 (5.0)
= 79◦

{Note that ‘opp’ and ‘adj’ must be the sides opposite and adjacent to the angle you
are trying to find.}

Back 733
Contents 
Question 6.7

H = height of West Tower + height of base of Cathedral − height of theodolite

= 66 m + 15 m − 1.5 m
= 79.5 m

θ = 2.27◦
H
tan θ =
D
Multiplying both sides by D gives

D tan θ = H

Dividing both sides by tan θ gives

H
D=
tan θ
79.5 m
=
tan 2.27◦
= 2006 m
≈ 2000 m

So you can estimate the distance of the theodolite from Ely Cathedral to be about
2 km.

Back 734
Contents 
Question 6.8
From Equation 6.10,

V = W tan θ

where W = 65 m and θ = 36◦ . So

V = 65 m × tan 36◦
= 47 m to two significant figures.

The vertical thickness of the stratum is 47 metres.

Back 735
Contents 
Question 6.9
From Equation 6.11,

r = h cos 45◦

where r is the required radius and h = 302 pm. So

r = 302 pm × cos 45◦

= 214 pm to three significant figures.

The radius of a lithium ion is 214 pm (i.e. 2.14 × 10−10 m)

Back 736
Contents 
Question 6.10
i = 45.0◦ r = 26.3◦ v1 = 3.00 × 108 m s−1
Snell’s law states that
sin i v1
=
sin r v2
We are trying to find v2 , the speed of light in glass.
sin i v1
Multiplying both sides of = by v2 and by sin r gives
sin r v2
v2 sin i = v1 sin r

Dividing both sides by sin i gives

sin r
v2 = v1
sin i
sin 26.3◦
= 3.00 × 108 m s−1 ×
sin 45.0◦
0.4431
= 3.00 × 108 m s−1 ×
0.7071
= 1.88 × 10 m s
8 −1

So the speed of light in glass is 1.88 × 108 m s−1 .

Back 737
Contents 
Question 6.11
From Equation 6.13
nλ
sin θn =
d
Reversing the equation and multiplying both sides by d gives

nλ = d sin θn

Dividing both sides by n

d sin θn
λ=
n

d = 1.64 × 10−6 m θn = 24.1◦ n=1

So
1.64 × 10−6 m × sin 24.1◦
λ=
1
= 6.70 × 10 m to three significant figures.
−7

Back 738
Contents 
Question 6.12
Let the distance to car ferry = d.
The length of car ferry = l = 86 m.
The angle subtended = θ = 3.5◦ .
Converting θ to radians:
360◦ = 2π radians
2π
so 1◦ = radians
360
2π
3.5◦ = 3.5 × radians = 0.0611 radians
360
From Equation 6.1
s
θ (in radians) =
r
In this case s ≈ l and r ≈ d so
l
θ≈
d
Multiplying both sides by d gives
θd≈l
Dividing both sides by θ gives
l
d=
θ
Back I 739
Contents 
So
86 m
d≈
0.0611
≈ 1408 m

The ferry is approximately 1.4 km away.

Back J 740
Contents 

α
α
α

β γ β γ β γ
(a) (b) (c)

β γ β α β
(d) (e) (f)

Figure 6.12: Triangles of various shapes.

Click on Back to return to text

Back 741
Contents 

h ′′
b ′′
h′
b′
h
b

θ θ θ
a a′ a ′′

Figure 6.14: Three similar right-angled triangles.

Click on Back to return to text

Back 742
Contents 
y

−3π −2π −π 0 π 2π 3π

θ /radians

−1
(a)

−3π −2π −π 0 π 2π 3π

θ /radians

−1
(b)

Figure 6.17: Graphs of (a) y = a sin θ, (b) y = a cos θ .

Click on Back to return to text

Back 743
Contents 
Question 7.1 (a)
Since 100 = 102 , log10 100 = 2.

Back 744
Contents 
Question 7.1 (b)
Since 0.001 = 10−3 , log10 0.001 = −3.

Back 745
Contents 
Question 7.1 (c)
√ √ 1
Since 10 = 101/2 , log10 10 =
2

Back 746
Contents 
Question 7.1 (d)
Since 1.329 = 100.1235 (from the section of text just above the question),
log10 1.329 = 0.1235.

Back 747
Contents 
Question 7.2 (a)
log10 2 = 0.3010

Back 748
Contents 
Question 7.2 (b)
log10 2000 = 3.301
{ Note that log10 2000 is exactly 3 greater than log10 2. This result will be
discussed further in Sections 7.2 and 7.3. }

Back 749
Contents 
Question 7.3 (a)
101.5 = 31.62

Back 750
Contents 
Question 7.3 (b)
p = 31.62
{ Because of the way in which log to base 10 is defined, this follows straight from
the answer to Question 7.3(a). }

Back 751
Contents 
Question 7.4 (a)
For human blood the hydrogen ion concentration is 4.0 × 10−8 mol dm−3 , so

4.0 × 10−8 mol dm−3

!
pH = − log10
mol dm−3
 
= − log10 4.0 × 10−8
= −(−7.4)
= 7.4

Back 752
Contents 
Question 7.4 (b)
For the hair shampoo, the hydrogen ion concentration is 3.2 × 10−6 mol dm−3 , so

3.2 × 10−6 mol dm−3

!
pH = − log10
mol dm−3
 
= − log10 3.2 × 10−6
= −(−5.5)
= 5.5

Back 753
Contents 
Question 7.5 (a)

log10 300 = log10 (3 × 100)

= log10 3 + log10 100 (from Equation 7.2)
= 0.4771 + log10 10 2

= 0.4771 + 2 (from Equation 7.1)

= 2.477 to four significant figures.

Back 754
Contents 
Question 7.5 (b)

log10 0.03 = log10 (3 ÷ 100)

= log10 3 − log10 100 (from Equation 7.3)
= 0.4771 − log10 10 2

= 0.4771 − 2 (from Equation 7.1)

= −1.523 to four significant figures.

Back 755
Contents 
Question 7.5 (c)
 
log10 9 = log10 32
= 2 log10 3 (from Equation 7.4)
= 2 × 0.4771
= 0.9542 to four significant figures.

Back 756
Contents 
Question 7.6
2.5 − 0.5 2.0
The gradient of the line = = = 2.0.
1.0 − 0.0 1.0
This is the result expected.
The intercept of the line on the vertical axis is approximately 0.5.
log10 π = 0.497 to three significant figures, so the result seems reasonable.

Back 757
Contents 
Question 7.7
Taking the log to base 10 of both sides of the equation y = 2x3 gives
 
log10 y = log10 2x3
= log10 2 + log10 x3 (from Equation 7.2)
= log10 2 + 3 log10 x (from Equation 7.4)

We can reverse the order of the two terms on the right-hand side to give

log10 y = 3 log10 x + log10 2

Comparison with the general equation of a straight-line graph, y = mx + c, reveals

that m = 3 and c = log10 2, so the gradient of the graph will be 3 and the intercept
on the vertical axis will be log10 2.

Back 758
Contents 
Question 7.8
n = n0 e at
Taking the log to base 10 of both sides of the equation gives
 
log10 n = log10 n0 e at
= log10 n0 + log10 e at (from Equation 7.2)
= log10 n0 + at log10 e (from Equation 7.4)

We can reverse the order of the two terms on the right-hand side to give

log10 n = at log10 e + log10 n0

= (a log10 e) t + log10 n0

Comparison with the general equation of a straight-line graph, y = mx + c, shows

that a graph of log10 n against t will be a straight line of gradient a log10 e and
intercept on the vertical axis of log10 n0 .

Back 759
Contents 
Question 7.9 (a)
ln 4 = 1.386

Back 760
Contents 
Question 7.9 (b)
The number whose natural logarithm is 4 is e 4 = 54.60.

Back 761
Contents 
Question 7.10
n = n0 e at
Taking the log to base e of both sides of the equation gives
 
ln n = ln n0 e at
= ln n0 + ln e at (from Equation 7.8)
= ln n0 + at (from Equation 7.7)

We can reverse the order of the two terms on the right-hand side to give

ln n = at + ln n0

Comparison with the general equation of a straight-line graph, y = mx + c, shows

that a graph of ln n against t will be a straight line of gradient a and intercept on
the vertical axis of ln n0 .

Back 762
Contents 

Stage Number k value

! !
N0 72
maximum total number of eggs if N0 = 72 k1 = log10 = log10 = 0.1498
N1 51
all pairs bred and laid 3 eggs (max-
imum possible)
! !
N1 51
maximum possible number of eggs N1 = 51 k2 = log10 = log10 = 0.0741
N2 43
from the 17 pairs that did breed
! !
N2 43
actual number of eggs laid N2 = 43 k3 = log10 = log10 = 0.4293
N3 ! 16 !
N3 16
number of eggs that hatched N3 = 16 k4 = log10 = log10 = 0.0280
N4 ! 15 !
N4
number of chicks that fledged N4 = 15 k5 = log10 = log10 =
N5 ! !
N0 72
number of owlets that survive to N5 = 9 ktotal = log10 = log10 = 0.9031
N5 9
form pairs

Table 7.1: k-values for various stages in the breeding of 24 pairs of owls in Wytham Wood in 1952–1953

Click on Back to return to text

Back 763
Contents 
log10 y
y
0.6

0.4

0.2

0
0.1 0.2 0.3 0.4 0.5 0.6
−0.2 log10 x

−0.4

−0.6
0 x
(a) (b)

Figure 7.3: Graphs of (a) y against x and (b) log10 y against log10 x for the equation
y = 3x−2 .

Click on Back to return to text

Back 764
Contents 
T
TE
30
Saturn
28

26

24

22

20

18

16

14

12
Jupiter
10
Figure 7.4: A graph of T/T E against a/aE where T is a
8
planet’s orbital period and a is the planet’s average distance
from the Sun. T E and aE are the values of T and a for the 6
Earth, and the values of T and a for other planets have been
4 Mars
divided by these so as to make the numbers plotted more Earth
manageable. 2 Venus
Mercury
0 2 4 6 8 10 a
aE
Click on Back to return to text

Back 765
Contents 

log10 T
TE

1.5
Saturn

1.0 Jupiter

0.5

Mars

Figure 7.5: A graph of log10 (T/T E )

−0.5 Earth 0.5 1.0 a
against log10 (a/aE ) where T is a log10 a
E
planet’s orbital period and a is the Venus
planet’s average distance from the
Sun. T E and aE are the values of T –0.5
and a for the Earth. Mercury

Click on Back to return to text

Back 766
Contents 

1000

disintegrations per minute (on a log scale)

100

10

Figure 7.7: A graph of disintegra-

tions per minute (on a log scale)
against time for the radioactive decay
of the excited state of barium-137.
1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
time/minutes
Click on Back to return to text

Back 767
Contents 
Question 8.1 (a)
Of the 52 cards in the pack, 13 are hearts. So according to Equation 8.2, the
13 1
probability of a card drawn at random being a heart is = .
52 4
{This result also follows from noting that there are 4 suits, each with the same
number of cards, so one-quarter will be hearts.}

Back 768
Contents 
Question 8.1 (b)
Of the 52 cards in the pack, 26 are red (13 hearts and 13 diamonds). So the
26 1
probability of a card drawn at random being red is = .
52 2
2 1
{Or 2 of the 4 suits are red, so the probability is = .}
4 2

Back 769
Contents 
Question 8.1 (c)
Of the 52 cards in the pack, 4 are aces (one for each suit). So the probability of a
4 1
card drawn at random being an ace is = .
52 13

Back 770
Contents 
Question 8.1 (d)
Of the 52 cards in the pack, 12 are picture cards (3 for each suit). So the
12 3
probability of a card drawn at random being a picture card is = .
52 13

Back 771
Contents 
Question 8.2 (a)
For any one toss the probability of heads is always the same: 12 .

Back 772
Contents 
Question 8.2 (b)
For the single toss of the third coin, the probability of getting heads is 12 and that is
unaffected by what has gone before. This is no different to tossing the same coin
three times in succession. Only foolish gamblers believe that because heads have
come up twice running the chances of tails coming up the next time are thereby
increased!

Back 773
Contents 
Question 8.3 (a)
If two coins are tossed simultaneously, there are four possible outcomes, all of
which are equally likely:

Outcome 1 H H
Outcome 2 H T
Outcome 3 T H
Outcome 4 T T
The outcome of two tails can occur in only one way, so the probability of getting
two tails is 14 .
This result can also be found from the multiplication rule:
the probability that the first coin will show tails is 12 ;
the probability that the second coin will show tails is 12 ;
so the probability of getting two tails is 1
2 × 1
2 = 14 .

Back 774
Contents 
Question 8.3 (b)
The probability of throwing a six with one dice is 16 . So the probability of getting a
pair of sixes when throwing two dice is 61 × 16 = 36
1
.

Back 775
Contents 
Question 8.4 (a)
Assuming the germination probabilities to be independent of one another, the
probability of seeds of both A and B germinating is 12 × 13 = 61 .

Back 776
Contents 
Question 8.4 (b)
Assuming the germination probabilities to be independent of one another, the
probability of the seeds of all three species germinating is 12 × 13 × 14 = 24
1
.

Back 777
Contents 
Question 8.4 (c)
Assuming the germination probabilities to be independent of one another,
the probability that a seed of A will not germinate is 12 ;
the probability that a seed of B will not germinate is 23 ;
the probability that a seed of C will not germinate is 34 ;
so the probability that none will germinate is 1
2 × 2
3 × 3
4 = 41 .

Back 778
Contents 
Question 8.5
1
The probability of drawing any one particular card from the pack is 52 . This is true
for each of the three named cards. So the probability of drawing the Jack of
diamonds or the Queen of diamonds or the King of diamonds is 52 1
+ 52
1
+ 52
1
= 523
.

Back 779
Contents 
Question 8.6
The situation is similar to the one described in Question 8.3. If two coins are tossed
simultaneously, there are four possible outcomes, all of which are equally likely:

Outcome 1 H H
Outcome 2 H T
Outcome 3 T H
Outcome 4 T T
The outcome of a head and a tail can occur in two ways, so the probability of
getting a head and a tail is 24 = 12 .
This result can also be found from a combination of the multiplication and
addition rules. For the combination of one head and one tail:
the probability that the coin on the left will be tails is 12 ;
the probability that the coin on the right will be heads is 12 ;
So the probability that the combination T H will occur is 1
2 × 1
2 = 41 .
By similar reasoning, the probability that the combination H T will occur is also 14 .
These possibilities are mutually exclusive, so the probability of getting one head
and one tails is 14 + 41 = 12 .

Back 780
Contents 
Question 8.7
The fraction of the atmosphere that is oxygen is
0.26 0.26
=
0.26 + 1 1.26
Expressed as a percentage to 2 significant figures, this fraction is 21%.

Back 781
Contents 
Question 8.8
For the 10 measurements in Table 8.4,
mean = 1.122 mm
standard deviation sn = 0.123 mm

Back 782
Contents 
Question 8.9 (a)
For nine measurements, the median is the 5th measurement in the list (in ascending
or descending order). This is 7.8 cm.

Back 783
Contents 
Question 8.9 (b)
70.4 cm
From Equation 8.3, the mean is = 7.82 cm.
9

Back 784
Contents 
Question 8.10
The best estimate that can be made from this data of the mean number, µ, of
flowers per plant in the colony is the mean of the sample, x. In this case,

x = 7.25 flowers

{Note that it is normal practice to quote means and medians in this way, even for
quantities, such as numbers of flowers, which cannot really be fractional!}
The best estimate that can be made of the standard deviation of the population is
the sample standard deviation sn−1 . In this case,

sn−1 = 1.94 flowers.

Back 785
Contents 

your guess is as
good as mine
impossible inevitable
outcome outcome

an extremely long shot virtually a certainty

0 1 1
2
increasingly unlikely increasingly likely

Figure 8.1: The scale of probabilities.

Click on Back to return to text

Back 786
Contents 
xi /nm di /nm di2 /10−5 nm2
2.458 0.0036 1.296
2.452 −0.0024 0.576
2.454 −0.0004 0.016
2.452 −0.0024 0.576
2.459 0.0046 2.116
2.455 0.0006 0.036
2.464 0.0096 9.216
2.453 −0.0014 0.196
2.449 −0.0054 2.916
2.448 −0.0064 4.096
n
X n
X n
X
xi = 24.544 nm di = 0 di2 = 21.04 × 10−5 nm2
i =1 i =1 i =1
x = 2.4544 nm di2 = 2.104 −5
q × 10 nm
2

sn = di2
= 4.587 × 10−3 nm
= 0.0046 nm
Table 8.3: Calculation of the standard deviation for the set of measurements
originally given in Table 8.2.

Return to Page 412

Back 787
Contents 
Question 9.1 (a)
The answer to this question depends on which significance level is used.
Employing the usual convention, i.e. rejecting the null hypothesis if P < 0.05, the
null hypothesis should be rejected on this occasion, since P < 0.01 means that P
must be less than 0.05. Therefore the alternative hypothesis should be accepted.
However, if it had been decided only to reject the null hypothesis if P were less
than 0.001, we would not be justified in categorically rejecting the null hypothesis
in this way.

Back 788
Contents 
Question 9.1 (b)
Employing the usual convention, the null hypothesis should be accepted, since
P > 0.05.

Back 789
Contents 
Question 9.1 (c)
This inequality is written in a way that is very unhelpful and ought to be avoided.
We are told that P > 0.01. But how much greater? If P > 0.05 then, employing the
usual convention, the null hypothesis must be accepted. However, if P lies
between 0.05 and 0.01 (i.e. 0.05 > P > 0.01) then, employing the usual
convention, the null hypothesis should be rejected and the alternative hypothesis
accepted. In the former situation, the result ought to have been given as P > 0.05;
in the latter it ought to have been given as either P < 0.05 or 0.05 > P > 0.01.

Back 790
Contents 
Question 9.2 (a)
Since the actual number of parasites per sheep is known, this data is at the interval
level.

Back 791
Contents 
Question 9.2 (b)
Since the sheep are classified into just two contrasting categories (‘parasitized’ and
‘unparasitized’) this data is best treated as being at the categorical level.
{Since there is an element of ranking here, you might have regarded this data as
being at the ordinal level. However, whether ‘unparasitized’ is ‘good’ or ‘bad’
does depend on whether you take the point of view of the sheep or the parasites!
‘Parasitized’ and ‘unparasitized’ might correspond to the clear-cut categories
‘susceptible to parasites’ and ‘resistant to parasites’. In general, ordinal level data
is subdivided into more than two classes.}

Back 792
Contents 
Question 9.2 (c)
Since degree of parasitization is recorded, but not precisely how many parasites
there were on each sheep, this data is at the ordinal level.

Back 793
Contents 
Question 9.3
The total number of plants in the next generation was 636 (i.e. 185 + 305 + 146). If the
ratio in a sample of 636 plants were 1 red-flowered : 2 pink-flowered : 1 white-flowered,
then there would be
636
= 159 red-flowered plants
4
636
= 318 pink-flowered plants
2
636
= 159 white-flowered plants.
4
These are therefore the ‘expected’ numbers. Drawing up a table, calculating each
(Oi − Ei )2
value and then summing these values to obtain the test statistic, χ2 :
Ei

Flower (Oi − Ei )2
Oi Ei (Oi − Ei ) (Oi − Ei )2
colour Ei
red 185 159 26 676 4.252
pink 305 318 –13 169 0.531
white 146 159 –13 169 1.063
total 636 636 0 5.846

Back I 794
Contents 
The number of degrees of freedom is given by
!
number of cells containing
−1
observed numbers
=3−1
=2

Reading across the row for 2 degrees of freedom in Table 9.3, it can be seen that the χ2
value of 5.846 corresponds to a significance level (P) of less than 0.1 but more than 0.05
(i.e. 0.1 > P > 0.05).
Since the probability that the entire population from which the sample of 636 plants was
drawn was in the ratio 1 red-flowered : 2 pink-flowered : 1 white-flowered is greater
than 0.05, the null hypothesis cannot be rejected at the 5% significance level. The
experimental data is therefore compatible with the prediction from genetics theory that
the ratio of plants in the next generation should be in the ratio 1 red-flowered : 2
pink-flowered : 1 white-flowered.

Back J 795
Contents 
Question 9.4
n
X
The values of (RA )i , (RB )i , Di , D2i and D2i are given below.
i =1

Vertical Rank (RA )i Mean water content/ (RB )i Di = (RA )i − (RB )i D2i
distance/cm % dry mass
0 1 76 1 0 0
4 2 83 3 −1 1
7 3.5 93 4 −0.5 0.25
9 5 80 2 3 9
7 3.5 102 6 −2.5 6.25
11 7 95 5 2 4
10 6 120 7 −1 1
13 8 130 8 0 0
n
X n
X
Di = 0 D2i = 21.5
i =1 i =1

{Since there are two vertical distances of 7 cm, both are given the rank
3+4
= 3.5 and the next vertical distance (9 cm) is given the rank 5.}
2

Back I 796
Contents 
n
X
Substituting D2i = 21.5 and n = 8 into Equation 9.2 gives
i =1

6 × 21.5
rs = 1 − = 0.744
8 × (82 − 1)
Reading across the row for 8 pairs of measurements in Table 9.8, it can be seen
that 0.05 > P > 0.01. Since P < 0.05, the null hypothesis must be rejected at the
5% significance level and the alternative hypothesis accepted. There is a
statistically significant positive correlation between mean soil water content and
vertical distance from ridge crest.
{Although mean soil water content was significantly correlated with both
horizontal and vertical distance from the nearest ridge crest, the former produced a
value of rs that was both higher and more significant than the latter (i.e.
rs = 0.905, P < 0.01 compared to rs = 0.744, P < 0.05). This was because
horizontal distance from the ridge crest had been arranged to increase regularly.}

Back J 797
Contents 
Question 9.5 (a)
These samples are unmatched. There is no logical link between any particular
plant growing in one reserve and any particular plant growing in the other reserve.

Back 798
Contents 
Question 9.5 (b)
These samples are matched. For each sampling station along the stream, the
number of nymphs is known for two species of Stonefly.

Back 799
Contents 
Question 9.6
In this case x1 = 7.7, x2 = 7.2, s1 = 2.7, s2 = 2.1, n1 = 18 and n2 = 15.
Substituting for s1 , s2 , n1 and n2 into Equation 9.5 gives

(18 − 1)(2.7)2 + (15 − 1)(2.1)2

(S c )2 =
(18 − 1) + (15 − 1)
(17 × 7.29) + (14 × 4.41)
=
17 + 14
= 5.989

Substituting for (S c )2 , n1 and n2 into Equation 9.4 gives

r
5.989 5.989
SED = + = 0.856
18 15
Substituting for x1 , x2 and SED into Equation 9.3 gives
7.7 − 7.2
t= = 0.584 to three places of decimals.
0.856
In this case, the number of degrees of freedom is

(18 − 1) + (15 − 1) = 31.

The value of t (i.e. 0.584) is smaller than any of the critical values in the row for
30 degrees of freedom (the nearest equivalent to 31) in Table 9.13. The probability

Back I 800
Contents 
of obtaining a value of t as large as this by chance if the null hypothesis were true
is therefore greater than 0.1 (i.e. P > 0.1), probably much greater. The difference
in number of flowers per plant growing either side of this ridge is not statistically
significant.
{Note: If you worked to a different number of significant figures in this question
you may have obtained a slightly different value for t. However, your conclusion
— that the difference in number of flowers per plant growing either side of this
ridge is not statistically significant — should be have been the same.}

Back J 801
Contents 
Degrees of P = 0.1 P = 0.05 P = 0.01 Degrees of P = 0.1 P = 0.05 P = 0.01
freedom freedom
1 2.706 3.841 6.635 16 23.542 26.296 32.000
2 4.605 5.991 9.210 17 24.769 27.587 33.409
3 6.251 7.815 11.341 18 25.989 28.869 34.805
4 7.779 9.488 13.277 19 27.204 30.144 36.191
5 9.236 11.070 15.086 20 28.412 31.410 37.566
6 10.645 12.592 16.812 21 29.615 32.671 38.932
7 12.017 14.067 18.475 22 30.813 33.924 40.289
8 13.362 15.507 20.090 23 32.007 35.172 41.638
9 14.684 16.919 21.666 24 33.196 36.415 42.980
10 15.987 18.307 23.209 25 34.382 37.652 44.314
11 17.275 19.675 24.725 26 35.563 38.885 45.642
12 18.549 21.026 26.217 27 36.741 40.113 46.963
13 19.812 22.362 27.688 28 37.916 41.337 48.278
14 21.064 23.685 29.141 29 39.087 42.557 49.588
15 22.307 24.996 30.578 30 40.256 43.773 50.892

Table 9.3: Critical values of χ2 for different degrees of freedom and at three levels of
significance. The null hypothesis is usually rejected if, for the appropriate number
of degrees of freedom, the calculated value of χ2 is greater than the value tabulated
at the P = 0.05 significance level.

Click on Back to return to text

Back 802
Contents 
Number of pairs P = 0.1 P = 0.05 P = 0.01 Note: (i) The null hypothesis is usually
of measurements rejected if, for the appropriate number
7 0.714 0.786 0.929 of pairs of measurements, the calculated
8 0.643 0.738 0.881 value of rS is greater than or equal to the
9 0.600 0.683 0.833 value tabulated at the P = 0.05 signifi-
10 0.564 0.648 0.794 cance level.
12 0.506 0.591 0.777 (ii) The lower part of Table 9.8 does not
14 0.456 0.544 0.715 have entries for odd numbers of pairs of
16 0.425 0.506 0.665 measurements. Should the data you are
18 0.399 0.475 0.625 analysing comprise (say) 17 pairs of mea-
20 0.377 0.450 0.591 surements, it is better to err on the side
22 0.359 0.428 0.562 of caution and compare your value of the
24 0.343 0.409 0.537 test statistic with the critical values for 16
26 0.329 0.392 0.515 pairs rather than those for 18 pairs. Be-
28 0.317 0.377 0.496 cause each critical value for 16 pairs of
30 0.306 0.364 0.478 measurements is higher than the corre-
Table 9.8: Critical values for the Spearman rank corre- sponding value for 18 pairs, this makes it
lation coefficient (rS ) for different numbers of pairs of less likely that you will mistakenly reject
measurements and at three levels of significance a true null hypothesis.

Click on Back to return to text

Back 803
Contents 
Water Rank Number of Rank Di = (RA )i − (RB )i D2i
speed/m s−1 (RA )i nymphs (RB )i
0.8 9 35 12 -3 9
1.1 11 28 11 0 0
0.5 5.5 11 6 -0.5 0.25
0.7 7.5 12 7 0.5 0.25
0.2 2.5 7 4 -1.5 2.25
0.4 4 5 1 3 9
0.5 5.5 6 2.5 3 9
1.3 12 21 9 3 9
0.9 10 23 10 0 0
1.7 13 43 13 0 0
0.2 2.5 10 5 -2.5 6.25
0.1 1 6 2.5 -1.5 2.25
0.7 7.5 19 8 -0.5 0.25
n
X n
X
Di = 0 D2i = 47.5
i =1 i =1

Table 9.10: Extension of Table 9.9 to include ranks ((RA )i and (RB )i ), differences between ranks (Di )
and values of D2i

Click on Back to return to text

Back 804
Contents 
Degrees of P = 0.1 P = 0.05 P = 0.01 Degrees of P = 0.1 P = 0.05 P = 0.01
freedom freedom
1 6.314 12.706 63.657 18 1.734 2.101 2.878
2 2.920 4.303 9.925 19 1.729 2.093 2.861
3 2.353 3.182 5.841 20 1.725 2.086 2.845
4 2.132 2.776 4.604 21 1.721 2.080 2.831
5 2.015 2.571 4.032 22 1.717 2.074 2.819
6 1.943 2.447 3.707 23 1.714 2.069 2.807
7 1.895 2.365 3.499 24 1.711 2.064 2.797
8 1.860 2.306 3.355 25 1.708 2.060 2.787
9 1.833 2.262 3.250 26 1.706 2.056 2.779
10 1.812 2.228 3.169 27 1.703 2.052 2.771
11 1.796 2.201 3.106 28 1.701 2.048 2.763
12 1.782 2.179 3.055 29 1.699 2.043 2.756
13 1.771 2.160 3.012 30 1.697 2.042 2.750
14 1.761 2.145 2.977 40 1.684 2.021 2.704
15 1.753 2.131 2.947 60 1.671 2.000 2.660
16 1.746 2.120 2.921 120 1.658 1.980 2.617
17 1.740 2.110 2.898 ∞ 1.645 1.960 2.576

Click on Back to return to text

Table 9.13: Critical values of t for the t-test for unmatched samples for different
degrees of freedom at three levels of significance. The null hypothesis is usually
rejected if the calculated value of t is greater than the value given for the P = 0.05
significance level at the appropriate number of degrees of freedom.
Back 805
Contents 
Question 10.1 (a)
The gradient of the tangent drawn to the graph of y = x2 at
y
x = 1 is
rise (3.0 − 0.0) 3.0 18
gradient = = = = 2.0
run (2.0 − 0.5) 1.5
16
So the gradient of the curve at x = 1 is 2.0 to two significant
figures.
14

12

10
tangent
at x = 3
8

2 tangent
at x = 1

0 1 2 3 4 5 x

Back 806
Contents 
Question 10.1 (b)
The graph shows y = x2 with a tangent drawn at x = 2. The
y
gradient of this tangent is
rise (12.0 − 0.0) 12.0 18
gradient = = = = 4.0
run (4.0 − 1.0) 3.0
16
So the gradient of the curve at x = 2 is 4.0 to two significant
figures.
14
{Note that drawing accurate tangents is difficult; values for
the gradient of y = x2 at x = 2 found by this method could 12
reasonably be anything between 3.5 and 4.5.
10
A comparison of the values for gradient at x = 1, x = 2 and
x = 3 shows that the gradient increases as x increases. This
8
is consistent with the observed increase in the gradient of
the graph as x increases.}
6
tangent
at x = 2
4

2 tangent
at x = 1

0 1 2 3 4 5 x

Back 807
Contents 
Question 10.2 (a)
y = x4 so C = 1 and n = 4
dy
= 1 × 4x3 = 4x3
dx
dy
When x = 4, = 4 × 43 = 44 = 256
dx
So at x = 4 the gradient of the graph is 256.

Back 808
Contents 
Question 10.2 (b)
y = 5x so C = 5 and n = 1 so
dy
= 5x1−1 = 5x0 = 5
dx
The gradient of the graph is 5 for all values of x.
{You may have been able to give this result without differentiating y = 5x, from
your knowledge of the gradient of straight-line graphs.}

Back 809
Contents 
Question 10.2 (c)
y = 3x2 so C = 3 and n = 2
dy
= 3 × 2x2−1 = 6x
dx
dy
When x = 4, = 6 × 4 = 24
dx
So at x = 4 the gradient of the graph is 24.

Back 810
Contents 
Question 10.2 (d)
y = 5 so C = 5 and n = 0
dy
= 5 × 0 × x−1 = 0
dx
The gradient of the graph is 0 for all values of x.
{You may have been able to give this result without differentiating y = 5.}

Back 811
Contents 
Question 10.3 (a)
1
y = √ = x−1/2 so C = 1 and n = − 12
x

dy 1 x−3/2 1
= − x−1/2−1 = − = − 3/2
dx 2 2 2x
This could also be written as
dy 1
=− √
dx 2x x
When x = 4,
dy 1 1 1
=− √ =− =−
dx 2×4× 4 2×4×2 16

So at x = 4 the gradient of the graph is −1/16.

Back 812
Contents 
Question 10.3 (b)
2
y= = 2x−2 so C = 2 and n = −2
x2
dy 4
= 2 × (−2)x−2−1 = −4x−3 = − 3
dx x
When x = 4,
dy 4 1 1
=− 3 =− 2 =−
dx 4 4 16

So at x = 4 the gradient of the graph is −1/16.

Back 813
Contents 
Question 10.4 (a)
x = t7
so
dx
= 7t 7−1 = 7t 6
dt

Back 814
Contents 
Question 10.4 (b)
C
E= = C r −1
r
so
dE C
= C × (−1) r −1−1 = −Cr −2 = − 2
dr r

Back 815
Contents 
Question 10.5
z = 4y 2 + y
Differentiating each of the terms separately gives
dz
= (4 × 2y 2−1 ) + (1 × y 1−1 )
dy
= 8y 1 + y 0
= 8y + 1

Back 816
Contents 
Question 10.6 (a)
x = 2t3 + 4t2 − 2t + 3
Differentiating this with respect to t gives
dx
= (2 × 3t2 ) + (4 × 2t) − 2 = 6t2 + 8t − 2
dt
Differentiating again gives

d2 x
= (6 × 2t) + 8 = 12t + 8
dt2

Back 817
Contents 
Question 10.6 (b)
2
z= = 2y−1
y
Differentiating with respect to y gives
dz 2
= 2 × (−1)y−1−1 = −2y−2 = − 2
dy y
Differentiating again gives

d2 z 4
2
= −2 × (−2)y−2−1 = 4y−3 = 3
dy y

Back 818
Contents 
Question 10.7 (a)
y = 2 e x so C = 2 and k = 1.
dy
= 2 × 1 e x = 2 e x = y (since y = 2 e x ).
dx

Back 819
Contents 
Question 10.7 (b)
z = e t/2 so C = 1 and k = 1
2

dz z
= 1
2 e t/2 = (since z = e t/2 ).
dt 2

Back 820
Contents 

Figure 10.4: A graph

to show the increasing
concentration of hypo-
bromite ions in a partic-
2.0 × 10−3 ular chemical reaction,
concentration of hypobromite ions/mol˚dm−3

at 25 ◦ C.

1.0 × 10−3

0 1000 2000 3000 4000 5000 6000

time/s

Click on Back to return to text

Back 821
Contents 
Box 10.4 Differentiating y = x2 − 4x + 3 from first principles

The graph of y = x2 − 4x + 3 is shown again in y

Figure 10.9 and, as in the previous differentiation
from first principles, P and Q represent any two
3
points on the curve.
Since both points lie on the curve, we can say
2
y = x − 4x + 3
2
(10.1)

and 1
Q
(y + ∆y) = (x + ∆x)2 − 4(x + ∆x) + 3 (10.6)
0
Multiplying out the brackets on the right-hand side 1 2 3 4 x
∆y
of Equation 10.6 gives
P
−1 ∆x
y + ∆y = x + 2x ∆x + (∆x) − 4x − 4∆x + 3
2 2

and rearranging gives

y + ∆y = (x2 − 4x + 3) + 2x ∆x − 4∆x + (∆x)2

Figure 10.9: Points P and Q on the curve y =
x2 − 4x + 3.

Back I 822
Contents 
Since y = x2 − 4x + 3 (from Equation 10.1), we can subtract y from the left-
hand side and (x2 − 4x + 3) from the right-hand side to give

∆y = 2x ∆x − 4∆x + (∆x)2

Dividing both sides by ∆x gives

∆y
= 2x − 4 + ∆x
∆x
In the limit as ∆x approaches zero, the final term on the right-hand side will
∆y dy
disappear, and will become equal to , so we can say
∆x dx
dy
= 2x − 4
dx

Return to Section 10.2.4

Back J 823
Contents 
dy
dx
dy
4 = 2x − 4
y dx

y = x2 − 4x + 3 d2y
3 3 dx2
d2y
= 2
dx2
2 2 2

1 1 1

0 0 0
1 2 3 4 x 1 2 3 4 x 1 2 3 4 x

−1 −1

(c)
−2
(a)

−3

−4

(b)

dy d2 y
Figure 10.10: Graphs of (a) y against x, (b) against x and (c) 2 against x for y = x2 − 4x + 3.
dx dx
Click on Back to return to text

Back 824
Contents 

75˚m

Figure 10.11: An object being dropped from the Clifton Suspension Bridge.

Click on Back to return to text

Back 825
Contents 

s v a

9.81ms–2

0 t 0 t 0 t
(a) (b) (c)

Figure 10.12: Graphs to show the variation of (a) distance, (b) speed and (c) acceleration with time for
an object dropped from a bridge. Note that distance from the bridge, speed and acceleration are all
measured in a downwards direction.

Click on Back to return to text

Back 826
Contents 
y
20

18

16 tangent

14

12

10

4
tangent

2
tangent

Figure 10.13: A graph of y = e x.

−3 −2 −1 0 1 2 3 x
Click on Back to return to text

Back 827
Contents 

Back J I 828
Contents 
Question A.1
adj vx
cos α = =
hyp v
so y

v x = v cos α
= 8.6 m s−1 × cos 42◦
= 6.4 m s−1 to two significant figures.

opp vy
sin α = =
hyp v vy
so v

vy = v sin α
α
= 8.6 m s−1 × sin 42◦
vx x
= 5.8 m s−1 to two significant figures.

Back 829
Contents 
Question A.2
F 2 = F 2x + Fy2
So y
q
F= F 2x + Fy2
q
= (4.0 N)2 + (3.0 N)2
= 5.0 N

opp
tan β =
adj Fy
Fy F
=
Fx
3.0 N
=
4.0 N Fx x
= 0.75

So β = tan−1 (0.75) = 37◦ to two significant

figures.
So the resultant vector F has a magnitude of
5.0 N and acts at an angle of 37◦ to the hori-
zontal axis.

Back 830
Contents 

Index

absolute value, 474 average, 410, 422

absolute zero of temperature, 285 Avogadro’s number, 68
accuracy, 407 axis (of graph), 240
addition rule (probability), 395
algebra, 131, 154 bar chart, 235
algebraic fractions, 183 base number, 40
al-Khwarizmi, 176 base units, 66
alternative hypothesis, 438 BEDMAS, 53
angle, 294 bell-shaped curve, 409, 419
angle of dip, 327 bequerel (unit), 288
ångström, 71 best-fit line, 250
arc, 302 brackets, 15, 53, 137, 192
arccos, 318 calculator use, 20, 43, 100, 316, 350,
arcsec, 301 373, 417, 430
arcsin, 318 cancellation, 29
arctan, 318 categorical level, 443
area, 91, 107, 276 Celsius degree, 67, 287
arithmetic mean, 410 centi (prefix), 71
Arrhenius equation, 379 checking, 198, 219

Back I 831
Contents 
chemical reactions, 379, 482 decimal notation, 10, 42, 390
χ2 -test, 446 decimal places, 11, 416, 450
χ2 -test (critical values), 802 degree (of arc), 296
chord, 483 degree (temperature), 67, 287
combining equations, 203 degrees of freedom, 451, 474
combining probabilities, 391 denominator, 24
common denominator, 28 density, 127
common logarithm, 347 dependent variable, 258
common population variance, 472 derivative, 484
commutative operation, 17, 136 derived units, 144
complex number, 13 descriptive statistics, 405
component (of vector), 515 difference of squares, 196
concentration, 125, 482 differentiation, 483
constant of proportionality, 266 differentiation from first principles, 486,
correlation, 459 822
correlation coefficient, 459 diffraction, 336, 406
cosine, 313 direct proportionality, 265
critical value, 440
critical values (tables), 802, 803, 805 e, 290, 505
cube root, 50 elimination, 204
cystic fibrosis, 392 energy, 223
equation, 131, 155
decay constant, 378 equation of straight-line graph, 272,
deci (prefix), 71 277
decibel, 80, 352 equivalent fraction, 25

Back J I 832
Contents 
estimated standard deviation of a pop- graph, 234
ulation, 430 Greek letters, 133, 660
estimating, 102 Green-winged Orchid, 434, 446, 471
expected numbers, 448 gun clinometer, 323
exponent, 40
exponential decay, 288, 370, 376, 505, half-life, 290, 378
510 hertz (unit), 144
exponential function, 505 histogram, 237
exponential growth, 292, 372, 379, 505 hyperbola, 282
expression, 156 hypotenuse, 310
extrapolation, 246 hypothesis, 435
hypothesis testing, 435
factor, 195
factorize, 195 imaginary number, 13, 201
femto (prefix), 69 improper fraction, 28
Fibonacci numbers, 14 independent outcomes (probability), 391
first derivative, 484 independent variable, 258
formulae, 132, 655 index (plural indices), 40
fractional exponent, 50, 492 inheritance, 392, 398
fractions, 24, 183, 390 integer, 10
function, 261, 493 intercept, 272
function notation, 495 interpolation, 245
interval level, 443
giga (prefix), 69 inverse cosine, 318
gradient, 252, 479 inverse proportionality, 282

Back J I 833
Contents 
inverse sine, 318 lowest common denominator, 28
inverse tangent, 318
inverse trigonometric function, 318 Ma (million years), 71
irrational number, 12 magnitude, 512
mass, 66
joule (unit), 144 matched samples, 470
mean, 410
kelvin (unit), 66 median, 424
Kepler’s third law, 368 mega (prefix), 69
kilo (prefix), 69 Mendel’s peas, 398
kilogram, 66 metre, 66
k-value analysis, 360 micro (prefix), 69
latitude, 298 milli (prefix), 69
Leibniz, Gottfried Wilhelm, 496 minute of arc, 300
level of measurement, 443 mixed number, 28
linear relationship, 250 mode, 423
linear scale, 78 modulus, 514
litre, 112 mole, 66
log–linear graph, 370 multiplication rule (probability), 391
log–log graph, 364 mutually exclusive outcomes (proba-
logarithm, 346 bility), 395
logarithm to base 10, 347 nano (prefix), 69
logarithm to base e, 373 Napier, John, 346
logarithmic scale, 78, 352 Napier’s bones, 346
longitude, 298

Back J I 834
Contents 
natural logarithm, 373 population mean, 428
negative angles, 321 positive correlation, 459
negative correlation, 459 power, 39
negative exponent, 41, 114, 492 powers of powers, 48
negative numbers, 15 powers of ten, 42, 59, 98
newton (unit), 144 precedence, 53
Newton, Sir Isaac, 496 precision, 406, 421
normal distribution, 409, 419 probability, 385, 440
null hypothesis, 437 problem solving, 217
number line, 10, 618 proportionality, 265
numerator, 24 Pythagoras, 310
Pythagoras’ Theorum, 310, 516
observed numbers, 448
order of arithmetic operations, 53 radian, 302
order of magnitude, 74 radioactive decay, 288, 291, 378, 510,
ordinal level, 443 767
origin (of graph), 246 random uncertainty, 406
ranked data, 443, 461, 465
parabola, 276, 501 rate of change, 479
pascal (unit), 144 ratio, 398
percentage, 26, 390 rational number, 12
pH scale, 353 real number, 13
photoelectric effect, 255, 274 rearranging equations, 155
pico (prefix), 69 reciprocal, 41
population, 427 recurring decimal, 12

Back J I 835
Contents 
repeated measurements, 406 simultaneous equations, 207
replicate samples, 458 sine, 313
resolving vectors, 515 sketch graph, 267
Richter scale, 81, 352 skewed distribution, 422
ridge-and-furrow, 434 small angle approximation, 338
right angle, 297 Snell’s law, 333
right-angled triangle, 310 Spearman rank correlation coefficient,
root, 50 460
rounding, 74, 85 Spearman rank correlation coefficient
rounding error, 94 (critical values), 803
speed, 113, 115, 512
sample, 427 square root, 21, 50, 159, 160
sample standard deviation, 429 standard deviation, 413
scalar, 512 standard error, 471
scientific notation, 60, 97 standard index form, 60
sea-floor spreading, 716 statistically significant, 441
second (of arc), 301 statistics, 405, 433
second (time), 66 straight-line graph, 248, 277
second derivative, 500 Student’s t-test, 469
semi-log graph paper, 372 Student’s t-test (critical values), 805
SI units, 66, 144 subject (of equation), 154
significance, 441 substitution, 141, 203
significance level, 440 subtend, 295
significant figures, 84, 93, 317, 416 symbols, 133
simplifying equations, 183 systematic uncertainty, 407

Back J I 836
Contents 
tangent (to a curved graph), 479 writing maths, 105
tangent (trigonometry), 313
tera (prefix), 69 zero, 23
term, 156
test of association, 435
test of difference, 435
test statistic, 439, 449, 463, 471
trigonometric ratios, 313
trigonometry, 312
true mean, 428
t-test, 469
t-test (critical values), 805

unit conversions, 106

units, 65, 92, 141
units (on axes of graphs), 240
unmatched samples, 470

vector, 512
velocity, 512
volume, 110

watt (unit), 144

wave motion, 321
weight, 68, 144
word equation, 131

Back J I 837
Contents 
End of document

Back J 838