Calculus For Computer Graphics 2Nd Edition John Vince Online Ebook Texxtbook Full Chapter PDF

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 69

Calculus for Computer Graphics 2nd

Edition John Vince


Visit to download the full and correct content document:
https://ebookmeta.com/product/calculus-for-computer-graphics-2nd-edition-john-vince
/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Calculus for Computer Graphics 3rd Edition John Vince

https://ebookmeta.com/product/calculus-for-computer-graphics-3rd-
edition-john-vince-2/

Calculus for Computer Graphics, 3rd Edition John Vince

https://ebookmeta.com/product/calculus-for-computer-graphics-3rd-
edition-john-vince/

Quaternions for Computer Graphics 2nd Edition John


Vince

https://ebookmeta.com/product/quaternions-for-computer-
graphics-2nd-edition-john-vince/

Vector Analysis for Computer Graphics 2nd Edition John


Vince

https://ebookmeta.com/product/vector-analysis-for-computer-
graphics-2nd-edition-john-vince/
Vector Analysis for Computer Graphics (Second Edition)
John Vince

https://ebookmeta.com/product/vector-analysis-for-computer-
graphics-second-edition-john-vince/

Mathematics for Computer Graphics (Undergraduate Topics


in Computer Science), 6th Edition 2022 John Vince

https://ebookmeta.com/product/mathematics-for-computer-graphics-
undergraduate-topics-in-computer-science-6th-edition-2022-john-
vince/

Foundation Mathematics for Computer Science: A Visual


Approach John Vince

https://ebookmeta.com/product/foundation-mathematics-for-
computer-science-a-visual-approach-john-vince/

The LATEX Graphics Companion Tools and Techniques for


Computer Typesetting 2nd Edition Michel Goossens

https://ebookmeta.com/product/the-latex-graphics-companion-tools-
and-techniques-for-computer-typesetting-2nd-edition-michel-
goossens/

Plan Graphics for the Landscape Designer With Section


Elevation and Computer Graphics 3rd Edition Tony
Bertauski

https://ebookmeta.com/product/plan-graphics-for-the-landscape-
designer-with-section-elevation-and-computer-graphics-3rd-
edition-tony-bertauski/
John Vince

Calculus for
Computer
Graphics
Second Edition
Calculus for Computer Graphics
John Vince

Calculus for Computer


Graphics
Second Edition

123
John Vince
Bournemouth University
Hereford, UK

ISBN 978-3-030-11375-9 ISBN 978-3-030-11376-6 (eBook)


https://doi.org/10.1007/978-3-030-11376-6

Library of Congress Control Number: 2019930968

1st edition: © Springer-Verlag London 2013


2nd edition: © Springer Nature Switzerland AG 2019
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
This book is dedicated to my wife Hei @ip.
Preface

Calculus is one of those subjects that appears to have no boundaries, which is why
some Calculus books are so large and heavy! So when I started writing the first
edition of this book, I knew that it would not fall into this category. It would be
around 200 pages long and take the reader on a gentle journey through the subject,
without placing too many demands on their knowledge of mathematics.
Apart from reviewing the original text and correcting a few typos, this second
edition incorporates 3 extra chapters, and all 175 illustrations are in colour. I have
also extended Chap. 9 on arc length to include parameterisation of curves.
The objective of the book remains the same: to inform the reader about functions
and their derivatives, and the inverse process: integration, which can be used for
computing area and volume. The emphasis on geometry gives the book relevance to
the computer graphics community, and hopefully will provide the mathematical
background for professionals working in computer animation, games and allied
disciplines to read and understand other books and technical papers where differ-
ential and integral notation is found.
The book divides into 16 chapters, with the obligatory Introduction and
Conclusion chapters. Chapter 2 reviews the ideas of functions, their notation and the
different types encountered in everyday mathematics. This can be skipped by
readers already familiar with the subject.
Chapter 3 introduces the idea of limits and derivatives, and how mathematicians
have adopted limits in preference to infinitesimals. Most authors introduce inte-
gration as a separate subject, but I have included it in this chapter so that it is seen as
an antiderivative, rather than something independent.
Chapter 4 looks at derivatives and antiderivatives for a wide range of functions
such as polynomial, trigonometric, exponential and logarithmic. It also shows how
function sums, products, quotients and function of a function are differentiated.
Chapter 5 covers higher derivatives and how they are used to detect a local
maximum and minimum.

vii
viii Preface

Chapter 6 covers partial derivatives, which although are easy to understand, have
a reputation for being difficult. This is possibly due to the symbols used, rather than
the underlying mathematics. The total derivative is introduced here as it is required
in a later chapter.
Chapter 7 introduces the standard techniques for integrating different types of
functions. This can be a large subject, and I have deliberately kept the examples
simple in order to keep the reader interested and on top of the subject.
Chapter 8 shows how integration reveals the area under a graph and the concept
of the Riemann sum. The idea of representing area or volume as the limiting sum of
some fundamental unit, is central to understanding Calculus.
Chapter 9 deals with arc length, and uses a variety of worked examples to
compute the length of different curves and their parameterisation.
Chapter 10 shows how single and double integrals are used to compute the
surface area for different objects. It is also a convenient point to introduce
Jacobians, which hopefully I have managed to explain convincingly.
Chapter 11 shows how single, double and triple integrals are used to compute the
volume of familiar objects. It also shows how the choice of a coordinate system
influences a solution’s complexity.
Chapter 12 covers vector-valued functions, and provides a short introduction to
this very large subject.
Chapter 13 shows how to calculate tangent and normal vectors for a variety of
curves and surfaces, which will be useful in shading algorithms and physically
based animation.
Chapter 14 shows how differential Calculus is used to manage geometric con-
tinuity in B-splines and Bézier curves.
Chapter 15 looks at the curvature of curves such as a circle, helix, parabola and
parametric plane curves. It also shows how to compute the curvature of 2D
quadratic and cubic Bézier curves.
I used Springer’s excellent author’s LaTeX development kit on my Apple iMac,
which is so fast that I could recompile the entire book in 3 or 4 s, just to change a
single character! This book contains over 170 colour illustrations to provide a
strong visual interpretation for derivatives, antiderivatives and the calculation of arc
length, curvature, tangent vectors, area and volume. I used Apple’s Grapher
application for most of the graphs and rendered images, and Pages for the diagrams.
There is no way I could have written this book without the Internet and several
excellent books on Calculus. One only has to Google ‘What is a Jacobian?’ to
receive over a 1000 entries in about 0.25 s! YouTube also contains some highly
informative presentations on virtually every aspect of Calculus one could want. So I
have spent many hours watching, absorbing and disseminating videos, looking for
vital pieces of information that are key to understanding a topic.
The books I have referred to include: Teach Yourself Calculus, by Hugh Neil,
Calculus of One Variable, by Keith Hirst, Inside Calculus, by George Exner, Short
Calculus, by Serge Lang and my all-time favourite: Mathematics from the Birth of
Numbers, by Jan Gullberg. I acknowledge and thank all these authors for the
influence they have had on this book. One other book that has helped me is Digital
Preface ix

Typography Using LaTeX by Apostolos Syropoulos, Antonis Tsolomitis and Nick


Sofroniou.
Writing any book can be a lonely activity, and finding someone willing to read
an early draft, and whose opinion one can trust is extremely valuable.
Consequently, I thank Dr. Tony Crilly for his valuable feedback after reading the
final manuscript. Tony identified flaws in my reasoning and inconsistent notation,
and I have incorporated his suggestions. However, I take full responsibility for any
mistakes that may have found there way into this publication.
Finally, I would like to thank Helen Desmond, Editor for Computer Science,
Springer UK, for her continuing professional support.

Breinton, Herefordshire Professor Emeritus John Vince


January 2019 M.Tech, Ph.D., D.Sc., C.Eng.
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 What is Calculus? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Where is Calculus Used in Computer Graphics? . . . . . . . . . . . . 2
1.3 Who Invented Calculus? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Expressions, Variables, Constants and Equations . . . . . . . . . . . 5
2.3 Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3.1 Continuous and Discontinuous Functions . . . . . . . . . . . 7
2.3.2 Linear Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3.3 Periodic Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.4 Polynomial Functions . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.5 Function of a Function . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.6 Other Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 A Function’s Rate of Change . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4.1 Slope of a Function . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4.2 Differentiating Periodic Functions . . . . . . . . . . . . . . . . 15
2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 Limits and Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Small Numerical Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3 Equations and Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.1 Quadratic Function . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.2 Cubic Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.3.3 Functions and Limits . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.3.4 Graphical Interpretation of the Derivative . . . . . . . . . . 26
3.3.5 Derivatives and Differentials . . . . . . . . . . . . . . . . . . . . 27
3.3.6 Integration and Antiderivatives . . . . . . . . . . . . . . . . . . 28

xi
xii Contents

3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.5 Worked Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.5.1 Limiting Value of a Quotient . . . . . . . . . . . . . . . . . . . 31
3.5.2 Limiting Value of a Quotient . . . . . . . . . . . . . . . . . . . 31
3.5.3 Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.5.4 Slope of a Polynomial . . . . . . . . . . . . . . . . . . . . . . . . 32
3.5.5 Slope of a Periodic Function . . . . . . . . . . . . . . . . . . . . 33
3.5.6 Integrate a Polynomial . . . . . . . . . . . . . . . . . . . . . . . . 33
4 Derivatives and Antiderivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2 Differentiating Groups of Functions . . . . . . . . . . . . . . . . . . . . . 36
4.2.1 Sums of Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2.2 Function of a Function . . . . . . . . . . . . . . . . . . . . . . . . 38
4.2.3 Function Products . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.2.4 Function Quotients . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.2.5 Summary: Groups of Functions . . . . . . . . . . . . . . . . . . 47
4.3 Differentiating Implicit Functions . . . . . . . . . . . . . . . . . . . . . . . 47
4.4 Differentiating Exponential and Logarithmic Functions . . . . . . . 50
4.4.1 Exponential Functions . . . . . . . . . . . . . . . . . . . . . . . . 50
4.4.2 Logarithmic Functions . . . . . . . . . . . . . . . . . . . . . . . . 52
4.4.3 Summary: Exponential and Logarithmic Functions . . . . 54
4.5 Differentiating Trigonometric Functions . . . . . . . . . . . . . . . . . . 54
4.5.1 Differentiating tan . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.5.2 Differentiating csc . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.5.3 Differentiating sec . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.5.4 Differentiating cot . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.5.5 Differentiating arcsin, arccos and arctan . . . . . . . . . . . . 58
4.5.6 Differentiating arccsc, arcsec and arccot . . . . . . . . . . . . 59
4.5.7 Summary: Trigonometric Functions . . . . . . . . . . . . . . . 60
4.6 Differentiating Hyperbolic Functions . . . . . . . . . . . . . . . . . . . . 62
4.6.1 Differentiating sinh, cosh and tanh . . . . . . . . . . . . . . . 63
4.6.2 Differentiating cosech, sech and coth . . . . . . . . . . . . . . 65
4.6.3 Differentiating arsinh, arcosh and artanh . . . . . . . . . . . 66
4.6.4 Differentiating arcsch, arsech and arcoth . . . . . . . . . . . 67
4.6.5 Summary: Hyperbolic Functions . . . . . . . . . . . . . . . . . 70
4.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5 Higher Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.2 Higher Derivatives of a Polynomial . . . . . . . . . . . . . . . . . . . . . 71
5.3 Identifying a Local Maximum or Minimum . . . . . . . . . . . . . . . 73
5.4 Derivatives and Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Contents xiii

5.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.5.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 79
6 Partial Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.2 Partial Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.2.1 Visualising Partial Derivatives . . . . . . . . . . . . . . . . . . . 84
6.2.2 Mixed Partial Derivatives . . . . . . . . . . . . . . . . . . . . . . 86
6.3 Chain Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6.4 Total Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6.5 Second-Order and Higher Partial Derivatives . . . . . . . . . . . . . . 91
6.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.6.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 91
7 Integral Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
7.2 Indefinite Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
7.3 Standard Integration Formulae . . . . . . . . . . . . . . . . . . . . . . . . . 94
7.4 Integration Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
7.4.1 Continuous Functions . . . . . . . . . . . . . . . . . . . . . . . . . 95
7.4.2 Difficult Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
7.4.3 Trigonometric Identities . . . . . . . . . . . . . . . . . . . . . . . 97
7.4.4 Exponent Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7.4.5 Completing the Square . . . . . . . . . . . . . . . . . . . . . . . . 100
7.4.6 The Integrand Contains a Derivative . . . . . . . . . . . . . . 102
7.4.7 Converting the Integrand into a Series
of Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.4.8 Integration by Parts . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7.4.9 Integration by Substitution . . . . . . . . . . . . . . . . . . . . . 113
7.4.10 Partial Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
8 Area Under a Graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8.2 Calculating Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8.3 Positive and Negative Areas . . . . . . . . . . . . . . . . . . . . . . . . . . 129
8.4 Area Between Two Functions . . . . . . . . . . . . . . . . . . . . . . . . . 130
8.5 Areas with the y-Axis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
8.6 Area with Parametric Functions . . . . . . . . . . . . . . . . . . . . . . . . 133
8.7 Bernhard Riemann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.7.1 Domains and Intervals . . . . . . . . . . . . . . . . . . . . . . . . 135
8.7.2 The Riemann Sum . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
8.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
xiv Contents

9 Arc Length and Parameterisation of Curves . . . . . . . . . . . . . . . . . . 139


9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
9.2 Lagrange’s Mean-Value Theorem . . . . . . . . . . . . . . . . . . . . . . 139
9.3 Arc Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
9.3.1 Arc Length of a Straight Line . . . . . . . . . . . . . . . . . . . 142
9.3.2 Arc Length of a Circle . . . . . . . . . . . . . . . . . . . . . . . . 142
9.3.3 Arc Length of a Parabola . . . . . . . . . . . . . . . . . . . . . . 143
3
9.3.4 Arc Length of y ¼ x2 . . . . . . . . . . . . . . . . . . . . . . . . . 147
9.3.5 Arc Length of a Sine Curve . . . . . . . . . . . . . . . . . . . . 148
9.3.6 Arc Length of a Hyperbolic Cosine Function . . . . . . . . 148
9.3.7 Arc Length of Parametric Functions . . . . . . . . . . . . . . 149
9.3.8 Arc Length of a Circle . . . . . . . . . . . . . . . . . . . . . . . . 150
9.3.9 Arc Length of an Ellipse . . . . . . . . . . . . . . . . . . . . . . 151
9.3.10 Arc Length of a Helix . . . . . . . . . . . . . . . . . . . . . . . . 152
9.3.11 Arc Length of a 2D Quadratic Bézier Curve . . . . . . . . 153
9.3.12 Arc Length of a 3D Quadratic Bézier Curve . . . . . . . . 155
9.3.13 Arc Length Parameterisation of a 3D Line . . . . . . . . . . 157
9.3.14 Arc Length Parameterisation of a Helix . . . . . . . . . . . . 160
9.3.15 Positioning Points on a Straight Line Using
a Square Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
9.3.16 Positioning Points on a Helix Curve Using
a Square Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
9.3.17 Arc Length Using Polar Coordinates . . . . . . . . . . . . . . 164
9.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
9.4.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 167
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
10 Surface Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.2 Surface of Revolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.2.1 Surface Area of a Cylinder . . . . . . . . . . . . . . . . . . . . . 171
10.2.2 Surface Area of a Right Cone . . . . . . . . . . . . . . . . . . . 171
10.2.3 Surface Area of a Sphere . . . . . . . . . . . . . . . . . . . . . . 173
10.2.4 Surface Area of a Paraboloid . . . . . . . . . . . . . . . . . . . 175
10.3 Surface Area Using Parametric Functions . . . . . . . . . . . . . . . . . 176
10.4 Double Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
10.5 Jacobians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
10.5.1 1D Jacobian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
10.5.2 2D Jacobian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
10.5.3 3D Jacobian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
10.6 Double Integrals for Calculating Area . . . . . . . . . . . . . . . . . . . 188
10.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
10.7.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 194
Contents xv

11 Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
11.2 Solid of Revolution: Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
11.2.1 Volume of a Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . 198
11.2.2 Volume of a Right Cone . . . . . . . . . . . . . . . . . . . . . . . 199
11.2.3 Volume of a Right Conical Frustum . . . . . . . . . . . . . . 200
11.2.4 Volume of a Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . 202
11.2.5 Volume of an Ellipsoid . . . . . . . . . . . . . . . . . . . . . . . . 203
11.2.6 Volume of a Paraboloid . . . . . . . . . . . . . . . . . . . . . . . 204
11.3 Solid of Revolution: Shells . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
11.3.1 Volume of a Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . 206
11.3.2 Volume of a Right Cone . . . . . . . . . . . . . . . . . . . . . . . 207
11.3.3 Volume of a Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . 207
11.3.4 Volume of a Paraboloid . . . . . . . . . . . . . . . . . . . . . . . 209
11.4 Volumes with Double Integrals . . . . . . . . . . . . . . . . . . . . . . . . 210
11.4.1 Objects with a Rectangular Base . . . . . . . . . . . . . . . . . 211
11.4.2 Rectangular Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
11.4.3 Rectangular Prism . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
11.4.4 Curved Top . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
11.4.5 Objects with a Circular Base . . . . . . . . . . . . . . . . . . . . 214
11.4.6 Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
11.4.7 Truncated Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
11.5 Volumes with Triple Integrals . . . . . . . . . . . . . . . . . . . . . . . . . 217
11.5.1 Rectangular Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
11.5.2 Volume of a Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . 218
11.5.3 Volume of a Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . 220
11.5.4 Volume of a Cone . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
11.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
11.6.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 223
12 Vector-Valued Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
12.2 Differentiating Vector Functions . . . . . . . . . . . . . . . . . . . . . . . 225
12.2.1 Velocity and Speed . . . . . . . . . . . . . . . . . . . . . . . . . . 226
12.2.2 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
12.2.3 Rules for Differentiating Vector-Valued Functions . . . . 228
12.3 Integrating Vector-Valued Functions . . . . . . . . . . . . . . . . . . . . 229
12.3.1 Velocity of a Falling Object . . . . . . . . . . . . . . . . . . . . 230
12.3.2 Position of a Moving Object . . . . . . . . . . . . . . . . . . . . 230
12.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
12.4.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 231
xvi Contents

13 Tangent and Normal Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233


13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
13.2 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
13.3 Tangent Vector to a Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
13.4 Normal Vector to a Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
13.5 Gradient of a Scalar Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
13.5.1 Unit Tangent and Normal Vectors to a Line . . . . . . . . 240
13.5.2 Unit Tangent and Normal Vectors to a Parabola . . . . . 242
13.5.3 Unit Tangent and Normal Vectors to a Circle . . . . . . . 244
13.5.4 Unit Tangent and Normal Vectors to an Ellipse . . . . . . 246
13.5.5 Unit Tangent and Normal Vectors to a Sine Curve . . . . 248
13.5.6 Unit Tangent and Normal Vectors to a cosh Curve . . . 249
13.5.7 Unit Tangent and Normal Vectors to a Helix . . . . . . . . 251
13.5.8 Unit Tangent and Normal Vectors to a Quadratic
Bézier Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
13.6 Unit Tangent and Normal Vectors to a Surface . . . . . . . . . . . . . 255
13.6.1 Unit Normal Vectors to a Bilinear Patch . . . . . . . . . . . 255
13.6.2 Unit Normal Vectors to a Quadratic Bézier Patch . . . . . 256
13.6.3 Unit Tangent and Normal Vector to a Sphere . . . . . . . 259
13.6.4 Unit Tangent and Normal Vectors to a Torus . . . . . . . . 261
13.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
13.7.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 263
14 Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
14.2 B-Splines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
14.2.1 Uniform B-Splines . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
14.2.2 B-Spline Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . 269
14.3 Derivatives of a Bézier Curve . . . . . . . . . . . . . . . . . . . . . . . . . 271
14.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
15 Curvature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
15.2 Curvature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
15.2.1 Curvature of a Circle . . . . . . . . . . . . . . . . . . . . . . . . . 278
15.2.2 Curvature of a Helix . . . . . . . . . . . . . . . . . . . . . . . . . . 279
15.2.3 Curvature of a Parabola . . . . . . . . . . . . . . . . . . . . . . . 281
15.2.4 Parametric Plane Curve . . . . . . . . . . . . . . . . . . . . . . . . 283
15.2.5 Curvature of a Graph . . . . . . . . . . . . . . . . . . . . . . . . . 286
15.2.6 Curvature of a 2D Quadratic Bézier Curve . . . . . . . . . 286
15.2.7 Curvature of a 2D Cubic Bézier Curve . . . . . . . . . . . . 288
15.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
15.3.1 Summary of Formulae . . . . . . . . . . . . . . . . . . . . . . . . 289
Contents xvii

16 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Appendix A: Limit of ðsin hÞ=h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Appendix B: Integrating cosn h . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Chapter 1
Introduction

1.1 What is Calculus?

Well this is an easy question to answer. Basically, Calculus has two parts: differential
and integral. Differential Calculus is used for computing a function’s rate of change
relative to one of its arguments. Generally, one begins with a function such as f (x),
and as x changes, a corresponding change occurs in f (x). Differentiating f (x) with
respect to x, produces a second function f  (x), which gives the rate of change of f (x)
for any x. For example, and without explaining why, if f (x) = x 2 , then f  (x) = 2x,
and when x = 3, f (x) is changing 2 × 3 = 6 times faster than x. Which is rather
neat!
In practice, one also writes y = x 2 , or even y = f (x), which means that differ-
entiating is expressed in a variety of ways:

dy d d
f  (x), , f (x), y,
dx dx dx

thus for y = f (x) = x 2 , we can write

dy d d
f  (x) = 2x, = 2x, f (x) = 2x, y = 2x.
dx dx dx

Integral Calculus reverses the operation, where integrating f  (x), produces f (x),
or something similar. But surely, Calculus can’t be as easy as this, you’re asking
yourself? Well, there are some problems, which is what this book is about. To begin
with, not all functions are easily differentiated, as they may contain hidden infinities
and discontinuities. Some functions are expressed as products or quotients, and many
functions possess more than one argument. All these, and other conditions, must be
addressed. Furthermore, integrating a function produces some useful benefits, such
as calculating the area under a graph, the length of curves, and the surface area and
volume of objects. But more of this later.

© Springer Nature Switzerland AG 2019 1


J. Vince, Calculus for Computer Graphics,
https://doi.org/10.1007/978-3-030-11376-6_1
2 1 Introduction

Fig. 1.1 Two abutting


curves without matching
slopes

But why should we be interested in rates of change? Well, say we have a function
that specifies the changing velocity of an object over time, then differentiating the
function gives the rate of change of the function over time, which is the object’s
acceleration. And knowing the object’s mass and acceleration, we can compute the
force responsible for the object’s acceleration. There are many more reasons for
having an interest in rates of change, which will emerge in the following chapters.

1.2 Where is Calculus Used in Computer Graphics?

If you are lucky, you may work in computer graphics without having to use Calculus,
but some people have no choice but to understand it, and use it in their work. For
example, we often join together curved lines and surfaces. Figure 1.1 shows two
abutting curves, where the join is clearly visible. This is because the slope information
at the end of the first curve, does not match the slope information at the start of the
second curve. By expressing the curves as functions, differentiating them gives their
slopes at any point in the form of two other functions. These slope functions can
also be differentiated, and by ensuring that the original curves possess the same
differentials at the join, a seamless join is created. The same process is used for
abutting two or more surface patches.
Calculus finds its way into other aspects of computer graphics such as digital
differential analysers (DDAs) for drawing lines and curves, interpolation, curvature,
arc-length parametrisation, fluid animation, rendering, animation, modelling, etc. In
later chapters I will show how Calculus permits us to calculate surface normals to
curves and surfaces, and the curvature of different curves.

1.3 Who Invented Calculus?

More than three-hundred years have passed since the English astronomer, physicist
and mathematician Isaac Newton (1643–1727) and the German mathematician Got-
tfried Leibniz (1646–1716) published their treaties describing Calculus. So called
‘infinitesimals’ played a pivotal role in early Calculus to determine tangents, area and
1.3 Who Invented Calculus? 3

volume. Incorporating incredibly small quantities (infinitesimals) into a numerical


solution, means that products involving them can be ignored, whilst quotients are
retained. The final solution takes the form of a ratio representing the change of a
function’s value, relative to a change in its independent variable.
Although infinitesimal quantities have helped mathematicians for more than two-
thousand years solve all sorts of problems, they were not widely accepted as a rig-
orous mathematical tool. In the latter part of the 19th century, they were replaced
by incremental changes that tend towards zero to form a limit identifying some
desired result. This was mainly due to the work of the German mathematician Karl
Weierstrass (1815–1897), and the French mathematician Augustin-Louis Cauchy
(1789–1857).
In spite of the basic ideas of Calculus being relatively easy to understand, it has
a reputation for being difficult and intimidating. I believe that the problem lies in
the breadth and depth of Calculus, in that it can be applied across a wide range
of disciplines, from electronics to cosmology, where the notation often becomes
extremely abstract with multiple integrals, multi-dimensional vector spaces and
matrices formed from partial differential operators. In this book I introduce the reader
to those elements of Calculus that are both easy to understand and relevant to solving
various mathematical problems found in computer graphics.
Perhaps you have studied Calculus at some time, and have not had the opportunity
to use it regularly and become familiar with its ways, tricks and analytical techniques.
In which case, this book could awaken some distant memory and reveal a subject
with which you were once familiar. On the other hand, this might be your first journey
into the world of functions, limits, differentials and integrals – in which case, you
should find the journey exciting!
Chapter 2
Functions

2.1 Introduction

In this chapter the notion of a function is introduced as a tool for generating one
numerical quantity from another. In particular, we look at equations, their variables
and any possible sensitive conditions. This leads toward the idea of how fast a function
changes relative to its independent variable. The second part of the chapter introduces
two major operations of Calculus: differentiating, and its inverse, integrating. This is
performed without any rigorous mathematical underpinning, and permits the reader
to develop an understanding of Calculus without using limits.

2.2 Expressions, Variables, Constants and Equations

One of the first things we learn in mathematics is the construction of expressions, such
as 2(x + 5) − 2, using variables, constants and arithmetic operators. The next step
is to develop an equation, which is a mathematical statement, in symbols, declaring
that two things are exactly the same (or equivalent). For example, (2.1) is the equation
representing the surface area of a sphere:

S = 4πr 2 (2.1)

where S and r are variables. They are variables because they take on different values,
depending on the size of the sphere. S depends upon the changing value of r , and to
distinguish between the two, S is called the dependent variable, and r the independent
variable. Similarly, (2.2) is the equation for the volume of a torus:

V = 2π 2 r 2 R (2.2)

© Springer Nature Switzerland AG 2019 5


J. Vince, Calculus for Computer Graphics,
https://doi.org/10.1007/978-3-030-11376-6_2
6 2 Functions

where the dependent variable V depends on the torus’s minor radius r and major
radius R, which are both independent variables. Note that both formulae include
constants 4, π and 2. There are no restrictions on the number of variables or constants
employed within an equation.

2.3 Functions

The concept of a function is that of a dependent relationship. Some equations merely


express an equality, such as 19 = 15 + 4, but a function is a special type of equa-
tion in which the value of one variable (the dependent variable) depends on, and is
determined by, the values of one or more other variables (the independent variables).
Thus, in the equation
S = 4πr 2

one might say that S is a function of r , and in the equation

V = 2π 2 r 2 R

V is a function of r and also of R.


It is usual to write the independent variables, separated by commas, in brackets
immediately after the symbol for the dependent variable, and so the two equations
above are usually written
S(r ) = 4πr 2

and
V (r, R) = 2π 2 r 2 R.

The order of the independent variables is immaterial.


Mathematically, there is no difference between equations and functions, it is sim-
ply a question of notation. However, when we do not have an equation, we can use the
idea of a function to help us develop one. For example, no one has been able to find
an equation that generates the nth prime number, but I can declare a function P(n)
that pretends to perform this operation, such that P(1) = 2, P(2) = 3, P(3) = 5,
etc. At least this imaginary function P(n), permits me to move forward and reflect
upon its possible inner structure.
A mathematical function must have a precise definition. It must be predictable,
and ideally, work under all conditions. √
We are all familiar with mathematical functions such as sin x, cos x, tan x, x,
etc., where x is the independent variable. Such functions permit us to confidently
write statements such as
2.3 Functions 7

sin 30◦ = 0.5


cos 90◦ = 0.0
tan 45◦ = 1.0

16 = ±4

without worrying whether they will always provide a correct answer.


We often need to design a function to perform a specific task. For instance, if I
require a function y(x) to compute x 2 + x + 6, the independent variable is x and the
function is written
y(x) = x 2 + x + 6

such that

y(0) = 02 + 0 + 6 = 6
y(1) = 12 + 1 + 6 = 8
y(2) = 22 + 2 + 6 = 12
y(3) = 32 + 3 + 6 = 18.

2.3.1 Continuous and Discontinuous Functions

Understandably, a function’s value is sensitive to its independent variables. A simple


square-root function, for instance, expects a positive real number as its independent
variable, and registers an error condition for a negative value. On the other hand, a
useful square-root function would accept positive and negative numbers, and output
a real result for a positive input and a complex result for a negative input.
Another danger condition is the possibility of dividing by zero, which is not
permissible in mathematics. For example, the following function y(x) is undefined
for x = 1, and cannot be displayed on the graph shown in Fig. 2.1.

x2 + 1
y(x) =
x −1
2
y(1) =
0
which is why mathematicians include a domain of definition in the specification of
a function:
x2 + 1
y(x) = for x = 1.
x −1
8 2 Functions

Fig. 2.1 Graph of


y = (x 2 + 1)/(x − 1)
showing the discontinuity at 20
x =1
10

-30 -20 -10 0 10 20 30

-10

-20

We can create equations or functions that lead to all sorts of mathematical anomalies.
For example, (2.3) creates the condition 0/0 when x = 4

x −4
y(x) = √ (2.3)
x −2
0
y(4) =
0
Similarly, mathematicians would write (2.3) as

x −4
y(x) = √ for x = 4.
x −2

Such conditions have no numerical value. However, this does not mean that these
functions are unsound—they are just sensitive to specific values of their independent
variable. Fortunately, there is a way of interpreting these results, as we will discover
in the next chapter.

2.3.2 Linear Functions

Linear functions are probably the simplest functions we will ever encounter and are
based upon equations of the form

y = mx + c.
2.3 Functions 9

Fig. 2.2 Graph of


y = 0.5x + 2 5

-5 0 5 10

-5

Fig. 2.3 Graph of


y = 5 sin x
5

4 -2 0 2 4

-5

For example, the function y(x) = 0.5x + 2 is shown as a graph in Fig. 2.2, where
0.5 is the slope, and 2 is the intercept with the y-axis.

2.3.3 Periodic Functions

Periodic functions are also relatively simple and employ the trigonometric functions
sin, cos and tan. For example, the function y(x) = 5 sin x is shown over the range
−4π < x < 4π as a graph in Fig. 2.3, where the 5 is the amplitude of the sine wave,
and x is the angle in radians.

2.3.4 Polynomial Functions

Polynomial functions take the form

f (x) = an x n + an−1 x n−1 + an−2 x n−2 + · · · b + a2 x 2 + a1 x + a0


10 2 Functions

Fig. 2.4 Graph of f (x) = f(x)


4x 4 − 5x 3 − 8x 2 + 6x − 12
20

10

-1 0 1 2

-10

where n takes on some value, and an are assorted constants. For example, the function
f (x) = 4x 4 − 5x 3 − 8x 2 + 6x − 12 is shown in Fig. 2.4.

2.3.5 Function of a Function

In mathematics we often combine functions to describe some relationship succinctly.


For example, the trigonometric function

f (x) = sin(2x + 1)

is a function of a function. Here we have 2x + 1, which can be expressed as the


function
u(x) = 2x + 1

and the original function becomes

f (u(x)) = sin(u(x)).

We can increase the depth of functions to any limit, and in the next chapter we
consider how such descriptions are untangled and analysed in Calculus.

2.3.6 Other Functions

You are probably familiar with other functions such as exponential, logarithmic,
complex, vector, recursive, etc., which can be combined together to encode simple
equations such as
e = mc2
2.3 Functions 11

or something more difficult such as

N −1
1 
A(k) = f j ω− jk for k = 0, 1, . . . , N − 1.
N j=0

2.4 A Function’s Rate of Change

Mathematicians are particularly interested in the rate at which a function changes


relative to its independent variable. Even you would be interested in this characteristic
in the context of your salary or pension annuity. For example, I would like to know
if my pension fund is growing linearly with time; whether there is some sustained
increasing growth rate; or more importantly, if the fund is decreasing! This is what
Calculus is about—it enables us to calculate how a function’s value changes, relative
to its independent variable.
The reason why Calculus appears daunting, is that there is such a wide range
of functions to consider: linear, periodic, complex, polynomial, imaginary, rational,
exponential, logarithmic, vector, etc. However, we must not be intimidated by such
a wide spectrum, as the majority of functions employed in computer graphics are
relatively simple, and there are plenty of texts that show how specific functions are
tackled.

2.4.1 Slope of a Function

In the linear equation


y = mx + c

the independent variable is x, but y is also influenced by the constant c, which


determines the intercept with the y-axis, and m, which determines the graph’s slope.
Figure 2.5 shows this equation with 4 different values of m. For any value of x, the
slope always equals m, which is what linear means.
In the quadratic equation
y = ax 2 + bx + c

y is dependent on x, but in a much more subtle way. It is a combination of two


components: a square law component ax 2 , and a linear component bx + c. Figure 2.6
shows these two components and their sum for the equation y = 0.5x 2 − 2x + 1.
For any value of x, the slope is different. Figure 2.7 identifies three slopes on the
graph. For example, when x = 2, y = −1, and the slope is zero. When x = 4, y = 1,
and the slope looks as though it equals 2. And when x = 0, y = 1, the slope looks
as though it equals −2.
12 2 Functions

Fig. 2.5 Graph of 4 m=2


y = mx + 2 for different m=1
values of m
3 m = 0.5
m = 0.25
2

-4 -3 -2 -1 0 1 2 3 4 5

-1

-2

Fig. 2.6 Graph of


y = 0.5x 2 − 2x + 1 3
showing its two components
2

y = 0.5x2
1

-3 -2 -1 0 1 2 3 4 5 6

-1 y = 0.5x2-2x + 1

-2
y = -2x + 1

Fig. 2.7 Graph of


y = 0.5x 2 − 2x + 1 3
showing three gradients
2

slope = -2 1 slope = 2

-3 -2 -1 0 1 2 3 4 5 6

-1
slope = 0
-2
2.4 A Function’s Rate of Change 13

Fig. 2.8 Linear relationship slope


between slope and x 3

-3 -2 -1 0 1 2 3 4 5 6

-1

-2

Even though we have only three samples, let’s plot the graph of the relationship
between x and the slope m, as shown in Fig. 2.8. Assuming that other values of slope
lie on the same straight line, then the equation relating the slope m to x is

m = x − 2.

Summarising: we have discovered that the slope of the function

f (x) = 0.5x 2 − 2x + 1

changes with the independent variable x, and is given by the function

f  (x) = x − 2.

Note that f (x) is the original function, and f  (x) (pronounced f prime of x) is the
function for the slope, which is a convention often used in Calculus.
Remember that we have taken only three sample slopes, and assumed that there
is a linear relationship between the slope and x. Ideally, we should have sampled the
graph at many more points to increase our confidence, but I happen to know that we
are on solid ground!
Calculus enables us to compute the function for the slope from the original func-
tion. i.e. to compute f  (x) from f (x):

f (x) = 0.5x 2 − 2x + 1 (2.4)



f (x) = x − 2. (2.5)

Readers who are already familiar with Calculus will know how to compute (2.5)
from (2.4), but for other readers, this is the technique:
14 2 Functions

1. Take each term of (2.4) in turn and replace ax n by nax n−1 .


2. Therefore 0.5x 2 becomes x.
3. −2x, which can be written −2x 1 , becomes −2x 0 , which is −2.
4. 1 is ignored, as it is a constant.
5. Collecting up the terms we have

f  (x) = x − 2.

This process is called differentiating a function, and is easy for this type of polyno-
mial. So easy in fact, we can differentiate the following function without thinking:

f (x) = 12x 4 + 6x 3 − 4x 2 + 3x − 8
f  (x) = 48x 3 + 18x 2 − 8x + 3.

This is an amazing relationship, and is one of the reasons why Calculus is so impor-
tant.
If we can differentiate a polynomial function, surely we can reverse the operation
and compute the original function? Well of course! For example, if f  (x) is given by

f  (x) = 6x 2 + 4x + 6 (2.6)

then this is the technique to compute the original function:


1
1. Take each term of (2.6) in turn and replace ax n by n+1 ax n+1 .
2 3
2. Therefore 6x becomes 2x .
3. 4x becomes 2x 2 .
4. 6 becomes 6x.
5. Introduce a constant C which might have been present in the original function.
6. Collecting up the terms we have

f (x) = 2x 3 + 2x 2 + 6x + C.

This process is called integrating a function. Thus Calculus is about differentiating


and integrating functions, which sounds rather easy, and in some cases it is true. The
problem is the breadth of functions that arise in mathematics, physics, geometry,
cosmology, science, etc. For example, how do we differentiate or integrate

sin x + cosh
x
f (x) = x
?
cos2 x − ln x 3

Personally, I don’t know, but hopefully, there is a solution somewhere.


2.4 A Function’s Rate of Change 15

Fig. 2.9 A sine curve over f(x)


the range 0◦ to 360◦ 2

-1

-2

Fig. 2.10 Sampled slopes of slope


a sine curve 2

-1

-2

2.4.2 Differentiating Periodic Functions

Now let’s try differentiating the sine function by sampling its slope at different points.
Figure 2.9 shows a sine curve over the range 0◦ –360◦ . When the scales for the vertical
and horizontal axes are equal, the slope is 1 at 0◦ and 360◦ . The slope is zero at 90◦
and 270◦ , and −1 at 180◦ . Figure 2.10 plots these slope values against x and connects
them with straight lines.
It should be clear from Fig. 2.9 that the slope of the sine wave does not change
linearly as shown in Fig. 2.10. The slope starts at 1, and for the first 20◦ , or so, slowly
falls away, and then collapses to zero, as shown in Fig. 2.11, which is a cosine wave
form. Thus, we can guess that differentiating a sine function creates a cosine function:

f (x) = sin x
f  (x) = cos x.

Consequently, integrating a cosine function creates a sine function. Now this analysis
is far from rigorous, but we will shortly provide one. Before moving on, let’s perform
a similar ‘guesstimate’ for the cosine function.
16 2 Functions

Fig. 2.11 The slope of a (x)


sine curve is a cosine curve 2

-1

-2

Fig. 2.12 Sampled slopes of slope


a cosine curve 2

-1

-2

Fig. 2.13 The slope of a f(x)


cosine curve is a negative 2
sine curve
1

-1

-2

Figure 2.11 shows a cosine curve, where the slope is zero at 0◦ , 180◦ and 360◦ .
The slope equals −1 at 90◦ , and equals 1 at 270◦ . Figure 2.12 plots these slope values
against x and connects them with straight lines. Using the same argument for the
sine curve, this can be represented by f  (x) = − sin x as shown in Fig. 2.13.
2.4 A Function’s Rate of Change 17

Summarising, we have:

f (x) = sin x
f  (x) = cos x
f (x) = cos x
f  (x) = − sin x

which illustrates the intimate relationship between the sine and cosine functions.
Just in case you are suspicious of these results, they can be confirmed by differ-
entiating the power series for the sine and cosine functions. For example, the sine
and cosine functions are represented by the series

x3 x5 x7
sin x = x − + − + ···
3! 5! 7!
x2 x4 x6
cos x = 1 − + − + ···
2! 4! 6!
and differentiating the sine function using the above technique for a polynomial we
obtain
x2 x4 x6
f  (x) = 1 − + − + ···
2! 4! 6!
which is the cosine function. Similarly, differentiating the cosine function, we obtain
 
 x3 x5 x7
f (x) = − x − + − + ···
3! 5! 7!

which is the negative sine function.


Finally, there is a series that when differentiated, remains the same:

x2 x3 x4
f (x) = 1 + x + + + + ···
2! 3! 4!
x2 x3 x4
f  (x) = 1 + x + + + + ···
2! 3! 4!
which is e x , and has a rate of growth equal to itself!

2.5 Summary

We have covered quite a lot in this chapter, but hopefully it was not too challenging,
bearing in mind the subject. We have covered the nature of simple functions and
noted that Calculus is interested in a function’s rate of change, relative to its inde-
18 2 Functions

pendent variable. Differentiating a function creates another function that describes


the function’s rate of change relative to its independent variable. For simple polyno-
mials, this is a trivial algebraic operation, which can even be undertaken by software.
For trigonometric functions, there is a direct relationship between the sine and cosine
functions.
Integration is the reverse process, where the original function is derived from a
knowledge of the differentiated form. Much more will be said of this process in later
chapters.
Chapter 3
Limits and Derivatives

3.1 Introduction

Some quantities, such as the area of a circle or an ellipse, cannot be written precisely,
as they incorporate π , which is irrational, but also transcendental; i.e. not a root
of a single-variable polynomial whose coefficients are all integers. However, an
approximate value can be obtained by devising a definition that includes a parameter
that is made infinitesimally small. The techniques of limits and infinitesimals have
been used in mathematics for over two-thousand years, and paved the way towards
today’s Calculus.
Although the principles of integral Calculus were being used by Archimedes (287–
212 B.C.) to compute areas, volumes and centres of gravity, it was Isaac Newton and
Gottfried Leibniz who are regarded as the true inventors of modern Calculus. Leibniz
published his results in 1684, followed by Newton in 1704. However, Newton had
been using his Calculus of fluxions as early as 1665. Since then, Calculus has evolved
conceptually and in notation.
Up until recently, Calculus was described using infinitesimals, which are num-
bers so small, they can be ignored in certain products. However, infinitesimals, no
matter how small they are, do not belong to an axiomatic mathematical system, and
eventually, Augustin-Louis Cauchy and Karl Weierstrass showed how they could be
replaced by limits. In this chapter I show how limits are used to measure a function’s
rate of change accurately, instead of using intelligent guess work. Limiting condi-
tions also permit us to explore the behaviour of functions that are discontinuous for
particular values of their independent variable. For example, rational functions are
often sensitive to a specific value of their variable, which gives rise to the meaningless
condition 0/0. Limits permit us to handle such conditions.
We continue to apply limiting conditions to identify a function’s derivative, which
provides a powerful analytical tool for computing the derivative of function sums,
products and quotients. We begin this chapter by exploring small numerical quantities
and how they can be ignored if they occur in certain products, but remain important
in quotients.

© Springer Nature Switzerland AG 2019 19


J. Vince, Calculus for Computer Graphics,
https://doi.org/10.1007/978-3-030-11376-6_3
20 3 Limits and Derivatives

3.2 Small Numerical Quantities

The adjective small is a relative term, and requires clarification in the context of
numbers. For example, if numbers are in the hundreds, and also contain some decimal
component, then it seems reasonable to ignore digits after the 3rd decimal place for
any quick calculation. For instance,

100.000003 × 200.000006 ≈ 20, 000

and ignoring the decimal part has no significant impact on the general accuracy of
the answer, which is measured in tens of thousands.
To develop an algebraic basis for this argument let’s divide a number into two
parts: a primary part x, and some very small secondary part δx (pronounced delta x).
In one of the above numbers, x = 100 and δx = 0.000003. Given two such numbers,
x1 and y1 , their product is given by

x1 = x + δx
y1 = y + δy
x1 y1 = (x + δx)(y + δy)
= x y + x · δy + y · δx + δx · δy.

Using x1 = 100.000003 and y1 = 200.000006 we have

x1 y1 = 100 × 200 + 100 × 0.000006 + 200 × 0.000003 + 0.000003 × 0.000006


= 20,000 + 0.0006 + 0.0006 + 0.00000000018
= 20,000 + 0.0012 + 0.00000000018
= 20,000.00120000018

where it is clear that the products x · δy, y · δx and δx · δy contribute very little to
the result. Furthermore, the smaller we make δx and δy, their contribution becomes
even more insignificant. Just imagine if we reduce δx and δy to the level of quantum
phenomenon, i.e. 10−34 , then their products play no part in every-day numbers. But
there is no need to stop there, we can make δx and δy as small as we like, e.g.
10−100,000,000,000 . Later on we employ the device of reducing a number towards zero,
such that any products involving them can be dropped from any calculation.
Even though the product of two numbers less than zero is an even smaller number,
care must be taken with their quotients. For example, in the above scenario, where
δy = 0.000006 and δx = 0.000003,

δy 0.000006
= =2
δx 0.000003
so we must watch out for such quotients.
3.2 Small Numerical Quantities 21

From now on I will employ the term derivative to describe a function’s rate of
change relative to its independent variable. I will now describe two ways of computing
a derivative, and provide a graphical interpretation of the process. The first way uses
simple algebraic equations, and the second way uses a functional representation.
Needless to say, they both give the same result.

3.3 Equations and Limits

3.3.1 Quadratic Function

Here is a simple algebraic approach using limits to compute the derivative of a


quadratic function. Starting with the function y = x 2 , let x change by δx, and let δy
be the corresponding change in y. We then have

y = x2
y + δy = (x + δx)2
= x 2 + 2x · δx + (δx)2
δy = 2x · δx + (δx)2 .

Dividing throughout by δx we have

δy
= 2x + δx.
δx
The ratio δy/δx provides a measure of how fast y changes relative to x, in increments
of δx. For example, when x = 10

δy
= 20 + δx,
δx
and if δx = 1, then δy/δx = 21. Equally, if δx = 0.001, then δy/δx = 20.001. By
making δx smaller and smaller, δy becomes equally smaller, and their ratio converges
towards a limiting value of 20.
In this case, as δx approaches zero, δy/δx approaches 2x, and is written

δy
lim = 2x.
δx→0 δx

Thus in the limit, when δx = 0, we create a condition where δy is divided by zero—


which is a meaningless operation. However, if we hold onto the idea of a limit,
as δx → 0, it is obvious that the quotient δy/δx is converging towards 2x. The
22 3 Limits and Derivatives

subterfuge employed to avoid dividing by zero is to substitute another quotient dy/d x


to stand for the limiting condition:

dy δy
= lim = 2x.
dx δx→0 δx

dy/d x (pronounced dee y dee x) is the derivative of y = x 2 , i.e. 2x. For instance,
when x = 0, dy/d x = 0, and when x = 3, dy/d x = 6. The derivative dy/d x, is the
instantaneous rate at which y changes relative to x.
If we had represented this equation as a function:

f (x) = x 2

then dy/d x is another way of expressing f  (x).


Now let’s introduce two constants into the original quadratic equation to see what
effect, if any, they have on the derivative. We begin with

y = ax 2 + b

and increment x and y:

y + δy = a(x + δx)2 + b
 
= a x 2 + 2x · δx + (δx)2 + b
 
δy = a 2x · δx + (δx)2 .

Dividing throughout by δx:


δy
= a(2x + δx)
δx
and the derivative is
dy δy
= lim = 2ax.
dx δx→0 δx

Thus we see the added constant b disappears (i.e. because it does not change), whilst
the multiplied constant a is transmitted through to the derivative.

3.3.2 Cubic Equation

Now let’s repeat the above analysis for y = x 3 :


3.3 Equations and Limits 23

y = x3
y + δy = (x + δx)3
= x 3 + 3x 2 · δx + 3x(δx)2 + (δx)3
δy = 3x 2 · δx + 3x(δx)2 + (δx)3 .

Dividing throughout by δx:

δy
= 3x 2 + 3x · δx + (δx)2 .
δx
Employing the idea of infinitesimals, one would argue that any term involving δx
can be ignored, because its numerical value is too small to make any contribution
to the result. Similarly, using the idea of limits, one would argue that as δx is made
increasingly smaller, towards zero, any term involving δx rapidly disappears.
Using limits, we have
δy
lim = 3x 2
δx→0 δx

or
dy δy
= lim = 3x 2 .
dx δx→0 δx

We could also show that if y = ax 3 + b then

dy
= 3ax 2 .
dx
This incremental technique can be used to compute the derivative of all sorts of
functions.
If we continue computing the derivatives of higher-order polynomials, we discover
the following pattern:

dy
y = x 2, = 2x
dx
dy
y = x 3, = 3x 2
dx
dy
y = x 4, = 4x 3
dx
dy
y = x 5, = 5x 4 .
dx
Clearly, the rule is
dy
y = xn, = nx n−1
dx
24 3 Limits and Derivatives

but we need to prove why this is so. The solution is found in the binomial expansion
for (x + δx)n , which can be divided into three components:
1. Decreasing terms of x.
2. Increasing terms of δx.
3. The terms of Pascal’s triangle.
For example, the individual terms of (x + δx)4 are:
Decreasing terms of x: x4 x3 x2 x1 x0
Increasing terms of δx: (δx)0 (δx)1 (δx)2 (δx)3 (δx)4
The terms of Pascal’s triangle: 1 4 6 4 1
which when combined produce

x 4 + 4x 3 (δx) + 6x 2 (δx)2 + 4x(δx)3 + (δx)4 .

Thus when we begin an incremental analysis:

y = x4
y + δy = (x + δx)4
= x 4 + 4x 3 (δx) + 6x 2 (δx)2 + 4x(δx)3 + (δx)4
δy = 4x 3 (δx) + 6x 2 (δx)2 + 4x(δx)3 + (δx)4 .

Dividing throughout by δx:

δy
= 4x 3 + 6x 2 (δx)1 + 4x(δx)2 + (δx)3 .
δx
In the limit, as δx slides to zero, only the second term of the original binomial
expansion remains:
4x 3 .

The second term of the binomial expansion (1 + δx)n is always of the form

nx n−1

which is the proof we require.

3.3.3 Functions and Limits

In order to generalise the above findings, let’s approach the above analysis using
a function of the form y = f (x). We begin by noting some arbitrary value of its
3.3 Equations and Limits 25

independent variable and note the function’s value. In general terms, this is x and
f (x) respectively. We then increase x by a small amount δx, to give x + δx, and
measure the function’s value again: f (x + δx). The function’s change in value is
f (x + δx) − f (x), whilst the change in the independent variable is δx. The quotient
of these two quantities approximates to the function’s rate of change at x:

f (x + δx) − f (x)
. (3.1)
δx
By making δx smaller and smaller towards zero, (3.1) converges towards a limiting
value expressed as
dy f (x + δx) − f (x)
= lim (3.2)
dx δx→0 δx

which can be used to compute all sorts of functions. For example, to compute the
derivative of sin x we proceed as follows:

y = sin x
y + δy = sin(x + δx).

Using the identity sin(A + B) = sin A · cos B + cos A · sin B, we have

y + δy = sin x · cos(δx) + cos x · sin(δx)


δy = sin x · cos(δx) + cos x · sin(δx) − sin x
= sin x(cos(δx) − 1) + cos x · sin(δx).

Dividing throughout by δx we have

δy sin x sin(δx)
= (cos(δx) − 1) + cos x.
δx δx δx

In the limit as δx → 0, (cos(δx) − 1) → 0 and sin(δx)/δx = 1 (See Appendix A),


and
dy
= cos x
dx
which confirms our ‘guesstimate’ in Chap. 2. Before moving on, let’s compute the
derivative of cos x.

y = cos x
y + δy = cos(x + δx).

Using the identity cos(A + B) = cos A · cos B − sin A · sin B, we have


26 3 Limits and Derivatives

y + δy = cos x · cos(δx) − sin x · sin(δx)


δy = cos x · cos(δx) − sin x · sin(δx) − cos x
= cos x(cos(δx) − 1) − sin x · sin(δx).

Dividing throughout by δx we have

δy cos x sin(δx)
= (cos(δx) − 1) − sin x.
δx δx δx

In the limit as δx → 0, (cos(δx) − 1) → 0 and sin(δx)/δx = 1 (See Appendix A),


and
dy
= − sin x
dx
which also confirms our ‘guesstimate’. We will continue to employ this strategy to
compute the derivatives of other functions later on.

3.3.4 Graphical Interpretation of the Derivative

To illustrate this limiting process graphically, consider the scenario in Fig. 3.1 where
the sample point is P. In this case the function is f (x) = x 2 and P’s coordinates
are (x, x 2 ). We identify another point R, displaced δx to the right of P, with coor-
dinates (x + δx, x 2 ). The point Q on the curve, vertically above R, has coordinates
2

x + δx, (x + δx) . When δx is relatively small, the slope of the line P Q approxi-
mates to the function’s rate of change at P, which is the graph’s slope. This is given
by

QR (x + δx)2 − x 2
slope = =
PR δx
x 2 + 2x(δx) + (δx)2 − x 2
=
δx
2x(δx) + (δx)2
=
δx
= 2x + δx.

We can now reason that as δx is made smaller and smaller, Q approaches P, and
slope becomes the graph’s slope at P. This is the limiting condition:

dy
= lim (2x + δx) = 2x.
dx δx→0
3.3 Equations and Limits 27

Fig. 3.1 Sketch of


f (x) = x 2 4
Q

P R
1

-1 0 1 2

-1

Thus, for any point with coordinates (x, x 2 ), the slope is given by 2x. For example,
when x = 0, the slope is 0, and when x = 4, the slope is 8, etc.

3.3.5 Derivatives and Differentials

Given a function f (x), d f /d x represents the instantaneous change of f for some


x, and is called the first derivative of f (x). For linear functions, this is constant,
for other functions, the derivative’s value changes with x and is represented by a
function.
The elements d f , dy and d x are called differentials, and historically, the derivative
used to be called the differential coefficient, but has now been dropped in favour of
derivative. One can see how the idea of a differential coefficient arose if we write,
for example:
dy
= 3x
dx
as
dy = 3x d x.

In this case, 3x acts like a coefficient of d x, nevertheless, we will use the word
derivative. It is worth noting that if y = x, then dy/d x = 1, or dy = d x. The two
differentials are individual algebraic quantities, which permits us to write statements
such as
dy 1
= 3x, dy = 3x d x, dx = dy.
dx 3x
For example, given
y = 6x 3 − 4x 2 + 8x + 6
28 3 Limits and Derivatives

then
dy
= 18x 2 − 8x + 8
dx
which is the instantaneous change of y relative to x. When x = 1, dy/d x = 18 −
8 + 8 = 18, which means that y is changing 18 times faster than x. Consequently,
d x/dy = 1/18.
Gottfried Leibniz developed what has become known as Leibniz notation for
differentiation, where
dy
dx
is a composite definition for a derivative. Leibniz also treated them individually as
infinitesimals, which is no longer the case. It was Joseph Lagrange who developed
the prime mark notation f  (x) to denote the first derivative, with extra prime marks
for higher derivatives.
Personally, I find that separating dy/d x into dy and d x has useful pedagogic uses,
even if it is not mathematically rigorous!

3.3.6 Integration and Antiderivatives

If it is possible to differentiate a function, it seems reasonable to assume the existence


of an inverse process to convert a derivative back to its associated function. Fortu-
nately, this is the case, but there are some limitations. This inverse process is called
integration and reveals the antiderivative of a function. Many functions can be paired
together in the form of a derivative and an antiderivative, such as 2x with x 2 , and
cos x with sin x. However, there are many functions where it is impossible to derive
its antiderivative in a precise form. For example, there is no simple, finite functional
antiderivative for sin(x 2 ) or (sin x)/x. To understand integration, let’s begin with a
simple derivative.
If we are given
dy
= 18x 2 − 8x + 8
dx
it is not too difficult to reason that the original function could have been

y = 6x 3 − 4x 2 + 8x.

However, it could have also been

y = 6x 3 − 4x 2 + 8x + 2
3.3 Equations and Limits 29

or
y = 6x 3 − 4x 2 + 8x + 20

or with any other constant. Consequently, when integrating the original function, the
integration process has to include a constant:

y = 6x 3 − 4x 2 + 8x + C.

The value of C is not always required, but it can be determined if we are given some
extra information, such as y = 10 when x = 0, then C = 10. 
The notation for integration employs a curly ‘S’ symbol , which may seem
strange, but is short for sum and will be explained later. So, starting with

dy
= 18x 2 − 8x + 8
dx
we rewrite this as
dy = (18x 2 − 8x + 8)d x

and integrate both sides, where dy becomes y and the right-hand-side becomes

(18x 2 − 8x + 8) d x

although brackets are not always used:



y= 18x 2 − 8x + 8 d x.

This equation reads: ‘y is the integral of 18x 2 − 8x + 8 dee x.’ The d x reminds us
that x is the independent variable. In this case we can write the answer:

dy
= 18x 2 − 8x + 8
dx
dy = 18x 2 − 8x + 8 d x

y = 18x 2 − 8x + 8 d x

= 6x 3 − 4x 2 + 8x + C

where C is some constant.


30 3 Limits and Derivatives

Another example:

dy
= 6x 2 + 10x
dx
dy = 6x 2 + 10x d x

y = 6x 2 + 10x d x

= 2x 3 + 5x 2 + C.

Finally,

dy
=1
dx
dy = 1 d x

y = 1 dx

= x + C.

The antiderivatives for the sine and cosine functions are written:

sin x d x = − cos x + C

cos x d x = sin x + C

which you may think obvious, as we have just computed their derivatives. How-
ever, the reason for introducing integration alongside differentiation, is to make you
familiar with the notation, and memorise the two distinct processes, as well as lay
the foundations for later chapters.

3.4 Summary

This chapter has shown how limits provide a useful tool for computing a function’s
derivative. Basically, the function’s independent variable is disturbed by a very small
quantity, typically δx, which alters the function’s value. The quotient

f (x + δx) − f (x)
δx
is a measure of the function’s rate of change relative to its independent variable. By
making δx smaller and smaller towards zero, we converge towards a limiting value
called the function’s derivative. Unfortunately, not all functions possess a derivative,
3.4 Summary 31

therefore we can only work with functions that can be differentiated. In the next
chapter we discover how to differentiate different types of functions and function
combinations.
We have also come across integration—the inverse of differentiation—and as we
compute the derivatives of other functions, the associated antiderivative will also be
included.

3.5 Worked Examples

3.5.1 Limiting Value of a Quotient

Find the limiting value of 3xx 2+x


8 2

−x 3
, as x → 0.
First, we simplify the quotient by dividing the numerator and denominator by x 2 :

x6 + 1
.
3−x

We can now reason that as x → 0, (x 6 + 1) → 1 and (3 − x) → 3, therefore,

x8 + x2
lim = 1
x→0 3x 2 − x 3 3

which is confirmed by the function’s graph in Fig. 3.2.

3.5.2 Limiting Value of a Quotient

x 2 −1
Find the limiting value of 3x 2 −2x−1
, as x → 0.

Fig. 3.2 Graph of


x 8 +x 2
f (x) = 3x 2 −x 3 4

-1 0 1

-1
Another random document with
no related content on Scribd:
The cabman strenuously assured him that this was the
Tarlington, and he decided that he had better go aboard and look for
Captain O’Shea. If a mistake had been made, it should be an easy
matter to find the landing-pier and the waiting tug.
No sooner had the king reached the deck than he was convinced
that he had been directed to the wrong steamer. The people who
stared at him were utter strangers. There was not a subject of
Trinadaro among them, nor did any of the officers of the ship step
forward to greet him. He was about to accost the nearest spectator
when an officious man dressed in seedy black confronted him,
flourished a formidable-appearing document under the royal nose,
and pompously affirmed:
“A writ from the judge duly appointed and authorized by the Lord
Chancellor to take cognizance of such cases, distraining Colonel
Osmond George Sydenham-Leach from attempting to quit the
jurisdiction of said court pending an inquisition de lunatico
inquirendo. Take it calm and easy, sir. This won’t interfere with your
liberty as long as you obey the writ.”
Another minion of the law, a fat man with a well-oiled voice,
thereupon formally took possession of the steamer, explaining that
because clearance papers had been issued for a voyage to
Trinadaro, the court held that a departure from England was actually
and speedily contemplated. The presence of Colonel Sydenham-
Leach on board in person was also evidence after the fact.
The blow was staggering, humiliating, incredibly painful. It shook
the amiable gentleman’s presence of mind to the very foundations.
To be interfered with as an alleged madman was enough to bewilder
the most sapient monarch that ever wielded sceptre. As a landed
proprietor, a retired officer of the militia, a Conservative in politics,
King Osmond had profound respect for the law and the constitution
of his native land. He was not one to defy a judicial writ or to grapple
with the situation in a high-handed manner. In other words, he was
rather Colonel Sydenham-Leach in this cruel crisis than the
sovereign ruler of the independent principality of Trinadaro.
No help or comfort was to be obtained from the company around
him. These spurious voyagers from the employment agency were
whispering uneasily among themselves and regarding the
unfortunate Osmond with suspicious glances. They had not
bargained to entangle themselves in the affairs of an alleged lunatic
on board of a ship which had been seized in the name of the law.
Ten shillings was not enough for this sort of thing.
“It don’t look right to me,” said one of them. “The job is on the
queer. I say we hook it before the bloomin’ bobbies come and put the
lot of us in jail.”
This sentiment expressed the general view of the situation, and
the counterfeit subjects of Trinadaro began to flock down the
gangway and scatter in a hunted manner among the gloomy
warehouses. Presently Colonel Sydenham-Leach was left alone with
the two court officers. Recovering somewhat of his composure and
dignity, he declared that he must consult with his legal advisers
before consenting to leave the ship. He clung to the hope that delay
might enable Captain O’Shea to come to his rescue, although he
was unwilling to try to send a message to the Tarlington. This might
reveal to the officers of the law that the wrong ship had been
detained, and put them on the track of the right one.
There was no legal reason why the luckless king should not
remain in the Tyneshire Glen until his lawyers could come and confer
with him, wherefore the captors grumblingly sat themselves down in
the cabin to wait. The king had nothing to say to them. He was
absorbed in his own unhappy reflections. His dreams had turned to
ashes. His island empire would know him not. He felt very old and
helpless, and sad.
Thus he sat and brooded for some time. At length he heard the
sound of men tramping across the deck above his head. He roused
himself to look in the direction of the door-way. A moment later it
framed the well-knit, active figure of Captain Michael O’Shea. Behind
him puffed stout Johnny Kent.
“’Twas a good guess, Your Majesty,” cried O’Shea. “We thought
you might have gone adrift and fetched up aboard this old steamer.
Who are your two friends?”
“Officers from the bench of one of the judges in lunacy,”
reluctantly admitted King Osmond. “They have served distraining
papers on me and on my ship.”
“On this ship?” exclaimed Johnny Kent. “How ridiculous! What’ll
we do with this pair of bailiffs, or whatever you call ’em, Cap’n Mike?
Make ’em eat their documents?”
“No; we will take the two meddlers along with us,” sweetly
answered O’Shea. “We can’t afford to leave them behind to tell how
it happened.”
“But they have all the power and authority of the British
government behind them,” spoke up King Osmond.
“And they have a long voyage ahead of them,” said O’Shea.
“Your Majesty can give them jobs in your own judicial department
and they will grow up with the country.”
“I cannot countenance such actions,” began the king, but Johnny
Kent interrupted to remark with much vehemence:
“Excuse us, Your Majesty, but this ain’t no time for arguments
about the British constitution. Cap’n Mike and me agreed to take you
and your ship to Trinadaro. It was a contract, and we propose to earn
our wages. If you won’t come easy and willin’, then we’ll just have to
call a couple of our men from the boat that’s waiting alongside and
escort you, anyhow. We aim to live up to our agreements.”
O’Shea wasted no more words. Suddenly grasping one of the
court officers by the back of the neck and the slack of his garments,
he propelled him rapidly toward the deck, fiercely admonishing him
to make no outcry unless he wished to be tossed overboard.
The other man had started to flee, but Johnny Kent caught him
in a few heavy strides, tucked him under one mighty arm, clapped a
hand over his mouth, and waddled with his burden to the nearest
cargo port.
“Drop them into the boat,” commanded O’Shea. “Ahoy, there,
below! Catch these two lads, and let them make no noise.”
The astonished King Osmond had followed the abductors out of
the cabin. Before he could renew the discussion Captain O’Shea,
breathing hard, but calm and smiling, faced him with the courteous
invitation:
“After you, Your Majesty. We are at your service. A few minutes
in the boat and you will be aboard the Tarlington and heading for the
open sea.”
It was obviously so futile to protest that the king meekly
descended to the boat, steadied by the helping hand of Johnny Kent.
The seamen shoved off and O’Shea steered for the long black hull of
the steamer visible a few hundred yards down-stream. Unable to
voice his confused emotions, the king suffered himself to be
conducted up the gangway of the Tarlington.
His loyal subjects, the real ones, cheered frantically at sight of
him. It was an ovation worthy of his station. He bowed and smiled
and was himself again. Already the recollection of his detention as a
madman seemed less distressing.
He felt the ship tremble under his feet as her engines began to
drive her toward the blessed sea and the long road to wave-washed
Trinadaro. Had it not been for the bold and ready conduct of his two
faithful mariners, he would now be a broken-spirited old man in
London, a butt of public ridicule. He went below to the state-rooms
which had been suitably fitted for his comfort and privacy, and
discovered that he was greatly wearied.
Before retiring he sent one of his secretaries to request Captain
O’Shea and Johnny Kent to give him the pleasure of their company
at breakfast next morning.
“That makes me feel a bit more cheerful,” said O’Shea to
himself. “Maybe he has decided to forgive us. We were guilty of high
treason, disobedience, and a few other things, in packing him off to
sea while he was trying to tell us he couldn’t go at all.”
The Tarlington was in blue water next morning when the captain
and the chief engineer bashfully entered the private dining-room of
His Majesty. The latter greeted them with marked affability, and said:
“I take great pleasure, my dear friends, in conferring on you the
insignia of the Grand Cross of Trinadaro as a recognition of your
invaluable loyalty and assistance. You will be entitled to call
yourselves barons of my realm by royal warrant. While I must
confess that I could not ordinarily approve of such summary methods
as you made use of——”
“It looks different now that old England is dropping astern,”
suggested O’Shea. “The British constitution doesn’t loom as big as it
did. Your own flag is at the mast-head, Your Majesty, and you can
make treaties if ye like. I thank you with all my heart for the reward
you have given me.”
“It pleases me a heap more to be a member of the nobility of
Trinadaro than to earn big wages for the voyage,” warmly assented
Johnny Kent. “I’ll be the only life-size baron in my neck of the woods
when I settle down on that farm in the State o’ Maine, eh, Cap’n
Mike?”
Freed of all anxieties and besetments, the royal passenger
resumed his labor of planning the occupations of his subjects. His
enthusiasm was delightful to behold. He seemed to grow younger
with every day of the voyage southward. His was to be a kingdom of
peace and good-will, of a benevolent ruler and a contented,
industrious people. He was the stanchest kind of a royalist, and
Trinadaro was to be a constitutional monarchy with an aristocracy
which should be recruited after the pioneering work had been
accomplished.
The relations between the king and his mariners twain became
those of pleasant intimacy. They came to know him much better
during the long weeks at sea, and felt toward him an affectionate,
tolerant respect.
The ship had crossed the equator and was ploughing through
the long blue surges of the South Atlantic when Captain O’Shea,
after working out the noon observations, informed the king:
“A couple of days more and we’ll begin to look for a sight of the
peaks of Trinadaro. If the weather holds calm, we can begin to put
the people and the cargo ashore right after that.”
“The peaks of Trinadaro!” fondly echoed Osmond I. “Do you
know, Captain O’Shea, I have wondered if you considered me a
crack-brained old fool. Many men in England think so, I am sure. I
know that my relatives do.”
“’Tis my opinion that ye wish to make folks happy and that you
will do no harm with your money,” was the reply. “And there’s few
rich men that can say the same. No; ’tis not crack-brained to want to
be a king. Power is what men desire, and they will trample on others
to get it. I have heard ye talk here on board ship, and I have admired
what you had to say. You will live your own life in your own way, but
ye will not forget to make this island of yours a place for men and
women to call home and to be glad that they have found it.”
“I thank you, Captain O’Shea,” said the other. “I cannot help
thinking now and then of what will be the fate of my principality when
death comes to me. If I am spared for ten or fifteen years longer, I
shall have time to set my affairs in order, to make Trinadaro self-
sustaining, to win the recognition of foreign governments, to arrange
for an administration to succeed my reign.”
“May you live to be a king until you are a hundred!” cried
O’Shea. “And a man who is as happy and contented as you are is
pretty sure of a ripe old age.”
“I hope that you and Mr. Kent will consent to sail under my flag
as long as I live,” earnestly said the king. “I have learned to depend
on you, and I need not tell you that the financial arrangement will be
more favorable than you could make elsewhere.”
“We have no notion of quitting your service,” replied O’Shea, with
a smile. “’Tis up to us to see the kingdom fairly under way before we
turn rovers again.”
It was early in the morning of the second day after this when the
officer on watch roused Captain O’Shea with the news that land had
been sighted on the starboard bow. The master of the Tarlington
stared through his binoculars and saw a black, jagged foreland of
rock lifting from the sea. He sent word to the passengers that
Trinadaro lay ahead of them.
King Osmond had left word that he was to be called whenever
the first glimpse of his island should be revealed. But he came not to
the bridge in response to the message from Captain O’Shea. In his
stead appeared his physician, with a demeanor terribly distressed.
His voice was unsteady as he said:
“It is my sad duty to inform you that His Majesty passed away
some time during the night. His heart simply ceased to beat. It had
been somewhat feeble and irregular of late, but the symptoms were
not alarming. His strength was overtaxed during those last weeks in
London.”
O’Shea bared his head and stood silent. The announcement was
very hard to believe. Pulling himself together, he murmured to the
chief officer:
“The king is dead. Please set the flag of Trinadaro at half-mast.”
As soon as the word was passed down to the engine-room
Johnny Kent sought the bridge and his eyes filled with tears as he
exclaimed:
“It don’t seem right, Cap’n Mike. I ain’t reconciled to it one mite.
He deserved to have what he wanted.”
“Yes, he had slipped his cable, Johnny. There are cruel tricks in
this game of life.”
“What will you do now?”
“I have had no time to think. But one thing is certain. I will carry
King Osmond to his island, and there we will bury him. ’Tis the one
place in all the world where he would want to rest. And the peaks of
Trinadaro will guard him, and the big breakers will sing anthems for
him, and he will be the king there till the Judgment Day.”
The Tarlington slowly approached the precipitous coast-line and
changed her course to pass around to the lee of the island. As the
deeply indented shore opened to view, and one bold headland after
another slid by, a comparatively sheltered anchorage was disclosed.
There, to the amazement of Captain O’Shea, rode two small
cruisers. One of them flew the red ensign of England, the other the
green and yellow colors of the navy of Brazil. He guessed their
errand before a British lieutenant came alongside the Tarlington in a
steam-launch and climbed the gangway which had been dropped to
receive him.
Gazing curiously at the silent company and the half-masted flag
of Trinadaro, he was conducted into the saloon, where Captain
O’Shea waited for him to state his business.
“This steamer belongs to Colonel Sydenham-Leach, I presume,”
said the visitor. “I should like to see him, if you please. Sorry, but I
have unpleasant news for him.”
“If it is King Osmond of Trinadaro ye mean, he is dead, God rest
his soul! He went out last night.”
“You don’t say! Please express my sympathy to the ship’s
company,” exclaimed the lieutenant. “How extraordinary! We
received orders by cable at Rio to proceed to Trinadaro in time to
intercept this vessel of yours.”
“And what were the orders, and why is that Brazilian man-of-war
anchored alongside of you?” asked O’Shea.
“It is all about the ownership of the island,” the lieutenant
explained. “Nobody wanted it for centuries, and now everybody
seems keen on getting hold of it. The English government suddenly
decided, after you sailed from London, that it might need Trinadaro
as a landing-base for a new cable between South America and
Africa, and sent us to hoist the flag over the place. Brazil heard of
the affair and sent a ship to set up a claim on the basis of an early
discovery. The Portuguese have presented their evidence, I believe,
because their people made some kind of a settlement at Trinadaro
once upon a time.”
“And the forsaken island was totally forgotten until poor King
Osmond got himself and his project into the newspapers,” slowly
commented O’Shea.
“That is the truth of the matter, I fancy.” The naval lieutenant
paused, and commiseration was strongly reflected in his manly face.
“Tell me,” said he, “what was the opinion at home about this King of
Trinadaro? He was a bit mad, I take it.”
“No more than you or me,” answered O’Shea. “He had a
beautiful dream, and it made him very happy, but it was not his fate
to see it come true. And no doubt it is better that he did not live to
know that the scheme was ruined. His island has been taken away
from him. It will be wrangled over by England and Brazil and the rest
of them, and there is no room for a king that hoped to enjoy himself
in his own way. The world has no place for a man like Colonel
Osmond George Sydenham-Leach, my dear sir.”
“Too bad!” sighed the lieutenant. “And what are your plans,
Captain O’Shea? Do you intend to make any formal claim in behalf
of the late king?”
“No. His dreams died with him. There is no heir to the throne. I’m
thankful that his finish was so bright and hopeful. There will be
funeral services and a burial to-morrow. I should take it as a great
favor if detachments from the British cruiser and the Brazilian war-
vessel could be present.”
“I will attend to it,” said the lieutenant.
When the coffin of King Osmond I was carried ashore it was
draped with the flag of Trinadaro, which he himself had designed.
Launches from the two cruisers towed sailing-cutters filled with
bluejackets, who splashed through the surf and formed in column led
by the bugles and the muffled drums. The parade wound along the
narrow valleys and climbed to the plateau on which the ruler had
planned to build his capital.
There the first and last King of Trinadaro was laid to rest, and the
guns of the cruisers thundered a requiem. The British lieutenant
counted the guns and turned to Captain O’Shea to say:
“It is the salute given only to royalty, according to the navy
regulations. It is the least we can do for him.”
“And it is handsomely done,” muttered the grateful O’Shea as he
brushed his hand across his eyes.
“Will you take your ship back to England?”
“Yes. I can do nothing else. ’Twill be a sad voyage, but God
knows best. As it all turned out, this king of ours had to die to win his
kingdom.”
When the mourners had returned to the Tarlington, Captain
O’Shea and Johnny Kent went into the chart-room and talked
together for some time. At length the simple-hearted chief engineer
said with a wistful smile:
“I’m glad we stood by and did what we could for him, ain’t you,
Cap’n Mike?”
“You bet I am, Johnny. He was a good man, and I loved him.
Here’s to His Majesty, King Osmond of Trinadaro! Even the pair of
court officers we kidnapped had come to be fond of him and wished
him no harm. There may be trouble waiting for us in London River on
account of them and the ship that took out no clearances. But we will
face the music. ’Tis not much to do for him that was so good to us.”
“Right you are, Cap’n Mike; but do you suppose we’ll go to jail?”
“No; for the blame will be laid to poor King Osmond, and the law
will hold him responsible for the acts of his agents. But we would not
mind going to jail for him.”
“Well, anyhow, they can never take his kingdom away from him,”
softly quoth Johnny Kent.
THE LINER “ALSATIAN”
Fifteen years ago the crack Atlantic liners were no larger than
ten thousand tons. Some of them are still in service, safe and
comfortable ships, quite fast enough for the traveller who is not bitten
with speed madness. When the Alsatian of the International Line
was new she attracted as much attention as one of the monsters of
to-day with its length of almost a fifth of a mile and horse-power to
stagger the imagination.
As she rode at anchor in the Mersey on a certain sailing day in
March, spick-and-span with fresh paint, brasswork sparkling in the
sunshine, flags snapping in the breeze, the Alsatian was a
handsome picture to greet the passengers who arrived in the special
train from London and were transferred on board in the paddle-wheel
tender. There were fewer than a hundred of them in the first cabin,
for the season of the year was between high tides of travel east and
west.
It was a tradition of the International Line that its steamers
should sail precisely on the stroke of the hour appointed. More than
five minutes’ delay was viewed by the port superintendents in
Liverpool and New York as a nautical crime. Therefore when noon
came and there was none of the activity of departure, the
passengers were curious. A loquacious young man, of the noisy
breed which makes the English say unkind things about American
tourists, ordered another cocktail of the smoking-room steward and
pettishly exclaimed:
“This right-on-the-minute business is all a bluff. The gangway
hasn’t been hoisted and the tender is still alongside. This ship is
nowhere near ready to start. Slow country—slow people, these
Britishers. We can show ’em a few things, bet your life.”
A nervous, thin-faced gentleman who had been fidgeting
between the deck and the smoking-room door chimed in to say:
“Confound it, I hate to be behind time! I can’t stand it! What’s the
matter with this steamer? Why don’t the officers tell us something?”
Several passengers listened deferentially to this jerky protest.
The speaker was immensely, notoriously rich, and, although
dyspepsia had played hob with his internal workings, and his temper
was chronically on edge, he was an enviable personage in the eyes
of many American citizens. Whether he toiled or loafed, his millions
were working night and day to earn more millions for him. It could
make no essential difference under heaven at what hour the Alsatian
should carry him out of Liverpool, for he could not be happy
anywhere; but the delay made him acutely miserable.
An old man with kindly, scrutinizing eyes laid down his cigar to
comment:
“My dear sir, I crossed the ocean in a sailing-packet some forty-
odd years ago, and we anchored in the channel two weeks waiting
for a fair wind, and were fifty-seven days to Sandy Hook.”
“Times have changed, thank God!” snapped the great Jenkins P.
Chase, of the bankrupt digestion.
“And changed not altogether for the better when it comes to all
this fuss and clatter to get somewhere else in a hurry, my friend. It is
a national disease,” was the smiling, tolerant reply.
Jenkins P. Chase glanced at his watch, muttered something, and
darted on deck as if a bee had stung him.
“Bet you the drinks he’s gone to find the captain and blow him
up,” admiringly cried the loquacious young man. “If Jenkins P. Chase
gets his dander up he’s liable to buy the ship and the whole blamed
line and run it to suit himself. He is the original live-wire. Most
wonderful man in the little old United States.”
In a rather secluded corner of the smoking-room sat two
passengers who had taken no part in the general conversation. One
might have suspected that all this fuss over a belated sailing caused
them mild amusement. The younger was of a cast of features
unmistakably Irish, with the combination of pugnacity and humor so
often discernible in men of that blood.
His companion was ruddy and big-bodied, his hair and mustache
well frosted by time. Said the latter, after due reflection:
“Hurry has killed a whole lot of people, Cap’n Mike. What’s the
matter with these peevish gents, anyhow? The company is givin’
them their board and they’re as comfortable as lords. I don’t care if
the steamer lays in port a week.”
“That Jenkins P. Chase is a horrible example, Johnny,” quoth
Captain Michael O’Shea. “’Tis his habit to go flyin’ about, and there is
no rest for him anywhere. If ye accumulate too much money, you
may get that way yourself.”
“I ain’t got a symptom,” said improvident old Johnny Kent. “I’ve
learned, for one thing, that it’s poor business to try to hurry the sea.
A ship must bide her time and sail when she’s ready.”
“But what ails this one, I wonder?” queried Captain O’Shea. “I
mistrust something is wrong. The skipper of her, and a grand man he
is, with his gold buttons and all, he went below a while ago, Johnny,
and he has not come back.”
They strolled outside, and being seafaring men of wide
experience, found significance in trifles which would have meant little
or nothing to a landsman. This was no ordinary delay. The whole
complex organization of the liner was disturbed.
“There is trouble amongst the crew,” observed O’Shea. Johnny
Kent halted near an engine-room skylight and cocked his head to
listen.
“The trouble is in this department,” said he.
Presently a tug-boat hastily cast off from the nearest quay and
churned her way out to the Alsatian. A dozen Liverpool policemen
scrambled aboard the liner and vanished between-decks. From the
depths below the water-line arose a hubbub of oaths and shouts.
A few minutes later two policemen reappeared dragging
between them to the gangway a shock-headed, muscular fellow in
blue dungarees. Although he made no resistance, they handled him
roughly and he was expeditiously handcuffed to a stanchion on the
deck of the tug. Immediately thereafter the sounds of disturbance
down below increased in violence, and swarming up ladders and
through passageways came a sooty, greasy crowd of stokers,
trimmers, and coal-passers.
Scrambling on board the tug, and taking her by storm, they
voiced their opinions of the Alsatian and the International Line in
language which caused the feminine passengers to clap their hands
to their ears and flee from the rail.
A junior officer with whom Captain O’Shea had scraped
acquaintance halted to explain, in passing:
“The blackguards went on strike for more pay and recognition of
their union. The company patched up the trouble yesterday, but the
beggars were stirred up again this morning by the chap the bobbies
put the irons on. He persuaded them to kick up a rumpus just before
sailing-time.”
“If they have signed articles, ’tis more like a mutiny than a strike,”
observed O’Shea.
“They know that right enough,” said the officer, “but they don’t
seem to care whether they are jugged for it or not. It’s an incident of
the general labor trouble in this port, I presume. The longshoremen’s
strike is not settled yet, you know.”
“And what will ye do for a fire-room gang?” O’Shea asked him.
“There was near a hundred and fifty of them that quit just now.”
“Hanged if I know,” sighed the officer as he walked away.
The tug was black with the mob of strikers, who were packed
wherever they could find standing-room. The police could do nothing
with them, and the distracted skipper of the tug decided to make for
a quay and get rid of his riotous cargo. The passengers of the
Alsatian surmised that sailing-day might be indefinitely postponed
and they bombarded the officers with excited demands for
information. Captain O’Shea and Johnny Kent, philosophers of sorts,
viewed the situation with good-natured composure, and were more
interested in the summons to the dining-saloon for luncheon than in
the strike of the fire-room gang.
“As long as I get three square meals per day and a dry bunk I
ain’t especially uneasy about anything,” remarked Johnny Kent as he
fondly scanned the elaborate menu card.
“Same here,” replied O’Shea. “But that jumpy gentleman,
Jenkins P. Chase, must be throwing assorted fits by this time.”
Facing them across the table was a blond, spectacled man with
a small, pointed beard, his appearance notably studious and precise.
Although he spoke English with cultivated ease and fluency, the ear
detected certain shades and intonations to indicate that he was a
German by birth. He was affable to his neighbors at table and
courteous to the steward who waited on him. Garrulous, sociable
Johnny Kent found him companionable, and ventured to inquire:
“Your first trip to America? Business or pleasure?”
“Both. I shall interest myself in studying scientific education in
the United States. I am a chemist by profession, and also a lecturer
on the subject before the classes of a university. Yes, it is my first
voyage to your wonderful country. Tell me, please, have you met the
famous Professor Crittenden, of Baltimore?”
Johnny Kent was about to proclaim that as a seafaring man he
was not in touch with scientists, but O’Shea, to prevent any such
disclosure, kicked him on the shin as a reminder that he was to
eschew personalities. It was not discreet to advertise themselves
and their affairs in the mixed company of the Atlantic liner. O’Shea
was aware that if Johnny Kent should once begin yarning about his
adventures it would be like pulling the cork from an overturned jug.
The marine engineer blushed guiltily, bent over to rub his bruised
shin, and briefly assured the blond scientist that he had not been
lucky enough to meet the distinguished Professor Crittenden, of
Baltimore.
“I was only last night reading his masterly paper on ‘The Action
of Diazobenzene Sulphonic Acid on Thymine, Uracil, and Cytosine,’”
politely returned the other. “It is as brilliant as his discussion of
imidechlorides.”
Johnny Kent threw up an arm as if to ward off a blow.
“If one of those words had hit me plumb and square, it would
have jolted me out of my chair!” he exclaimed. “I could feel the wind
of ’em.”
The studious stranger smiled and apologized for talking shop.
“Those strikers—will the company be able to fill their places?”
said he, addressing O’Shea.
“Perhaps a crew can be scraped up ashore. If not, we will have
to shift to another steamer. Firemen are an ugly, cross-tempered lot
to handle, so I am told.”
“Have you been much on the ocean? Do you know much about
ships?”
“I have made a voyage or two as a passenger,” O’Shea assured
him. “’Tis a hard life in the stoke-hole of a big steamer, I imagine.”
The scientist returned emphatically:
“I have no sympathy with them; none whatever. Lacking
intelligence, fitness, they must labor for those who have earned or
won the right to rule them.”
“’Tis your opinion that might makes right?” spoke up O’Shea.
“Always, everywhere!” declared the scientist. “The mind is the
man. The founders of your government proclaimed the fallacy that all
men are equal, but your strong men know better, and they ride and
exploit your masses.”
“It’s the best country God ever made,” cried Johnny Kent with
some heat.
“I beg your pardon”; and the chemist bowed. “It was a rudeness
for me to speak so.”
As they left the table he gave them his card with a touch of
formality, and they discovered that his name was Professor Ernst
Wilhelm Vonderholtz.
Three hours later the passengers were notified that the Alsatian
would be ready to sail next morning. It was learned that the company
had been able to recruit an unexpectedly large number of
unemployed firemen among the boarding-houses and taverns of the
Liverpool water-front. They were willing to take the places of the
strikers, and it was hoped that the liner could be sent to sea with a
fairly complete complement of men. Apparently the strikers had been
poorly advised and led, for they were beaten with no great
inconvenience to the management of the company.
As soon as the Alsatian had lifted anchor and was steaming out
of the Mersey the passengers ceased grumbling, and settled into the
comfortable, somnolent routine of a modern transatlantic voyage. A
party of poker-players mobilized in the smoking-room. The ladies
reclined all in a row in their steamer-chairs on the lee side of the
deck, like so many shawl-wrapped mummies. The spoiled American
child whanged the life out of the long-suffering piano in the music-
room. A few conscientious persons undertook to walk so many miles
around the deck each day. There was much random conversation, a
spice of flirtation, and a vast deal of eating and sleeping. That
hectoring gentleman Jenkins P. Chase spent most of the time in his
own rooms, where he was ministered to by his physician, his
secretary, and his valet.
Captain O’Shea and Johnny Kent enjoyed the luxury of such a
voyage as this. There was no responsibility to burden them on the
bridge or in the engine-room. No one guessed that they were
uncommonly capable mariners, accustomed to command. Professor
Ernst Wilhelm Vonderholtz seemed to find their company congenial,
and tried to make them talk about themselves. His curiosity was
politely dissembled, but O’Shea took note of it and built up an
elaborate fiction to the effect that he was a pavement contractor in
New York with friends at Tammany Hall, while Johnny Kent found
genuine satisfaction in posing as a retired farmer from the State of
Maine. It occurred to O’Shea to remark to his comrade as they were
undressing in their room on the second night at sea:
“The chemical professor suspects we are not what we seem.
And he is anxious to fathom us.”
“Oh, pooh! He’s one of them high and lofty thinkers that wouldn’t
bother his head about ignorant, every-day cusses like us,” sleepily
replied Johnny Kent as he kicked off his shoes.
“You fool yourself,” and O’Shea spoke with decision. “He is full of
big words and things that I do not pretend to understand at all, but he
is not wrapped up in them entirely, like most of the professors and
such. There is a pair of keen eyes behind those gold-rimmed
spectacles of his, Johnny, and he is not missing anything that goes
on.”
“I take notice that he ain’t overlookin’ that handsome school-
teacher that’s been studyin’ abroad for a year. His eyes are sharp
enough to sight her whenever she comes on deck. And she ain’t
hostile to him, either.”
“I grant ye that, you sentimental old pirate,” said O’Shea, “but I
am not a match-maker, and ’tis no concern of mine. What I am
wondering is whether the man is really a university professor bent on
‘investigating the scientific education of the United States.’”
“You’re welcome to sit up and hatch mysteries by yourself,”
grumbled the other. “I want to go to sleep. What’s the clew to all this,
Cap’n Mike? What makes you so darned suspicious?”
“’Tis no more than a hunch, Johnny. I’m Irish, and my people feel
things in the air. We don’t have to be told. This Professor Ernst
Wilhelm Vonderholtz does not ring true. There is a flaw in him
somewhere.”
“Well, we’re sort of travellers in disguise ourselves, ain’t we,
Cap’n Mike? I feel plumb full of false pretences. The pot calls the
kettle black. How about that?”
“’Tis our own business,” snapped O’Shea.
“So is his,” briefly concluded Johnny Kent as he crawled into the
bunk. No more than five minutes later he was snoring with the
rhythm and volume of a whistling buoy in a swinging sea. O’Shea lay
awake for some time, trying to fit his uneasy surmises together, or to
toss them aside as so much rubbish. He had not heard the banshee
cry, but a vague conjecture had fastened itself in his mind that
something was fated to go wrong with this voyage of the Alsatian.
And without tangible cause or reason, he found this foreboding
interwoven with the presence on board of the affable, mild-
mannered, studious Professor Ernst Wilhelm Vonderholtz.
Sailormen are notably superstitious, and O’Shea had been
schooled to beware of cross-eyed Finns in the forecastle and black
cats in the cabin. But surely no tradition of the sea held it an ill omen
to have on board a blond scientist with gold-rimmed spectacles and
a well-cut beard who was seeking information among the technical
schools and universities of the United States.
“He has it in his head that Johnny Kent and I are seafarin’ men
by trade, and he wants to make sure of it for some reason of his
own,” reflected O’Shea. “It has strained me imagination to lie to him
and get away with it. As for Johnny, he would rather talk farming than
anything else in the world, so he will pass for a genuine hayseed in
any company.”
They were deprived of the pleasant society of Professor
Vonderholtz next day, for he boldly monopolized the school-teacher,
Miss Jenness, who seemed not in the least bored by his assiduous
attentions. Elderly ladies watched them with open interest, and
diagnosed it as one of those swift and absorbing steamship
romances.
For three days out of Liverpool the Alsatian moved uneventfully
over the face of the waters. The weather was bright, the sea smooth.
The scratch crew of firemen toiled faithfully in the torrid caverns far
below, and the mighty engines throbbed unceasingly to whirl the twin
screws that pushed the foaming miles astern. On the bridge the
captain and his officers went cheerfully about their tasks, thankful for
clear skies and a good day’s run.
It was after midnight, and the Alsatian was in mid-ocean, when a
few of the first-cabin passengers heard what sounded to their drowsy
ears like several pistol shots. There are many noises aboard a
steamship that are unfamiliar to the landsman. Excepting Captain
O’Shea and Johnny Kent, such of the passengers as had been
awakened paid so little heed to the sounds that they soon went to
sleep again.
The two seafarers slumbered lightly, as is the habit of men used
to turning out to stand watch. And they were not likely to mistake the
report of a revolver for any sound to be expected in the routine of
things on shipboard. O’Shea leaned over from the upper berth and
asked in low tones:
“Are ye awake, Johnny?”
“Sure I am. Did you hear the rumpus?”
“Yes. At first I thought I was dreaming we were aboard the old
Fearless with Jiminez, the big black nigger from Venezuela, taking
pot shots at me. What did ye make of it? It sounded like pretty lively
gun-play to me.”
“It wasn’t no ordinary sailors’ fracas,” hoarsely whispered Johnny
Kent. “Several of those shots was fired for’ard, and others came
from below, about amidships. We heard ’em through the bulkheads.”
“And there was some running to and fro on deck,” said O’Shea,
“by men with no shoes on. I heard their bare feet slapping the planks
over me head.”
“We haven’t been boarded by pirates, and, anyhow, pirates are
out of date in the Atlantic trade, Cap’n Mike. The ship hasn’t
stopped. It would have waked me in a jiffy if her engines had quit
poundin’ along, even for a minute.”
“I thought I heard yells, faint and far away, from men in trouble,
but ’tis all quiet now, Johnny.”
“Too darn quiet. The vessel has slowed down a trifle, by six or
eight revolutions, but she’s joggin’ along all serene. Shall we take a
turn on deck and look around?”

You might also like