Visit https://ebookultra.

com to download the full version and

explore more ebooks

Separation of Variables for Partial Differential

Equations An Eigenfunction Approach 1st Edition
George Cain

_____ Click the link below to download _____

https://ebookultra.com/download/separation-of-
variables-for-partial-differential-equations-an-
eigenfunction-approach-1st-edition-george-cain/

Explore and download more ebooks at ebookultra.com

Here are some suggested products you might be interested in.
Click the link to download

Partial differential equations An introduction 2nd Edition

Strauss W.A.

https://ebookultra.com/download/partial-differential-equations-an-
introduction-2nd-edition-strauss-w-a/

Handbook of Differential Equations Stationary Partial

Differential Equations Volume 6 1st Edition Michel Chipot

https://ebookultra.com/download/handbook-of-differential-equations-
stationary-partial-differential-equations-volume-6-1st-edition-michel-
chipot/

Partial Differential Equations A Unified Hilbert Space

Approach 1st Edition Rainer Picard

https://ebookultra.com/download/partial-differential-equations-a-
unified-hilbert-space-approach-1st-edition-rainer-picard/

Applied Partial Differential Equations John Ockendon

https://ebookultra.com/download/applied-partial-differential-
equations-john-ockendon/
Stochastic Partial Differential Equations 2nd Edition Chow

https://ebookultra.com/download/stochastic-partial-differential-
equations-2nd-edition-chow/

Solution techniques for elementary partial differential

equations Third Edition Constanda

https://ebookultra.com/download/solution-techniques-for-elementary-
partial-differential-equations-third-edition-constanda/

Introduction to Partial Differential Equations Rao K.S

https://ebookultra.com/download/introduction-to-partial-differential-
equations-rao-k-s/

Control theory of partial differential equations 1st

Edition Guenter Leugering

https://ebookultra.com/download/control-theory-of-partial-
differential-equations-1st-edition-guenter-leugering/

Partial Differential Equations 2nd Edition Lawrence C.

Evans

https://ebookultra.com/download/partial-differential-equations-2nd-
edition-lawrence-c-evans/
Separation of Variables for Partial Differential Equations
An Eigenfunction Approach 1st Edition George Cain
Digital Instant Download
Author(s): George Cain, Gunter H. Meyer
ISBN(s): 9781584884200, 1584884207
Edition: 1
File Details: PDF, 9.72 MB
Year: 2005
Language: english
 Separation
of Variables
for Partial
Differential
Equations
An Eigenfunction
Approach

STUDIES IN ADVANCED MATHEMATICS

Separation
of Variables
for Partial
Differential
Equations
An Eigenfunction
Approach
Studies in Advanced Mathematics

Titles Included in the Series

John P. D'Angelo, Several Complex Variables and the Geometry of Real Hypersurfaces
Steven R. Bell, The Cauchy Transform, Potential Theory, and Conformal Mapping
John J. Benedetto, Harmonic Analysis· and Applications
John J. Benedetto and Michael l¥. Fraz.ier, Wavelets: Mathematics and Applications
Albert Boggess, CR Manifolds and the Tangential Cauchy-Riemann Complex
Keith Bums and Marian Gidea, Differential Geometry and Topology: With a View to Dynamical Systems
George Cain and Gunter H. Meyer, Separation of Variables for Partial Differential Equations: An
Eigenfunction Approach
Goo11g Chen and Jianxi11 Zhou, Vibration and Damping in Distributed Systems
Vol. I: Analysis, Estimation, Attenuation, and Design
Vol. 2: WKB and Wave Methods, Visualization, and Experimentation
Carl C. Cowen and Barbara D. MacCluer, Composition Operators on Spaces of Analytic Functions
Jewgeni H. Dshalalow, Real Analysis: An Introduction to the Theory of Real Functions and Integration
Dean G. Duffy, Advanced Engineering Mathematics with MATLAB®, 2nd Edition
Dean G. Duffy, Green's Functions with Applications
Lawrence C. Evans and Ronald F. Gariepy, Measure Theory and Fine Properties of Functions
Gerald B. Folland, A Course in Abstract Harmonic Analysis
Josi Garcfa-Cuerva, Eugenio Hernd.ndez, Fernando Soria, and Josi-Luis Torrea,
Fourier Analysis and Partial Differential Equations
Peter 8. Gilkey. Invariance Theory, the Heat Equation,_and the Atiyah·Singer Index Theorem,
2nd Edition
Peter B. Gilke.v, John V. Leahy, and Jeonghueong Park, Spectral Geometry, Riemannian Submersions,
and the Gromov-Lawson Conjecture
Alfred Gray, Modem Differential Geometry of Curves and Surfaces with Mathematica, 2nd Edition
Eugenio Hemd.ndez and Guido Weiss. A First Course on Wavelets
Kenneth B. Howell, Principles of Fourier Analysis
Steven G. Krantz, The Elements of Advanced Mathematics, Second Edition
Steven G. Krantz., Partial Differential Equations and Complex Analysis
Steven G. Krantz. Real Analysis and Foundations, Second Edition
Kenneth L. Kurt/er, Modern Analysis
Michael Pedersen, Functional Analysis in Applied Mathematics and Engineering
Ciark Robinson, Dynamical Systems: Stability, Symbolic Dynamics, and Chaos, 2nd Edition
Jolm Rya11. Clifford Algebras in Analysis and Related Topics
Joh11 Sclierk. Algebra: A Computational Introduction
Pai·ei Soffn. Karel Segeth, and lvo Doletel, High-Order Finite Element Method
Andr<i Unterberger and Harald Upmeier, Pseudodifferential Analysis on Symmetric Cones
James S. lfolker, Fast Fourier Transforms, 2nd Edition
James S. H'Cilker, A Primer on Wavelets and Their Scientific Applications
Gilbert G. U'i?lter and Xiaoping Shen, Wavelets and Other Orthogonal Systems, Second Edition
Nik Weaver. Mathematical Quantization
Kehe Zhu. An Introduction to Operator Algebras
Separation
of Variables
for Partial
Differential
Equations
An Eigenfunction
Approach

George Cain
Georgia Institute of Technology
Atlanta, Georgia, USA
Gunter H. Meyer
Georgia Institute of Technology
A.r!anta, Georgia, USA

Boc<1 Raton London New York

Published in 2006 by
Chapman & HaIVCRC
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2006 by Taylor & Francis Group, LLC

Chapman & HalVCRC is an imprint of Taylor & Francis Group

No claim to original U.S. Government works

Printed in the United States of America on acid-free paper
10987654321

International Standard Book Number-IO: 1-58488-420-7 (Hardcover)

International Standard Book Number-13: 978-1-58488-420-0 (Hardcover)
Library of Congress Card Number 2005051950

This book contains information obtained from authentic and highly regarded sources. Reprinted material is
quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts
have been made to publish reliable data and infonnation, but the author and the publisher cannot assume
responsibility for the validity of all materials or for the consequences of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic.
mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com
(hltp://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive,
Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration
for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate
system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks. and are used only
for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Cain, George L.
Separation of variables for partial differential equations : an eigenfunction approach I George Cain,
Gunter H. Meyer.
p. cm. -- (Studies in advanced mathematics)
Includes bibliographical references and index.
ISBN 1-58488-420-7 (alk. paper)
l. Separation of variables. 2. Eigenfunctions. L Meyer, Gunter H. IL Title. III. Series.

QA377.C247 2005
5 l 5'.353--dc22 2005051950

informa Taylor & Francis Group

is the Academic Division of lnforma plc.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
Acknowledgments

\Ve would like to thank our editor Sunil Nair for welcoming the project and
for his willingness to stay with it as it changed its scope and missed promised
deadlines. ·
We also wish to express our gratitude to Ms. Annette Rohrs of the School
of Mathematics of Georgia Tech who transformed decidedly low-tech scribbles
into a polished manuscript. Without· her talents, and patience, we would· not
have completed the book.
Preface

3-=paration of variables is a solution method for partial differential equations.

\\liile its beginnings date back to work of Daniel Bernoulli (1753), Lagrange
, 1759), and d'Alembert (1763) on wave motion (see [2]), it is commonly asso-
,ciated with the name of Fourier (1822), who developed it for his research on
·:-Jnductive heat transfer. Since Fourier's time it has been an integral part of
e!:gineering mathematics, and in spite of its limited applicability and heavy com-
petition from numerical methods for partial differential equations, it remains a
-;;.·ell-known and widely used technique in applied mathematics.
Separation of variables is commonly considered an analytic solution
::nethod that yields the solution of certain partial differential equations in terms
cf an infinite series such as a Fourier series. While it may be straightforward to
·:uite formally the series solution, the question in what sense it solves the prob-
lem is not readily answered without recourse to abstract mathematical analysis.
A modern treatment focusing in part on the theoretical underpinnings of the
method and employing the language and concepts of Hilbert spaces to analyze
che infinite series may be fonnd in the text of MacCluer [15]. For many problems
•he formal series can be shown to represent an analytic solution of the differ-
ential equation. As a tool of analysis, however, separation of variables with its
:nfirtite series solutions is not needed. Other mathematical methods exist which
guarantee the existence and uniqueness of a solution of the problem under much
::nore general conditions than those required for the applicability of the method
;:,f separation of variables.
In this text we mostly ignore infinite series solutions and their theoretical
and practical complexities. We concentrate instead on the first N terms of the
series which are all that ever are computed in an engineering application. Such
a partial sum of the infinite series is an approximation to the analytic solution
c,f the original problem. Alternatively, it can be viewed as the exact analytic
solution of a new problem that approximates the given problem. This is the
point of view taken in this book.
Specifically, we view the method of separation of variables in the following
context: mathematical analysis applied to the given problem guarantees the
existence and uniqueness of a solution u in some infinite dimensional vector space
o:,f functions X, but in general provides no means to compute it. By modifying
the problem appropriately, however, an approximating problem results which
has a computable closed form solution UN in a subspace M of X. If fl!! is

vii
viii PREFACE

suitably chosen, then UN is a good approximation to the unknown solution u.

As we shall see, M will be defined such that UN is just the partial sum of the
first N terms of the infinite series traditionally associated with the method of
separation of variables.
The reader may recognize this view as identical to the setting of the finite
element, collocation, and spectral methods that have been developed for the
numerical solution of differential equations. All these methods differ in how the
subspace Mis chosen and in what sense the original problem is approximated.
These choices dictate how hard it is to compute the approximate solution UN
and how well it approximates the ai!alytic solution u.
Given the almost universal applicability of numerical methods for the solu-
tion of partial differential equations, the question arises whether separation of
variables with its severe restrictions on the type of equation and the geometry of
the problem is still a viable tool and deserves further exposition. The existence
of this text ·reflects our view that the method of separation of variables still
belongs to the core of applied mathematics. There are a number of reasons.
Closed form (approximate) solutions show structure and exhibit explicitly
the influence of the problem parameters on the solution. We think, for example,
of the decomposition of wave motion into standing waves, of the relationship
between driving frequency and resonance in sound waves, of the influence of
diffusivity on the rate of decay of temperature in a heated bar, or of the gen-
eration of equipotential and stream lines for potential flow. Such structure and
insight are not readily obtained frnm purely numerical solutions of the underly-
ing differential equation. Moreover, optimization, control, and inverse problems
tend to be easier to solve when an analytic representation of the (approximate)
solution is available. In addition, the method is not as limited in its applicability
as one might infer from more elementary texts on separation of variables. Ap-
proximate solutions are readily computable for problems with time-dependent
data, for diffusion with convection and wave motion with dissipation, problems
seldom seen in introductory textbooks. Even domain restrictions can sometimes
be overcome with embedding and domain decomposition techniques. Finally,
there is the class of singularly perturbed and of higher dimensional problems
where numerical methods are not easily applied while separation of variables
still yieltb an analytic approximate solution.
Our rationale for offering a new exposition of separation of variables is then
twofold. First, although quite common in more advanced treatments (such as
[15]), interpreting the separation of variables solution as an eigenfunction expan-
sion is a point of view r.arely taken when introducing the method to students.
Usually the formalism is based on a product solution for the partial differential
equation, and this limits the applicability of the method to homogeneous partial
differential equations. When source terms do appear, then a reformulation of
problems for the heat and wave equation with the help of Duhamel's superposi-
tion principle and an approximation of the source term in the potential equation
with the help of an eigenfunction approximation become necessary. In an expo-
sition based from the beginning on an eigenfunction expansion, the presence of
source terms in the differential equation is only a technical, but not a conceptual
PREFACE ix

complication, regardless of the type of equation under consideration. A concise

algorithmic approach results.
Equally important to us is the second reason for a new exposition of the
method of separation of variables. We wish to emphasize the power of the
method by solving a great variety of problems which often go well beyond the
usual textbook examples. Many of the applications ask questions which are
not as easily resolved with numerical methods as with analytic approximate
solutions. Of course, evaluation of these approximate solutions usually relies on
numerical methods to integrate, solve linear systems or nonlinear equations, and
to find values of special functions, but these methods by now may be considered
universally available "black boxes." We are, however, mindful of the gap between
the concept of a solution in principle and a demonstrably computable solution
and try .to convey our experience with how well the eigenfunction approach
actually solves the sample problems.
The method of separation of variables from a spectral expansion view is
presented in nine chapters.
Chapter 1 collects some background information on the three dominant equa-
tions of this text, the potential equation, the heat equation, and the wave equa-
tion. We refer to these results when applying and analyzing the method of
separation of variables.
Chapter 2 contains a discussion of orthogonal projections which are used time
and again to approximate given data functions in a specified finite-dimensional
but parameter-dependent subspace.
Chapter 3 introduces the subspace whose basis consists of the eigenfunctions
of a so-called Sturm-Liouville problem associated with the application under
consideration. These are the eigenfunctions of the title of this text. We cite
results from the Sturm-Liouville theory and provide a table of eigenvalues and
eigenfunctions that arise in the method of separation of variables.
Chapter 4 treats the case in which the eigenfunctions are sine and cosine
functions with a common period. In this case the projection into the subspace is
closely related to the Fourier series representation of the data functions. Precise
information about the convergence of the Fourier series is known. We cite those
results which are helpful later on for the application of separation of variables.
Chapter 5 constitutes the heart of the text. We consider a partial differen-
tial equation in two independent variables with a source term and subject to
boundary and initial conditions. We give the algorithm for approximating such
a problem and for solving it in a finite-dimensional space spanned by eigen-
functions determined by the "spacial part" of the equation and its boundary
conditions. We illustrate in broad outline the application of this approach to
the heat, wave, and potential equations.
Chapter 6 gives an expansive exposition of the algorithm for the one-dimen-
sional heat equation. It contains many worked examples with comments on
the numerical performance of the method, and concludes with a rudimentary
analysis of the error in the approximate solution.
Chapter 7 parallels the previous chapter but treats the wave equation.
x PREF...\ CE

Chapter 8 deals with the potential equation. It describes how one can pr~
condition the data of problems with smooth solutions in order not to introdi.:ce
artificial discontinuities into the separation of variables solution. We solve P'=-
tential problems with various boundary conditions and conclude with a calcu-
lation of eigenfunctions for the two-dimensional Laplacian.
Chapter 9 uses the eigenfunctions of the preceding chapter to find eigenfunc-
tion expansion solutions of two- and three-dimensional heat, wave, and potenti.a::
equations.
This text is written for advanced undergraduate and graduate students :.::.
science and engineering with previous exposure to a course in engineering math-
ematics, but not necessarily separation of variables. Basic prerequisites beyond
calculus are familiarity with linear algebra, the concept of vector spaces of func-
tions, norms and inner products, the ability to solve linear inhomogeneous first
and second order ordinary differential equations, and some contact with practi-
cal applications of partial differential equations.
The book contains more material than can (and should) be taught in a cour~
on separation of variables. We have introduced the eigenfunction approach to
our own students based on an early version of this text. We covered parts of
Chapters 2-4 to lay the groundwork for an extensive discussion of Chapter 5.
The remainder of the term was filled by working through selected examples
involving the heat, wave, and potential equation. We believe that by term ·s
end the students had an appreciation that they could solve realistic problems.
Since we view Chapters 2-5 as suitable for teaching separation of variables,
we have included exercises to help deepen the reader's understanding of the
eigenfunction approach. The examples of Chapters 6-8 and their exercise sets
generally lend themselves for project assignments.
This text will put a bigger burden on the instructor to choose topics and
guide students than more elementary texts on separation of variables that start
with product solutions. The instructor who subscribes to the view put forth
in Chapter 5 should find this text workable. The more advanced applications,
such as interface, inverse, and multidimensional problems, as well as the the
more theoretical topics require more mathematical sophistication and may be
skipped without breaking continuity.
The book is also meant to serve as a reference text for the method of separa-
tion of variables. We hope the many examples will guide the reader in deciding
whether and how to apply the method to any given problem. The examples
should help in interpreting computed solutions, and should give insight into
those cases in which formal answers are useless because of lack of convergence
or unacceptable oscillations. Chapters 1 and 9 are included to support the
reference function. They do not include exercises.
We hasten to add that this text is not a complete reference book. We
do not attempt to characterize the equations and coordinate systems where
a separation of variables is applicable. We do not even mention the various
coordinate systems (beyond cartesian, polar, cylindrical, and spherical) in which
the Laplacian is separable. We _have not scoured the literature for new and
innovative applications of separation of variables. Moreover, the examples we do
PREFACE xi

include are often meant to show structure rather than represent reality because
in general little attention is given to the proper scaling of the equations.
There does not appear to exist any other source that could serve as a prac-
tical reference book for the practicing engineer or scientist. We hope this book
will alert the reader that separation of variables has more to offer than may be
apparent from elementary texts.
Finally, this text does not mention the implementation of our formulas and
calculations on the computer, or do we provide numerical algorithms or pro-
grams. Yet the text, and in particular our numerical examples, could not
have been presented without access to symbolic and numerical packages such as
!viaple, Mathematica, and Matlab. We consider our calculations and the graph-
ical representation of their results routine and well within the competence of
today's students and practitioners of science and engineering.
Contents

Acknowledgments v

Preface vii

1 Potential, Heat, and Wave Equation 1

1.1 Overview . . . . . . . . . . . . . .
1.2 Classification of second order equations 2
1.3 Laplace's and Poisson's equation 3
1.4 The heat equation 11
1.5 The wave equation . 18

2 Basic Approximation Theory 25

2.1 Norms and inner products . . 26
2.2 Projection and best approximation 29
2.3 Important function spaces 34

3 Sturm-Liouville Problems 45
3.1 Sturm-Liouville problems for q/' = µ¢ 45
3.2 Sturm-Liouville problems for £¢ = µ¢ . . . 53
3.3 A Sturm-Liouville problem with an interface 59

4 Fourier Series 67
4.1 Introduction . 67
4.2 Convergence . 68
4.3 Convergence of Fourier series 74
4.4 Cosine and sine series . . . . 77
4.5 Operations on Fourier series . 80
4.6 Partial sums of the Fourier series and the Gibbs phenomenon 84

5 Eigenfunction Expansions for Equations in Two Independent

Variables 95
XiY CONTENTS

6 One-Dimensional Diffusion Equation 115

6.1 Applications of the eigenfunction expansion method 115
Example 6.1 How many terms of the series solution are enough? 115
Example 6.2 Determination of an unknown diffusivity from mea-
sured data .. 119
Example 6.3 Thermal waves 120.
Example 6.4 Matching a temperature history 125
Example 6.5 Phase shift for a thermal wave . 129
Example 6.6 Dynamic determination of a convective heat transfer
coefficient from measured data . . . . . . . . . . . . 131
Example 6.7 Radial heat flow in a sphere 134
Example 6.8 A boundary layer problem . 137
Example 6.9 The Black-Scholes equation 139
Example 6.10 Radial heat flow in a disk . 142
Example 6.11 Heat flow in a composite slab 146
Example 6.12 Reaction-diffusion with blowup 148
6.2 Convergence of UN(x, t) to the analytic solution : 151
6.3 Influence _of the boundary conditions and Duhamel 's solution 155

7 One-Dimensional Wave Equation 161

7.1 Applicatfons of the eigenfunction expansion method - : 161
Example 7.1 A vibrating string with initial displacement . 161
Example 7.2 A vibrating string with initial velocity . . 166
Example 7.3 A forced wave and resonance . 168
Example 7.4 Wave propagation in a resistive medium. 171
Example 7.5 Oscillations of a hanging chain 175
Example 7.6 Symmetric pressure wave in a sphere 177
Example 7.7 Controlling the shape of a wave . . . 180
Example 7.8 The natural frequencies. of a uniform beam 182
Example 7.9 A system of wave equations . . . . . 185
7.2 Convergence of UN(x, t) to the analytic solution . 188
7.3 Eigenfunction expansions and Duhamel's
principle . 190

8 Potential Problems in the Plane 195

8.1 Applications of the eigenfunction expansion method . 195
Example 8.1 The Dirichlet problem for the Laplacian on a rectanglel95
Example 8.2 Preconditioning for general boundary data . . 201
Example 8.3 Poisson's equation with Neumann boundary data 213
Example 8.4 A discontinuous potential 215
Example 8.5 Lubrication of a plane slider bearing . 218
Example 8.6 Lubrication of a step bearing . . 220
Example 8.7 The Dirichlet problem on an £-shaped domain 221
Example 8.8 Poisson's equation in polar coordinates . . . . 225
Example 8.9 Steady-state heat flow around an insulated pipe I 230
Example 8.10 Steady-state heat flow around an insulated pipe II 232
CONTENTS xv

Example 8.11 Poisson's equation on a triangle . 234

8.2 Eigenvalue problem for the two-dimensional Laplacian 237
Example 8.12 The eigenvalue problem for the Laplacian on a
rectangle . . . . . . . . . . 237
Example 8.13 The Green's function for the Laplacian on a square 239
Example 8.14 The eigenvalue problem for the Laplacian on a disk 243
Example 8.15 The .eigenvalue problem for the Laplacian on the
surface of a sphere . . . . . . . . . 244
8.3 Convergence of UN(x, y) to the analytic solution 247

9 Multidimensional Problems 255

9.1 Applications of the eigenfunction expansion method 255
Example 9.1 A diffusive pulse test . . . . . . . 255
Example 9.2 Standing waves on a circular membrane 258
Example 9.3 The potential inside a charged sphere 260
Example 9.4 Pressure in a porous slider bearing . . . 261
9.2 The eigenvalue problem for the Laplacian in ffi:. 3 . . . 265
Example 9.5 An eigenvalue problem for quadrilaterals 265
Example 9.6 An eigenvalue problem for the Laplacian in a cylinder266
Example 9. 7 Periodic heat flow in a cylinder 267
Example 9.8 An eigenvalue problem for the Laplacian in a sphere 269
Example 9.9 The eigenvalue problem for Schrodinger's equation
with a spherically symmetric potential well . . . . 271

Bibliography 277

Index 279
Chapter 1

Potential, Heat, and Wave

Equation

This chapter provides a quick look into the vast field of partial differential
equations. The main goal is to extract some qualitative results on the three
dominant equations of mathematical physics, the potential, heat, and wave
equation on which our attention will be focused throughout this text.

1.1 Overview
When processes that change smoothly with two or more independent variables
are modeled mathematically, then partial differential equations arise. Most
common are second order equations of the general form
M M
Lu= L aijUxix, +L biuxi +cu = F, xE DC IRM (1.1)
i,j=l i=l

where the coefficients and the source term may depend on the independent vari-
ables {x 1, ... , x M}, on u, and on its derivatives. D is a given set in !RM (whose
boundary will be denoted by [) D). The equation may reflect conservation and
balance laws, empirical relationships, or may be purely phenomenological. Its
solution is used to explain, predict, and control processes in a bewildering array
of applications ranging from heat, mass, and fluid flow, migration of biological
species, electrostatics, and molecular vibration to mortgage banking.
In (1.1) C. is known as a partial differential operator that maps a smooth
function u to the function C.u. Throughout this text a smooth function denotes
a function with as many continuous derivatives as are necessary to carry out
the operations to which it is subjected. lu = F is the equation to be solved.
Given a partial differential equation and side constraints on its solution,
typically initial and boundary conditions, it becomes a question of mathematical
analysis to establish whether the problem has a solution, whether the solution
2 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

is unique, and whether the solution changes continuously with the data of the
problem. If that is the case, then the given problem for (1.1) is said to be well
posed; if not then it is ill posed. We note here that the data of the problem are
the coefficients of£, the source term F, any side conditions imposed on u, and
the shape of D. However, dependence on the coefficients and on the shape of D
will be ignored. Only continuous dependence with respect to the source term
and the side conditions will define well posedness for our purposes.
The technical aspects of in what sense a function u solves the problem and
in what sense it changes with the data of the problem tend to be abstract and
complex and constitute the mathematical theory of partial differential equations
(e.g., [5]). Such theoretical studies are essential to establish that equation (1.1)
and its side conditions are a consistent description of the processes under consid-
eration and to characterize the behavior of its solution. Outside mathematics
the validity of a mathematical model is often taken on faith and its solution
is assumed to exist on "physical grounds." There the emphasis is entirely on
solving the equation, analytically if possible, or approximately and numerically
otherwise. Approximate solutions are the subject of this text.

1.2 Classification of second order equations

The tools for the analysis and solution of (1.1) depend on the structure of the
coefficient matrix
A= {aij}
in (1.1). By assuming that ux,x 1 = ux 1 x, we can always write A in such a way
that it is symmetric. For example, if the equation which arises in modeling a
process is

then it will be rewritten as

so that
A= G ~)
We can now introduce three broad classes of differential equations.

Definition The operator £ given by

M M
Lu =Li,j=l
aijUxtx) +L
i::::l
biuXi +cu

is
i) Elliptic at x = (x 1 , ••. , XM) if all eigenvalues of the symmetric matrix A are
nonzero and have the same algebraic sign,
1.3. LAPLACE'S AND POISSON'S EQUATION 3

ii) Hyperbolic at i if all eigenvalues of A are nonzero and one has a different
algebraic sign from all others,
ili) Parabolic at x if A has a zero eigenvalue.

If A depends on u and its derivatives, then £ is elliptic, etc. at a given point

relative to a specific function u. If the operator£ is elliptic at a point then (1.1)
is an elliptic equation at that point. (As mnemonic we note that for M = 2 the
level sets of
\AG~),(~~)) =constant,
where (i, if') denotes the dot product of x and y, are elliptic, hyperbolic, and
parabolic under the above conditions on the eigenvalues of A). The lower order
~erms in ( 1.1) do not affect the type of the equation, but in particular applica-
tions they can dominate the behavior of the solution of ( 1.1).
Each class of equations has its own admissible side conditions to make (Ll)
well posed, and all solutions of the same class have, broadly speaking, common
characteristics. We shall list some of them for the three dominant equations
of mathematical physics: Laplace's equation, the heat equation, and the wave
equation.

1.3 Laplace's and Poisson's equation

The most extensively studied example of an elliptic equation is Laplace's equa-
~ion
£u = \1 ·\Ju= 0
which arises in potential problems, steady-state heat conduction, irrotational
flow, minimal surface problems, and myriad other applications. The operator
£u is known as the Laplacian and is generally denoted by

£u = \1 ·\Ju= \7 2 u = 6.u.
The last form is common in the mathematical literature and will be used con-
sistently throughout this text. The Laplacian in cartesian coordinates
M

~u. = LUxixi
i=l

assumes the forms

i.i In polar coordinates (r, ())

ii) In cylindrical coordinates (r, (), z)

4 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

iii) In spherical coordinates (r, (), ¢)

2 1 1
f).u = Urr + -r Ur+ ~(sin</iuq,).p
rsm.,,
+ - 2- . -2-
rsm</i
uee.

For special applications other coordinate systems may be more advantageous

and we refer to the literature (see, 'e.g., [13]) for the representation of f).u in ad-
ditional coordinate systems. For cartesian coordinates we shall use the common
notation
x= (x,y), x=(x,y,z)
for i! E 1R2 and lR3, respectively.
The generalization of Laplace's equation to

f).u=F (1.2)

for a given source term Fis known as Poisson's equation. It will be the dominant
elliptic equation in this text.
In general equation (1.2) is to be solved for i! E D where D is an open set
in JRM. For the applications in this text D will usually be a bounded set with a
sufficiently smooth boundary 8D. On 8D the solution u has to satisfy boundary
conditions. We distinguish between three classes of boundary data for (1.2) and
its generalizations.
i) The Dirichlet problem (also known as a problem of the first kind)

u = g(i!), i! E 8D.

ii) The Neumann problem (also known as a problem of the second kind)

au
an=gx,
(-) i! E 8D

where ~ = 'Vu · ii.(i!) is the normal derivative of u, i.e., the directional

derivative of u in the direction of the outward unit normal ii.(i!) to D at
i! E 8D.
iii) The Robin problem (also known as a problem of the third kind):

au + a2u = g (-)
°''on x ' i! E 8D,

where, at least in this text, a 1 and a 2 are piecewise nonnegative constants.

We shall call general boundary value problems for (1.2) potential problems.
The differential equation and the boundary conditions are called homogeneous
if the function u(i!) = 0 can satisfy them. For example, (1.2) is homogeneous if
=
F(i!) 0, and the boundary data are homogeneous if g(i!) 0. =
If D is a bounded open set which has a well-defined outward normal at every
point of 8 D, then for continuous functions F on D and g on {) D a classical
solution of these three problems is a function u which is twice continuously
differentiable in D and satisfies (1.2) at every point of D. In addition, the
J.3. LAPLACE'S AND POISSON'S EQUATION 5

dassical solution of the Dirichlet problem is required to be continuous on the

dosed set fJ =DU oD and equal tog on oD. For problems of the second and
rhird kind the classical solution also needs continuous first derivatives on fJ in
order to satisfy the given boundary condition on oD.
The existence of classical solutions is studied in great generality in [6]. It is
known that for continuous F and g and smooth oD the Dirichlet problem has a
classical solution, and that the Robin problem has a classical solution whenever
F and g are continuous and
°'1°'2 > 0.
A classical solution for the Neumann problem is known to exist for continuous
F and g provided
f, F(x)dx
D
= (
lav
g(S)ds.

Why this compatibility condition arises is discussed below.

Considerable effort has been devoted in the mathematical literature to ex-
tending these existence results to domains with corners and edges where the
normal is not defined, and to deriving analogous results when F and g (and the
coefficients in ( 1.1)) are not necessarily continuous. Classical solutions no longer
exist but so-called weak solutions can be defined which solve integral equations
derived from (1.1) and the boundary conditions. This general existence theory
is also presented in [6].
In connection with separation of variables we shall be concerned only with
bounded elementary domains like rectangles, wedges, cylinders, balls, and shells
where the boundaries are smooth except at isolated corners and edges. At such
points the normal is not defined. Likewise, isolated discontinuities in the data
functions F and g may occur. We shall assume throughout the book (with
optimism, or on physical grounds) that the given problems have weak solutions
which are smooth and satisfy the differential equation and boundary conditions
at all points where the data are continuous.
Our separation of variables solution will be an approximation to the analytic
solution found by smoothing the data F and g. Such an approximation can only
be meaningful if the solution of the original boundary value problem depends
continuously on the data, in other words, if the boundary value problem is well
posed. We shall examine this question for the Dirichlet and Neumann problem.
Dirichlet problem: The Dirichlet problem for Poisson's equation is the
most thoroughly studied elliptic boundary value problem. We shall assume
that F is continuous for x E D and g is continuous for x E {) D so that the
problem
£::,.u = F(x), xE D
u=g(x), xEoD
has a classical solution. Any approximating problem formulated to solve the
Dirichlet problem analytically should likewise have a classical solution. As
discussed in Chapter 8, this may require preconditioning the problem before
applying separation of variables.
6 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

We remark that especially for establishing the existence of a solution it often

is advantageous to split the solution

where

b.u1 = 0, xED
U1 g(x),
= xE8D
b.u2 = F(x), xED
Uz = 0, xE8D

because different mathematical tools are available for Laplace's equation with
nonzero boundary data and for Poisson's equation with zero boundary data
(which, in an abstract sense, has a good deal in common with the matrix problem
Au= b). Splittings of this type will be used routinely in Chapter 8. Of course,
if g is defined and continuous on all of fJ and twice continuously differentiable
in the open set D, then it is usually advantageous to introduce the new function

w=u-g

and solve the Dirichlet problem

b.w = F - b.g, xE D
w =0, xE8D
without splitting.
Given a classical solution we now wish to show that it depends continuously
on F and g. To give meaning to this phrase we need to be able to measure
change in the functions F and g. Here this will be done with respect to the so-
called supremum norm. We recall from analysis that for any function G defined
on a set D c RM

sup [G(i7)[ =least upper bound on the set of values {[G(i7)[ : if ED}.
fjED

A common notation is
llGI[ =sup [G(i7)[
fjED

which is called the supremum norm of G and which is just one example of the
concept of a norm discussed in Chapter 2.
If fJ is a closed and bounded set and G is continuous in fJ, then G must
take on its maximum and minimum on fJ so that

llGI[ =sup [G(il)I = m8-2'[G(il)[.

fjED fjED
i!.3. LAPLACE'S AND POISSON'S EQUATION 7

Continuous dependence of a classical solution (with respect to the supremum

:?ODrm) is given if for every £ > 0, there is a J > 0 such that

whenever
llFll + llgll '.": J.
Here

!lull =ma!' iu(x)I, llFll =sup IF(x)I, and llgll = xrp.Eaaxo lg(x)l-
XED :lED

Continuous dependence on the data in this sense, and uniqueness follow from
foe maximum principle for elliptic equations. Since it is used later on and always
provides a quick check on computed or approximate solutions of the Dirichlet
problem, and since it is basically just the second derivative test of elementary
calculus, we shall briefly discuss it here.

Theorem 1.1 The maximum principle

Let u be a smooth solution of Poisson's equation

l::.u=F, xED. (1.3)

Assume that
F >0 for all x ED.
Then u(x) cannot assume a relative maximum in D.

Proof. If u has a relative maximum at some xo E D, then the second derivative

test requires that ux,x,(x0 ) '.":: 0 for all i which would contradict .6.u(x0 ) =
F(xo) > o.
We note that if fJ is bounded, then a classical solution u of the Dirichlet
problem must assume a maximum at some point in fJ. Since this point cannot
lie in D, it must lie on fJD. Hence u(x) '.":: max;;awg(iJ) for all x E fJ. We
also note that if F < 0, then -u satisfies the above maximum principle which
translates into u(x) ;:> min;;EaD g(iJ).
Stronger statements, extensions to the general elliptic equation (1.1), and
more general boundary conditions may be found in most texts on partial differ-
ential equations (see, e.g., [6]).

Theorem 1.2 A sobtion of the Dirichlet problem depends continuously on the

data F and g.

Proof. Suppose that D is such that a < .xk < b for some k, 1 < k < M where
a and b denote finite lower and upper bounds on the kth coordinate of E D. x
Define the function
¢(£) = (llFll + <) (xk; a) 2
8 CHAPTER l. POTENTIAL, HEAT, AND WAVE EQUATION

where E > 0 is arbitrary and where llFll = supxED JF(x)I and 11911 = maxxEi:W l9(i
are assumed to be finite. Then

6 [,P(:f) ± u(x)] ~ E >0

and by the maximum principle [¢ ± u] is bounded above by its value on 8D so
that
±u(x) :S .P(x) ± u(x) :S CllFll + E) (b-2 a) 2 + 11911-
Since E is arbitrary, it follows that for all xE D
lu(x)I :S llFll (b ~ a) 2 + 11911
which implies
(1.4)

This inequality establishes continuous dependence since llFll _, 0 and 11911 _, 0

imply that max 0 lu(x)I _, 0.

Corollary 1.3 The solution of the Dirichlet problem is unique.

Proof. The difference between two solutions satisfies the Dirichlet problem

b..u = 0 in D
u = 0 on 8D

which by Theorem 1.2 has only the zero solution.

Corollary 1.4 Let F ~ 0. Then the solution of the Dirichlet problem assumes
its maximum on 8D.

Proof. Let E > 0 be arbitrary. Then the solution u, of

b..u = F +E inD
u=9 on av

assumes its maximum on 8D by Theorem 1.1. By Theorem 1.2

(b - a) 2
lu,(x) - u(x)I :SE - 2- .

Since E is arbitrary, u cannot exceed maxyEiJD 9(iJ) by a nonzero amount at any

point in D; hence u(x) :S maxgEi:W 9(fj).
Similarly, if F :S 0, then u assumes its minimum on 8D. Consequently, the
solution of Laplace's equation

b..u=O, xED
1.3. LAPLACE'S AND POISSON'S EQUATION 9

:::nust assume its maximum and minimum on EJD.

Looking ahead, we see that in Chapter 8 the solution u of the Dirichlet
problem for (1.2) will be approximated by the computable solution UN of a
~elated Dirichlet problem

xE D (1.5)
x E EJD.
The existence of UN is given because it will be found explicitly. Uniqueness of
rhe solution guarantees that no other solution of (1.5) exists. It only remains
to establish in what sense UN approximates the analytic solution u. But it is
clear from Theorem 1.2 that for all x ED

(1.6)

wher~ the constant K depends only on the geometry of D. Thus the error in
the approximation depends on how well FN and 9N approximate the given data
F and g. These issues are discussed in Chapters 3 and 4.
When the Dirichlet problem does not have a classical solution because the
data are not smooth, then continuous dependence for weak solutions must be
established. It generally is possible to show that if the data tend to zero in a
mean square sense, then the weak solution of the Dirichlet problem tends to
zero in a mean square sense. This translates into mean square convergence of
UN to u. The analysis of such problems becomes demanding and we refer to [6]
for details. A related result for the heat equation is discussed in Section 6.2.
Neumann problem: In contrast to the Dirichlet problem, the Neumann
problem for Poisson's equation is not well posed because if u is a solution then
u + c for any constant c is also a solution, hence a solution is not unique. But
there may not be a solution at all if the data are inconsistent. Suppose that
u is a solution of the Neumann problem; then it follows from the divergence
theorem that

l F(X)dx = l t:,.udx = l '17. 'lludx

= r
./!'JD
'llu(S). n(S)ds= r
./8D
g(S)ds,

where ii is the outward unit normal on EJD. Hence a necessary condition for the
existence of a solution is the compatibility condition

{ F(x)dx = { g(S)ds. (1.7)

.Iv .!av
Ifwe interpret the Neumann problem for Poisson's equation as a (scaled) steady-
state heat transfer problem, then this compatibility condition simply states that
the energy generated (or destroyed) in D per unit time must be balanced exactly
by the energy flux across the boundary of D. Were this not the case D would
warm up or cool down and could not have a steady-state temperature. Equation
10 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION I

(1.7) implies that the solution does not change continuously with the data since.
F and g cannot be changed independently. If, however, (1.7) does hold, then the
Neumann problem is known to have a classical solution which is unique up to ,
an additive constant. In practice the solution is normalized by assigning a value
to the additive constant, e.g., by requiring u(x0 ) = 1 for some fixed x 0 E D.
The Neumann problems arise frequently in applications and can be solved with
separation of variables. This requires care in formulating the approximating
problem because it, too, must satisfy the compatibility condition (1.7) We shall
address these issues in Chapter 8.
We note that problems of the third kind formally include the Dirichlet and
Neumann problem. The examples considered later on are simply assumed (on
physical grounds) to be well posed. Uniqueness, however, is easy to show with
the maximum principle, provided

As stated above, in this case we have a classical solution. Indeed, if u 1 and u 2

are two classica! solutions of .the Robin problem, then

W = U1 - Uz

satisfies
~w =0, xE D
ow
0<1 on + 0<2W = 0, x E oD.
From the maximum principle we know that w must assume its maximum and
minimum.on oD. Suppose that w has a positive maximum at io E oD; then
the boundary condition implies that

ow
- (-
xo ) = --
°'2 w (-
xo ) < 0
on °'1
so that w has a strictly positive directional derivative along the inward unit
normal -ii. This contradicts that w(£0 ) is a maximum of w on D. Hence w
cannot have a positive maximum on D. An analogous argument rules out a
negative minimum so that w = 0 is the only possibility.
Finally, let us illustrate the danger of imposing the wrong kind of boundary
conditions on Laplace's equation.
For any positive integer k let us set

Ek = (2k + l)7r

and consider the problem

~u = 0, (x,y) E (0, 1) x (0, 1)

u =0 on x = 0, 1
IA. THE HEAT EQUATION 11

and
u(x,O) = 0 for x E (0, 1)

uy(x, 0) =Ek sin..:'.. for x E (0, 1).

Ek
We verify that
uk(x, y) = E% sin..:'.. sinh .!
Ek Ek
is a solution of this problem. By inspection we see that

while
lu G, I 1) = E~ sinh ~ _, oo ask-. oo.

Hence the boundary data tend to zero uniformly while the solution blows up.
Thus the problem cannot be well posed.
In general it is very dangerous to impose simultaneously Dirichlet and Neu-
mann data (called Cauchy data) on the solution of an elliptic problem on a
portion of 8D even if the application does furnish such data. The resulting
problem, even if formally solvable, tends to hav: an unstable solution.

1.4 The heat equation

The best known example of a parabolic equation is the 111-dimensional heat
equation
.Cu= Ll.u - u, = F(x, t), xED, t>to (1.8)
defined for the (M + 1)-dimensional variable (x, t) = (x 1 , .•. , XM, t). It is cus-
tomary to use t for the (111 + 1)st component because usually (but not always)
time is a natural independent variable in the derivation of (1.8). For example,
(1.8) is the mathematical model for the (scaled) temperature u in a homogeneous
body D which changes through conduction in space and time. F represents a
heat source when F :::; 0 and a sink when F > 0. In this application the equa-
tion follows from the principle of conservation of energy and Fourier's law of
heat conduction (sec, e.g., [7)). However, quite diverse applications lead to (1.8)
and to generalizations which formally look like (1.1). A parabolic equation like
(1.1) with variable coefficients and additional terms is usually called the dif-
'1sion equation. We shall consider here the heat equation (1.8) because the
qualitative behavior of its solution is generally a good guide to the behavior of
the solution of a general diffusion equation. -
We observe that a steady-state solution of (1.8) with a time-independent
source term is simply the solution of Poisson's equation (1.2). Hence it is consis-
tent to impose on (1.8) the same types of boundary conditions on 8D discussed
in Section 1.3 for Poisson's equation. Thus we speak of a Dirichlet, Neumann,
12 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

or Robin problem (also known as a reflection problem) for (1.8). In addition,

the application will usually provide an initial condition

u(x, to) = uo(i), x E fJ

at some time t 0 (henceforth set to t 0 = 0).
It is possible in applications that the domain D for the spacial variable
x changes with time. However, separation of variables will require a time-
independent domain. Hence D will be a fixed open set in IPl.M with boundary
&D. The general formulation of an initial/boundary value problem for ( 1.8) is

Cu= F(x,t), x ED, t>0 (1.9a)

with initial condition
u(x, O) = uo(i), xE fJ (i.9b)
and boundary condition
&u
oq &n + a2u = g (- )
x, t , x E &D, t>0 (1.9c)

for a1, 0<2 ;::: 0 and a1 + a2 > 0. Let us define the set

Qr={(x,t):xED, O<t'.'OT},

and the so-called parabolic boundary

&Qr = { (x, 0) : x E fJ} u { (x, t) : x E &D, 0 < t '.'OT},

where T > 0 is an arbitrary but fixed final time. Then a solution of (1.9) has
to satisfy (in some sense) (1.9a) in Qr and the boundary and initial conditions
on &Qr.
A classical solution of (1.9) is a function which is smooth in Qr, which is
continuous on Qr U &Qr (together with its spacial derivatives if a 1 f= 0) and
which satisfies the given data at every point of &Qr.
It is common for diffusion problems that the initial and boundary condi-
tions are not continuous at all points of the parabolic boundary. In this case
u cannot be continuous on &Qr and one again has to accept suitably defined
weak solutions which only are required to solve the diffusion equation and ini-
tial/boundary conditions in an integral equation sense.
First and foremost in the discussion of well posedness for the heat equation
is the question of existence of a solution. For the one-dimensional heat equation
one can sometimes exhibit and analyze a solution in terms of an exponential
integral - see the discussion of (1.13), (1.14) below - but in generirl~is
question is resolved with fairly abstract classical and functional analysis. As in
the case of Poisson's equation it is possible to split the problem by writing

where
(x,t) E Qr
1.4. THE HEAT EQUATION 13

°'1
8u1 + 0<2U1 =
Bn (- t ) ,
g X, xEaD, 0<t:5.T

u1(x,O) = 0, x E [J
and
.Cu2 = F(x, t), (x,t) E Qr

xECID, 0<t:5.T

u2(x, O) = uo(x),
and employ special techniques to establish the existence of u 1 and u 2 . In par-
ticular, the problem for Uz in an abstract sense has a lot in common with an
n-dimensional first order system
du
dt - A(t)u = G(t), u(O) = uo

and can be analyzed within the framework of (abstract) ordinary differential

equations. We refer to the mathematics literature, notably [14], for an exten-
sive discussion of classical and weak solutions of boundary value problems for
linear and nonlinear diffusion equations. In general, it" is safe to assume that if
the boundary and initial data are continuous on the parabolic boundary, then
all three types of initi!J-1/boundary value problems for the heat equation on a
reasonable domain have unique classical solutions. Note that for the Neumann
problem the heat equation does not require a compatibility condition linking
the source F and the flux g.
If the data are discontinuous only at t = 0 (such as instantaneously heating
an object at t > 0 on CID above its initial temperature), then the solution will
be discontinuous at t = 0 but be differentiable for t > 0. It is useful to visualize
such problems as the limit of problems with continuous but rapidly changing
data near t = 0.
Continuous dependence of classical solutions for initial/boundary value prob-
lems on the data, and the uniqueness of the solution can be established with
generalizations of the maximum principle discussed above for Poisson's equa-
tion. For example, we have the following analogues of Theorems 1.1 and 1.2.

Theorem 1.5 The maximum principle

Let u be a smooth solution of

.Cu= F, (x,t) E Qr .

If F > 0, then u cannot have a maximum in Qr.

Proof. If u had a maximum at some point (£0 , t 0 ) E Qr, i.e., £ 0 lies in the
open set D, then necessarily

ux,x,(£0 , t 0 ) :5. 0 and u,(£0 , to)= 0 if to< Tor u,(xo, to) 2: 0 if to= T.
14 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

In either case we could not satisfy

.Cu(xo, to) = F(xo, to) > 0.

Arguments analogous to those applied above to Poisson's equation establish

that if D is bounded and F ::'.: 0, then u must assume its maximum on iJQy and
if F :S 0, then u must assume its minimum on iJQy.
Continuous dependence of the solution of the Dirichlet problem for the heat
equation on the data with respect to the sup norm is now defined as before.
There is continuous dependence if for any f > 0, there exists a 6 > 0 such that

fluff :Sf

whenever ff Pff + max{ffgff, ffuoff} :S 6. Here

fluff = II!ax fu(x, t)f,

QT

ff Pff =sup fF(x, t)f, ffgff = ~~x fg(x, t)f, ffuoff = m11x fuo(x)f.
QT D
O:St~T

Theorem 1.6 The solution of the Dirichlet problem for (1.8) depends contin-
uously on the data F, g, and u 0 .

Proof. We again assume that the kth coordinate Xk satisfies a :S Xk :S b for all
x ED. For arbitrary f > 0 define

</J(x) = (ff Fii + E) (xk; a) 2 .

Then
£[¢(£) ±u(x,t)];::: f >o
and by Theorem 1.5 [4' ± u] is bounded by its maximum on iJQy. Hence

±u(x, t) :S <P(x) + u(x, t) :S (ff Fii + E) + max{ffgff, ffuoff}.

Continuous dependence now follows exactly as in the proof of Theorem 1.1 and
we have the following analogue of estimate ( 1.5):

(b a) 2
fluff :S ffFff ~ + max{ffgff, ffuoff}. (1.10)

~imilarly, we can conclude that the solution of the Dirichlet problem for the
heat equation is unique and that the solution of the heat equation with F 0 =
(the homogeneous heat equation) must take on-its maximum and minimum on
iJQy.
In addition to boundary value problems we also can consider a pure initial
value problem for the heat equation

.Cu= fi.u - u, = 0, XE !KM, t>O

J.4. THE HEAT EQUATION 15

u(x, O) = uo(x).
It can be verified that the problem is solved by the formula

u(x, t) = J.
IRM
s(x - fj, t)u0 (fj)dfj (1.11)

where
- 1 _jg!
s(x, t) = (47rt)M/2 e " -
Here (x,x) denotes the dot product for vectors in IRM- s(x, t) is known as the
fundamental solution of the heat equation. A simple calculation shows that s is
infinitely differentiable with respect to each component X; and t for t > 0 and
that
!is(x, t) - s,(x, t) = 0 fort> 0.
It follows from (Lll) that u(x, t) is infinitely differentiable with respect to all
variables for t > 0 provided only that the resulting integrals remain defined
and bounded. In particular, if u0 is a bounded piecewise continuous function
defined on IRM, then the solution to the initial value problem exists and is
infinitely differentiable for all X; and all t > 0. It is harder to show that u(x, t)
is continuous at (£0 , 0) at all points x 0 where u 0 is continuous and that u(x, t)
assumes the initial value u 0 (x0 ) as (x, t) _, (x0 , 0). We refer to [5) for a proof
of these results. Note that for discontinuous u 0 the expression (1.ll) is only a
weak solution because u(x, t) is not continuous at t = 0.
We see from (1.11) that if for any E > 0

( -) _ {1 for \\x\\::; E
uo x - 0 otherwise,
then u(x, t) > 0 for t > 0 at all x E !RM- In other words, the initial condition
spreads throughout space infinitely fast. This property is a consequence of the
mathematical model and contradicts the observation that heat does not flow
infinitely fast. But in fact, the change in the solution (1.11) at \\x\\ » E remains
unmeasurably small for a certain time interval before a detectable heat wave
arrives so that defacto the wave speed is finite. We shall examine this issue at
length in Example 6.3 where the speed of an isotherm is found numerically.
The setting of diffusion in all of IRM would seem to preclude the application of
(1.ll) to practical problems such as heat flow in a slab or bar. But (1.11) is not
as restrictive as it might appear. This is easily demonstrated if M = 1. Suppose
that u 0 is odd with respect to a given point x 0 • i.e., u 0 (x 0 + x) = -u 0 (x 0 - x);

u(x 0 1:
then with the obvious changes of variables

+ x, t) = s(x 0 +x - y, t)u 0 (y)dy = 1: s(x - z, t)u 0 (z + x 0 )dz

= -1
00
s(x - z, t)uo(xo - z)dz = 1-oo s(x - x 0 + y)u 0 (y)dy

=- 1:
-00

s(xo - x -y)uo(y)dy = -u(x 0

00

- x, t)
16 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION,

we see that u(x, t) is odd in x with respect to x 0 for all t. Since u(x, t) is
smooth for t > 0, this implies that u(x 0 , t) = 0. Furthermore, if uo is periodic
with period w, then a similar change of variables techniques establishes that
u(x, t) is periodic in x with period w. Hence if for an integer n
. mrx
u 0 (x ) =smy,

then u 0 is odd with respect to x 0 = 0 and x 0 = L and the corresponding solution

un(x, t) given by (1.ll) satisfies

Un(O, t) = Un(L, t) = 0 fort::: 0.

It follows by superposition that if

N
x = "
Uo () '' mrx
L.,°'nsiny (1.12)
n=l

for scaJars {&n}, then uN(x, t) given by (1.ll) is the unique classical solution
of the initial/boundary value problem

LU = Uxx - Ut = 0 - (1.13)
u(O, t) = u(L, t) = 0
u(x, 0) = ua(x).

Given any smooth function u 0 defined on [O, £] it can be approximated by a

finite sum (1.12) as discussed in Chapters 2 and 4. The corresponding solution
uN(x, t) will be an approximate solution of problem (1.13). It can be shown by
direct integration of (1.ll) that this approximate solution is in fact identical to
that obtained in Chapter 6 with our separation of variables approach. Note that
if ua(O) i= 0 or ua(L) i= 0, then (1.13) does not have a classical solution and the
maximum principle cannot be used to analyze the error u(x, t) - uN(x, t). Now
continuous dependence in a mean square sense must be employed to examine the
error. A simple version of the required arguments is given in Section 6.2 where
error bounds for the separation of variables solution for the one-dimensional heat
equation are considered. Of course, homogeneous boundary conditions are not
realistic, but as we show time and again throughout the text, nonhomogeneous
data can be made homogeneous at the expense of adding a source term to the
heat equation. Thus instead of (1.13) one might have the problem

LU= Uxx - Ut = F(x, t) (1.14)

u(O, t) = u(L, t) = 0
u(x,O) = 0.

We now verify by direct differentiation that if U(x, t, T) is a solution of

LU= Uxx(x,t,T)- U,(x,t,T) = 0

1.4. THE HEAT EQUATION 17

= U(L, t, r) = 0
U(O, t, r)
U(x,r,r) = -F(x,r),
where r is a nonnegative parameter, then

u(x, t) = 1' U(x, t, r)dr (1.15)

solves (1.14). The solution method leading to (1.15) is known as Duhamel's

principle and can be interpreted as the superposition of solutions to the heat
equation when the source is turned on only over a differential time interval dt
centered at r. The solution U(x, t, r) is given by (1.ll) as

U(x, t, r) = - I: s(x - y, t - r)F(y, r)dy.

If F(x, r) can be approximated by a sum of sinusoidal functions of period 2£,

then one can carry out all integrations analytically and obtain an approximate
solution of (1.14). It again is identical with that found in Chapter 6. Note that
the sum of the solutions of (1.13) and (1.14) solves the inhomogeneous heat
equation with nonzero initial condition.
Similar results can be derived for an even initial function u 0 which allows
the treatment of flux data at x = 0 or x = L. But certainly, this text promotes
the view that the approach presented in Chapter 6 provides an easier and more
general method for solving such boundary value problem than Duhamel's prin-
ciple because the integration of (1.ll) for a sinusoidal input and of (1.15) is
replaced by an elementary differential equations approach.
Let us conclude our discussion of parabolic problems with a quick look at
the so-called backward heat equation

£u = b.u + u, = 0.

u(x,O) =u 0 (£).
Suppose we wish to find u(:i', T) for T > 0. If we set r = T - t and w(x, r) =
u(x, T - r), then the problem is equivalent to finding w(x, 0) of the problem

£w =bow - w, = 0
w(x, T) = uo(x).
In a thermal setting this implies that from_ knowledge of the temperature at
some future time T we wish to find the temperature today. Intuition tells us
that if T is large and u 0 is near a steady-state temperature, then all initial
temperatures w(x, 0) will decay to near u 0 . In other words, small changes in u 0
could be consistent with large changes in w(£, 0), suggesting that the problem
is not well posed when the heat equation is integrated backward in time (or the
backward heat equation is integrated forward in time).
18 CHAPTER l. POTENTIAL, HEAT, AND WAVE EQUATION

A well-known example is furnished by the function

u(x, t) = ee'/' 2 sin(x/c), E > 0,

which solves the backward heat equation.
We see that
Jim u(x, 0)
,~o
=0
but
!~ u(7rc/2, t) = oo for any t > 0,

so there is no continuity with respect to the initial condition.

In general, any mathematical model leading to the backward heat equation
which is to be solved forward in time will need to be treated very carefully.
The comments at the end of Example 6.4 provide a further illustration of the
difficulty of discovering the past from the present. Of course, if the backward
heat equation is to be solved backward in time, as in the case of the celebrated
Black-Scholes equation for financial options, then a time reversal will yield the
usual well posed forward problem (see Example 6.9).

1.5 The wave equation

The third dominant equation of mathematical physics is the so-called wave
equation
1
£u =e 6u - ~Utt = 0, iEDCIRM, tE(-00,00). (1.16)

The equation is usually associated with oscillatory phenomena and shows mark-
edly different properties compared to Poisson's and the heat equation. Equation
(1.16) is an example of a hyperbolic equation. Here the (M + 1) x (M + 1) matrix
A has the form
A= c~ !1)
where IM is the M-dimensional i<lentity matrix.
It is easy to show that (1.16) allows wave-like solutions. For example, let f
be an arbitrary twice continuously differentiable function of a scalar variable y.
Let ii be a (Euclidean) unit vector in !RM and define

y = (ii,x) - ct

where (ii, x) is the dot product of n and x. Then differentiation shows that

u(x, t) = f( (n, x) - ct)

is a solution of (1.16). Suppose that c,t > 0, then the set {x : (n,x) -
ct = constant} is a plane in !RM traveling in the direction of ii with speed
J.5. THE WAVE EQUATION 19

c, and f( (ii., x) - ct) describes a wave with constant value on this plane. For
example, if
j(y) = eiY,

then
/((ii., x) - ct) = ei((n,x)-ct]
is known as a plane wave. Similarly, g( (ii., x) +ct) describes a wave traveling in
the direction of -n with speed c.
Our aim is to discuss again what constitutes well posed problems for (1.16).
We begin by exhibiting a solution which is somewhat analogous to the solution
of the one-dimensional heat equation discussed at the end of Section 1.4.
Ifwe set x = (ii.,x), then the solutions f(x-ct) and g(x+ct) of (1.16) solve
the one-dimensional wave equation
1
£u '=' Uxx - ;;2 Utt = 0 (1.17)

which, for example, describes the motion of a vibrating uniform string. Here
u(x, t) is the vertical displacement of the string from its equilibrium position.
We show next that any smooth solution of _(1.17) must be of the form

u(x, t) = f(x - ct)+ g(x +ct),

i.e., the superposition of a right and left traveling wave. This observation follows
if we introduce new variables
E=x-ct
1)=x+ct
and express the wave equation in the new variables. The chain rule shows that

so that by direct integration

u(E, 7J) = f(~) + g(7J) = f(x - ct)+ g(x +ct)

for arbitrary continuously differentiable functions f and g.

For a vibrating string it is natural to impose an initial displacement and
Yelocity of the string so that (1.17) is angmented with the initial conditions

u(x, 0) = uo(x) ( 1.18)

u,(x,O) = u 1 (x)
where u 0 and u 1 are given functions. Equations (1.18) constitute Cauchy data.
Let us suppose first that these functions are smooth and given on ( -oo, oo ).
Then we can construct a solution of (1.17), (1.18). We set

u(x, t) = f(x - ct) + g(x +ct)

20 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

and determine f and g so that u satisfies the initial conditions. Hence we need

u0 (x) = f(x) + g(x)

u 1 (x) = -cf'(x) + cg'(x).
Integration of the last equation leads to

-11"
c xo
ui(s)ds=-f(x)+g(x)+K

where xo is some arbitrary but fixed point in (-oo, oo) and f( = f(xo) - g(xo).
When we solve algebraically for f(x) and g(x), we obtain

f(x) = ~ [uo(x) -
.
~
C
{" u1(s)ds +
lxo
K]

Hence

u(x,t) = -l [uo(x- ct)+ uo(x +ct)]+ -1 [ - 1x-ct u (s)ds + 1x+ct u 1(s)dsl

1
2 2c xo xo

which simplifies to

u(x, t) =
1
2 [uo(x - ct)+ uo(x +ct)]+
1
2C
;·x+ct u (s)ds.
1 (1.19)
x-ct

The expression (1.19) is known as d'Alembert's solution for the initial value
problem of the one-dimensional wave equation. If u 0 is twice continuously dif-
ferentiable and u 1 is once continuonsly differentiable on ( -oo, oo), then the
d' Alembert solution is a classical solution of the initial value problem for all fi-
nite t and x. Moreover, it is unique because u has to be the superposition of two
traveling waves and the d'Alembert construction determines f and g uniquely.
Moreover, if we set

\\ukll = sup \uk(x)\, k = 0, 1,

(-00,00)

then
lu(x, t)i ~ lluoll + tllu1 I
which implies continuous dependence for all t ~ T where T is an arbitrary but
fixed time. Hence the initial value problem (1.17), (l.18) is well posed.
We observe from (1.19) that the value of u(x 0 , t 0 ) at a given point (x 0 , to)
depends only on the initial value u 0 at x 0 - ct 0 and x 0 + ct 0 and on the initial
value u 1 over the interval [xa - eta, xa +eta]. Thus, if
l.5. THE WAVE EQUATION 21

and xo > t, then regardless of the form of the data on the set lxl < e we have

u(x 0 , t) = {o ,
fort<=
c
..!..
2c J
-£
u 1 (s)ds for t > ~
c .
Hence these initial conditions travel with speed c to the point Xo but in general
~(xo, t) will not decay to zero as t ..... oo. This is a peculiarity of the M-
.iimensional wave equation for M = 1 and all even M [5]. If uo and u 1 do
not have the required derivatives but (1.19) remains well defined, then (1.19)
represents a weak solution of (1.17), (1.18). We shall comment on this aspect
when discussing a plucked string in Example 7.1.
As in the case of the heat kernel solution we can exploit symmetry properties
of the initial conditions to solve certain initial/boundary value problems for
the one-dimensional wave equation with d'Alembert's solution. For example,
suppose that u 0 and u 1 are smooth and odd with respect to the point x 0 ; then
the d'Alembert solution is odd with respect to x 0 •
To see this suppose that

u(x,t)= 1x+ct u (s)ds

x-ct
1

and that u 1 is odd with respect to the point x 0 , i.e., u 1 (x 0 +x) = -u 1 (x 0 -x).
Then with y = xo - s and r = Xo + y we obtain

u(xo + x, t) = 1xo+x+ct u, (s)ds 1-x-ct (xo - y)dy

xo+x-ct
= -
-x+ct
u1

= 1-x-ct u,(xo+y)dy= 1xo-x-ct u1(r)dr=-u(xo-x,t).

-x+ct xo-x+ct

Since by hypothesis

uo(xo +x - ct)+ uo(xo + x +ct)= -uo(xo - x +ct) - uo(xo - x.- ct),

we conclude that the d' Alembert solution is odd with respect to the point x 0 and
hence equal to zero at x 0 for all t. It follows that the boundary value problem
I
lu '= Uxx - ;::2" Utt = 0

u(O, t) = u(L, t) = 0
N
u(x, 0) = uo(x) = L &n sin n~x
n=l

N
u,(x, 0) = u1(x) = "'\"""' • mr
~ f3n sin L
n=l
22 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

for constant {&n} and {~n} is solved by the d'Alembert solution (1.19) because
u 0 an:d u 1 a.re defined on ( -oo, oo) and odd with respect to x = 0 and x = L.
As we remarked in Section 1.4, inhomogeneous boundary conditions can
often be made homogeneous at the expense of adding a source term to the
differential equation. This leads to problems of the type
1
LU= Uxx - - Utt = F(x, t)
C2

u(O, t) = u(L, t) = 0
u(x, 0) = u,(x, 0) = 0.
Now a Duhamel superposition principle can be applied. It is straightforward to

l
show that the function
u(x,t) = U(x,t,T)dT

solves our problem whenever U(x, t, T) is the solution of

1
[,U = Uxx - ;::2 Utt = 0

U(O, t, T) = U(L, t, T) = 0
U(x,T,T)=O
U,(x,T,T) = -c2 F(x,T)
where T is a parameter. It follows that if F is of the form

then the problem has the d' Alembert solution

U(x, t, T) = - -
C

2
1 x+c(t-T)

:L-C(l-T)
FN(s, T)ds.

All integrations can be carried out analytically and the resultant solution UN (x, t)
can be shown to be identical to the separation of variables solution found in
Chapter 7 when an arbitrary source term F is approximated by a trigonometric
sumFN.
Let us now turn to the general initial/boundary value problem of the form

l:.u - Utt= F(i, t), i ED, t>0 (1.20)

u(i, 0) = uo(i)
u,(i,0) = u 1 (i)
J.5. THE WAVE EQUATION 23

au + 0<2U = g (-X, t ),
0<1 [}n x E BD, t>0
where D is a given domain in !RM. For convenience we have set c = 1 which can
always be achieved by scaling time. We point out that if D = !RM and we have
a pure initial value problem, then it again is possible to give a formula for u(x, t)
analogous to the d'Alembert solution of the one-dimensional problem (see [5,
Chapter 2]), but for a true initial/boundary value problem the existence of a
solution will generally be based on abstract theory. (We note in this context
=
that if g 0 on [JD, then, again in a very general sense, the problem has a lot
in common with the ordinary differential equation
cf2u
dt 2 - A(t)u = G(t), u(O) = u0 , u'(O) = ui-)

We shall henceforth assume that the existence theory of [5] applies so that we
can concentrate on uniqueness of the solution and on its continuous dependence
on the data of the problem.
Uniqueness and continuous dependence follow from a so-called energy esti-
mate. If u is a smooth solution of (1.20) with g 0 on BD, then =
u,.Cu = Ut/:;.U - UtUtt = u,F(x, t)

so that
l [V' · u, V'u - V'u, · '\i'u - u,u,,] dx = l u,F di.

We now apply the divergence theorem and obtain

_dd { ~ [V'u· V'u+ut] dx- j u,~u dS= - { u,Fdx. (1.21)

tJD 2 hw vn JD
For a smooth solution the boundary data
Bu
°'1 [Jn + 0<2U = 0 with 0<1 + 0<2 =. 1

imply

so that
+ n2)u, Bu -0<2u,u - Bu (Bu) .
(n1
an = °'1
an an t
Since IUtl' I :S ¥- + t;-, we obtain from (1.21) the estimate
(1.22)

where
24 CHAPTER 1. POTENTIAL, HEAT, AND WAVE EQUATION

and
llF(.,t)ll = (L F (x,t)dxf
2
2

The inequality (1.22) can be written as

E'(t) = E(t) + llF(-, t)ll2 - g(t)

2
where g is some unknown nonnegative function. This differential equation has
the analytic solution

E(t) = E(O)e' + l et-r [llF(·;r)ll2 - g(r)] dr

from which we obtain the so-called Gronwall inequality

E(t) S E(O)e' + l et-r [llF(·;r)ll2] dr (1.23)

where

E(O) = ~ [L \luo · \luo + ui] di!+ ~ iD [a2u~ + a ~~) 1 (

2
] ds

is known from the initial data.

We have the following two immediate consequences of (1.23).

Theorem 1. 7 The solution of the initial/boundary value problem (1.20) is uniqw

Proof. The difference w of two solutions satisfies ((20) with

F = g = uo = u1 = 0.
This implies that E(O) = 0 so that (1.23) assures that E(t) = 0 for all t. Then
by Schwarz's inequality (see Theorem 2.4)

lw(x, t)l2 = Ila' 2

w 5 (x, .s)dsl St fa' w;(x, s)ds,
from which follows that

llw(., t)ll 2 = L w(x, t) 2 dx S 2t l E(s)ds = 0.

Hence w = 0 in the mean square sense for all t which implies that a classical
solution is identically zero.
Theorem l. 7 assures that the separation of variables solution constructed in
Section 7.1 is the only solution of the approximating problem. Similar argu-
ments are used in Section 7.2 to show for a vibrating string that this approxi-
mate solution converges to the analytic solution of the original problem as the
approximations of the data are refined.
Chapter 2

Basic Approximation
Theory

This chapter will review the abstract ideas of approximation that will be used
in the sequel. Let X be a linear space (sometimes called a vector space) over the
field S of real or complex numbers. The elements of X ~re called vectors and
those of S are called scalars. The vector spaces appearing in this book are Fl,,
and Cn with elements x = (x1, ... , Xn), jj, etc., or spaces of real- or complex-
valued functions/, g, etc. d-efined on a real interval or, more generally, a subset
of Euclidean n-space. In all spaces 0 denotes the zero vector. The scalars of S
are denoted by a., /3, or a, b, etc.
We now recall a few definitions from linear algebra which are central in our
discussion of the approximation of functions.

Definition Given a collection C = { ip 1, ip 2, ... , 'l'N} of vectors in a linear space

X, then
M='span{ip1, ... ,'f'N}
is the set of all linear combinations {a.1<p1 + · · · + O'.N'f'N} of the elements of C.

Note that M is a subspace of X because it is closed under vector addition

and scalar multiplication.

Definition A collection C {<p1, '1'2, ... , <p N} of vectors in a linear space is

linearly independent if
N
L O'.j'f'j =0 only if a. 1 = a. 2 = · · · = a.N = 0.
j=l

A sequence of vectors {'Pn} is linearly independent if any finite collection C

drawn from the sequence is linearly independent.
The definition implies that in a finite set of linearly independent vectors no
one element can be expressed as a linear combination of the remaining elements.

25
26 CHAPTER 2. BASIC APPROXIMATION THEORY

Definition Let C = { cp 1 , ... , 'PN} be a collection of N linearly independent

vectors of X with the property that every element of X is a linear combination
of the {'PJ} (i.e., span{cp 1 , ... ,cpN} = X). Then the set {cpJ} is a basis of X,
and X has dimension N.

The theorems of linear algebra assure that every basis of an N-dimensional

vector space consists of N elements, but not every vector space has a finite-
dimensional basis. Spaces containing sets of countably many linearly indepen-
dent vectors are called infinite dimensional.

2.1 Norms and inner products

Basic t 0 the idea of approximation is the concept of a distance between vectors
f and an approximation g, or the "size" of the vector f - g. This leads us to
the the idea of a norm for assessing the size of vectors.

Definition Let x· be a vector space. A norm on X is a real-valued function

F : X --+ R such that for every f, g E X and every scalar a E S, it is true that
i) F(f) # o if f # O;
ii) F(af) =Jal F(f);
iii) F(f + g) :0: F(f) + F(g).
Proposition 2.1 F(f) 2: 0.

Proof. From ii) with a= 0, we know that F(O) = 0. Thus

0 = F(f + (- f)) :0: F(f) + F(- f) = 2F(f).

In other words, F(f) 2: 0.

The value of the norm function F is almost always written as llfll · A vector
space X together with a norm on X is called a normed linear space. The
inequality iii) is commonly called the triangle inequality.
The concept of a norm is an abstraction of the usual length of a vector in
Euclidean three-space and provides a measure of the "distance" llf - 911 between
two vectors f, g E X. Thus in the space R3 of triples x = (x1, x2, X3) of real
numbers with the customary definitions of addition and scalar multiplication,
the function
11£11 = Jxi + x~ + x§
is a norm (see Example 2.6a). The distance induced by-this norm is the everyday
Euclidean distance.
In a linear space X, our approximation problem will be to find a member
fM of a given subspace JV[ C X that is closest to a given vector fin the sense
that II/ - /Mil :0: II/ - mil for all m E M. We shall be concerned only with
finite-dimensional subspaces M.
2.1. NORMS AND INNER PRODUCTS 27

We know from elementary geometry that in ordinary Euclidean three-space,

the closest point on a line or a plane containing the origin to a given point x in
the space is the perpendicular projection of x onto the line or plane. This is the
idea that we shall abstract to our general settin·g. For this, we need to extend
the notion of the "dot," or scalar, product.

Definition An inner product on a vector space X is a scalar-valued function

G : X x X --> S on ordered pairs of elements of X such that

i) G(f, f) ;:>- 0 and G(f, f) = 0 if and only if f = O;

ii) G(f, g) = G(g, f) [,6 denotes the complex conjugate of ,8];
iii) G(af,g) = aG(J,g); and

iv) G(f + g, h) = G(f, h) + G(g, h).

We shall usually denote G(f, g) by (f, g). A vector space together with .an
inner product defined on it is called an inner product space.
The next proposition is easy to verify.

Proposition 2.2 An inner product has the following properties:

i) (f, ag) =Ci. (f, g),

ii) (f,g + h) = (f,g) + (f, h),
iii) ( 0, g) = 0.
Definition Two vectors f and g in an inner product space are said to be
orthogonal if (f, g) = 0.

Example 2.3 a) In real Euclidean n-space Rn, the usual dot product

(x,fj) =x·iJ= L,xjyj,

j=l

where x = (x1, x2, ... , Xn) and iJ = (Y1, y2, ... , Yn) is an inner product.
b) In R2, for x = (x 1,x2) and iJ = (y 1,yz) define (x,fj) by (x,fj) =Ax· i],
where A is the matrix
A=(2 1)
. 1 2
and i1 · iJ is the usual dot product of i1 and ii. Then (x, fj) is an inner product.
First consider

(x,x) = Ax·x= (2x1 +x2,x1 +2x2) · (x1,x2) = 2 (xf +x2x1 +x~)

= 2 [(xi+ ~2 )2 + ~x~].
28 CHAPTER 2. BASIC APPROXIMATION THEORY

It is clear that (x, x) :'.'.: 0 and (x, x) = 0 if and only if x = (0, 0).
To see that (x, flJ = (y, x), simply compute both inner products. The re-
maining two properties are evident.
c) On the space of all continuous functions (real- or complex-valued) defined
on the reals having period 2£, it is easy to verify that
L

(f,g) = J f(t)g(t)dt
-L

is an inner product.

Theorem 2.4 In an inner product space, l(f,g)I '.::: j(f,J)~.

Proof. If(!, g) = 0, then the proposition is obviously true, so assume(!, g) of 0.

Let a be a complex number. Then

(f + ag, f + ag) :'.'.: 0.

Now
(f + ag, f + ag) = (!, f) +Ci(!, g) + a(g, f) + lnl 2 (g, g).
Next, let a= t(f,g), where tis any real number. Then-·

(f + ag, f + ag) = (!, f) +a(!, g) + o:(g,f) + (g, g)

= -u, !) + 2t l(f, g)l 2 + t 2 l(f,g)l 2 (g, g) ::=:: 0.

This expression is quadratic in t and so the fact that it is never negative means
that
4 l(f, g)l 4 - 4 l(f,g)l 2 (!, !) (g,g) ::; 0.
In other words,
l(f,g)l 2 ::; (!,!) (g,g),
which completes the proof.
The inequality
l(f,g)I < J(f,f) ~
is known as Schwarz's inequality.

Corollary 2.5 Suppose X is an inner product space. Then the function F

defined by F(f) = j(f,l) is a norm on X.

Proof. The proofs that F(f) :'.'.: 0 and F( a!) = lnl F(f) are simple and omitted.
We prove the triangle inequality.

F(f + g) 2 = (f +g,f +g) = (f,f) + (f,g) + (g,f) + (g,g)

= F(f) 2 + 2Re((f,g)) + F(g) 2 ::; F(f) 2 + 2 l(f, g)I + F(g) 2
'.::: F(/) 2 + 2F(f)F(g) + F(g) 2 = (F(f) + F(g)) 2 .
Another Random Scribd Document
with Unrelated Content
took to their heels and left them far behind. They turned out of the
way once or twice for water, but kept the same general course, and
on the afternoon of the third day brought their pursuers within sight
of the landmark the trapper had described to them.
 CHAPTER VII.
THE WILD MAN OF THE WOODS.

D ICK LEWIS had warned Eugene that constant vigilance was to

be the price of his success, and among other things had told him
that, before his expedition was ended, he might discover that horses
were the most obstinate and contrary things in the world. The drove
could probably be driven very easily, he said, until they reached the
entrance to the gully; and then, if they found that they were
expected to enter it, the chances were that they would take to their
heels, leaving their pursuers nothing to do but to make the best of
their way back to the Fort. Eugene, remembering the warning, took
measures to prevent so abrupt an ending to their hunt. He sent
Archie out on one flank and Fred on the other, with instructions to
head off the horses, should they attempt to escape, while he himself
followed after them as before. But it happened that no very serious
difficulty was experienced, and everything would have worked
smoothly but for an unlooked-for incident that occurred, and which
they could not have guarded against, even had they been expecting
it.
The horses had been without water all day, and knowing
probably that there was none to be found nearer than the valley
toward which they were heading, they kept straight on at a rapid
walk, turning neither to the right nor left, and in an hour more were
fairly between the two high hills which marked the entrance to the
gully. These hills were two or three miles apart. A few hundred yards
farther on the gully proper began, and was not more than a quarter
of a mile wide. Once in there, they were safe. Then the work would
begin in earnest. The spring which watered the valley was located at
the other end of the gully, twenty miles away in the mountains, and
Eugene, who was the acknowledged leader of the expedition, had
told his companions that the drove must not be allowed to stop
there to drink. The wild steeds had travelled at least twenty-five
miles that day, with no nourishment except the few mouthfuls of
grass they had been able to snatch as they passed along, and some
of them acted as if they would like to stop and rest. If the boys
camped that night as usual, the horses would doubtless pass on to
the spring, refresh themselves with a good drink, eat their fill of the
luxuriant grass growing about, take a good rest, and by the time
their pursuers reached the valley, which would be on the afternoon
of the following day, the wild nags would be ready for a long and
hard test of endurance. In order to prevent this the council of war,
which Eugene called, decided that no camp should be made that
night. They would follow the drove through the gully, drive it on past
the spring, and while two of their number made the camp and took
rest and refreshment, the other would keep the horses in motion.
Archie thought their plans had been laid with considerable skill.
He was on the right flank of the drove, which the boys called the
post of honor, for there was some responsibility connected with it. If
the horses became contrary and declined to enter the gully, the
probabilities were that they would attempt to run by it instead of
turning back, and in that event Archie was expected, if he could not
head them, to capture one of the drove; for the boys were resolved
that they would not go back to the Fort without catching something,
and in order to make the agreement more binding, they had shaken
hands on it.
Archie was prepared for anything that might happen. He had put
himself in the lightest possible running order, by giving his bundle
and weapons to his companions, and discarding his jacket and hat.
He had tied a handkerchief over his head to keep his hair out of his
eyes, and rode along with his lariat in his hand and his gaze
fastened upon the leader of the drove. In his eagerness to approach
as close to him as he could, he kept his horse in a fast walk, and in
this way gradually drew ahead of some of the laggards of the drove,
who, fearing that they were about to be cut off from their
companions, broke into a gallop. This set the whole drove in motion.
They went ahead at the top of their speed, holding straight for the
gully, and Archie, believing that he should have no use for his lasso
that day, was about to tie it fast to the horn of his saddle, when he
saw the leading horses suddenly swerve from their course, and then
stand motionless, gazing at some object before them that had
attracted their attention.
Archie looked, and saw something crouching behind a tree on
one side of the gully. He put his hand to his side where he usually
carried his field-glasses, but he had given them with the rest of his
accoutrements into the keeping of his friends. Then he shaded his
eyes with his hand and looked again, and after a few seconds’ close
inspection made out that the object was a human being, and not a
wild animal, as he had at first supposed. And a most forlorn looking
specimen it was too, unless his eyes greatly deceived him. It wore
the most dilapidated suit of clothes that Archie had ever seen; its
feet looked like two small barrels, being wrapped in blankets or
hides, no doubt, and from under something on its head, which might
once have been called a hat, the long unkempt hair was sticking in
every direction. Archie did not get a good glimpse of its face, and
indeed he did not think to try. His mind was fully occupied with the
horses. Forgetting every thing except that the frightened animals
were on the point of taking to their heels, and thereby causing him
and his companions the loss of three days’ hard work, Archie called
to the man, for such he believed the apparition to be.
“Say, you!” he shouted. “Couldn’t you step back out of sight a
moment, long enough to allow these horses to pass you?”
The figure complied with a readiness that was surprising. As if
frightened by the sound of his voice, it disappeared on the instant,
and made a most precipitate retreat up the hillside, if one might
judge by the shower of stones that came rattling down among the
bushes. The noise frightened the horses worse than ever. The
leaders turned back upon those behind them, and for a few seconds
they were mixed up in great confusion, some pressing one way and
some the other. Archie believing that the hunt was virtually over, and
that he and his friends were to have nothing but their trouble for
their pains, determined to make one bold stroke to save their
fortunes. In response to a touch from the spurs his horse sprang
forward, the wild steeds parting right and left before him, and with a
few rapid bounds carried his rider into the very midst of the drove.
Fred and Eugene had stopped at the first sign of commotion
among the wild horses, not knowing how to account for it and
utterly at a loss what to do. Being willing to leave the management
of affairs entirely in Archie’s hands, they sat in their saddles and
watched his motions with the greatest interest. When they saw him
dashing into the drove, and his lasso flying from his hand, uncoiling
itself as it went, their joy knew no bounds.
“The hunt is over,” shouted Eugene, in great glee. “Dick says he
doesn’t miss his aim once in a hundred times.”
“Then this throw must be the unlucky one,” said Fred, in reply,
“for he has certainly missed.”
When Fred spoke Archie’s horse was standing motionless, and
his rider was hauling in his lasso hand over hand. If he had failed in
one thing he was successful in another, for his sudden charge had
assisted the horses to decide a point they were unable to decide for
themselves. It showed them that the only safe way of retreat was
toward the gully, and into it they went with all the speed of which
they were capable, every one of them giving a wide berth to the tree
behind which had stood the object that occasioned their alarm. Fred
and Eugene, although greatly disappointed, had seen a sight they
would not have missed for a good deal. They had seen a lasso
thrown at a wild horse, and that was something to put into the next
letter they wrote to their friend Wilson. When they came up with
Archie they found him bent half double, holding his horse by the
bridle and peeping up under the bushes which covered the side of
the gully.
“Well, you didn’t catch him, did you?” exclaimed Eugene.
 “No, but I did my best, and if my lariat had been a little longer,
I’d have had a different story to tell, for I made as straight a throw
for his head as I ever made in my life. But I am encouraged after all.
I know that my horse has had good training and can be depended
on. When the lasso left my hand he stopped as if he had been shot.
If I can only get another chance like that the bay is ours.”
“Has one of them gone up there?” asked Fred, seeing that
Archie was still peering under the bushes.
“What was it that made such a commotion among them?”
inquired Eugene.
“The horses have all gone on down the gully, but the thing that
frightened them went up here,” replied Archie.
“What was it?”
“The Wild Man of the Woods.”
“The what?”
“Well, I wish you had seen him—for I don’t know what else to
call him. He stood here behind this tree, and the sight of him turned
the horses and frightened them so badly that I was afraid they were
going to run away from us.”
“Was it a man?”
“I judge so. He certainly was not an animal, for he was dressed;
but he acted like an animal, for when I spoke to him and asked him
to get back out of sight, he went up the hill on all-fours like a
streak.”
“There’s something been up there, sure enough,” said Eugene.
“The fresh dirt on those stones shows that they have but recently
been dislodged from their bed.”
“What could it have been?” asked Fred, greatly astonished.
“Ask me something hard,” replied Archie. “If there had been a
menagerie along here lately, I should say that one of the gorillas had
stolen a suit of the keeper’s clothes and decamped. I have heard of
wild men, but I never saw one before, and I have no desire to make
his acquaintance. Whoever he is he had better visit some trader
pretty soon, or go to hunting furs, for his clothes will not last him
much longer.”
“Well, what’s to be done about it?”
“Nothing,” answered Eugene. “We’re hunting wild horses, and I
suggest that we leave the wild men to take care of themselves.”
“That’s my idea,” said Archie, springing into his saddle; “so let’s
jog along and keep the drove moving. The sun says it is pretty
nearly supper time; so I think I’ll take a bite. But, fellows, I should
really like to know what that thing was.”
The others said they would like to know too. There was much
speculation indulged in, and Archie was required to describe the
object and its actions again and again; and after each description his
companions shook their heads, as if to say that the matter was too
deep for them, and applied themselves with renewed energy to the
bread and meat and flasks of cold tea which they drew from the
pockets of their shooting-jackets. They could not explain it after half
an hour’s debate, and they finally came to the conclusion that they
did not know anything about it and never would. After that nothing
was said on the subject, although they did not cease to think about
it, and each boy smiled at his neighbor when he saw him looking
toward the top of the cliffs which hung over the gully, for he knew
that he was looking to see if there were any signs of the wild man.
By the time their supper was over (they had eaten it as they
rode along), they again came up with the wild horses, which, having
recovered from their fright, were moving steadily on down the gully,
stopping only to take occasional bites at the grass which grew in
little bunches at the foot of the cliffs. It was almost dark now, and
the boys, as they twisted about in their hard saddles, trying to find a
comfortable position for their aching limbs, thought of the twenty
long miles yet before them, and wished for a roaring fire and a soft,
warm blanket. And these twenty miles were to be ridden during the
night, through a gorge with which they were entirely unacquainted.
There might be a thousand perils and obstacles in their way. Some
savage beast, like Old Davy, might have an ambush in there
somewhere, or their path might lead along the edge of some deep
long chasm, where a single slip on the part of their horses would
send them to destruction. Archie was somewhat hardened to such
things, but his companions were not, and when it is known that
Eugene was the one who planned and suggested this night ride, it
will be seen that he was resolved to accomplish something.
In less than twenty minutes after the sun went down a deep
gloom began to settle over the gorge, and in twenty minutes more it
was so dark in there that the boys could not discern the nearest
objects. The wild horses were close before them, following a well-
beaten path in Indian file, but the boys could not see them. They
were obliged to trust entirely to the animals they rode, and these, in
turn, trusted to the leader of the drove. The boys beguiled the weary
hours with song and story, awaking a thousand echoes in the gorge
and no doubt startling more than one wild beast which was going his
nightly rounds. But, what was very surprising, the wild horses, after
the first few minutes, seemed to pay no attention to their voices.
Only once were they frightened, and that was when a violent
rustling in a thicket of bushes at the foot of the cliff, drew a couple
of shots from Eugene’s rifle. Then they galloped on in advance, but
were again overtaken at the end of half an hour, and their pursuers
kept close company with them until daylight—so close, in fact, that
Eugene’s nag received one or two admonitory kicks from the last
horse in the line.
The night passed at length, to the boys’ great relief, and the
morning sun began to gild the summits of the surrounding
mountains. His rays gradually found their way into the gorge which
just here was as straight as if it had been cut out for a railroad, and
about a mile in advance of them the boys saw the valley into which
it opened. The horses saw it too, and gradually quickened their pace
to a rapid walk, then to a trot, and finally to a gallop; but their
pursuers kept close behind, Archie leading the way. Their object now
was to drive them beyond the spring, which Dick told them they
would find at the end of the gorge; and Archie, thinking that he
might possibly have another opportunity to try his lasso on the bay,
handed his bundle and weapons to his friends, and pressed close
upon the rear of the drove. Furthermore, he anticipated the very
thing that happened—a momentary confusion about the spring,
which Eugene, from the information the trapper had given him, was
able to describe so accurately that Archie knew just where to look to
find it, and a chance to use his lasso, which, if quickly improved,
would save them a day or two of hard work. Archie noticed that the
horses were not nearly as wild as they had appeared to be when
they first found them. They had grown more and more accustomed
to the presence of their pursuers as the days went by, and now,
having passed the night in their immediate company, they seemed to
pay no more attention to them than they did to the members of the
drove. Archie could not decide whether this indifference was the
result of fatigue or increasing confidence; but whatever it was, he
was ready to take advantage of it. In order to see just how far he
could go, he turned his horse out on one side, and to his great
gratification and the no small surprise of his companions, galloped
along side by side with the rear horses in the drove, which, if they
felt any displeasure at this familiarity, showed it only by laying back
their ears and biting at his horse.
Archie gradually urged his nag forward, and before the valley
was reached he had passed more than half the drove, and was
riding within fifty yards of the bay which looked over his shoulder at
him occasionally, but did not increase his speed or show any alarm.
He was heading straight for the spring. Archie began to tremble with
excitement. One after the other the members of the drove were
passed, and at length only three horses were running between
Archie and the leader. Just then the bay turned quickly out of the
path, and in a moment more was standing knee-deep in the spring
which bubbled out from under the cliff.
 Archie Captures the Wild Horse.

And now happened the event for which Archie was looking and
for which he was prepared. The other horses crowded in upon the
bay—his own would have followed had he not restrained him—and
in less time than it takes to tell it the spring was full of rearing,
kicking, biting animals. But they were not permitted to remain there
long. The leader began to assert his authority. Raising his head with
an angry snort, he sent his heels right and left to such good purpose
that he quickly cleared a space about him. The instant that his head
was elevated was enough for Archie. The lariat left his hand, true to
its aim this time, and settling down over the bay’s neck was quickly
drawn tight. Finding himself fast, the horse started to run, but the
other end of the lasso was securely fastened to Archie’s saddle, and
his own horse bracing himself to meet the shock, the captured steed
was thrown flat on his side. The others snorted with terror and took
to their heels in short order.
When Fred and Eugene arrived upon the ground they saw
something exciting.
 CHAPTER VIII.
THE EMIGRANT TRAIN.

“H URRAH for you, Archie!” shouted Eugene, as he and Fred

came galloping up. “You’ve made sure work this time, haven’t
you?”
“I told you that if I got another fair chance at him he was mine,”
replied Archie. “You had a suspicion that I was shooting with a long
bow, last night, when I told you that I had seen a horse help his
master catch cattle, didn’t you, Fred?”
“No,” returned Featherweight, quickly. “I only thought it
something wonderful.”
“I told you I had seen a horse hold a steer down while his owner
butchered him, didn’t I?”
“Yes,” answered Fred.
“Well, now you will see my horse hold this fellow down while I
put a bit in his mouth. Look out, there,” he added, as the boys,
having dismounted, ran up to watch his movements; “don’t come
too near his heels, for he can out-kick a mule. Keep a good hold of
your own horses too, or they’ll go off to join the drove.”
Archie’s nag, as his owner afterward declared, had shown
himself to be a “perfect trump.” He seemed to know what the boy
intended to do as well as the boy knew it himself. When the lariat
settled down over the bay’s head and the latter was about to run off,
he planted his fore-feet firmly on the ground to stop him, and the
bay was thrown on his side, as we have described. Archie’s horse,
without waiting for the word from his rider, quickly backed up until
the lariat was drawn tight, and thus the bay was held as securely as
if he had been tied to the ground.
Archie, finding that his horse could be depended on, quickly
dismounted, slipped the bridle off his head, and when Fred and
Eugene came up he was kneeling on the bay’s neck, trying to force
the bits into his mouth.
“You’re not going to ride him now, are you?” exclaimed Fred,
amazed at Archie’s apparent recklessness.
“Certainly I am,” was the reply. “He’s got to be backed some
time, and it might as well be now as to-morrow.”
“How are you going to get a saddle on him? You can’t put the
girth around him while he is lying down.”
“I don’t want any saddle!”
“Why, he’ll throw you and break your neck!”
“If he does it shan’t cost him anything,” answered Archie, as he
buckled the throat-latch about the bay’s neck. “Now, Fred, slack up
on that lariat, please, and give us room according to our size and
importance. We shall need plenty of space to spread ourselves in!”
Fred, in compliance with the request, unfastened the rawhide
rope from Archie’s saddle, and then he and Eugene, still holding fast
to their horses and leading Archie’s, retreated quickly to the foot of
the cliffs. As soon as Archie saw that they were out of the way, he
removed his weight from the bay’s neck, and in an instant the
captured horse was on his feet. How Archie managed to get upon
his back as he was rising was a great mystery to the two boys who
were looking on—it was done so quickly. But he was there, and he
had secured a firm seat, too.
The bay was no sooner on his feet than he began to make the
most desperate attempts to shake off his rider, but all to no purpose.
When he reared, Archie pounded him between the ears with his
clenched hand; and when he kicked, he used his spurs with such
hearty good-will that the horse was glad to desist. During all the
terrific struggle Archie was as cool as a cucumber and was never
once moved from his seat. Failing in his efforts to throw his rider, the
bay tried to run, and this Archie not only permitted, but encouraged.
There were two well-beaten paths at this point, one running to the
right and the other to the left around the valley, and when the bay
showed a desire to try his speed his rider turned him into one of
these paths and touched him gently with the lariat, which he had
managed to gather up, and now held in a coil in his right hand. The
horse responded on the instant, and with a few rapid bounds carried
Archie out of sight behind the trees. The boys stood listening to the
sound of his hoofs for a moment, and then turned and looked at
each other.
“That is something I never saw done before,” said Eugene, at
length.
“And if anybody had told me that it could be done, I wouldn’t
have believed him,” added Fred. “He will probably go on around the
valley, and as that is a good ten-mile ride, we shall have plenty of
time to make the camp and get something to eat before he returns.”
The boys watered the horses at the little brook that ran from the
spring, staked out their own animals (having no lasso with which to
picket Archie’s horse they hobbled him with a halter so that he could
not stray far away) and then went to work to prepare breakfast.
From one of the bundles Fred brought to light a small sheet-iron
camp-kettle, a coffee-pot and a hatchet. While he was undoing
another, which contained the small supply of bacon and hard-tack
they had remaining, Eugene picked up the hatchet, and in a few
minutes had split up dry wood enough to start a roaring fire. The
camp-kettle was filled with water from the spring, several slices of
bacon placed around the fire on spits to roast, and then the two
boys threw themselves on their blankets to take a few minutes’ rest
after their twenty-four hours in the saddle, and to await Archie’s
return.
He came much sooner than they expected, and they saw at a
glance that that ten-mile gallop had taken considerable of the wild
spirit out of the bay. He looked as if he had been driven through a
stream of water; his breast was flecked with foam, and, although he
still kept his ears laid back close to his head and showed a good deal
of the white of his eye, he did not kick and plunge as he had done at
starting, and seemed quite willing to stop when his rider gave him
the word.
“Well?” said both the boys at once.
There was a good deal of meaning in that little word, and Archie
rightly interpreted it. Fred and Eugene were anxious to know if their
new horse was swift enough to beat the black.
“I don’t know,” replied Archie, drawing his shirt-sleeve across his
forehead; “I am not yet prepared to say. The horse was tired when
he got here, and as the path around the valley is rocky and full of
logs and brush, he didn’t have a chance to show himself. But he
moves as if he were set on springs and clears a good stretch of
country at a jump, and there is no telling what he may do after he
has had a good rest. Want to try him, either of you?”
“Well, n—no,” replied Fred, with so comical an expression that
Archie and Eugene laughed outright. “I believe I’ve ridden far
enough for one day.”
“Then I’ll walk him up and down till he gets cooled off a little,”
said Archie. “After that we’ll tie him to a tree and blanket him. In the
meantime, if you don’t know what else to do, you might cook some
breakfast.”
“We’ve cooked one breakfast and eaten it,” said Eugene. “We
were too hungry to wait for you.”
“We’ll have to send somebody out on a foraging expedition
pretty soon,” said Fred, holding up all that was left of the bacon.
“To-morrow will see the last of our provisions, unless we put
ourselves on short rations.”
Fred set about his cooking operations again, while Eugene
reclined on his blanket and watched Archie as he rode the bay slowly
back and forth. Having once been thoroughly broken to saddle, the
animal was still “bridle-wise,” but, although he understood every
command of his rider, he did not like to yield obedience. He
preferred freedom to slavery and needed constant watching, for he
was ready to take advantage of the smallest chance that was given
him for escape. He was a very handsome, stylish animal, and it was
no wonder that Colonel Gaylord did not like to give him up. When
the bacon and coffee were ready, Archie hitched his horse in a
thicket of evergreens, where he was effectually protected from the
cold wind that was blowing, threw a blanket over him and went to
breakfast.
The boys spent one day and night in their camp in the valley,
during which the bay received the best of care and fully recovered
from his fatigue, although he did not appear to become reconciled to
captivity. He was a vicious brute, and so quick with his heels that it
was a matter of some difficulty to handle him, and Archie was the
only one who cared to attempt it. The wild horses seemed to miss
him. They frequently came about the camp and called to him, and at
such times he struggled so desperately to escape that the boys were
alarmed lest he should injure himself.
Early on the morning following their arrival in the valley, they
packed their bundles and took to their saddles for the return
journey, Eugene leading Archie’s old horse, and Archie himself riding
the bay. They now had a good view of the gorge, and it was so full
of obstructions—tangled thickets, fallen trees and boulders that had
tumbled from the cliffs above—and the path was so narrow and
winding, that it was a mystery to them how the horses could have
found their way through there in the dark. They stopped once for a
short rest and lunch, and about three o’clock in the afternoon
arrived within sight of the prairie. They thought of the wild man, but
saw no signs of him.
The way was now clear, and being impatient to reach the Fort
and exhibit their prize, they put their horses into a lope, and by the
time the sun began to dip behind the mountains they had made
fifteen miles farther on their journey. They followed a course a few
miles to the eastward of the route they had first travelled, in order to
find water. They knew just where to look for it, for the trapper had
been very explicit in his instructions, and Eugene had communicated
them to his companions while they were yet fresh in his memory, so
that there might be no chance for any mistakes. The willows, which
pointed out the position of the stream, came in sight at last, and
then the boys halted and held a consultation. A thin cloud of smoke
slowly rising above the tops of the trees showed them that their
camping-ground was already occupied.
“Ask me something hard,” said Archie, when his companions,
after looking at the smoke through their field-glasses, turned to him
as if to inquire what he thought about it. “I am sure I don’t know
who they are.”
“What shall we do?” asked Fred. “I don’t like the idea of going
into a stranger’s camp, for I have not forgotten Frank’s experience
with Black Bill and his party.”
“But Dick assured me that we had nothing to fear from any one
we might meet,” said Eugene. “He told me, too, that if we got into
any difficulty he and old Bob would be around.”
“Perhaps Dick and Bob are there now,” said Archie.
“That’s so. We can go down and see, and if they are strangers
we need not intrude upon them. We can go a mile or two farther up
the stream.”
The boys rode down the swell, and in a few minutes reached the
willows that lined both banks of the stream. No sooner had they
entered them than a chorus of angry yelps and growls arose in
advance, and three or four gaunt, savage-looking hounds came
toward them with open mouths.
“Begone, you brutes!” shouted a voice. “Begone, I say! Come
on, strangers, whoever you be. They won’t pester you.”
 The boys kept on, and when they reached the other side of the
willows, came in sight of the camp. Two covered wagons were
drawn up, one on each side of the fire, and a span of small, shaggy
mules and a yoke of very poor oxen were grazing close by on the
prairie. About the fire were gathered a party of men, women and
children, half a score in number, all but two of whom the boys at
once put down as emigrants. Three of these emigrants were men—
one an old gray-headed patriarch, and the others sturdy young
fellows, one of whom must have been some relation to the old man,
if there was any faith to be put in the resemblance they bore to each
other.
The two who were not emigrants reminded the boys of that
brace of worthies who had ridden into their camp on the night
Walter was spirited away—Parks and Reed—who were now safe in
irons in the Fort. The dress and accoutrements of these men
proclaimed them to be hunters, but their faces told a different story.
They did not seem at all pleased to see the boys, but the old man
welcomed him cordially.
“How do, strangers?” said he. “Alight and hitch.”
“We’re obliged to you,” said Archie, “but we have no wish to
intrude. We thought, when we saw the smoke of your fire, that
perhaps we should find some of our friends here. We’ll make our
camp a little farther up the stream.”
“No intrusion at all,” said the old man. “You are almost the only
white folks we’ve seen for weeks till we fell in with our friends here
——” nodding his head toward the two hunters—“and we’d like to
talk to you.”
Archie hesitated. If the old man wanted him and his companions
to camp with him, it was plain that the hunters did not, and he was
well enough acquainted with men of their class to know that it was
sometimes a dangerous proceeding to act in opposition to their
wishes. But then these men did not belong to the train. Archie
judged by what the old man said that he and his company had fallen
in with them accidentally during their journey across the prairie, and
if that was the case, they had nothing to say.
“Better get down,” urged the emigrant. “We’ve got grub enough
to feed you. We’re on our way to Fort Bolton; where might you be a
travellin’ to?”
“We’re bound for the same place.”
“Then camp with us to-night and to-morrow we’ll travel in
company.”
The promise of a good supper was a very tempting one to the
hungry boys, whose commissary would have been exhausted long
ago if they had given their appetites full swing, and they knew that if
they accepted the emigrant’s invitation their wants would be
abundantly supplied. There was a coffee-pot on the coals, something
that looked like biscuits in a pan beside the fire, and from one of the
trees hung a joint of fresh meat, from which had been cut a
sufficient number of steaks to fill a large frying-pan. The fact that
the tin plates and cups that were scattered around were not as clean
as some they had seen, and that the emigrant’s wife went about her
domestic duties with a pipe in her mouth, did not take the sharp
edge off their appetites, as it might have done a few months before.
So they decided to remain. The old man took their bundles,
accompanied them when they went out to picket their horses, asked
them more questions than he would give them time to answer, and
finally went back to camp to make sure that his wife had obeyed
orders and added more steaks to those already in the pan.
Fred and Eugene, having picketed their own horses, went to
assist Archie, the bay having suddenly taken it into his head that he
would not allow himself to be staked out. The boys had a lively
battle with him, and while it was going on, one of the hunters came
out.
“Look a yer! who be you an’ what brought you yer?” he
demanded.
 Archie, who had quite as much on his hands as he could attend
to, was not in just the right mood to answer such questions,
especially when propounded, as this one was, with all the insolence
the man could throw into his tones. He paid no attention to it until
the man said, pointing his finger at him:
“You heard me, I reckon! What’s your name, an’ what brought
you yer?”
“There, he’s safe enough,” said Fred, when he had driven down
the iron picket-pin with the hatchet.
“We’ll not trust to that alone,” said Archie. “We’ll hobble him,
too. Now, my friend, in reply to your question I have to say that I
left all my cards at Fr’isco; but when I return there, as I expect to do
in the course of a month or so, I will send you one, if you will be
kind enough to give me your address.”
The man stared, and then looked down at the ground in a brown
study. He could not make head or tail of what Archie said; but when
Eugene spoke, he began to have a vague idea that the boys were
making game of him.
“We’ll send it by special train, too,” said Eugene, who was highly
indignant over the man’s insulting manner.
“I don’t want no sass,” said the latter. “I axed you a fair
question.”
“But you didn’t ask it in a fair way,” said Archie. “Now, then,
stand clear of his heels and I will let him go,” he added, as Eugene
buckled the halter-strap around the bay’s leg below his knee. “I think
we shall find him here when we want him.”
The horse was doubly secured now. He was fastened by a lariat
to an iron pin, which was driven as deeply into the ground as
repeated blows from the hatchet could send it, and his head was
tied down to his feet, so that even if he succeeded in pulling the pin
out he could not run away. The boys had found that these
precautions were absolutely necessary. They had been so long about
this work that the fresh steaks had had plenty of time to cook, and
when they went back to camp the host told them that supper was
ready, and backed up his welcome announcement by grasping the
frying-pan by its long handle and passing it to each of his guests in
succession, who helped themselves by using their hunting-knives in
lieu of forks. The host, not being so particular, fished his share out
with his fingers. The emigrant’s wife passed them each a biscuit with
a hand which looked as though it had never seen water; a tin cup,
with which one of the youngsters had been amusing himself, was
filled with coffee and handed over with the remark: “I reckon you
three haint afeared of one another;” and everything being thus
satisfactorily arranged and everybody provided for, the host found a
seat and applied himself to the task of learning a full history of
everything our heroes had done since they were born. After asking a
few questions he suddenly paused, having thought of something.
“I declare, I’m losin’ all my manners since I left the settlements,”
said he. “I’m Reuben Holmes, gentlemen, all the way from Pike,
Missouri. That chap there,” he added, pointing to one of the young
men, “he’s Reuben Holmes, junior. He’s my son. Know him,
strangers.”
The boys, with great difficulty controlling their desire to laugh,
bowed to Reuben Holmes, junior, who did not know enough to
acknowledge it in any way, and the old man went on.
“This ’yer’s my old woman; that big gal is my darter, and them’s
my young children. That chap is Simon Cool, my hired help. He’s
going out to the mines to run a quartz-mill for me. Them two,”
waving a biscuit toward the two hunters, “are Zack and Sile. I don’t
know their other names, but they’re friends of our’n. They met us on
the prairie, and finding that we were a trifle out of our reckoning,
they took us in tow and are going to show us the way to where we
want to go. We’re bound for the mines, and I am going to set up a
quartz-mill there.”
Having thus, as he imagined, placed our heroes quite at their
ease, the old man took a bite out of his biscuit, first dipping it into
the frying-pan, and went on with his questioning. The boys good-
naturedly replied, paying no heed to the winks, nods and grunts of
disbelief which the two hunters constantly exchanged with Simon
Cool, and finally the emigrant came down to their recent exploits.
“Where you been lately and what a doing?” he asked.
“We have just come from the mountains,” was the reply of
Archie, who acted as spokesman.
“Humph!” grunted Zack.
“What were you doing there?”
“Hunting wild horses.”
“Did you ever hear tell of the like of that, Zack?” asked Silas.
“I never did,” was the answer. “They’re nice lookin’ chaps to go a
huntin’ wild hosses, haint they now?”
“An’ the mountains is a likely place to find ’em, too,” added Silas.
“Is that frisky critter of your’n out there a wild hoss?” asked the
old man.
“He is.”
“Well, I do think in my soul! And you did ketch him, didn’t you?
You say you’re going to Bolton to-morrow?”
“Yes.”
“How far is it from here?”
“About three days’ journey, the way you will travel. We, on
horseback, could accomplish it in less time.”
“Which way is it from here?”
“Off there.”
“Why how does that come?” said the emigrant, with some
surprise. “We’ve been travelling that way for the last two days,”
pointing in just the opposite direction.
 “Then you have lost your course, or else somebody has been
misleading you.”
“Don’t listen to ’em, old man,” said Zack. “Tain’t noways likely
that me an’ Sile could get lost.”
“We don’t suppose you could,” said Eugene, who had been made
so angry by the hunters’ sly winks and nods that he could scarcely
restrain himself. “You know where the Fort is as well as we do.”
“I reckon we do, an’ better,” said Zack.
Simon Cool, for some reason or other, had been making
persevering, but unavailable, efforts to turn the conversation into
another channel, and Archie, seeing that Eugene was treading on
dangerous ground, joined in with Simon, and finally succeeded in
getting the old man started on the subject of mining, on which he
could talk like a book. Archie listened until he had finished his
supper, and then went out to look at his horse. He glanced over his
shoulder as he went, and saw that Simon and the two hunters were
sitting with their heads close together, as if engaged in consultation.
“There’s something afoot,” thought Archie, “and I have half a
mind to saddle up and clear out. Here comes Eugene. I’ll see what
he thinks about it.”
 CHAPTER IX.
AN ENEMY IN CAMP.

“I HEARD Dick talking about Pikes the other day,” said Eugene,
who had followed Archie in order to lend his assistance in case
the bay had succeeded in freeing himself from his halter or lariat.
“Did he refer to such people as our friends here?”
“He did,” answered Archie. “As I understand it, these Pikes are
one of the results of our late war. A good many of the people in the
border states were completely broken up, their houses being
burned, their crops destroyed and their stock killed or driven off. It
takes time and patience to accumulate property by farming, and it is
hard for a man to begin over again where he began twenty or forty
years ago. There are not many who have the courage to do it. These
people had heard wonderful stories of the mines and the sudden
fortunes sometimes made there, and believing that since they had to
dig in the ground for a living, it would be easier to dig gold and
silver than potatoes, they emigrated by regiments. But where one
family made a permanent settlement in California, a dozen turned
back. But still some of them were not satisfied, and after staying a
year or two in their old homes, they would bundle up again and go
back to California; and from roaming about so much, they fell into
the habit of leading a wandering, gipsy life. They are not satisfied
with a permanent settlement anywhere. How the name they bear
came to be applied to them I don’t know, unless it was because the
first of them came from Pike county, Missouri. This man is a genuine
Pike. There is no telling how many times he has wandered back and
forth over these prairies, but he hasn’t learned much during his
wanderings. I have crossed the prairie only once, and I know that he
is a hundred miles north of where he belongs.”
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookultra.com