Differentiable Manifolds A Theoretical Physics Approach 2nd Edition Gerardo F. Torres Del Castillo All Chapter Instant Download

Full download test bank at ebook textbookfull.
com
Differentiable Manifolds A
Theoretical Physics Approach 2nd
Edition Gerardo F. Torres Del
CLICK LINK TO DOWLOAD
https://textbookfull.com/product/differentiab
le-manifolds-a-theoretical-physics-
approach-2nd-edition-gerardo-f-torres-del-
castillo/
textbookfull
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Crafting a Republic for the World Scientific Geographic

and Historiographic Inventions of Colombia Lina Del
Castillo
https://textbookfull.com/product/crafting-a-republic-for-the-
world-scientific-geographic-and-historiographic-inventions-of-
colombia-lina-del-castillo/
Skin Cancer Management A Practical Approach 2nd Edition

Deborah F. Macfarlane
https://textbookfull.com/product/skin-cancer-management-a-
practical-approach-2nd-edition-deborah-f-macfarlane/
Coastal Wetlands: An Integrated Ecosystem Approach

Gerardo Perillo
https://textbookfull.com/product/coastal-wetlands-an-integrated-
ecosystem-approach-gerardo-perillo/
Theoretical and computational physics of gas discharge

phenomena a mathematical introduction 2nd Edition
Sergej T. Suržikov
https://textbookfull.com/product/theoretical-and-computational-
physics-of-gas-discharge-phenomena-a-mathematical-
introduction-2nd-edition-sergej-t-surzikov/
Introduction to Graphene-Based Nanomaterials: From
Electronic Structure to Quantum Transport 2nd Edition
Luis E. F. Foa Torres
https://textbookfull.com/product/introduction-to-graphene-based-
nanomaterials-from-electronic-structure-to-quantum-transport-2nd-
edition-luis-e-f-foa-torres/
Theoretical Concepts in Physics An Alternative View of

Theoretical Reasoning in Physics 3rd Edition Malcolm S.
Longair
https://textbookfull.com/product/theoretical-concepts-in-physics-
an-alternative-view-of-theoretical-reasoning-in-physics-3rd-
edition-malcolm-s-longair/
Type 2 Fuzzy Logic in Control of Nonsmooth Systems

Theoretical Concepts and Applications Oscar Castillo
https://textbookfull.com/product/type-2-fuzzy-logic-in-control-
of-nonsmooth-systems-theoretical-concepts-and-applications-oscar-
castillo/
Calculus On Manifolds: A Modern Approach To Classical

Theorems Of Advanced Calculus First Edition Michael
Spivak
https://textbookfull.com/product/calculus-on-manifolds-a-modern-
approach-to-classical-theorems-of-advanced-calculus-first-
edition-michael-spivak/
Coherent Quantum Physics A Reinterpretation of the

Tradition Texts and Monographs in Theoretical Physics
1st Edition Arnold Neumaier
https://textbookfull.com/product/coherent-quantum-physics-a-
reinterpretation-of-the-tradition-texts-and-monographs-in-
theoretical-physics-1st-edition-arnold-neumaier/
Gerardo F. Torres del Castillo
Differentiable
Manifolds
A Theoretical Physics Approach
Second Edition
Differentiable Manifolds
A Theoretical Physics Approach
Second Edition
Instituto de Ciencias
Benemérita Universidad
Autónoma de Puebla
Puebla, Puebla, Mexico
ISBN 978-3-030-45192-9 ISBN 978-3-030-45193-6 (eBook)

https://doi.org/10.1007/978-3-030-45193-6
Mathematics Subject Classification (2010): 53-XX, 58-XX, 58AXX
1st edition: © Springer Science+Business Media, LLC 2012

2nd edition: © Springer Nature Switzerland AG 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
This book is published under the imprint Birkhäuser, www.birkhauser-science.com by the registered
company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface to the Second Edition
As in the first edition, I have tried to maintain the presentation at an elementary

level, adding more details throughout the book. Along all the chapters, my intention
has been to avoid the structure definition-theorem-definition-theorem, that is fre-
quently adopted in the books on this subject, giving examples that help to under-
stand the concepts as they are introduced. Instead of attempting to cover each theme
completely, I have tried to show how the definitions really work and, in various
cases, how the definitions arise.
This second edition is not essentially different in structure or content from the
first one; the principal changes are the following:
• The detailed solution to all the exercises is given.
• The number of examples and exercises has been increased.
• The double and the dual numbers are employed in some examples and exercises.
Among the topics added in this new edition, there is a section on the moment
map and an appendix dealing with fiber bundles. I have taken the opportunity to
correct some errors discovered in the first edition.
I want to thank Birkhäuser, especially Samuel DiBella, for inviting me to write
this second edition, and Chris Eder and Christopher Tominich, who supervised the
last stages of the process. I also like to thank the reviewers of this new edition for
their helpful comments and Dr. Iraís Rubalcava-García for her help with the new
figures added to the book.
Puebla, Puebla, México Gerardo F. Torres del Castillo

January 2020
v
Preface to the First Edition
The aim of this book is to present in an elementary manner the basic notions related
with the differentiable manifolds and some of their applications, especially in
physics. The book is aimed at advanced undergraduate or graduate students in
physics or mathematics, assuming a working knowledge of calculus in several
variables, linear algebra, and differential equations. For the last chapter, which deals
with Hamiltonian mechanics, it is convenient to have some knowledge of analytical
mechanics. Most of the applications of the formalism considered here are related to
differential equations, differential geometry, and Hamiltonian mechanics; they may
serve as an introduction to the specialized treatises on these subjects.
One of the aims of this book is to emphasize the connections among the areas of
mathematics and physics where the formalism of the differentiable manifolds is
applied. The themes treated in the book are somewhat standard, but the examples
developed here go beyond the elementary ones, trying to show how the formalism
works in actual calculations. Some results not previously presented in book form
are also included, most of them related to the Hamiltonian formalism of classical
mechanics. Whenever possible, coordinate-free definitions or calculations are pre-
sented; however, when it is convenient or necessary, computations using bases or
coordinates are given, not underestimating their importance.
Along the book there is a collection of exercises, of various degrees of difficulty,
which form an essential part of the book. It is advisable that the reader attempts to
solve them and to fill in the details of the computations presented in the book.
The basic formalism is presented in Chapters 1 and 3 (differentiable manifold,
differentiable mappings, tangent vectors, vector fields, and differential forms) after
which the reader can continue with Chapters 5 and 6, if he is interested in the
applications to differential geometry and general relativity (even though in the
definitions of a Killing vector field, and of the divergence of a vector field given in
Chapter 6, the definition of the Lie derivative, presented in Chapter 2, is required).
Chapter 7 deals with Lie groups and makes use of concepts and results presented in
Chapter 2 (one-parameter groups and Lie derivatives). Chapters 2 and 4 are related
with differential equations and can be read independently, after Chapters 1 and 3.
Finally, for Chapter 8, which deals with Hamiltonian mechanics, the material of
vii
viii Preface to the First Edition
Chapters 1, 2, and 3 is necessary and, for some sections, Chapters 6 and 7 are also
required.
Some of the subjects that are not treated here are the integration of differential
forms, cohomology theory, fiber bundles, complex manifolds, manifolds with
boundary, and infinite-dimensional manifolds.
This book has been gradually developed starting from the first version in Spanish
(with the title Notas sobre variedades diferenciables) written around 1981, at the
Centro de Investigación y de Estudios Avanzados del IPN, in Mexico, D.F. The
previous versions of the book have been used by the author and some colleagues in
courses addressed to advanced undergraduate and graduate students in physics and
mathematics.
I would like to thank Gilberto Silva Ortigoza, Merced Montesinos, and the
reviewers for helpful comments, and I thank Bogar Díaz Jiménez for his valuable
help with the figures. I also thank Jessica Belanger, Tom Grasso, and Katherine
Ghezzi at Birkhäuser for their valuable support.
Puebla, Puebla, México Gerardo F. Torres del Castillo

June 2011
Contents
1 Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Differentiable Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Tangent Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3 Vector Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4 1-Forms and Tensor Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2 Lie Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.1 One-Parameter Groups of Transformations and Flows . . . . . . . . . . 37
2.2 Lie Derivative of Functions and Vector Fields . . . . . . . . . . . . . . . 47
2.3 Lie Derivative of 1-Forms and Tensor Fields . . . . . . . . . . . . . . . . 50
3 Differential Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.1 The Algebra of Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2 The Exterior Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4 Integral Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4.1 The Rectification Lemma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4.2 Distributions and the Frobenius Theorem . . . . . . . . . . . . . . . . . . . 84
4.3 Symmetries and Integrating Factors . . . . . . . . . . . . . . . . . . . . . . . 98
5 Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.1 Covariant Differentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.2 Torsion and Curvature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.3 The Cartan Structural Equations . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.4 Tensor-Valued Forms and Covariant Exterior Derivative . . . . . . . . 135
6 Riemannian Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.1 The Metric Tensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.2 The Riemannian Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
6.3 Curvature of a Riemannian Manifold . . . . . . . . . . . . . . . . . . . . . . 176
ix
x Contents
6.4 Volume Element, Divergence, and Duality of Differential

Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
6.5 Elementary Treatment of the Geometry of Surfaces . . . . . . . . . . . 196
7 Lie Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
7.1 Basic Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
7.2 The Lie Algebra of the Group . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
7.2.1 The Structure Constants . . . . . . . . . . . . . . . . . . . . . . . . . . 216
7.2.2 Lie Group Homomorphisms . . . . . . . . . . . . . . . . . . . . . . . 217
7.2.3 The SUð2Þ-SOð3Þ Homomorphism . . . . . . . . . . . . . . . . . . 219
7.2.4 Lie Subgroups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
7.3 Invariant Differential Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
7.3.1 The Maurer–Cartan Equations . . . . . . . . . . . . . . . . . . . . . 225
7.3.2 Finding the Group from the Structure Constants . . . . . . . . 225
7.3.3 Invariant Forms on Subgroups of GLðn; RÞ . . . . . . . . . . . . 228
7.4 One-Parameter Subgroups and the Exponential Map . . . . . . . . . . . 230
7.4.1 A Torsion-Free Connection . . . . . . . . . . . . . . . . . . . . . . . 239
7.5 The Lie Algebra of the Right-Invariant Vector Fields . . . . . . . . . . 240
7.6 Lie Groups of Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . 242
7.6.1 The Adjoint Representation . . . . . . . . . . . . . . . . . . . . . . . 250
7.6.2 The Coadjoint Representation . . . . . . . . . . . . . . . . . . . . . . 252
8 Hamiltonian Classical Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.1 The Cotangent Bundle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
8.2 Hamiltonian Vector Fields and the Poisson Bracket . . . . . . . . . . . 261
8.3 The Phase Space and the Hamilton Equations . . . . . . . . . . . . . . . 271
8.4 Geodesics, the Fermat Principle, and Geometrical Optics . . . . . . . 278
8.5 Dynamical Symmetry Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
8.6 The Moment Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
8.7 The Rigid Body and the Euler Equations . . . . . . . . . . . . . . . . . . . 302
8.8 Time-Dependent Formalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Appendix A: Lie Algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

Appendix B: Invariant Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Appendix C: Fiber Bundles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
Chapter 1
Manifolds
The main goal of the theory of differentiable manifolds is to extend many of the con-
cepts and results of the multivariate calculus to sets that do not possess the structure
of a normed vector space. Recall that the differentiability of a function of Rn to Rm
means that around each interior point of its domain the function can be approximated
by a linear transformation, but this requires the notions of linearity and distance,
which are not present in an arbitrary set.
The essential idea in the definition of a manifold should already be familiar from
analytic geometry, where one represents the points of the Euclidean plane by a pair
of real numbers (e.g., Cartesian or polar coordinates). Roughly speaking, a manifold
is a set whose points can be labeled by coordinates.
In this chapter and the following two, the basic formalism applicable to any
finite-dimensional manifold is presented, without imposing any additional structure.
In Chaps. 5 and 6 we consider manifolds with a connection and a metric tensor,
respectively, which are essential in differential geometry. In Chap. 7 we give an
elementary treatment of Lie groups. Chapters 4 and 8 contain applications of the
formalism to differential equations and classical mechanics, respectively.
1.1 Differentiable Manifolds
As mentioned above, the basic idea in the definition of a differentiable manifold is

probably familiar to the reader; in analytic geometry, the points of the Euclidean
plane, for instance, are represented by pairs of real numbers, which allows us to
make use of algebra or calculus to study the Euclidean geometry. In elementary
mechanics, we employ coordinate systems (Cartesian, circular cylindrical, spherical,
and others) to represent the position of the particles forming a mechanical system,
and something similar is applied in electromagnetism and quantum mechanics (in
the latter, the state of a particle is frequently given by a “wave function,” which
© Springer Nature Switzerland AG 2020 1
G. F. Torres del Castillo, Differentiable Manifolds,
https://doi.org/10.1007/978-3-030-45193-6_1
2 1 Manifolds
is a complex-valued function of the coordinates employed to label the points of

the space). Another example is found in thermodynamics, where the equilibrium
states of a thermodynamical system are characterized by a set of coordinates (called
“thermodynamic variables” or “state variables” in this context). In all these cases,
we replace the points of interest by n-tuples of real numbers. For instance, when we
think of a thermodynamic process, such as the quasistatic compression of a sample
of gas, we usually think of a curve in R2 , or some other space Rn , representing the
process. We make use of coordinates so commonly that, very often, when we think
of the Euclidean space, we think of some space Rn , as if they were the same object
(which, by the way, are not).
In the study of differentiable manifolds these identifications are established in
a precise manner and the precision given by this language helps to clarify many
of the applications of calculus found in physics and other areas of mathematics.
Roughly speaking, a differentiable manifold is a set whose points can be labeled by
a set of variables (or parameters) which take “continuous” values. Thanks to the use
of coordinates we can express the functions of interest in terms of real variables,
allowing us to employ the usual calculus of several variables. The definition of a
differentiable manifold guarantees that if a function is differentiable when expressed
in terms of some coordinates, then it is also differentiable in terms of other coordinate
systems; this is achieved by restricting the admissible coordinate systems.
Coordinate Charts
Let M be a set. A chart (or local chart) on M is a pair (U, φ) such that U is a subset
of M and φ is a one-to-one map from U onto some open subset of Rn (see Fig. 1.1).
A chart on M is also called a coordinate system on M. Defining a chart (U, φ) on a
set M amounts to labeling each point p ∈ U by means of n real numbers since φ( p)
belongs to Rn , and therefore consists of n real numbers that depend on p; that is,
φ( p) is of the form
φ( p) = x 1 ( p), x 2 ( p), . . . , x n ( p) . (1.1)
This relation defines the n functions x 1 , x 2 , . . . , x n , which will be called the coordi-
nate functions or, simply coordinates, associated with the chart (U, φ). The fact that
φ is a one-to-one mapping ensures that two different points of U differ, at least, in
the value of one of the coordinates.
We would also like close points to have close coordinates, but that requires some
notion of nearness in M, which can be given by the definition of a distance between
points of M or, more generally, by assigning a topology to M. We are not assuming
that the reader is acquainted with the basic concepts of topological spaces and in
most applications we will be dealing with sets possessing a natural notion of nearness
(see, however, the comment after Exercise 1.6). Hence, we shall not make use of the
concepts required for an adequate general discussion. For a more rigorous treatment
see, e.g., Crampin and Pirani (1986), Isham (1999), Boothby (2002), Lee (2002) and
Conlon (2008).
The coordinates associated with any chart (U, φ) must be functionally inde-
pendent among themselves, since the definition of a chart requires that φ(U )(≡
1.1 Differentiable Manifolds 3
Fig. 1.1 A coordinate

system in a set M; with the
aid of φ, each point of U U
corresponds to some point of
Rn . The image of U under φ M
must be an open subset of
φ
some Rn
Rn
{φ( p) | p ∈ U }) be an open subset of Rn . If, for instance, the coordinate x n could be

expressed as a function of x 1 , x 2 , . . . , x n−1 , then the points φ( p)( p ∈ U ) would lie
in a hypersurface of Rn , which is not an open subset of Rn .
Frequently, we make use of more than one coordinate system. In some cases it may
be convenient to replace a given coordinate system by another in order to simplify
a problem of interest (this happens very often in analytical mechanics, electromag-
netism, and quantum mechanics, where one makes use of spherical, cylindrical, or
another set of curvilinear coordinates instead of the Cartesian ones in order to sim-
plify the relevant equations); in other cases the use of more than one coordinate
system may be necessary if a given chart (U, φ) on M does not cover all of M, as in
the case of the circle or the sphere, where at least two charts are necessary to cover
all the points of M (see Example 1.5, below). Then, some points of M may belong
to the domain of more than one chart. In order to have a well-defined meaning of
the differentiability of the mappings between manifolds we shall demand that the
admissible coordinate systems be related to one another by means of differentiable
functions.
A function F : Rn → Rm given by F(q) = f 1 (q), f 2 (q), . . . , f m (q) is differen-
tiable of class C k if the real-valued functions f 1 , f 2 , . . . , f m have kth continuous par-
tial derivatives; two charts on M, (U, φ) and (V, χ ), are said to be C k -related (or C k -
compatible) if U ∩ V = ∅ (the empty set), or if φ ◦ χ −1 : χ (U ∩ V ) → φ(U ∩ V )
and χ ◦ φ −1 : φ(U ∩ V ) → χ (U ∩ V ), whose domains are open in Rn , are dif-
ferentiable of class C k (see Fig. 1.2). If x 1 , x 2 , . . . , x n are the coordinates asso-
ciated with (U, φ) and y 1 , y 2 , . . . , y n are the coordinates associated with (V, χ ),
the fact that (U, φ) and (V, χ ) be C k -related amounts to the fact that, for all
p ∈ U ∩ V , y 1 ( p), y 2 ( p), . . . , y n ( p) be differentiable functions of class C k of
x 1 ( p), x 2 ( p), . . . , x n ( p), and conversely.
A C k subatlas on M is a collection of charts on M, {(Ui , φi )}, such that for any
pair of indices i, j, (Ui , φi ) and (U j , φ j ) are C k -related and M = U1 ∪ U2 ∪ · · · (so
that each point of M is in the domain of at least one chart). In other words, a C k
subatlas on M is a set of coordinate systems whose domains cover all of M and are
pairwise C k -related. The collection of all the charts C k -related with the charts of a
C k subatlas, on M, form a C k atlas on M (see the examples below).
4 1 Manifolds
U
V
φ χ
M
Rn χ (U ∩V )
χ ◦ φ −1
φ (U ∩V )
φ ◦ χ −1
Rn
Fig. 1.2 Two coordinate systems whose domains have a nonempty intersection. A point p belonging
to U ∩ V corresponds to two points of Rn , φ( p) and χ( p); the charts (U, φ) and (V, χ) are C k -
related if the maps φ( p) → χ( p) and χ( p) → φ( p) are differentiable functions of class C k
Definition 1.1 A C k manifold of dimension n is a set M with a C k atlas; if k 1, it

is said that M is a differentiable manifold. If k = 0, it is said that M is a topological
manifold.
Example 1.2 (The standard structure of Rn as a manifold.) The space Rn has a

natural coordinate system: each point p ∈ Rn is an n-tuple of real numbers, p =
(a 1 , a 2 , . . . , a n ), and we can take the same point (a 1 , a 2 , . . . , a n ) as φ( p), that is,
φ( p) = (a 1 , a 2 , . . . , a n ). Then, we can readily verify that this is indeed a coordinate
system defined in all of Rn since φ is the identity mapping, which is one-to-one and
the image of Rn under this mapping is again Rn , which is an open subset of Rn .
This coordinate chart, (Rn , id) (here and henceforth, id denotes the identity map),
by itself, forms a C ∞ subatlas. The infinite collection of all the coordinate systems
C ∞ -related with this chart forms a C ∞ atlas with which Rn is a C ∞ manifold of
dimension n. When we consider Rn as a differentiable manifold, it is understood that
this is its atlas.
In a similar manner, any open subset, U , of Rn is a C ∞ manifold of dimension
n with the atlas containing the chart (U, id). An example that will appear later is
the disk D ≡ {(a, b) ∈ R2 | a 2 + b2 < 1}, which is a C ∞ manifold of dimension two
with the atlas containing the chart (D, id).
Example 1.3 (The Euclidean plane as a manifold.) An example closely related

to the previous one is given by the Euclidean plane (or the Euclidean space of any
dimension). As is well known in elementary analytic geometry, choosing two straight
lines on the Euclidean plane (the coordinate axes) that meet at a right angle, each
point p of the plane corresponds to an ordered pair of real numbers (x( p), y( p))
(the Cartesian coordinates of p with respect to the axes chosen). In this case the
coordinates are defined in terms of lengths, but it should be clear that, in all cases,
the coordinates need not be directly related to lengths or angles. The coordinate chart
(U, φ), where U is the entire Euclidean plane and φ( p) ≡ (x( p), y( p)), is a C ∞
subatlas in the Euclidean plane, and the set of all coordinate charts C ∞ -related with
this chart forms a C ∞ atlas with which the Euclidean plane is a C ∞ manifold of
dimension two.
1.4 In the set of 2 × 2 real matrices with determinant equal to 1, M =

Example
{ ac db | a, b, c, d ∈ R, ad − bc = 1}, we can introduce the coordinate chart (U, φ),

with U ≡ { ac db | a, b, c, d ∈ R, ad − bc = 1, a = 0}, which is a proper subset of
M, and
φ( ac db ) ≡ (a, b, c).
Then we find that φ(U ) = {(x, y, z) ∈ R3 | x = 0}, which is an open subset of R3 .

(Note that there is no restriction on the values of b and c; taking d = (bc + 1)/a one
assures that the determinant ad − bc is equal to 1.) The fact that

x y
φ −1 (x, y, z) = (1.2)
z (1 + yz)/x
shows that φ is one-to-one.

Since U does not coincide with M, in order to have a subatlas on M, at least another
coordinate chart is needed. Taking into account that the entries a and c cannot be
simultaneously equal to zero (since ad− bc must be equal to 1), we define a second
coordinate chart, (V, χ ), with V ≡ { ac db | a, b, c, d ∈ R, ad − bc = 1, c = 0},
and
χ ( ac db ) ≡ (a, c, d).
We find that χ (V ) = {(x, y, z) ∈ R3 | y = 0}, which is an open subset of R3 , and we

have
x (x z − 1)/y
χ −1 (x, y, z) = . (1.3)
y z
Then, U ∪ V = M and

a b
U ∩V ={ c d | a, b, c, d ∈ R, ad − bc = 1, a, c = 0}.
The two charts (U, φ) and (V, χ ) are C ∞ -related. In fact, making use of (1.2) and
(1.3) we have

x (x z − 1)/y
(φ ◦ χ −1 )(x, y, z) = φ( y z ) = (x, (x z − 1)/y, y),
which is a differentiable function (of class C ∞ ) in χ (U ∩ V ) = {(x, y, z) ∈ R3

| x, y = 0} and

(χ ◦ φ −1 )(x, y, z) = χ ( x y
z (1 + yz)/x ) = (x, z, (1 + yz)/x),
6 1 Manifolds
(0, 0, . . . , 1)
(a1 , a2 , . . . , an+1 )
(b1 , b2 , . . . , 0)
Fig. 1.3 The stereographic projection establishes a one-to-one correspondence between the points
of the n-sphere, excluding the “north pole” (0, 0, . . . , 1), and the points of the plane x n+1 = 0. The
point (a 1 , a 2 , . . . , a n+1 ) is a point of the n-sphere different from (0, 0, . . . , 1)
which is a differentiable function (of class C ∞ ) in φ(U ∩ V ) = {(x, y, z) ∈ R3

| x, z = 0}. The collection of all the coordinate systems C ∞ -related with these two
charts forms a C ∞ atlas with which M is a C ∞ manifold of dimension three.
Example 1.5 (Stereographic projection.) Almost all the points of the n-sphere

Sn ≡ (a 1 , . . . , a n+1 ) ∈ Rn+1 | (a 1 )2 + · · · + (a n+1 )2 = 1
(n 1) can be put into a one-to-one correspondence with the points of Rn by

means of the stereographic projection defined in the following way. Any point
(a 1 , . . . , a n+1 ) ∈ Sn , different from (0, 0, . . . , 1), can be joined with (0, 0, . . . , 1)
by means of a straight line that intersects the hyperplane x n+1 = 0 at some point
(b1 , . . . , bn , 0) (see Fig. 1.3). The condition that the three points (a 1 , . . . , a n+1 ),
(0, 0, . . . , 1), and (b1 , . . . , bn , 0) lie on a straight line amounts to
(b1 , . . . , bn , 0) − (0, 0, . . . , 1) = λ (a 1 , . . . , a n+1 ) − (0, 0, . . . , 1) , (1.4)
for some λ ∈ R. By considering the last component in the vector equation (1.4)
we have 0 − 1 = λ(a n+1 − 1); hence, λ = 1/(1 − a n+1 ). Substituting this value
of λ into (1.4) we find that the mapping φ : Sn \ {(0, 0, . . . , 1)} → Rn defined by
φ(a 1 , . . . , a n+1 ) ≡ (b1 , . . . , bn ) is given by
1
φ(a 1 , . . . , a n+1 ) = (a 1 , . . . , a n ). (1.5)
1 − a n+1
The pair (U, φ), with U ≡ Sn \ {(0, 0, . . . , 1)}, is a chart of coordinates, since φ is
injective and φ(U ) = Rn (which is an open set in Rn ).
In a similar manner, joining the points of Sn with (0, 0, . . . , −1) by means
of straight lines, another projection is obtained, χ : Sn \ {(0, 0, . . . , −1)} → Rn ,
given by
1
χ (a 1 , . . . , a n+1 ) = (a 1 , . . . , a n ) (1.6)
1 + a n+1
so that (V, χ ), with V ≡ Sn \ {(0, 0, . . . , −1)}, is a second chart of coordinates

which is C ∞ -related with (U, φ). In effect, from (1.5) and (1.6) we find that
1 1 n

φ −1 (b1 , . . . , bn ) = n 2b , . . . , 2bn , −1 + (bi )2 ,
1+ i=1 (b )
i 2
i=1
(1.7)
−1 1 1 n

χ (b , . . . , b ) =
1 n
n 2b , . . . , 2bn , 1 − (b ) ,
i 2
1+ i=1 (b )
i 2
i=1
and therefore
(b1 , . . . , bn )
(χ ◦ φ −1 )(b1 , . . . , bn ) = (φ ◦ χ −1 )(b1 , . . . , bn ) = n .
i=1 (b )
i 2
We have U ∩ V = Sn \ {(0, 0, . . . , 1), (0, 0, . . . , −1)}; hence φ(U ∩ V ) = χ (U ∩

V ) = Rn \ {(0, 0, . . . , 0)}, where the compositions χ ◦ φ −1 and φ ◦ χ −1 are differ-
entiable of class C ∞ . Since Sn = U ∪ V , the charts (U, φ) and (V, χ ) form a C ∞
subatlas for Sn . The set of all the coordinate charts C ∞ -related with these two charts
form a C ∞ atlas for Sn with which Sn is a manifold of dimension n.
Let M be a manifold. A subset A of M is said to be open if for any chart (U, φ)

belonging to the atlas of M, the set φ(A ∩ U ) is open in Rn .
Exercise 1.6 Show that the collection τ of open subsets of a manifold M is a topology
of M; that is, show that M and the empty set belong to τ , that the union of any family
of elements of τ belongs to τ , and that the intersection of any finite family of elements
of τ belongs to τ . We say that this topology is induced by the manifold structure
given in M.
When a given set, M, already possesses a topology and one wants to give it the
structure of a manifold in such a way that the topology induced by the manifold
structure coincides with the topology originally given, one demands that for each
chart (U, φ), in the atlas of M, the map φ be continuous and has a continuous
inverse; as a consequence, U must be an open set of M. (A map is continuous if and
only if the preimage of any open set is open.)
The Cartesian product of two differentiable manifolds, M and N , acquires the
structure of a differentiable manifold in a natural way. If {(Ui , φi )} and {(V j , ψ j )}
are subatlases of M and N , respectively, one can verify that {(Ui × V j , ρi j )} is a
subatlas for M × N , with ρi j ( p, q) ≡ (x 1 ( p), . . . , x n ( p), y 1 (q), . . . , y m (q)), where
(x 1 ( p), . . . , x n ( p)) = φi ( p) and (y 1 (q), . . . , y m (q)) = ψ j (q).
Differentiability of Maps
If f is a real-valued function defined on a differentiable manifold M, f : M → R,
and (U, φ) is a chart belonging to the atlas of M, the composition f ◦ φ −1 is a real-
valued function defined on an open subset of Rn , which may be differentiable or not
(see Fig. 1.4). The differentiability of the composition f ◦ φ −1 does not depend on
8 1 Manifolds
Fig. 1.4 With the aid of a M

coordinate system on M, a
real-valued function f U f
defined on M is represented
by the function
φ R
f ◦ φ −1 : Rn → R
Rn
f ◦ φ −1
the chart chosen since the charts of the atlas of M are C k -related (for some k 1).
From the identities
f ◦ φ −1 = ( f ◦ χ −1 ) ◦ (χ ◦ φ −1 ), f ◦ χ −1 = ( f ◦ φ −1 ) ◦ (φ ◦ χ −1 )
it follows that f ◦ φ −1 is differentiable if and only if f ◦ χ −1 is. Hence, it makes

sense to state the following definition. Let M be a differentiable C k manifold. A
function f : M → R is differentiable of class C r (r k) if f ◦ φ −1 is differentiable
of class C r for every chart (U, φ) in the atlas of M.
Example 1.7 Let M be the set of matrices with the structure of differentiable man-
ifold considered in Example 1.4. The function f : M → R defined by

a b
f( c d )≡a+d
(i.e., f is the trace) is differentiable of class C ∞ . According to the definition above,

we only have to show that the compositions f ◦ φ −1 and f ◦ χ −1 , where φ and χ
are the coordinates defined by (1.2) and (1.3), are infinitely differentiable functions
from open subsets of R3 to R. In fact, we obtain
1 + yz
( f ◦ φ −1 )(x, y, z) = f ( xz y
(1 + yz)/x )=x+ ,
x
which is differentiable (of class C ∞ ) in φ(U ) = {(x, y, z) ∈ R3 | x = 0}, and

(x z − 1)/y
( f ◦ χ −1 )(x, y, z) = f ( xy z ) = x + z,
which is differentiable (of class C ∞ ) in χ (V ) = {(x, y, z) ∈ R3 | y = 0}.
For a coordinate system (U, φ) belonging to the atlas of M, and a real-valued

function f : M → R, letting F ≡ f ◦ φ −1 , we have [see (1.1)]
f ( p) = ( f ◦ φ −1 )(φ( p)) = F(φ( p)) = F(x 1 ( p), x 2 ( p), . . . , x n ( p)),

ψ N
U V
φ χ
Rm
Rn χ ◦ ψ ◦ φ −1
Fig. 1.5 The map ψ : M → N is locally represented by the map χ ◦ ψ ◦ φ −1 . ψ is differentiable

if the compositions χ ◦ ψ ◦ φ −1 are differentiable for any pair of charts (U, φ) on M, and (V, χ)
on N
for p ∈ U , which we shall write as f = F(x 1 , x 2 , . . . , x n ); in this manner, the func-

tion f is expressed in terms of a real-valued function defined in (a subset of) Rn . For
instance, the function f defined in Example 1.7 can be expressed as f = x + 1+yz x
, in
terms of the coordinate functions (x, y, z) corresponding to the chart (U, φ) defined
in Example 1.4. (Note that the expression f = x + 1+yz x
is valid only in the points
of U .)
Exercise 1.8 Let M be a C k manifold. Show that the coordinates associated with
any chart 1in the 2atlas of Mn are differentiable functions of class C k . (Hint: if
φ( p) = x ( p), x ( p), . . . , x ( p) , then x = π ◦ φ where π i : Rn → R is defined
i i
by π i (a 1 , a 2 , . . . , a n ) = a i .)
If M is a C k manifold and N is a C l manifold, a map ψ from M into N is
differentiable of class C r (with r min{k, l}) if for any pair of charts (U, φ) on M and
(V, χ ) on N , the map χ ◦ ψ ◦ φ −1 is differentiable of class C r ; that is, ψ : M → N
is differentiable if, for p ∈ M, the coordinates of ψ( p) depend differentiably on the
coordinates of p (see Fig. 1.5). In fact, if x 1 , x 2 , . . . , x n are the coordinates associated
with the chart (U, φ) on M and y 1 , y 2 , . . . , y m are the coordinates associated with
the chart (V, χ ) on N , we have

y 1 (ψ( p)), . . . , y m (ψ( p)) = χ ψ( p) = (χ ◦ ψ ◦ φ −1 ) φ( p)

= (χ ◦ ψ ◦ φ −1 ) x 1 ( p), . . . , x n ( p) .
A diffeomorphism ψ is a one-to-one map from a differentiable manifold M to a

differentiable manifold N such that ψ and ψ −1 are differentiable; two differentiable
manifolds M and N are diffeomorphic if there exists a diffeomorphism ψ from M
onto N .
10 1 Manifolds
Example 1.9 The hyperboloid N ≡ {(x, y, z) ∈ R3 | x 2 + y 2 − z 2 = −1, z > 0} is

a two-dimensional manifold in the following natural way. The pair (N , χ ) with
χ : N → R2 given by χ (a, b, c) = (a, b) is a chart of coordinates covering all of N .
Hence, this single chart constitutes a C ∞ subatlas for N , which defines a C ∞ atlas
for N . D will denote the manifold already mentioned in Example 1.2.
We define a map ψ : D → N by means of the equations
2X 2Y
x ◦ψ = , y◦ψ = ,
1 − X2 − Y 2 1 − X2 − Y 2
where (x, y) are the coordinate functions of the chart (N , χ ) (that is, χ ( p) =
(x( p), y( p))) and (X, Y ) are the coordinate functions of the chart (D, φ), with
φ(a, b) = (a, b) (that is, X (a, b) = a, Y (a, b) = b). In order to verify that ψ is
differentiable it suffices to consider the composition χ ◦ ψ ◦ φ −1 , which is given by

(χ ◦ ψ ◦ φ −1 )(a, b) = χ (ψ(a, b)) = x(ψ(a, b)), y(ψ(a, b))

2a 2b
= , .
1 − a 2 − b2 1 − a 2 − b2
Since, by the definition of D, a 2 + b2 < 1, we see that the composition χ ◦ ψ ◦ φ −1

is differentiable of class C ∞ , thus showing that ψ is differentiable. Furthermore, ψ
is a diffeomorphism; from the expression of χ ◦ ψ ◦ φ −1 obtained above we find that
its inverse is given by the function

a b
(a, b) → √ , √ ,
1+ 1 + a 2 + b2 1 + 1 + a 2 + b2
and this is also differentiable.

Exercise 1.10 Show that the set of diffeomorphisms of a manifold onto itself forms
a group with the operation of composition.
Let M be a C k manifold of dimension n. A subset N of M is a submanifold of
M, of dimension m (m n), if there exists a C k subatlas of M, {(Ui , φi )}, such that
φi (N ∩ Ui ) = {(a 1 , a 2 , . . . , a n ) ∈ Rn | a m+1 = a m+2 = · · · = a n = 0}.
Let π be the canonical projection from Rn onto Rm given by π(a 1 , a 2 , . . . , a n ) =

(a 1 , a 2 , . . . , a m ). The collection {(N ∩ Ui , π ◦ φi )} is a C k subatlas on N , and N
becomes a C k manifold of dimension m with the atlas generated by this subatlas;
in other words, N is a submanifold of dimension m if there exist coordinate sys-
tems (U, φ) on M such that if U intersects N , then N ∩ U = { p ∈ U | x m+1 ( p) =
x m+2 ( p) = · · · = x n ( p) = 0}, where x 1 , x 2 , . . . , x n are the coordinates associated
to (U, φ).
With the aid of the following proposition we can construct or identify many
examples of submanifolds.
Proposition 1.11 Let f 1 , f 2 , . . . , f m be real-valued differentiable functions defined

on M. The set N ≡ { p ∈ M | f 1 ( p) = f 2 ( p) = · · · = f m ( p) = 0} is a submani-
fold of dimension n − m of M if for any chart (U, φ) of the atlas of M such that U
intersects N , the matrix with entries Di ( f j ◦ φ −1 )|φ( p) (1 i n, 1 j m) is
of rank m for p ∈ N . (Di stands for the ith partial derivative.)
Proof Let p ∈ N and let (U, φ) be a chart on M with p ∈ U . Assuming that the
determinant of the square matrix Di ( f j ◦ φ −1 )|φ( p) (1 i, j m) is different from
zero (which can be achieved by appropriately labeling the coordinates if necessary)
and denoting by x 1 , x 2 , . . . , x n the coordinates associated with (U, φ), the relations
y1 ≡ f 1, y2 ≡ f 2, ..., ym ≡ f m ,
(1.8)
y m+1 ≡ x m+1 , ..., yn ≡ x n
define a coordinate system in some subset V of U , that is, the x i can be written as
differentiable functions of the y i . In the coordinates y i the points p of N satisfy
y 1 ( p) = y 2 ( p) = · · · = y m ( p) = 0. Therefore, N is a submanifold of M of dimen-
sion n − m.
Example 1.12 Let M = R3 and N ≡ { p ∈ R3 | f ( p) = 0} with f =x +
2
y 2 − z,

−1
where (x, y, z) are the natural coordinates of R . The matrix Di ( f◦ φ )|φ( p) ,
3
mentioned in Proposition 1.11, is the row matrix 2x( p) 2y( p) − 1 , whose rank
is equal to 1 at all the points of N (actually, it is equal to 1 everywhere). Thus, we
conclude that N is a submanifold of R3 of dimension two. However, in order to see
in detail how the proof of the proposition works, we shall explicitly show that N
satisfies the definition of a submanifold given above.
It is convenient to relabel the coordinates, so that f takes the form f = y 2 +
z − x, because in that way the first entry of the matrix Di ( f ◦ φ −1 )|φ( p) is always
2
different from zero. Then, following the steps in the proof of Proposition 1.11, we
introduce the coordinate system (u, v, w) by means of [see (1.8)]
u = f = y 2 + z 2 − x, v = y, w = z,
on all of R3 . From these expressions and their inverses, x = v2 + w2 − u, y = v,

z = w, we see that the two coordinate systems are C ∞ -related, and in terms of the
coordinate system (u, v, w), each point p ∈ N satisfies u( p) = 0.
Exercise 1.13 Show that if x 1 , x 2 , . . . , x n are the natural coordinates of Rn (that
2
is, if p = (a 1 , a 2 , . . . , a n ) ∈ Rn , x i ( p) ≡ a i ), then Sn−1 ≡ { p ∈ Rn | x 1 ( p) +
2 2 n 2
x ( p) + · · · + x ( p) = 1} is a submanifold of Rn of dimension n − 1.
Definition 1.14 Let M be a C k manifold. A differentiable curve, C, of class C r , in
M, is a differentiable mapping of class C r from an open subset of R into M; that is,
C : I → M is a differentiable curve of class C r in M if I is an open subset of R and
φ ◦ C is a differentiable map of class C r for every chart (U, φ) of the atlas of M (see
Fig. 1.6).
12 1 Manifolds
C φ
Rn
R φ ◦C
t0
Fig. 1.6 A curve in M and its image in a coordinate system. C is differentiable if φ ◦ C is differ-
entiable for any chart (U, φ) on M
Example 1.15 The mapping

cos t 2 sin t
C(t) ≡
− 21 sin t cos t
is a differentiable curve of class C ∞ in the manifold defined in Example 1.4. Indeed,

making use of the two charts defined there we see that

cos t 2 sin t
(φ ◦ C)(t) = φ( − 21 sin t cos t
) = (cos t, 2 sin t, − 21 sin t) (1.9)
and
cos t 2 sin t
(χ ◦ C)(t) = χ ( − 1 sin t cos t
) = (cos t, − 21 sin t, cos t). (1.10)
2
The functions φ ◦ C and χ ◦ C are differentiable mappings from R into R3 of class

C ∞ , and, therefore, C is a differentiable curve.
In what follows it will be assumed that all the objects dealt with (manifolds, maps,
curves, etc.) are of class C ∞ .
The set of all differentiable functions from M to R will be denoted by C ∞ (M).
This set is a ring with the operations given by
( f + g)( p) ≡ f ( p) + g( p)
(a f )( p) ≡ a f ( p) (1.11)
∞
( f g)( p) ≡ f ( p)g( p) for f, g ∈ C (M), a ∈ R and p ∈ M.
If ψ is a differentiable map from M to a differentiable manifold N and f ∈

C ∞ (N ), the pullback of f under ψ, ψ ∗ f , is defined by
ψ ∗ f ≡ f ◦ ψ. (1.12)
From the relation (ψ ∗ f ) ◦ φ −1 = ( f ◦ χ −1 ) ◦ (χ ◦ ψ ◦ φ −1 ) it follows that ψ ∗ f ∈

C ∞ (M). That is, ψ ∗ : C ∞ (N ) → C ∞ (M) (ψ ∗ is applied to functions defined on N
to produce functions defined on M; hence the name pullback for ψ ∗ ).
Exercise 1.16 Show that ψ ∗ (a f + bg) = aψ ∗ f + bψ ∗ g and ψ ∗ ( f g) = (ψ ∗ f )
(ψ ∗ g) for f, g ∈ C ∞ (N ) and a, b ∈ R.
Exercise 1.17 Show that a map ψ : M → N is differentiable if and only if ψ ∗ f ∈
C ∞ (M) for f ∈ C ∞ (N ).
Exercise 1.18 Show that if ψ1 : M1 → M2 and ψ2 : M2 → M3 are differentiable
maps, then (ψ2 ◦ ψ1 )∗ = ψ1 ∗ ◦ ψ2 ∗ .
It is useful to observe that a map, ψ : M → N , is fully determined by the knowl-
edge of the functions ψ ∗ y j , where the y j form a coordinate system of N (this is
similar to the fact that a linear map is fully defined by knowing the image of the
vectors forming a basis).
1.2 The Tangent Space
If C is a differentiable curve in M and f ∈ C ∞ (M), then C ∗ f = f ◦ C is a dif-

ferentiable function from an open subset I ⊂ R into R (see Fig. 1.7). If t0 ∈ I , the
tangent vector to C at the point C(t0 ), denoted by Ct0 , is defined by

d ∗ f C(t) − f C(t0 )
Ct0 [ f ] ≡ (C f ) = lim . (1.13)
dt t0 t→t0 t − t0
Hence, Ct0 is a map from C ∞ (M) into R with the properties (see Exercise 1.16)
d ∗
Ct0 [a f + bg] = C (a f + bg)
dt t0

d
= (a C ∗ f + b C ∗ g)
dt t0

= a Ct0 [ f ] + b Ct0 [g], for f, g ∈ C ∞ (M), a, b ∈ R,
and
d ∗
Ct0 [ f g] = C ( f g)
dt t0
d ∗
= (C f )(C ∗ g)
dt t0

= f C(t0 ) Ct0 [g] + g C(t0 ) Ct0 [ f ], for f, g ∈ C ∞ (M).
The real number Ct0 [ f ] is the rate of change of f along C around the point C(t0 ).
The properties of the tangent vector to a curve lead to the following definition.
14 1 Manifolds
f ◦C
M
f
C
U
R R
t0
φ
Rn
Fig. 1.7 Composition of a curve in M with a real-valued function f . The derivative of f ◦ C,

which is a function from R into R, represents the directional derivative of f along C
Definition 1.19 Let p ∈ M. A tangent vector to M at p is a map, v p , of C ∞ (M) in

R such that
v p [a f + bg] = av p [ f ] + bv p [g]
(1.14)
v p [ f g] = f ( p) v p [g] + g( p) v p [ f ],
for f, g ∈ C ∞ (M), a, b ∈ R.
For a constant function, c (denoting by c both the function and its value, i.e.,
c( p) = c for all p ∈ M), we have
v p [c] = v p [c · 1] = c v p [1] = c v p [1 · 1]
= c(1 · v p [1] + 1 · v p [1]) = 2c v p [1] = 2v p [c],
therefore,
v p [c] = 0. (1.15)
The tangent space to M at p, denoted by T p M (or by the symbols T p (M) and

M p ), is the set of all the tangent vectors to M at p. The set T p M is a real vector space
with the operations defined by
(v p + w p )[ f ] ≡ v p [ f ] + w p [ f ],
(1.16)
(av p )[ f ] ≡ a(v p [ f ]),
for v p , w p ∈ T p M, f ∈ C ∞ (M), and a, b ∈ R. Hence, 0 p , the zero vector of T p M,

satisfies 0 p [ f ] = 0 for f ∈ C ∞ (M).
If (U, φ) is a chart on M, with coordinates x 1 , x 2 , . . . , x n and p ∈ U , the tangent
vectors, (∂/∂ x 1 ) p , (∂/∂ x 2 ) p , . . . , (∂/∂ x n ) p , are defined by
1.2 The Tangent Space 15

∂
[ f ] ≡ Di ( f ◦ φ −1 )|φ( p) , for f ∈ C ∞ (M), (1.17)
∂xi p
where Di denotes the partial derivative with respect to the ith argument; that is,

∂ 1
[ f ] = lim ( f ◦ φ −1 ) x 1 ( p), . . . , x i ( p) + t, . . . , x n ( p)
∂xi p t→0 t

− ( f ◦ φ −1 ) x 1 ( p), . . . , x i ( p), . . . , x n ( p) . (1.18)
Using the definition (1.17) one readilyverifies that, in effect, (∂/∂ x i ) p satisfies the
conditions (1.14) and therefore ∂/∂ x i p ∈ T p M.
Taking f = x j in (1.18) and noting that

(x j ◦ φ −1 ) x 1 ( p), x 2 ( p), . . . , x n ( p) = (x j ◦ φ −1 ) φ( p) = x j ( p)
and, similarly,

−1
x j ( p) if i = j
(x ◦ φ ) x ( p), x ( p), . . . , x ( p) + t, . . . , x ( p) =
j 1 2 i n
x j ( p) + t if i = j
(for t sufficiently small, so that all the points belong to U ), we find that

∂ j 0 if i = j
[x ] =
j
δi ≡ (1.19)
∂xi p 1 if i = j.
n
The set (∂/∂ x i ) p i=1 is linearly independent since if a i ∂/∂ x i p = 0 p (here
and in what follows, any index that appears twice, once as a subscript and once as a
superscript, implies a sum over all the values of the index, for instance, a i ∂/∂ x i p =
n
i=1 a (∂/∂ x ) p ), then using (1.19) we have
i i

∂ j
0 = 0 p [x j ] = a i [x j ] = a i δi = a j .
∂xi p
n
Proposition 1.20 If (U, φ) is a chart on M and p ∈ U , the set (∂/∂ x i ) p i=1
is a
basis of T p M and
∂
v p = v p [x i ] (1.20)
∂xi p
for v p ∈ T p M.
Proof We only have to prove that any tangent vector to M at p can be expressed
as a linear combination of the vectors (∂/∂ x i ) p . Let f ∈ C ∞ (M). The composition
F ≡ f ◦ φ −1 is a real-valued function defined on φ(U ), which is an open set of Rn .
16 1 Manifolds
−1

For an arbitrary point q ∈ U , we have f (q) = ( f ◦ φ ) ◦ φ(q) = F φ(q) and,
similarly, f ( p) = F φ( p) . According to the mean value theorem for functions from
Rn in R, for a real-valued differentiable function, F, defined in some open subset of
Rn , given two points (a 1 , . . . , a n ) and (b1 , . . . , bn ) such that the straight line segment
joining them is contained in the domain of F, we have
F(b1 , . . . , bn ) − F(a 1 , . . . , a n ) = (bi − a i ) Di F|(c1 ,...,cn ) , (1.21)
where (c1 , . . . , cn ) is some point on the straight line segment joining the
points (a 1 , . . . , a n ) and (b1 , . . . , bn ) [i.e., (c1 , . . . , cn ) = (1 − t0 )(a 1 , . . . , a n )
+ t0 (b1, . . . , bn ), for some
t0 ∈ (0, 1)]. Applying the formula
1 (1.21) with (a , . . . ,
1
a ) = x ( p), . . . , x ( p) = φ( p) and (b , . . . , b ) = x (q), . . . , x (q) = φ(q)

n 1 n 1 n n
we obtain
F φ(q) = F φ( p) + [x i (q) − x i ( p)] Di F|(c1 ,...,cn ) . (1.22)
Taking p fixed, the real numbers Di F|(c1 ,...,cn ) depend on q and will be denoted by
gi (q); then (1.22) amounts to
f (q) = f ( p) + x i (q) − x i ( p) gi (q)
or, since q is an arbitrary point in a neighborhood of p,
f = f ( p) + x i − x i ( p) gi . (1.23)
Using (1.14), (1.15), and the expression (1.23), taking into account that f ( p) and
the x i ( p) are real numbers, while f , x i , and gi are real-valued functions defined in
a neighborhood of p, for any v p ∈ T p M we have
v p [ f ] = v p [ f ( p)] + [x i ( p) − x i ( p)] v p [gi ] + gi ( p) v p [x i − x i ( p)]

= gi ( p) v p [x i ],

but gi ( p) = Di F|φ( p) = ∂/∂ x i p [ f ] [see (1.17)]. Therefore

∂
v p [ f ] = v p [x ]
i
[f]
∂xi p
and, since f is arbitrary, we obtain the expression (1.20). As a corollary of this result
we find that the dimension of T p M coincides with the dimension of M.
According to (1.20), the tangent vector to a differentiable curve C in M (C : I →

M), at the point C(t0 ) is given by

∂
Ct0 = Ct0 [x i ] .
∂xi C(t0 )
But, from (1.13), Ct0 [x i ] = d(x i ◦ C)/dt|t0 ; therefore

d(x i ◦ C) ∂
Ct0 = . (1.24)
dt t0 ∂ x i C(t0 )
Exercise 1.21 Let v p ∈ T p M. Show that there exists a curve C such that v p = Ct0 .
(In fact, there are an infinite number of such curves.)
If (V, χ ) is a second chart on M with coordinate functions y 1 , y 2 , . . . , y n , and

n
p ∈ U ∩ V , then we have another basis for T p M given by (∂/∂ y i ) p i=1 . From
(1.20) we see that
∂ ∂ ∂
= [x ]
j
.
∂ yi p ∂ yi p ∂x j p
It is convenient to write (∂ f /∂ x i ) p instead of (∂/∂ x i ) p [ f ], keeping in mind the

j
definition (1.17), so that Eq. (1.19) becomes (∂ x j /∂ x i ) p = δi and the foregoing
relation can be expressed in the simpler form

∂ ∂x j ∂
= (1.25)
∂ yi p ∂ yi p ∂x j p
and, similarly,
∂ ∂ yi ∂
= , (1.26)
∂x j p ∂x j p ∂ yi p
n n
which means that the two bases of T p M, (∂/∂ x i ) p i=1 and (∂/∂ y i ) p i=1 , are
j
related by means of the n × n matrix with entries ci ( p) = ∂ x j /∂ y i p , whose inverse

is the matrix with entries c̃kj ( p) = ∂ y k /∂ x j p .
Example 1.22 In the case of the curve considered in Example 1.15, denoting by
x 1 , x 2 , x 3 the coordinates corresponding to the chart (U, φ) [see (1.1)], Eq. (1.9)
amounts to (x 1 ◦ C)(t) = cos t, (x 2 ◦ C)(t) = 2 sin t, (x 3 ◦ C)(t) = − 21 sin t. Hence,
according to (1.24), the tangent vector to this curve is given by

∂ ∂ 1 ∂
Ct0 = − sin t0 + 2 cos t0 − cos t0 ,
∂x1 C(t0 ) ∂x2 C(t0 ) 2 ∂x3 C(t0 )
except at the points where cos t0 = 0, which are not covered by this coordinate
system.
Similarly, denoting by y 1 , y 2 , y 3 the coordinates corresponding to the chart
(V, χ ), from (1.10) and (1.24) we obtain

∂ 1 ∂ ∂
Ct0 = − sin t0 − cos t0 − sin t0 ,
∂ y1 C(t0 ) 2 ∂ y2 C(t0 ) ∂ y3 C(t0 )
18 1 Manifolds
vp
ψ (p) ψ∗p (v p )
p
M
N
Fig. 1.8 If ψ : M → N is a differentiable mapping from M into N , its Jacobian, or differential,

maps tangent vectors to M into tangent vectors to N
which is valid if sin t0 = 0.

We can see that Eq. (1.24) reduces to the usual expression for the tangent vector
to a curve in R3 given in the textbooks on multivariate calculus. A curve in R3 must
be of the form C(t) = ( f (t), g(t), h(t)), where f, g, h are real-valued functions
with some common domain I ⊂ R (since C(t) must be a point of R3 ). Denoting
by x 1 , x 2 , x 3 the natural coordinates of R3 (see Example 1.2) we have x 1 ◦ C =
f, x 2 ◦ C = g, x 3 ◦ C = h, and, therefore, Eq. (1.24) gives

d f ∂ dg ∂ dh ∂
Ct0 = + + ,
dt t0 ∂ x 1 C(t0 ) dt t0 ∂ x 2 C(t0 ) dt t0 ∂ x 3 C(t0 )
identifying the basis vectors (∂/∂ x 1 )C(t0 ) , (∂/∂ x 2 )C(t0 ) , (∂/∂ x 3 )C(t0 ) with the stan-
dard basis vectors i, j, k, respectively.
The Jacobian of a Differentiable Map
Let M and N be two differentiable manifolds and let ψ : M → N be a differentiable
map. The map ψ induces a linear transformation between the tangent spaces T p M
and Tψ( p) N called the Jacobian (or differential) of ψ at p, denoted by ψ∗ p (or by
dψ p ) (Fig. 1.8). If v p ∈ T p M, ψ∗ p (v p ) is defined as the tangent vector to N at ψ( p)
such that for f ∈ C ∞ (N )
ψ∗ p (v p )[ f ] ≡ v p [ψ ∗ f ]. (1.27)
Exercise 1.23 Show that if v p ∈ T p M, then ψ∗ p (v p ) ∈ Tψ( p) N and that ψ∗ p is linear.

If (x 1 , x 2 , . . . , x n ) is a coordinate system on M about the
pointi p and (y , y , . . . ,
1 2
y ) is a coordinate system on N about ψ( p), since ψ∗ p ∂/∂ x p ∈ Tψ( p) N , using

m
(1.20) we obtain the relation

∂ ∂ ∂
ψ∗ p = ψ∗ p [y ]
j
.
∂xi p ∂xi p ∂y j ψ( p)

But from the definitions (1.27) and (1.12), ψ∗ p ∂/∂ x i p [y j ] = ∂/∂ x i p [ψ ∗ y j ] =

∂/∂ x i p [y j ◦ ψ]; therefore

∂ ∂(y j ◦ ψ) ∂
ψ∗ p = . (1.28)
∂xi p ∂xi p ∂y j ψ( p)

In other words, the matrix with entries ∂(y j ◦ ψ)/∂ x i p represents the linear trans-
n m
formation ψ∗ p with respect to the bases (∂/∂ x i ) p i=1 and (∂/∂ y j )ψ( p) j=1 (com-
pare with the usual definition of the Jacobian matrix in the calculus of several vari-
ables).
Some examples of the computation of the Jacobian of a map are given below
(see Eq. (1.33) and Example 1.32), and we shall find other examples in the following
chapters.
If ψ1 : M1 → M2 and ψ2 : M2 → M3 are differentiable maps between differ-
entiable manifolds, then, for v p ∈ T p M1 and f ∈ C ∞ (M3 ), using (1.27) and Exer-
cise 1.18, we have
(ψ2 ◦ ψ1 )∗ p (v p )[ f ] = v p [(ψ2 ◦ ψ1 )∗ f ] = v p [(ψ1 ∗ ◦ ψ2 ∗ ) f ]

= v p [ψ1 ∗ (ψ2 ∗ f )] = ψ1∗ p (v p )[ψ2 ∗ f ]

= ψ2∗ψ1 ( p) ψ1∗ p (v p ) [ f ],
i.e.,
(ψ2 ◦ ψ1 )∗ p = ψ2∗ψ1 ( p) ◦ ψ1∗ p . (1.29)
This relation is called the chain rule.

If ψ : M → N is a differentiable map between differentiable manifolds and C :
I → M is a curve in M, the composition ψ ◦ C is a curve in N . According
to (1.13)
and (1.27), the tangent vector to ψ ◦ C at the point (ψ ◦ C)(t0 ) = ψ C(t0 ) satisfies
d
(ψ ◦ C)t0 [ f ] = ( f ◦ ψ ◦ C)|t0 = Ct0 [ f ◦ ψ]
dt
= Ct0 [ψ ∗ f ] = ψ∗C(t0 ) (Ct0 )[ f ], for f ∈ C ∞ (N ).
Hence
(ψ ◦ C)t0 = ψ∗C(t0 ) (Ct0 ), (1.30)
which means that the tangent vectors to the image of a curve C under the map ψ are
the images of the tangent vectors to C under the Jacobian of ψ (see Fig. 1.9).
The Tangent Bundle of a Manifold
The tangent bundle of a differentiable manifold M, denoted
by T M, is the set of
all tangent vectors at all points of M; that is, T M = p∈M T p M. The canonical
projection, π , from T M on M is the mapping that associates to each element of
T M the point of M at which it is attached; that is, if v p ∈ T p M, then π(v p ) = p.
Therefore, π −1 ( p) = T p M (see Fig. 1.10).
The tangent bundle has the structure of a differentiable manifold induced by
the structure of M in a natural way. If (U, φ) is a coordinate system on M, each
20 1 Manifolds
ψ
Ct0
ψ∗C(t0 ) (Ct0 )
M N
C
ψ ◦C
R
t0
Fig. 1.9 The tangent vectors of the curve ψ ◦ C are obtained applying the Jacobian of ψ to the
tangent vectors of C
Fig. 1.10 The tangent Tp M

bundle of M is formed by the
union of the tangent spaces
to M at all the points of M.
Each tangent space to M is
represented here by a vertical TM
line
p
M
v p ∈ π −1 (U ) is a linear combination of the vectors (∂/∂ x i ) p , with real coefficients

that depend on v p . Hence, we can write

∂
v p = q̇ (v p )
i
, (1.31)
∂xi p
which defines n functions q̇ i : π −1 (U ) → R. (This notation comes from that com-

monly employed in Lagrangian mechanics, when M is the configuration space of a
mechanical system.) From (1.19) we also have
q̇ i (v p ) = v p [x i ]. (1.32)
−1 ∗ i
n functions q : π (U ) → R, by q ≡ x ◦ π = π x , the pair
i i i
Defining
−1
the
π (U ), φ , with

φ(v p ) ≡ q 1 (v p ), . . . , q n (v p ), q̇ 1 (v p ), . . . , q̇ n (v p ) ,
is a chart on T M (see Fig. 1.11). (The image of π −1 (U ) under φ is φ(U ) × Rn ,

which is an open subset of R2n , and the injectivity of φ follows from the injectivity
Another random document with
no related content on Scribd:
Story of Philip Brusque.
(Continued from page 50.)
CHAPTER III.
More particulars of Philip’s early life.
Our story, thus far, has shown us that absolute liberty cannot be
enjoyed except by an individual in solitude, where he has no
intercourse with his fellowmen. It shows us that as soon as
individuals, even supposing that there are only two of them, come to
live together, some rules, by which they may regulate their conduct,
become absolutely necessary. In other words, people cannot live
together in society without government; even two persons on an
island find that, to prevent quarrelling, they must define their mutual
rights and privileges; or, in other words, they must enact laws; and
these laws, we perceive, are restraints upon natural or absolute
liberty. The farther progress of our story will show how an increasing
community, with more varied interests, requires a more extended
and minute code of laws.
But before I proceed further, let me tell you something more of
Philip Brusque’s early history. He was the son of a brickmaker of St.
Addresse, a small village in France, near the flourishing seaport of
Havre, which you know is situated at the mouth of the Seine. Philip
was early taught to read and write, but he paid little attention to these
things in his boyhood. He was more fond of action than study. He
spent a great part of his time in wandering through the deep dells
that surrounded his native village, or in walking along the high chalky
bluff that formed the neighboring sea-shore. Here he particularly
loved to spend his time, looking out over the sea for many leagues,
and tracing the progress of the ships, bearing the flags of many
nations, that ploughed their way upon the bosom of the Atlantic.
In this way, he formed habits of reflection; and though he loved
stirring excitements, still Philip was a thinking youth. At the same
time he was of a sanguine temper, ardent in his feelings, loving and
hating strongly, and readily believing what his wishes and his hopes
prompted. Thus he grew up to the age of twenty, without a settled
profession, sometimes working at his father’s trade, and sometimes
serving as mate of a small vessel that plied between Havre and
Bordeaux.
About this period, the public mind in France had begun to be
agitated by the coming tempest of the revolution. In every city,
village, and hamlet, the people were talking about government,
liberty, and the rights of man. The people of France had long been
subject to kings, who had claimed a right to reign over them, even
without their consent, and they had reigned in such a manner as to
make the people miserable. The people were now examining into
this claim of their kings, and they had already discovered that it was
founded in injustice. Unhappily, they fell under the guidance of
bloody and selfish men, and for many years the sufferings of France
in her struggle for liberty and human rights, were greater than they
had been under the despotism of her worst kings.
Philip Brusque engaged very ardently in the political discussions
that resulted in the revolution, and when Paris became the great
theatre of action, he resolved to quit St. Addresse, and proceed to
the metropolis, to take his share in the great drama that he felt was
about to be acted. He took leave of his parents, and went to bid
adieu to Emilie Bonfils, whom he had long loved, and to whom he
was affianced. The parting was tender, for Emilie was well worthy of
the affection of the gallant youth, and her fears were now excited for
the fate of her lover. He was not only to leave her, but he was to be
exposed to the convulsions, which already, like the heavings and
swellings which portend the earthquake, began to be realized
throughout France. But Philip’s mind was too much influenced with
the spirit of the time, which, like the hot sirocco of the desert,
seemed to sweep over the land, to be delayed or dissuaded. He
gave his Emilie a long and ardent salute, and on foot wended his
way to Paris.
I have told enough of what followed, for the purposes of my story.
Philip’s active mind and devoted spirit raised him to a certain degree
of power and distinction in the revolution; he rode for a time on the
storm, and shared in the scenes of blood and horror. He was indeed
accessory to many of the atrocious executions, which, in a spirit of
madness and fury, were decreed and sanctioned by the leaders. But
in all this, Philip was rather insane than selfish. Indeed, he was
intoxicated by the whirl of events, and he yielded to the current. At
length, he became sensible of his error, but before he had the
opportunity of atoning for it, he was obliged to fly for his life. He
wished to see his aged parents, and his mind turned more than once
to his gentle, confiding Emilie, at the village of St. Addresse. But
there were many reasons for his not going to see them before his
departure. The first was, that it was not safe, either for himself or
them; and the next was, that he now began to consider his hands
sullied with the blood of his fellow-men, in such a manner as to make
him unfit for the pure affections either of his parents or his affianced
Emilie. Indeed, such was the idea he had formed of the latter, and
such was the true affection and reverence that he entertained
towards her, and such, at the same time, was his feeling of
repentance and remorse, that he shrank from the idea of attaching
her to one like himself, and dragging her down from the dignity of
truth and purity, to the lot of one who was sullied with crime.
Accordingly, he wrote a letter to his parents and Emilie, explaining
his feelings and designs, and bade farewell to his country, as we
have seen. The letter he wrote did not reach its destination, but,
falling into the hands of Robespierre and his associates, became the
source of bitter persecution to those for whom it was intended.
CHAPTER IV.
A Ship appears in view.—Pirates ashore.—A scene at night.—
Recognition of an old Friend.—Alarming Discoveries.—A
fearful Plot.—An Explosion.—Arrival of about seventy persons
at Fredonia.
We return to Brusque on the island of Fredonia. A few weeks after

the adoption of the constitution as before related, a fine vessel, in full
sail, appeared near the island. Brusque and Piquet saw it with a
mixture of emotions. She seemed to be crowding all her sails, and
sweeping before a brisk breeze. When first seen, masts and sails
only were visible, but now her full hull was in view. At length, she
came so near that both Brusque and his companion could distinctly
see the people on board.
The scene recalled the mind of Brusque to his home and his
country. The ship bore aloft the flag of France, and stirred within him
feelings that he could not well define. There are few that can forget
the land of their birth, particularly if parents, and one loved more
warmly than kindred, be there. Brusque’s mind touched on all these
points, and tears filled his eyes. “I am an outcast,” said he, “and
France rejects me. I am unworthy of my parents, and, more than all,
unworthy of Emilie. I must teach my heart to forget; and yet I fear it
will not forget, till it ceases to feel.” With these words he sat down
upon the hill, folded his arms, and with a melancholy countenance
gazed at the ship as she now seemed flying past the island.
At this moment, a new object attracted his attention; this was
another vessel, of small bulk, but with a prodigious spread of
canvass, pursuing the first-mentioned ship. She seemed, like the
sea-eagle, to have a vast expanse of wing in proportion to her body.
On she flew, and was soon near the object of her pursuit. Brusque
and his companion watched the scene with interest. Both saw that
the pursuing vessel was a pirate ship, and that in a few minutes a
desperate conflict must follow.
The pirate had now come abreast of the island, being at the
distance of not more than three miles. Brusque saw a white roll of
smoke uncoil itself at her side, and in a few seconds the booming
voice of the cannon broke over the island. At the same time, the ball
was seen to strike the water beyond the ship, and dipping at short
distances, made the spray shoot high into the air. Another and
another shot followed from the pirate in quick succession. These
were at length returned by the ship. The two now approached. Peal
after peal rung on the air. They were both completely wrapt in
smoke. Yet still the firing continued. At length there was a dreadful
volley as of a broadside, a thickening of the smoke, and then a
fearful silence. Slowly the coiling vapor was lifted up, and the two
ships were in view. All eyes seemed directed to the larger ship. Her
masts and the cloud of canvass swayed heavily from side to side.
Finally, they sank lower and lower, and with a heavy crash fell into
the waves.
The deck was now a scene of confusion. The pirate approached,
and was soon grappled to the ship. Swiftly a few of her men leaped
upon the deck. There was a short struggle, and all was still. “They
have yielded like a pack of cowardly hounds!” said Brusque to his
companion. “Nay,” said the fisherman, “they fought bravely. That
piratical craft has five hands to her one, for she has more than a
hundred men on board. The other is but a merchant vessel, and had
not twenty seamen. The greater part of the men who fought are
passengers, and they fought bravely. Beside, there were women
among them!”
“How do you know that?” said Brusque, quickly.
“I saw them,” said Piquet, “as the vessel passed.”
“What is to be done?” said Brusque, jumping up.
“What can you do?” said the other.
“What can I do?” said Brusque; “good God, I can do nothing: and
women on board! women to fall into the hands of these pirates! It is
too dreadful to think of. I will go down to the shore.”
“Stay,” said the fisherman; “if you show yourself we are both lost.
The ship cannot be taken away, but must remain. It is likely the
pirates will come ashore before they leave. It is now near sunset. Let
us wait for events.”
“You are right, you are right!” said Brusque. “We will watch till
evening. Perhaps something may turn up, by which we may aid the
captives. And yet I know not what we can do. We have no weapons,
no boat. Still, what we can do, we will do.”
With these resolutions, Brusque and his companion went to their
cave, and laid their plans. Considering it extremely probable that the
pirates would come ashore, they concluded to watch and wait for
circumstances. Agreeing to take separate stations, and meet again
at midnight, they parted, it being now dark.
Brusque had not waited long before he heard the regular dipping
of oars in the direction of the pirate ship, and soon saw a boat with
about twenty men approaching the shore. Getting into the cover of
some bushes, he waited till they reached the shore. They were soon
followed by another party of an equal number. Drawing their boats
upon the beach, and leaving a single sailor as a guard, the whole
party moved up to a little grassy hill. Here some sat down, and
others stood around. The leader of the party gave directions to six of
his men to go in search of water; taking two officers with him, he
stepped aside, leaving the rest to themselves. While they were
talking and laughing, the captain and his two friends sat down close
to the bushes where Brusque lay concealed, and began to talk over
the events of the battle.
The question was soon started as to the disposal of the ship and
her inmates. It was agreed by all that the vessel must be scuttled.
“Shall the people go down with her?” asked one of the officers.
“What think you, Jaques?” said the captain. “As to the sailors, and
those rascally passengers that entered into the fight, let them die,”
said Jaques. “It’s the fortune of war, and I shall care as little for their
death as for the bursting of so many bubbles. But the women——”
“Well, what of the women?” said the captain.
“Why,” said Jaques, “one of them is very pretty, and one of them
is very old, and I do not like to be concerned in drowning either a
pretty woman or an old one. They are very likely to haunt a man after
death. Beside, there are thirty women in all; it will be too bad to tip
them all into the sea.”
“Well,” said the captain, “what is your plan?”
“Well,” said Jaques, “I propose that we pick out the prettiest for
ourselves, and send the rest ashore here to take care of themselves.
They can set up a petticoat republic, or any other government they
please.”
This plan occasioned a hearty laugh, but still it seemed to be
approved. The party soon broke up and joined the rest. Brusque had
heard the whole of their conversation, and, after a short time, crept
from his hiding-place, and set out to join the fisherman at the cave.
On his way he fell in with one of the pirates who was in search of
water. He had no chance to conceal himself, but as it was dark, he
spoke to the man, as if he were one of his comrades. “Have you
found any water?” said he. “Not a drop,” said the other. “Well, go with
me,” said Brusque, “and I will take you to a spring. I have been on
this island before. A long time ago, on a voyage we stopped here,
and I remember that between these two hills there was a fine
spring.”
“Indeed,” said the other, “is it you, Tom? Really, I did not know
you; your voice is strangely changed.” “I’ve got a cold,” said
Brusque, coughing. “But we are near the place, I think. It’s so dark
we may not be able to find it. However, we can but try. Yes, here is
the spot—I remember it by this tall palm-tree. I can see the shape of
it against the sky, and know it is the same. The spring is within ten
feet of this place. Aye, here it is! How delightful it will be to get a
drink of fresh water, just from the ground. It’s as good to drink direct
from mother earth, as in infancy to draw milk from a mother’s
breast.”
“Get out, you sentimental dog!” said the other. “It’s treason to
remind a pirate of his mother. Good God, I never dare to think of
mine.”
“Is she living?” said Brusque.
“Is she living? How dare you speak to me of my mother? Is she
living? Good God, I know too well that she is living. Tell me, Tom,
and tell me truly!—suppose your mother was in that ship, what would
you do? Nay, more,—suppose your sister were there, pure as an
angel from heaven, and as beautiful too? Yes, and suppose your
aged father, bowed with toil and care and sorrow, and gray with
years, were also in that ship? And suppose you were the pirate that
had aided in their capture? What would you do?”
“Tell me, in the name of Heaven, tell me your name!” said
Brusque, in great agitation.
“You know my name is François——” The man hesitated.
“Yes, indeed, I do know your name; you are François Bonfils. You
are the brother of Emilie—and here before you is Philip Brusque!”
The pirate started at this, and drawing a pistol from his belt, stood
in an attitude of defiance. At the same time he said, “Am I betrayed?
What means this? Are you not Tom Garson, of our ship?” Brusque
hastened to explain, and in few words told his story to François. It
was a scene of mutual agitation and explanation. Each had many
questions to ask, but these were deferred that they might consider
what was to be done. For the sake of conversing freely, they retired
to Brusque’s cave, where they both agreed to attempt the rescue of
the people on board the ship. Piquet soon arrived, and he joined
heartily in the enterprise. Several plans were discussed, but none
seemed feasible. At length, François spoke as follows:
“I am afraid that we are too sanguine. There are two hundred men
belonging to the pirate. They are desperate freebooters, and armed
to the teeth. Like all rogues, they are suspicious and watchful. We
cannot hope to surprise or deceive them. The captured vessel is a
trading ship, from St. Domingo. She is filled with people that have
fled from an insurrection of the negroes there. There are about thirty
females, several children, and thirty or forty men. They are guarded
by ten of our marines, and are kept under the hatches. We must
convey instructions to them to be on the lookout for relief, that they
may exert themselves if any opportunity should offer. We must blow
up the pirate ship, and I will do it, and share the fate of the rest, if
need be.”
“Nay,” said Brusque, “this is a mad and desperate scheme. Let us
think of something more feasible.”
“It is time,” said François, “for me to return to the captain. I shall
be missed and suspected. I will take care to be in the watch of the
merchant ship to-morrow night. You, Brusque, are a good swimmer.
The vessel is not more than two miles out. You must come at twelve
o’clock, and I will see that a rope is over the stem. You must climb
up, and enter the dead-lights, which shall be prepared. You must
then wait till Heaven send you some opportunity for exertion.
Mention me not to my parents or Emilie, if I perish. It will be better for
them to mourn over an uncertainty, than the memory of a pirate son
or brother. Farewell!” Saying this, and wringing Brusque’s hand
convulsively, the pirate departed.
I shall pass over the scene of riot which took place among the
pirates on the island, next day, as well as the anxiety of Brusque and
his friend Piquet. Night at length came, and at the appointed hour
Brusque repaired to the shore, and began to swim toward the vessel,
as directed by François. It was dark, and the water was ruffled, but
he could see the vessel floating like a dusky shade upon the water,
and being steady of limb and stout of heart, and withal an excellent
swimmer, he soon neared the vessel. Cautiously and slowly
approaching the stern, he at length descried a tall sentinel standing
on the deck, and thought he could make out the figure of François.
He then drew close, and at length was able to find the promised
rope. Climbing up by this, he swung himself to the window, which
was cautiously opened from within. It was too dark to see any one,
but he entered the cabin and sat down. Pretty soon a boat started
from the side of the ship, and looking through the window, Brusque
saw it set off toward the pirate vessel. He thought he could trace in
the athletic form of the man who guided the helm of the boat the
form of François, and he began to think seriously that he intended to
put his plan into execution. He was the more fearful of this from
having observed that all the pirates had left the island, and he
suspected that the opportunity of thus blowing the whole into air was
too powerful a temptation for the almost maddened mind of François.
Pondering upon the awful chances of such an event, and of the
action that must follow on the part of the ship’s crew and passengers
for liberation, should it take place, he sat for some time in silence. At
length, a hand was laid upon his arm, and he was told to follow.
Being led across the cabin, he was taken into a small state-room,
where there was a light. His guide left him here alone. Soon a man
entered, who announced himself as the captain. He said he had
received an intimation that an effort would be made for their relief,
but he knew nothing more. Brusque now entered into a detail of the
circumstances which we have related, and expressed his conviction
that the pirate vessel would be blown up. He advised the captain
quietly to apprize all the men on board of the prospect before them,
and to see that they were ready to second any effort that should be
made. This plan was adopted, and accordingly, about twenty-five
men got together in the cabin, each having provided himself with
some club, or spar, or other weapon. The captain alone had a sword
and pistol, which he had found concealed in a drawer, and which had
escaped the search of the pirates.
Brusque now took place on the transom of the vessel, where he
could have a full view of the pirate ship. He sat long, earnestly
watching the object of his attention. He hardly knew whether to fear
or hope for the awful explosion that he anticipated. The sudden
transition of two hundred breathing men from life to death, from the
full flush of riotous passion and crime into the presence of their God,
was a thought too horrible to be dwelt upon. Yet, here were other
men, and helpless women and children, whose only chance for life
or escape from a fate worse than death, seemed to depend upon
that fearful catastrophe. Dwelling upon these agitating topics,
Brusque sat in the darkness, gazing upon the pirate ship. In his
anxiety, seconds seemed to lengthen into minutes, and minutes into
hours. His impatience almost mastered him. His heart beat audibly,
and his brain seemed swelled to bursting. He was on the point of
starting up to relieve his feelings, when he saw a stream of light like
a rocket shoot out from the side of the pirate vessel. In an instant,
another and another followed, and then one wide flash enveloped
the whole firmament. In the midst of the sea of fire that seemed
thrown into the sky, were the fragments of the ship, the wheels of
cannon, and the mangled forms of men, seeming like demons, lit up
in the red and ghastly glare.
This mighty blaze was almost instantly followed by total darkness,
by a heavy sound, and by a rocking of the ship, as if struck by a
gale. In an instant, the men within, rushed against the hatches, and
with one united effort threw them open. Starting to the deck, they
soon levelled four of the sentinels with their weapons, and the rest,
in the sudden panic, leaped into the sea.
The inmates of the ship now found themselves restored to liberty,
as if by the hand of enchantment. Passing from the deepest
despondency, they indulged in the most violent transports of joy.
Brusque made himself known to his parents, and he and Emilie
found out each other in the darkness. I need not tell the rest, till we
get into another chapter; and that must be deferred to our next
number.
(To be continued.)
The Spectre of the Brocken.
I will now tell you of certain strange appearances, which are

sometimes produced by clouds, operating like mirrors, and reflecting
upon the sky the images of things on the earth.
In Germany, there is a range of elevations, called the Hartz
Mountains. The Brocken is the loftiest peak, and is said to be about
three fourths of a mile high. The view from the top of it is so
extensive as to embrace a tract of land inhabited by more than five
millions of people.
Now these reflecting clouds of which I have spoken, sometimes
collect around this mountain, and bear a very distinct though
shadowy image of whatever may be on the summit of the Brocken,
when the sun is rising. It is remarkable that this image is greatly
magnified, so that if a man is on the mountain, his figure upon the
cloud is as tall as a steeple. The best account of this wonderful
spectacle is given by a very learned Frenchman, called Hauy. He
visited the place in 1797. I give his own account of what he saw,
which is as follows:
“After having come here for the thirteenth time, I was at length so
fortunate as to have the pleasure of seeing the spectre. The sun
rose about four o’clock, and the atmosphere was quite serene. I was
looking round to see whether the atmosphere would permit me to
have a free prospect of the southwest, when I observed at a very
great distance, toward one of the other mountains, what seemed like
a human figure, of a monstrous size. A violent gust of wind having
almost carried off my hat, I clapped my hand to my head, and the
colossal figure did the same.
“The pleasure which I felt at this discovery can hardly be
described; for I had already walked many a weary step, in the hopes
of seeing this shadowy image, without being able to gratify my
curiosity. I immediately made another movement by bending my
body, and the colossal figure before me repeated it. I was desirous of
doing the same thing once more, but my colossus had vanished. I
remained in the same position, waiting to see whether it would
return, and, in a few minutes, it again made its appearance on the
mountain.
“I paid my respects to it a second time, and it did the same to me.
I then called the landlord of the Brocken, and, having both taken the
same position, we looked towards the mountain, but saw nothing.
We had not, however, stood long, when two colossal figures were
formed in the same situation, which repeated our compliments by
bending their bodies as we did, after which they vanished.
“We retained our position, kept our eyes fixed on the same spot,
and, in a little while, the two figures again stood before us, and were
joined by a third, which was most likely the double reflection of one
of us. Every movement that we made by bending our bodies these
figures imitated, but with this difference, that the phenomenon was
sometimes weak and faint, and sometimes strong and well defined.”
There are many other interesting stories relating to these
reflecting clouds, but I have not room to tell them here. You will find
them in one of Parley’s books, entitled, “Wonders of the Earth, Sea,
and Sky,” from which I have been permitted to copy this account and
the engraving that accompanies it.
Trifles.
“Father, didn’t you say the world was round?”

“Yes, my son.”
“Well, how can it come to an end if it’s round?”
“William, I wish you wouldn’t talk with your mouth so full of
victuals.”
“John, I wish you wouldn’t go to balls and parties—it is very bad

indeed.”
“Father, didn’t you and mother go to balls and parties, when you
were young?”
“Yes, my son—but we have seen the folly of it.”
“Well, I want to see the folly of it too, father!”
View of Malta.
The Travels, Adventures, and Experiences of
Thomas Trotter.
(Continued from page 47.)
CHAPTER II.
A Wreck at Sea.—Mother Carey’s Chickens.—A Gale of Wind.—
Singular Phenomenon of the Corpo Santo.—Arrival at the
Straits of Gibraltar.—Wonderful Fortifications of that place.
When we had sailed about half way across the Atlantic, we fell in
with the wreck of a vessel. All her masts were gone, and the sea was
breaking over her in every part. We could not discover her name, nor
to what nation she belonged. When a ship meets with a wreck at
sea, it is customary to set the wrecked vessel on fire, or blow her up
with gunpowder, lest any other vessel should run foul of her in the
night; a casualty which has caused the destruction of many ships,
that have never been heard of afterwards. The wreck we met with
lay so low in the water that we found it impossible to get at her for
this purpose. So the most we could do was steer clear of her. She
was surrounded by a great shoal of black-fish.
Now and then the solitude of the ocean was enlivened by the
sight of a little dark-colored bird, about the size of a swallow, called
the Stormy Petrel, but among sailors known by the name of Mother
Carey’s Chicken. These birds are met with in every part of the
ocean, thousands of miles from the land. They fly very swiftly, and
come fluttering about the ship, but seldom light on the rigging or
deck. The sailors have many superstitious notions concerning them,
and always look out for a storm after their appearance; but I never
found there was any dependence to be placed on such
prognostications. They believe also that these birds never set foot on
land, that they lay their eggs at sea, and hatch them under their
wings. But these stories are all fables. The petrels lay their eggs on
the shore, among the rocks and sand. Their nests are often found in
the Bahama Islands.
We had now got about two thirds of the way across the ocean,
when the wind died away, and we lay two or three days becalmed.
The sea was as quiet as a mill-pond, and as smooth as glass. The
captain did nothing but fret and fidget, for the master of a ship cannot
endure any delay on his voyage. About the third day there rose a
heavy swell of the sea, which caused the vessel to roll from side to
side in a manner most uncomfortable to us all. I was surprised at
this, as there was no wind to agitate the water; but the captain
informed me that when a gale of wind is approaching, the swell
always comes before the wind. He now told us to look out for a
heavy blow. The mercury in the barometer had fallen suddenly,
which is a pretty sure indication of a storm at hand. By-and-by, a
mass of thick, heavy clouds began to rise in the west, and soon the
heavens were completely overspread. The surface of the water
quickly became agitated by ripples, and the swell increased. The
wind now began to snuffle, then to blow in heavy gusts and sing
through the cordage in a most alarming style. We close-reefed the
topsails and scudded before it. The gale came on harder and harder,
and the seas rolled around us in a most terrific manner. Now and
then the crest of a mountainous wave would dash over the stern and
sweep the deck fore and aft. At such times the sailors were obliged
to cling fast to the spars and rigging, to save themselves from being
washed overboard.
In the midst of the gale I was astonished at the sight of a
wonderful flame of fire that came hovering round the ship. It was a
bright, thin, quivering mass of light, as big as a man’s head,
somewhat like the sun when seen through a fog or thin haze. From
what quarter it came I could not discern—whether from the clouds or
the sea, but the captain said it appeared to gather in the air. It
hovered over us for some minutes, and then settled on one of the
lifts or ropes which sustain the upper yards. There it remained two or
three minutes, after which it glided down the stay to the bowsprit,
and then disappeared. I must confess I was greatly amazed at this
strange phenomenon, which, happening in the midst of a terrible
storm, was certainly enough to frighten any common person. The
captain, however, told me not to be alarmed, for such appearances,
though not very common, were yet too well known at sea to cause
any fear to an experienced mariner. This strange luminous body is
called by the sailors a corposant, a corruption of the Portuguese
words corpo santo, “holy body.” It is a sort of meteor, engendered
probably from electrical matter in the air, and never appears but in
heavy gales of wind. Sometimes two of them appear together. After
their disappearance, the sailors believe the strength of the gale to be
broken. In fact, within an hour after the appearance of this, which I
saw, the wind began to lull, and ere long subsided to a moderate
breeze, so that we considered ourselves out of danger, and stood on
our course.
About a week after this, just as I had waked in the morning, I was
aroused by the cry of “Land!” I ran upon deck, and saw what no man
can see for the first time without feelings of indescribable enthusiasm
—the shores of the old world! We were directly abreast of the straits
of Gibraltar. Europe and Africa lay before me, and the sun was rising
behind the lofty ridge of the Atlas mountains. Were I to live a
thousand years, I should never forget this moment, nor the
overpowering emotions that took possession of me at the sight. Few
prospects in the world can be more imposing. The stern and craggy
cliffs of the Spanish coast; the towering wood-crowned peaks of the
African mountains; the noble strait that separates these two famous
quarters of the globe; and the grand and interesting historical
recollections connected with the spot—all combine to fill the mind of
the spectator with the most thrilling emotions. Long did I gaze on the
noble scene without the power to utter a word, as the sun broke from
the mass of rich blue clouds that hung round the head of Mount
Atlas, and poured his golden light on the shaggy masses of forest in
Africa and the rugged and frowning cliffs of Spain. To see such a
prospect once is an epoch in a man’s life; the vivid and overpowering
feelings of the moment are never to be experienced a second time.
As we sailed up the strait, I had leisure to view the shore on both
sides by the help of a telescope. The Spanish coast is rocky, and
generally barren, but in many spots I was able to discern little
patches of green cultivation, scattered about in the valleys between
the dark rock. The African shore is almost entirely covered with
woods up to the mountain-tops. Here and there I could see a wreath
of white smoke slowly curling upward from the thick woods. These
were made by the Moors, who were stripping the cork trees of their
bark. Farther up the strait, we came in sight of the famous fortress of
Gibraltar. It is an enormous rock, connected with the Spanish shore
by a low, flat beach. The rock is cut and tunnelled into immensely
long caverns and galleries, with embrasures for cannon, and is
fortified in every part so strongly as to be considered impregnable. It
was taken from the Spaniards by the English, more than a century
ago, but at that time it was very poorly fortified. The English, finding it
so well situated for guarding the entrance of the strait, expended
vast sums of money in strengthening it, and would never give it up to
the Spaniards. It has sustained many hard sieges since that period,
but has hitherto resisted every attack. There is always a strong
garrison of troops kept here, and the harbor is a regular station for
ships of war. A considerable town has grown up near the rock, and a
good deal of trade is carried on by the merchants of Gibraltar.
Vessels from all the Mediterranean ports bring their goods to this
place, and American vessels carry the productions of our continent
to exchange for them; so that an establishment designed at first only
for a military fortress, has become a flourishing commercial mart.
Boston vessels commonly carry to Gibraltar cargoes of flour,
tobacco, coffee, tar, pipe-staves, &c., and take the Spanish wines
and fruits in return. Sometimes, after disposing of their cargoes at
Gibraltar, they take in ballast and sail for the Cape Verd Islands,
where they load with salt and return home.
CHAPTER III.
Voyage along the coast of Spain.—Prospect of Sicily.—Account of
an Island thrown up from the bottom of the sea by a Volcano.

Differentiable Manifolds A Theoretical Physics Approach 2nd Edition Gerardo F. Torres Del Castillo All Chapter Instant Download

Uploaded by

Copyright:

Available Formats

Differentiable Manifolds A Theoretical Physics Approach 2nd Edition Gerardo F. Torres Del Castillo All Chapter Instant Download

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Differentiable Manifolds A Theoretical Physics Approach 2nd Edition Gerardo F. Torres Del Castillo All Chapter Instant Download

Uploaded by

Copyright:

Available Formats

Full download test bank at ebook textbookfull.

CLICK LINK TO DOWLOAD

Crafting a Republic for the World Scientific Geographic

Skin Cancer Management A Practical Approach 2nd Edition

Coastal Wetlands: An Integrated Ecosystem Approach

Theoretical and computational physics of gas discharge

Theoretical Concepts in Physics An Alternative View of

Type 2 Fuzzy Logic in Control of Nonsmooth Systems

Calculus On Manifolds: A Modern Approach To Classical

Coherent Quantum Physics A Reinterpretation of the

ISBN 978-3-030-45192-9 ISBN 978-3-030-45193-6 (eBook)

1st edition: © Springer Science+Business Media, LLC 2012

As in the ﬁrst edition, I have tried to maintain the presentation at an elementary

Puebla, Puebla, México Gerardo F. Torres del Castillo

Puebla, Puebla, México Gerardo F. Torres del Castillo

6.4 Volume Element, Divergence, and Duality of Differential

Appendix A: Lie Algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

1.1 Differentiable Manifolds

As mentioned above, the basic idea in the definition of a differentiable manifold is

is a complex-valued function of the coordinates employed to label the points of

Fig. 1.1 A coordinate

{φ( p) | p ∈ U }) be an open subset of Rn . If, for instance, the coordinate x n could be

Definition 1.1 A C k manifold of dimension n is a set M with a C k atlas; if k 1, it

Example 1.2 (The standard structure of Rn as a manifold.) The space Rn has a

Example 1.3 (The Euclidean plane as a manifold.) An example closely related

 1.4 In the set of 2 × 2 real matrices with determinant equal to 1, M =

Then we find that φ(U ) = {(x, y, z) ∈ R3 | x = 0}, which is an open subset of R3 .

shows that φ is one-to-one.

We find that χ (V ) = {(x, y, z) ∈ R3 | y = 0}, which is an open subset of R3 , and we

which is a differentiable function (of class C ∞ ) in χ (U ∩ V ) = {(x, y, z) ∈ R3

which is a differentiable function (of class C ∞ ) in φ(U ∩ V ) = {(x, y, z) ∈ R3

(n 1) can be put into a one-to-one correspondence with the points of Rn by

(b1 , . . . , bn , 0) − (0, 0, . . . , 1) = λ (a 1 , . . . , a n+1 ) − (0, 0, . . . , 1) , (1.4)

so that (V, χ ), with V ≡ Sn \ {(0, 0, . . . , −1)}, is a second chart of coordinates

We have U ∩ V = Sn \ {(0, 0, . . . , 1), (0, 0, . . . , −1)}; hence φ(U ∩ V ) = χ (U ∩

Let M be a manifold. A subset A of M is said to be open if for any chart (U, φ)

Fig. 1.4 With the aid of a M

it follows that f ◦ φ −1 is differentiable if and only if f ◦ χ −1 is. Hence, it makes

(i.e., f is the trace) is differentiable of class C ∞ . According to the definition above,

which is differentiable (of class C ∞ ) in φ(U ) = {(x, y, z) ∈ R3 | x = 0}, and

which is differentiable (of class C ∞ ) in χ (V ) = {(x, y, z) ∈ R3 | y = 0}.

For a coordinate system (U, φ) belonging to the atlas of M, and a real-valued

f ( p) = ( f ◦ φ −1 )(φ( p)) = F(φ( p)) = F(x 1 ( p), x 2 ( p), . . . , x n ( p)),

Fig. 1.5 The map ψ : M → N is locally represented by the map χ ◦ ψ ◦ φ −1 . ψ is differentiable

for p ∈ U , which we shall write as f = F(x 1 , x 2 , . . . , x n ); in this manner, the func-

A diffeomorphism ψ is a one-to-one map from a differentiable manifold M to a

Example 1.9 The hyperboloid N ≡ {(x, y, z) ∈ R3 | x 2 + y 2 − z 2 = −1, z > 0} is

Since, by the definition of D, a 2 + b2 < 1, we see that the composition χ ◦ ψ ◦ φ −1

and this is also differentiable.

φi (N ∩ Ui ) = {(a 1 , a 2 , . . . , a n ) ∈ Rn | a m+1 = a m+2 = · · · = a n = 0}.

Let π be the canonical projection from Rn onto Rm given by π(a 1 , a 2 , . . . , a n ) =

Proposition 1.11 Let f 1 , f 2 , . . . , f m be real-valued differentiable functions defined

on all of R3 . From these expressions and their inverses, x = v2 + w2 − u, y = v,

Example 1.15 The mapping

is a differentiable curve of class C ∞ in the manifold defined in Example 1.4. Indeed,

The functions φ ◦ C and χ ◦ C are differentiable mappings from R into R3 of class

If ψ is a differentiable map from M to a differentiable manifold N and f ∈

From the relation (ψ ∗ f ) ◦ φ −1 = ( f ◦ χ −1 ) ◦ (χ ◦ ψ ◦ φ −1 ) it follows that ψ ∗ f ∈

1.2 The Tangent Space

1.4 In the set of 2 × 2 real matrices with determinant equal to 1, M =

But, from (1.13), Ct0 [x i ] = d(x i ◦ C)/dt|t0 ; therefore

If (V, χ ) is a second chart on M with coordinate functions y 1 , y 2 , . . . , y n , and