Tensor Algebra and Tensor Analysis for

Engineers With Applications to

Continuum Mechanics Fourth Edition
Mikhail Itskov
Visit to download the full and correct content document:
https://textbookfull.com/product/tensor-algebra-and-tensor-analysis-for-engineers-with
-applications-to-continuum-mechanics-fourth-edition-mikhail-itskov/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Tensor Algebra and Tensor Analysis for Engineers With

Applications to Continuum Mechanics Third Edition
Mikhail Itskov

https://textbookfull.com/product/tensor-algebra-and-tensor-
analysis-for-engineers-with-applications-to-continuum-mechanics-
third-edition-mikhail-itskov/

Tensor Analysis for Engineers 2nd Edition Mehrzad

Tabatabaian

https://textbookfull.com/product/tensor-analysis-for-
engineers-2nd-edition-mehrzad-tabatabaian/

Tensor Calculus for Engineers and Physicists Emil De

Souza Sánchez Filho

https://textbookfull.com/product/tensor-calculus-for-engineers-
and-physicists-emil-de-souza-sanchez-filho/

Notes on Tensor Products and the Exterior Algebra

Patrick Brosnan

https://textbookfull.com/product/notes-on-tensor-products-and-
the-exterior-algebra-patrick-brosnan/
Principles of Tensor Calculus Tensor Calculus First
Edition Taha Sochi

https://textbookfull.com/product/principles-of-tensor-calculus-
tensor-calculus-first-edition-taha-sochi/

Fundamentals of Tensor Calculus for Engineers with a

Primer on Smooth Manifolds 1st Edition Uwe Mühlich
(Auth.)

https://textbookfull.com/product/fundamentals-of-tensor-calculus-
for-engineers-with-a-primer-on-smooth-manifolds-1st-edition-uwe-
muhlich-auth/

Tensor Calculus and Applications: Simplified Tools and

Techniques 1st Edition Bhaben Chandra Kalita

https://textbookfull.com/product/tensor-calculus-and-
applications-simplified-tools-and-techniques-1st-edition-bhaben-
chandra-kalita/

Mathematical Analysis of Continuum Mechanics and

Industrial Applications III Proceedings of the
International Conference CoMFoS18 Hiromichi Itou

https://textbookfull.com/product/mathematical-analysis-of-
continuum-mechanics-and-industrial-applications-iii-proceedings-
of-the-international-conference-comfos18-hiromichi-itou/

Size Dependent Continuum Mechanics Approaches Theory

and Applications 1st Edition Esmaeal Ghavanloo

https://textbookfull.com/product/size-dependent-continuum-
mechanics-approaches-theory-and-applications-1st-edition-esmaeal-
ghavanloo/
Mathematical Engineering

Mikhail Itskov

Tensor Algebra and

Tensor Analysis for
Engineers
With Applications to Continuum
Mechanics
Fourth Edition
Mathematical Engineering

Series editors
Claus Hillermeier, Neubiberg, Germany
Jörg Schröder, Essen, Germany
Bernhard Weigand, Stuttgart, Germany
More information about this series at http://www.springer.com/series/8445
Mikhail Itskov

Tensor Algebra and Tensor

Analysis for Engineers
With Applications to Continuum Mechanics
Fourth Edition

123
Mikhail Itskov
Department of Continuum Mechanics
RWTH Aachen University
Aachen
Germany

ISSN 2192-4732 ISSN 2192-4740 (electronic)

Mathematical Engineering
ISBN 978-3-319-16341-3 ISBN 978-3-319-16342-0 (eBook)
DOI 10.1007/978-3-319-16342-0

Library of Congress Control Number: 2015934223

Springer Cham Heidelberg New York Dordrecht London

© Springer International Publishing Switzerland 2015
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.

Printed on acid-free paper

Springer International Publishing AG Switzerland is part of Springer Science+Business Media

(www.springer.com)
Preface to the Fourth Edition

In this edition some new examples dealing with the inertia tensor and the propa-
gation of compression and shear waves in an isotropic linear-elastic medium are
incorporated. Section 3.3 is completely revised and enriched by an example of thin
membranes under hydrostatic pressure. The so derived Laplace law is illustrated
there by a thin wall vessel of torus form under internal pressure. In Chap. 8
I introduced a section concerned with the deformation of a line, area and volume
element and some accompanying kinematic identities. Similar to the previous
edition some new exercises and solutions are added.

Aachen, December 2014 Mikhail Itskov

vii
Preface to the Third Edition

This edition is enriched by some new examples, problems and solutions, in par-
ticular, concerned with simple shear. I also added an example with the derivation
of constitutive relations and tangent moduli for hyperelastic materials with the
isochoric-volumetric split of the strain energy function. Besides, Chap. 2 is com-
pleted with some new figures, for instance, illustrating spherical coordinates. These
figures have again been prepared by Uwe Navrath. I also gratefully acknowledge
Khiêm Ngoc Vu for careful proofreading of the manuscript. At this opportunity,
I would also like to thank Springer-Verlag and in particular Jan-Philip Schmidt for
the fast and friendly support in getting this edition published.

Aachen, February 2012 Mikhail Itskov

ix
Preface to the Second Edition

This second edition is completed by a number of additional examples and exercises.

In response to comments and questions of students using this book, solutions of
many exercises have been improved for better understanding. Some changes and
enhancements are concerned with the treatment of skew-symmetric and rotation
tensors in the first chapter. Besides, the text and formulae have been thoroughly
reexamined and improved where necessary.

Aachen, January 2009 Mikhail Itskov

xi
Preface to the First Edition

Like many other textbooks the present one is based on a lecture course given by the
author for master students of the RWTH Aachen University. In spite of a somewhat
difficult matter those students were able to endure and, as far as I know, are still
fine. I wish the same for the reader of the book.
Although the present book can be referred to as a textbook one finds only little
plain text inside. I tried to explain the matter in a brief way, nevertheless going into
detail where necessary. I also avoided tedious introductions and lengthy remarks
about the significance of one topic or another. A reader interested in tensor algebra
and tensor analysis but preferring, however, words instead of equations can close
this book immediately after having read the preface.
The reader is assumed to be familiar with the basics of matrix algebra and
continuum mechanics and is encouraged to solve at least some of the numerous
exercises accompanying every chapter. Having read many other texts on mathe-
matics and mechanics, I was always upset vainly looking for solutions to the
exercises which seemed to be the most interesting for me. For this reason, all the
exercises here are supplied with solutions amounting a substantial part of the book.
Without doubt, this part facilitates a deeper understanding of the subject.
As a research work this book is open for discussion which will certainly con-
tribute to improving the text for further editions. In this sense, I am very grateful for
comments, suggestions and constructive criticism from the reader. I already expect
such criticism, for example, with respect to the list of references which might be far
from complete. Indeed, throughout the book I only quote the sources indispensable
to follow the exposition and notation. For this reason, I apologize to colleagues
whose valuable contributions to the matter are not cited.
Finally, a word of acknowledgment is appropriate. I would like to thank Uwe
Navrath for having prepared most of the figures for the book. Further, I am grateful
to Alexander Ehret who taught me the first steps as well as some “dirty” tricks in
LaTeX, which were absolutely necessary to bring the manuscript to a printable

xiii
xiv Preface to the First Edition

form. He and Tran Dinh Tuyen are also acknowledged for careful proofreading and
critical comments to an earlier version of the book. My special thanks go to
Springer-Verlag and in particular to Eva Hestermann-Beyerle and Monika Lempe
for their friendly support in getting this book published.

Aachen, November 2006 Mikhail Itskov

Contents

1 Vectors and Tensors in a Finite-Dimensional Space . . . . . . . . . . . 1

1.1 Notion of the Vector Space . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Basis and Dimension of the Vector Space . . . . . . . . . . . . . . . 3
1.3 Components of a Vector, Summation Convention . . . . . . . . . . 5
1.4 Scalar Product, Euclidean Space, Orthonormal Basis . . . . . . . . 6
1.5 Dual Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Second-Order Tensor as a Linear Mapping. . . . . . . . . . . . . . . 13
1.7 Tensor Product, Representation of a Tensor
with Respect to a Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.8 Change of the Basis, Transformation Rules . . . . . . . . . . . . . . 21
1.9 Special Operations with Second-Order Tensors . . . . . . . . . . . . 22
1.10 Scalar Product of Second-Order Tensors . . . . . . . . . . . . . . . . 28
1.11 Decompositions of Second-Order Tensors . . . . . . . . . . . . . . . 30
1.12 Tensors of Higher Orders. . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2 Vector and Tensor Analysis in Euclidean Space . . . . . . . . ...... 37

2.1 Vector- and Tensor-Valued Functions,
Differential Calculus . . . . . . . . . . . . . . . . . . . . . . . . ...... 37
2.2 Coordinates in Euclidean Space, Tangent Vectors . . . . ...... 39
2.3 Coordinate Transformation. Co-, Contra-
and Mixed Variant Components . . . . . . . . . . . . . . . . ...... 43
2.4 Gradient, Covariant and Contravariant Derivatives . . . ...... 45
2.5 Christoffel Symbols, Representation of the Covariant
Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 51
2.6 Applications in Three-Dimensional Space: Divergence
and Curl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 54

3 Curves and Surfaces in Three-Dimensional Euclidean Space. . . . . 69

3.1 Curves in Three-Dimensional Euclidean Space . . . . . . . . . . . . 69
3.2 Surfaces in Three-Dimensional Euclidean Space . . . . . . . . . . . 76
3.3 Application to Shell Theory . . . . . . . . . . . . . . . . . . . . . . . . . 84

xv
xvi Contents

4 Eigenvalue Problem and Spectral Decomposition

of Second-Order Tensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.1 Complexification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.2 Eigenvalue Problem, Eigenvalues and Eigenvectors. . . . . . . . . 99
4.3 Characteristic Polynomial. . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.4 Spectral Decomposition and Eigenprojections . . . . . . . . . . . . . 104
4.5 Spectral Decomposition of Symmetric Second-Order
Tensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 109
4.6 Spectral Decomposition of Orthogonal and Skew-Symmetric
Second-Order Tensors . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 112
4.7 Cayley-Hamilton Theorem . . . . . . . . . . . . . . . . . . . . . . . . .. 116

5 Fourth-Order Tensors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 121

5.1 Fourth-Order Tensors as a Linear Mapping . . . . . . . . . . .... 121
5.2 Tensor Products, Representation of Fourth-Order Tensors
with Respect to a Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.3 Special Operations with Fourth-Order Tensors . . . . . . . . . . . . 125
5.4 Super-Symmetric Fourth-Order Tensors . . . . . . . . . . . . . . . . . 128
5.5 Special Fourth-Order Tensors . . . . . . . . . . . . . . . . . . . . . . . . 130

6 Analysis of Tensor Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

6.1 Scalar-Valued Isotropic Tensor Functions. . . . . . . . . . . . . . . . 135
6.2 Scalar-Valued Anisotropic Tensor Functions. . . . . . . . . . . . . . 139
6.3 Derivatives of Scalar-Valued Tensor Functions . . . . . . . . . . . . 142
6.4 Tensor-Valued Isotropic and Anisotropic Tensor Functions . . . 152
6.5 Derivatives of Tensor-Valued Tensor Functions . . . . . . . . . . . 159
6.6 Generalized Rivlin’s Identities . . . . . . . . . . . . . . . . . . . . . . . 164

7 Analytic Tensor Functions. . . . . . . . . . . . . . . . . . . . .......... 169

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 169
7.2 Closed-Form Representation for Analytic Tensor
Functions and Their Derivatives . . . . . . . . . . . . .......... 173
7.3 Special Case: Diagonalizable Tensor Functions. . .......... 176
7.4 Special Case: Three-Dimensional Space. . . . . . . .......... 179
7.5 Recurrent Calculation of Tensor Power Series
and Their Derivatives . . . . . . . . . . . . . . . . . . . .......... 185

8 Applications to Continuum Mechanics . . . . . . . . . . . . ......... 191

8.1 Deformation of a Line, Area and Volume Element ......... 191
8.2 Polar Decomposition of the Deformation Gradient ......... 193
8.3 Basis-Free Representations for the Stretch
and Rotation Tensor . . . . . . . . . . . . . . . . . . . . . ......... 194
8.4 The Derivative of the Stretch and Rotation Tensor
with Respect to the Deformation Gradient . . . . . . ......... 197
Contents xvii

8.5 Time Rate of Generalized Strains . . . . . . . . . . . . . . . . . . . . . 201

8.6 Stress Conjugate to a Generalized Strain . . . . . . . . . . . . . . . . 204
8.7 Finite Plasticity Based on the Additive Decomposition
of Generalized Strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

9 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
9.1 Exercises of Chap. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
9.2 Exercises of Chap. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
9.3 Exercises of Chap. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
9.4 Exercises of Chap. 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
9.5 Exercises of Chap. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
9.6 Exercises of Chap. 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
9.7 Exercises of Chap. 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
9.8 Exercises of Chap. 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Chapter 1
Vectors and Tensors in a Finite-Dimensional
Space

1.1 Notion of the Vector Space

We start with the definition of the vector space over the field of real numbers R.

Definition 1.1 A vector space is a set V of elements called vectors satisfying the
following axioms.

A. To every pair, x and y of vectors in V there corresponds a vector x + y, called

the sum of x and y, such that
(A.1) x + y = y + x (addition is commutative),
(A.2) (x + y) + z = x + ( y + z) (addition is associative),
(A.3) there exists in V a unique vector zero 0, such that 0 + x = x, ∀x ∈ V,
(A.4) to every vector x in V there corresponds a unique vector −x such that
x + (−x) = 0.
B. To every pair α and x, where α is a scalar real number and x is a vector in V,
there corresponds a vector αx, called the product of α and x, such that
(B.1) α (βx) = (αβ) x (multiplication by scalars is associative),
(B.2) 1x = x,
(B.3) α (x + y) = αx + α y (multiplication by scalars is distributive with respect
to vector addition),
(B.4) (α + β) x = αx + βx (multiplication by scalars is distributive with respect
to scalar addition),
∀α, β ∈ R, ∀x, y ∈ V.

Examples of vector spaces.

(1) The set of all real numbers R.
(2) The set of all directional arrows in two or three dimensions. Applying the usual
definitions for summation, multiplication by a scalar, the negative and zero vector
(Fig. 1.1) one can easily see that the above axioms hold for directional arrows.

© Springer International Publishing Switzerland 2015 1

M. Itskov, Tensor Algebra and Tensor Analysis for Engineers,
Mathematical Engineering, DOI 10.1007/978-3-319-16342-0_1
2 1 Vectors and Tensors in a Finite-Dimensional Space

x +y =y +x

x x
−x

y
vector addition negative vector

2.5x

2x

zero vector

multiplication by a real scalar

Fig. 1.1 Geometric illustration of vector axioms in two dimensions

(3) The set of all n-tuples of real numbers R:

⎧ ⎫
⎪
⎪ a1 ⎪
⎪
⎪
⎪ ⎪
⎨ 2⎪
a ⎬
a= . .
⎪
⎪ ⎪
⎪
⎪ . ⎪
⎪
⎩ ⎪ ⎭
an

Indeed, the axioms (A) and (B) apply to the n-tuples if one defines addition,
multiplication by a scalar and finally the zero tuple, respectively, by
⎧ ⎫ ⎧ ⎫ ⎧ ⎫
⎪
⎪ a1 + b1 ⎪
⎪ ⎪
⎪ αa1 ⎪
⎪ ⎪
⎪ 0⎪⎪
⎪
⎪ ⎪ ⎪ ⎪ ⎪ ⎪
⎨ a2 + b2 ⎪
⎬ ⎪
⎨ αa2 ⎪
⎬ ⎨0⎪
⎪ ⎬
a+b= . , αa = . , 0= . .
⎪
⎪ ⎪
⎪ ⎪
⎪ ⎪ ⎪ ⎪
⎪
⎪ . ⎪
⎪ ⎪
⎪ . ⎪⎪
⎪
⎪
⎪
⎪ .⎪⎪
⎩ ⎭ ⎩ ⎭ ⎩ ⎪ ⎭
an + bn αan 0

(4) The set of all real-valued functions defined on a real line.

1.2 Basis and Dimension of the Vector Space 3

1.2 Basis and Dimension of the Vector Space

Definition 1.2 A set of vectors x 1 , x 2 , . . . , x n is called linearly dependent if there

exists a set of corresponding scalars α1 , α2 , . . . , αn ∈ R, not all zero, such that
n
αi x i = 0. (1.1)
i=1

Otherwise, the vectors x 1 , x 2 , . . . , x n are called linearly independent. In this case,

none of the vectors x i is the zero vector (Exercise 1.2).

Definition 1.3 The vector

n
x= αi x i (1.2)
i=1

is called linear combination of the vectors x 1 , x 2 , . . . , x n , where αi ∈ R (i =

1, 2, . . . , n).

Theorem 1.1 The set of n non-zero vectors x 1 , x 2 , . . . , x n is linearly dependent if

and only if some vector x k (2 ≤ k ≤ n) is a linear combination of the preceding ones
x i (i = 1, . . . , k − 1).

Proof If the vectors x 1 , x 2 , . . . , x n are linearly dependent, then

n
αi x i = 0,
i=1

where not all αi are zero. Let αk (2 ≤ k ≤ n) be the last non-zero number, so that
αi = 0 (i = k + 1, . . . , n). Then,

k k−1
−αi
αi x i = 0 ⇒ x k = xi .
αk
i=1 i=1

Thereby, the case k = 1 is avoided because α1 x 1 = 0 implies that x 1 = 0

(Exercise 1.1). Thus, the sufficiency is proved. The necessity is evident.

Definition 1.4 A basis in a vector space V is a set G ⊂ V of linearly independent

vectors such that every vector in V is a linear combination of elements of G. A vector
space V is finite-dimensional if it has a finite basis.

Within this book, we restrict our attention to finite-dimensional vector spaces.

Although one can find for a finite-dimensional vector space an infinite number of
bases, they all have the same number of vectors.
4 1 Vectors and Tensors in a Finite-Dimensional Space

Theorem 1.2 All the bases of a finite-dimensional vector space V contain the same
number of vectors.
Proof Let G = g 1 , g 2 , . . . , g n and F = f 1 , f 2 , . . . , f m be two arbitrary bases
of V with different numbers of elements, say m > n. Then, every vector in V is a
linear combination of the following vectors:

f 1 , g1 , g2 , . . . , gn . (1.3)

These vectors are non-zero and linearly dependent. Thus, according to Theorem 1.1
we can find such a vector g k , which is a linear combination of the preceding ones.
Excluding this vector we obtain the set G  by

f 1 , g 1 , g 2 , . . . , g k−1 , g k+1 , . . . , g n

again with the property that every vector in V is a linear combination of the elements
of G  . Now, we consider the following vectors

f 1 , f 2 , g 1 , g 2 , . . . , g k−1 , g k+1 , . . . , g n

and repeat the excluding procedure just as before. We see that none of the vectors
f i can be eliminated in this way because they are linearly independent. As soon as
all g i (i = 1, 2, . . . , n) are exhausted we conclude that the vectors

f 1 , f 2 , . . . , f n+1

are linearly dependent. This contradicts, however, the previous assumption that they
belong to the basis F.
Definition 1.5 The dimension of a finite-dimensional vector space V is the number
of elements in a basis of V.
Theorem 1.3 Every set F = f 1 , f 2 , . . . , f n of linearly independent vectors in
an n-dimensional vectors space V forms a basis of V. Every set of more than n
vectors is linearly dependent.
Proof The proof of this theorem is similar to the preceding one. Let G = g 1 , g 2 ,
. . . , g n be a basis of V. Then, the vectors (1.3) are linearly dependent and non-
zero. Excluding a vector g k we obtain a set of vectors, say G  , with the property
that every vector in V is a linear combination of the elements of G  . Repeating this
procedure we finally end up with the set F with the same property. Since the vectors
f i (i = 1, 2, . . . , n) are linearly independent they form a basis of V. Any further
vectors in V, say f n+1 , f n+2 , . . . are thus linear combinations of F. Hence, any set
of more than n vectors is linearly dependent.
Theorem 1.4 Every set F = f 1 , f 2 , . . . , f m of linearly independent vectors in
an n-dimensional vector space V can be extended to a basis.
1.2 Basis and Dimension of the Vector Space 5

Proof If m = n, then F is already a basis according to Theorem 1.3. If m < n,

then we try to find n − m vectors f m+1 , f m+2 , . . . , f n , such that all the vectors f i ,
that is, f 1 , f 2 , . . . , f m , f m+1 , . . . , f n are linearly independent and consequently
form a basis. Let us assume, on the contrary, that only k < n − m such vectors can
be found. In this case, for all x ∈ V there exist scalars α, α1 , α2 , . . . , αm+k , not all
zero, such that

αx + α1 f 1 + α2 f 2 + . . . + αm+k f m+k = 0,

where α = 0 since otherwise the vectors f i (i = 1, 2, . . . , m + k) would be

linearly dependent. Thus, all the vectors x of V are linear combinations of f i
(i = 1, 2, . . . , m + k). Then, the dimension of V is m + k < n, which contradicts
the assumption of this theorem.

1.3 Components of a Vector, Summation Convention

Let G = g 1 , g 2 , . . . , g n be a basis of an n-dimensional vector space V. Then,

n
x= x i g i , ∀x ∈ V. (1.4)
i=1

Theorem 1.5 The representation (1.4) with respect to a given basis G is unique.

Proof Let
n n
x= x gi
i
and x = y i gi
i=1 i=1

be two different representations of a vector x, where not all scalar coefficients x i and
y i (i = 1, 2, . . . , n) are pairwise identical. Then,
n n n
0 = x + (−x) = x + (−1) x = x i gi + −y i g i = x i − y i gi ,
i=1 i=1 i=1

where we use the identity −x = (−1) x (Exercise 1.1). Thus, either the numbers
x i and y i are pairwise equal x i = y i (i = 1, 2, . . . , n) or the vectors g i are lin-
early dependent. The latter one is likewise impossible because these vectors form a
basis of V.

The scalar numbers x i (i = 1, 2, . . . , n) in the representation (1.4) are called

components of the vector x with respect to the basis G = g 1 , g 2 , . . . , g n .
6 1 Vectors and Tensors in a Finite-Dimensional Space

The summation of the form (1.4) is often used in tensor algebra. For this reason
it is usually represented without the summation symbol in a short form by
n
x= x i gi = x i gi (1.5)
i=1

referred to as Einstein’s summation convention. Accordingly, the summation is

implied if an index appears twice in a multiplicative term, once as a superscript and
once as a subscript. Such a repeated index (called dummy index) takes the values
from 1 to n (the dimension of the vector space in consideration). The sense of the
index changes (from superscript to subscript or vice versa) if it appears under the
fraction bar.

1.4 Scalar Product, Euclidean Space, Orthonormal Basis

The scalar product plays an important role in vector and tensor algebra. The properties
of the vector space essentially depend on whether and how the scalar product is
defined in this space.
Definition 1.6 The scalar (inner) product is a real-valued function x· y of two vectors
x and y in a vector space V, satisfying the following conditions.
C. (C.1) x · y = y · x (commutative rule),

(C.2) x · ( y + z) = x · y + x · z (distributive rule),

(C.3) α (x · y) = (αx) · y = x · (α y) (associative rule for the multiplication

by a scalar), ∀α ∈ R, ∀x, y, z ∈ V,

(C.4) x · x ≥ 0 ∀x ∈ V, x · x = 0 if and only if x = 0.

An n-dimensional vector space furnished by the scalar product with properties (C.1–
C.4) is called Euclidean space En . On the basis of this scalar product one defines the
Euclidean length (also called norm) of a vector x by
√
x = x · x. (1.6)

A vector whose length is equal to 1 is referred to as unit vector.

Definition 1.7 Two non-zero vectors x and y are called orthogonal (perpendicular),
denoted by x⊥ y, if

x · y = 0. (1.7)

Of special interest is the so-called orthonormal basis of the Euclidean space.

1.4 Scalar Product, Euclidean Space, Orthonormal Basis 7

Definition 1.8 A basis E = {e1 , e2 , . . . , en } of an n-dimensional Euclidean space

En is called orthonormal if

ei · e j = δij , i, j = 1, 2, . . . , n, (1.8)

where

1 for i = j,
δij = δ =
ij
δ ij = (1.9)
0 for i = j

denotes the Kronecker delta.

Thus, the elements of an orthonormal basis represent pairwise orthogonal unit
vectors. Of particular interest is the question of the existence of an orthonormal
basis. Now, we are going to demonstrate that every set of m ≤ n linearly inde-
pendent vectors in En can be orthogonalized and normalized by means of a linear
transformation (Gram-Schmidt procedure). In other words, starting from linearly
independent vectors x 1 , x 2 , . . . , x m one can always construct their linear combi-
nations e1 , e2 , . . . , em such that ei · e j = δij (i, j = 1, 2, . . . , m). Indeed, since
the vectors x i (i = 1, 2, . . . , m) are linearly independent they are all non-zero (see
Exercise 1.2). Thus, we can define the first unit vector by
x1
e1 = . (1.10)
x1

Next, we consider the vector

e2 = x 2 − (x 2 · e1 ) e1 (1.11)
 
orthogonal to e1 . This holds for the unit vector e2 = e2 /e2  as well. It is also seen
  
that e2  = e2 · e2 = 0 because otherwise e2 = 0 and thus x 2 = (x 2 · e1 ) e1 =
(x 2 · e1 ) x 1 −1 x 1 . However, the latter result contradicts the fact that the vectors
x 1 and x 2 are linearly independent.
Further, we proceed to construct the vectors

e
e3 = x 3 − (x 3 · e2 ) e2 − (x 3 · e1 ) e1 , e3 =  3  (1.12)
e 
3

orthogonal to e1 and e2 . Repeating this procedure we finally obtain the set of ortho-
normal vectors e1 , e2 , . . . , em . Since these vectors are non-zero and mutually orthog-
onal, they are linearly independent (see Exercise 1.6). In the case m = n, this set
represents, according to Theorem 1.3, the orthonormal basis (1.8) in En .
8 1 Vectors and Tensors in a Finite-Dimensional Space

With respect to an orthonormal basis the scalar product of two vectors x = x i ei

and y = y i ei in En takes the form

x · y = x 1 y1 + x 2 y2 + · · · + x n yn . (1.13)

For the length of the vector x (1.6) we thus obtain the Pythagoras formula

x = x 1 x 1 + x 2 x 2 + · · · + x n x n , x ∈ En . (1.14)

1.5 Dual Bases

Definition 1.9 Let G = g 1 , g 2 , . . . , g n be a basis in the n-dimensional Euclidean

space En . Then, a basis G  = g 1 , g 2 , . . . , g n of En is called dual to G, if

j
g i · g j = δi , i, j = 1, 2, . . . , n. (1.15)

In the following we show that a set of vectors G  = g 1 , g 2 , . . . , g n satisfying the

conditions (1.15) always exists, is unique and forms a basis in En .
Let E = {e1 , e2 , . . . , en } be an orthonormal basis in En . Since G also represents
a basis, we can write
j j
ei = αi g j , g i = βi e j , i = 1, 2, . . . , n, (1.16)

j j
where αi and βi (i = 1, 2, . . . , n) denote the components of ei and g i , respectively.
Inserting the first relation (1.16) into the second one yields

j j
g i = βi αkj g k , ⇒ 0 = βi αkj − δik g k , i = 1, 2, . . . , n. (1.17)

Since the vectors g i are linearly independent we obtain

j
βi αkj = δik , i, k = 1, 2, . . . , n. (1.18)

Let further

g i = αij e j , i = 1, 2, . . . , n, (1.19)

where and henceforth we set e j = e j ( j = 1, 2, . . . , n) in order to take the advantage

of Einstein’s summation convention. By virtue of (1.8), (1.16) and (1.18) one finally
finds
j j j j
g i · g j = βik ek · αl el = βik αl δkl = βik αk = δi , i, j = 1, 2, . . . , n. (1.20)
1.5 Dual Bases 9

Next, we show that the vectors g i (i = 1, 2, . . . , n) (1.19) are linearly independent

and for this reason form a basis of En . Assume on the contrary that

ai g i = 0,

where not all scalars ai (i = 1, 2, . . . , n) are zero. Multiplying both sides of this
relation scalarly by the vectors g j ( j = 1, 2, . . . , n) leads to a contradiction. Indeed,
using (1.170) (see Exercise 1.5) we obtain

0 = ai g i · g j = ai δ ij = a j , j = 1, 2, . . . , n.

The next important question is whether the dual basis is unique. Let G  = g 1 , g 2 , . . . ,
g n and H = h1 , h2 , . . . , hn be two arbitrary non-coinciding bases in En , both
dual to G = g 1 , g 2 , . . . , g n . Then,

hi = h ij g j , i = 1, 2, . . . , n.

Forming the scalar product with the vectors g j ( j = 1, 2, . . . , n) we can conclude

that the bases G  and H coincide:

δ ij = hi · g j = h ik g k · g j = h ik δ kj = h ij ⇒ hi = g i , i = 1, 2, . . . , n.

Thus, we have proved the following theorem.

Theorem 1.6 To every basis in an Euclidean space En there exists a unique dual
basis.
Relation (1.19) enables to determine the dual basis. However, it can also be obtained
without any orthonormal basis. Indeed, let g i be a basis dual to g i (i = 1, 2, . . . , n).
Then

g i = g ij g j , g i = gij g j , i = 1, 2, . . . , n. (1.21)

Inserting the second relation (1.21) into the first one yields

g i = g ij gjk g k , i = 1, 2, . . . , n. (1.22)

Multiplying scalarly with the vectors gl we have by virtue of (1.15)

δli = g ij gjk δlk = g ij g jl , i, l = 1, 2, . . . , n. (1.23)

   
Thus, we see that the matrices gk j and g k j are inverse to each other such that
   
−1
g k j = gk j . (1.24)
10 1 Vectors and Tensors in a Finite-Dimensional Space

Now, multiplying scalarly the first and second relation (1.21) by the vectors g j and
g j ( j = 1, 2, . . . , n), respectively, we obtain with the aid of (1.15) the following
important identities:

g ij = g ji = g i · g j , gij = gji = g i · g j , i, j = 1, 2, . . . , n. (1.25)

By definition (1.8) the orthonormal basis in En is self-dual, so that

j
ei = ei , ei · e j = δi , i, j = 1, 2, . . . , n. (1.26)

With the aid of the dual bases one can represent an arbitrary vector in En by

x = x i g i = xi g i , ∀x ∈ En , (1.27)

where

x i = x · g i , xi = x · g i , i = 1, 2, . . . , n. (1.28)

Indeed, using (1.15) we can write

x · g i = x j g j · g i = x j δ ij = x i ,
j
x · g i = x j g j · g i = x j δi = xi , i = 1, 2, . . . , n.

The components of a vector with respect to the dual bases are suitable for calculating
the scalar product. For example, for two arbitrary vectors x = x i g i = xi g i and
y = y i g i = yi g i we obtain

x · y = x i y j gij = xi y j g ij = x i yi = xi y i . (1.29)

The length of the vector x can thus be written by

  
x = xi x j g ij = x i x j gij = xi x i . (1.30)

Example 1.1 Dual basis in E3 . Let G = g 1 , g 2 , g 3 be a basis of the three-

dimensional Euclidean space and
 
g = g1 g2 g3 , (1.31)

where [• • •] denotes the mixed product of vectors. It is defined by

[abc] = (a × b) · c = (b × c) · a = (c × a) · b, (1.32)

where “×” denotes the vector (also called cross or outer) product of vectors. Consider
the following set of vectors:
1.5 Dual Bases 11

g 1 = g −1 g 2 × g 3 , g 2 = g −1 g 3 × g 1 , g 3 = g −1 g 1 × g 2 . (1.33)

It is seen that the vectors (1.33) satisfy conditions (1.15), are linearly independent
(Exercise 1.11) and consequently form the basis dual to g i (i = 1, 2, 3). Further, it
can be shown that
 
g 2 = gij  , (1.34)

where |•| denotes the determinant of the matrix [•]. Indeed, with the aid of (1.16)2
we obtain
   j

g = g 1 g 2 g 3 = β1i ei β2 e j β3k ek
   
j j  
= β1i β2 β3k ei e j ek = β1i β2 β3k eijk = β ij  , (1.35)

where eijk denotes the permutation symbol (also called Levi-Civita symbol). It is
defined by
 
eijk = eijk = ei e j ek
⎧
⎨ 1 if ijk is an even permutation of 123,
= −1 if ijk is an odd permutation of 123, (1.36)
⎩
0 otherwise,
where the orthonormal vectors e1 , e2 and e3 are numerated in such a way that they
form a right-handed system. In this case, [e1 e2 e3 ] = 1.
On the other hand, we can write again using (1.16)2

3
gij = g i · g j = βik β kj .
k=1

The latter sum can be represented as a product of two matrices so that

   j   j T
gij = βi βi . (1.37)

Since the determinant of the matrix product is equal to the product of the matrix
determinants we finally have

   j 2
gij  = β  = g 2 . (1.38)
i

With the aid of the permutation symbol (1.36) one can write
 
g i g j g k = eijk g, i, j, k = 1, 2, 3, (1.39)
12 1 Vectors and Tensors in a Finite-Dimensional Space

which by (1.28)2 yields an alternative representation of the identities (1.33) as

g i × g j = eijk g g k , i, j = 1, 2, 3. (1.40)

Similarly to (1.35) one can also show that (see Exercise 1.12)
 
g 1 g 2 g 3 = g −1 (1.41)

and
 ij 
g  = g −2 . (1.42)

Thus,
  eijk
gi g j gk = , i, j, k = 1, 2, 3, (1.43)
g

which yields by analogy with (1.40)

eijk
gi × g j = g , i, j = 1, 2, 3. (1.44)
g k

Relations (1.40) and (1.44) permit a useful representation of the vector product.
Indeed, let a = a i g i = ai g i and b = b j g j = b j g j be two arbitrary vectors in E3 .
Then, in view of (1.32)
 1 2 3
a a a 
 
a × b = a g i × b g j = a b eijk gg = g  b1 b2 b3  ,
i j i j k
 g1 g2 g3 
 
 a1 a2 a3 
1  
a × b = ai g i × b j g j = ai b j eijk g −1 g k =  b1 b2 b3  . (1.45)
g g g g 
1 2 3

For the orthonormal basis in E3 relations (1.40) and (1.44) reduce to

ei × e j = eijk ek = eijk ek , i, j = 1, 2, 3, (1.46)

so that the vector product (1.45) can be written by

 
 a1 a2 a3 
 
a × b =  b1 b2 b3  , (1.47)
 e1 e2 e3 

where a = ai ei and b = b j e j .
1.6 Second-Order Tensor as a Linear Mapping 13

1.6 Second-Order Tensor as a Linear Mapping

Let us consider a set Linn of all linear mappings of one vector into another one within
En . Such a mapping can be written as

y = Ax, y ∈ En , ∀x ∈ En , ∀A ∈ Linn . (1.48)

Elements of the set Linn are called second-order tensors or simply tensors. Linearity
of the mapping (1.48) is expressed by the following relations:

A (x + y) = Ax + A y, ∀x, y ∈ En , ∀A ∈ Linn , (1.49)

A (αx) = α (Ax) , ∀x ∈ En , ∀α ∈ R, ∀A ∈ Linn . (1.50)

Further, we define the product of a tensor by a scalar number α ∈ R as

(αA) x = α (Ax) = A (αx) , ∀x ∈ En (1.51)

and the sum of two tensors A and B as

(A + B) x = Ax + Bx, ∀x ∈ En . (1.52)

Thus, properties (A.1), (A.2) and (B.1–B.4) apply to the set Linn . Setting in (1.51)
α = −1 we obtain the negative tensor by

−A = (−1) A. (1.53)

Further, we define a zero tensor 0 in the following manner

0x = 0, ∀x ∈ En , (1.54)

so that the elements of the set Linn also fulfill conditions (A.3) and (A.4) and accord-
ingly form a vector space.
The properties of second-order tensors can thus be summarized by

A + B = B + A, (addition is commutative), (1.55)

A + (B + C) = (A + B) + C, (addition is associative), (1.56)

0 + A = A, (1.57)

A + (−A) = 0, (1.58)

α (βA) = (αβ) A, (multiplication by scalars is associative), (1.59)

14 1 Vectors and Tensors in a Finite-Dimensional Space

1A = A, (1.60)

α (A + B) = αA + αB, (multiplication by scalars is distributive

with respect to tensor addition), (1.61)

(α + β) A = αA + βA, (multiplication by scalars is distributive

with respect to scalar addition), ∀A, B, C ∈ Linn , ∀α, β ∈ R. (1.62)

Example 1.2 Vector product in E3 . The vector product of two vectors in E3 repre-
sents again a vector in E3

z = w × x, z ∈ E3 , ∀w, x ∈ E3 . (1.63)

According to (1.45) the mapping x → z is linear (Exercise 1.16) so that

w × (αx) = α (w × x) ,
w × (x + y) = w × x + w × y, ∀w, x, y ∈ E3 , ∀α ∈ R. (1.64)

Thus, it can be described by means of a tensor of the second order by

w × x = Wx, W ∈ Lin3 , ∀x ∈ E3 . (1.65)

The tensor which forms the vector product by a vector w according to (1.65) will be
denoted in the following by ŵ. Thus, we write

w × x = ŵx. (1.66)

Clearly

0̂ = 0. (1.67)

Example 1.3 Representation of a rotation by a second-order tensor. A rotation of a

vector a in E3 about an axis yields another vector r in E3 . It can be shown that the
mapping a → r (a) is linear such that

r (αa) = αr (a) , r (a + b) = r (a) + r (b) , ∀α ∈ R, ∀a, b ∈ E3 . (1.68)

Thus, it can again be described by a second-order tensor as

r (a) = Ra, ∀a ∈ E3 , R ∈ Lin3 . (1.69)

This tensor R is referred to as rotation tensor.

1.6 Second-Order Tensor as a Linear Mapping 15

Fig. 1.2 Finite rotation of a

vector in E3
e

x y
ω

a∗
a r (a)

Let us construct the rotation tensor which rotates an arbitrary vector a ∈ E3 about
an axis specified by a unit vector e ∈ E3 (see Fig. 1.2). Decomposing the vector a by
a = a∗ + x in two vectors along and perpendicular to the rotation axis we can write
 
r (a) = a∗ + x cos ω + y sin ω = a∗ + a − a∗ cos ω + y sin ω, (1.70)

where ω denotes the rotation angle. By virtue of the geometric identities

 
a∗ = (a · e) e = (e ⊗ e) a, y = e × x = e × a − a∗ = e × a = êa, (1.71)

where “⊗” denotes the so-called tensor product (1.83) (see Sect. 1.7), we obtain

r (a) = cos ωa + sin ω êa + (1 − cos ω) (e ⊗ e) a. (1.72)

Thus the rotation tensor can be given by

R = cos ωI + sin ω ê + (1 − cos ω) e ⊗ e, (1.73)

where I denotes the so-called identity tensor (1.92) (see Sect. 1.7).
Another useful representation for the rotation tensor can be obtained utilizing the
fact that x = y × e = −e × y. Indeed, rewriting (1.70) by

r (a) = a + x (cos ω − 1) + y sin ω (1.74)

and keeping (1.71)2 in mind we receive

 2
r (a) = a + sin ω êa + (1 − cos ω) ê a. (1.75)
Another random document with
no related content on Scribd:
 A Sanitary Drinking Tube

To make this tube, procure a piece of ³⁄₄-in. gauge glass, about 10

in. long, and bend it by heating the glass with a blowtorch 2¹⁄₂ in.
from one end, as shown. If this is done in a place where no air will
strike the glass while it is heated and the tube is then allowed to cool
slowly, it will not crack. It is used by putting the short end under the
faucet and drawing on it as on a pipe.
 A Comb Cleaner

The Twisted Wires Form an Excellent Device for Cleaning a Comb

A very handy device for cleaning combs can be made from a piece
of spring sheet brass, 4¹⁄₂ in. long and about 3 in. wide. Notches are
cut in the ends of the brass with a hacksaw, making the projections
as wide as the saw cut. The brass is then bent into shape with a
special clamp made for the purpose, which consists of three pieces
of wood, two being cut to form a curved slot, then fastened to the
third piece. The brass is sprung into the slot, and then fine wire is
stretched between the ends in the notches. The wire should be very
fine and two strands twisted together and run through the notches.
 Concealing the House Key
The time-honored custom of concealing the house key under the
door mat, or in the letter box, when the family has not enough keys
to go around, is so well known that an unauthorized person seeking
to enter the house would look in these places first of all.

A simple and effective hiding place for the key can be quickly and
easily made with the aid of an auger and two pieces of tin. Pick out
an obscure section of the porch railing, and in the edge of this bore a
³⁄₄-in. hole, about ¹⁄₄ in. deeper than the length of the key. Make a
piece of tin into a cylinder, the same length as the key, so that the
latter will slide easily into the hole. At one end of this cylinder solder
a 1-in. disk of tin, which will make it appear as in the illustration.
If the key is placed in the cylinder and the latter pushed into the
hole until it is flush with the surface, it will scarcely be noticed by
anyone not in the secret, and by painting it the same color as the
railing it will become still more inconspicuous.—Contributed by Frank
L. Matter, Portland, Ore.
 A Mysterious Watch

A very interesting experiment may be made with the ordinary

dollar watch in illustrating the law of the pendulum. A pendulum 39.1
in. long will make 60 one-way swings per minute, the number of
swings varying inversely as the square root of the length. By actual
count it was found that the balance wheel of the watch in question
made 240 one-way swings per minute, which is just 4 times as fast
as the 39-in. pendulum. Therefore, according to the foregoing law, a
pendulum ¹⁄₁₆ as long, or about 2¹⁄₂ in., would swing in unison with
the wheel of the watch. The question then arises as to what would
happen if the watch itself were suspended so as to swing as a
pendulum of the latter length. The experiment was made as
illustrated, with the result that the watch keeps on swinging
continuously. The swing amounts to about ³⁄₈ in., and appears so
vigorous that it is almost incomprehensible that the small spring in a
watch should be able to maintain so much weight in continuous
motion for 24 hours.
 Bait Casting for
Game Fish
by Stillman Taylor

There are many enjoyable phases of the fisher’s art, but bait casting
from the free reel probably has a wider appeal than any other
branch of angling. The pleasure of handling a short bait-casting rod
is in itself a good sport, even when casting done in the back yard is
made for distance and accuracy only. Get two or three enthusiastic
casters together, and you may have an interesting little tournament,
held on a vacant lot, on the lake, or on the greensward of a city park.
There is plenty of action in casting from the reel, and it is its variety
that has made angling of this type so universally popular. True,
considerable practice is required before the caster is able to shoot
an accurate plug far off, but the knack may be gained after
reasonable application. The handling of the short rod differs from all
other angling methods—it is an active sport rather than a
contemplative recreation—and when the fun of handling a good outfit
is combined with the sport of coaxing out a black bass, shooting a
wooden minnow among the lily pads for pickerel, or casting the bait
for those pirates of our fresh-water lakes, the wall-eyed pike and the
muskellunge, the angler gets a taste of fishing—plus.

The Outfit and Its Selection

A good firearm is necessary to the good marksman and a well-

made, nicely balanced rod, to which is clamped a smooth-running
reel, is the essential equipment of an able angler. Bait casting is, of
course, more a matter of skill than of outfit, but that is no reason why
the angler should be handicapped by poor tackle, although large
expenditure is not necessary in obtaining a satisfactory outfit. The
casting rod may be from 5 to 6 ft. long. Where the caster regards
length of cast as the important factor, the 5-ft. length will probably
give the best satisfaction. However, the superiority of one over the
other in this regard is rather to be found in the elasticity and quality
of the rod than its length. The skill of the caster is, moreover, the
ruling factor. The 5¹⁄₂-ft. rod is better than the 5-ft. length for fishing
purposes, and the 6-ft. rod is more resilient than the 5¹⁄₂-ft. tip. If the
greatest range is not of more importance than the landing of the fish,
the longer rod will prove the most pleasant to handle.
The split-bamboo rod has many friends, and a well-made rod of
this material is a thoroughly satisfactory casting tool. A fine-quality
casting rod of split bamboo costs from $10 to $25. A fairly good rod
may be bought for $2 or a trifle less, and a smooth-running reel for
another $2, while a line for it will bring the total up to about $5. An
outfit of this kind purchased with discrimination, while not embodying
everything that a critical angler would desire, will cast a long and
accurate line and stand up under the strain of handling our heaviest
game fish. The one-piece rod is classed by many anglers as the
ideal casting rod. The two-piece rod is handier to pack and is more
generally used.
The chief fault in bait-casting rods is the stiff action of the tip
section. For tournament casting, a fairly stiff rod is desirable. The
same requirement holds good for muskellunge and other heavy
fishing. A flexible rod affords more pleasure in playing black bass
and other active “gamey” fish. However, the best way to test out a
rod is by casting with it.

A Serviceable Reel Costing $4, a High-Grade Reel Costing $13.50, and a Fine
Tournament Casting Reel Costing $26, are Shown from Left to Right. The
Center One Has a Level-Winding Device and a Thumb-Click Drag on the
Rims of the Plates. The Tournament Casting Reel Has a Cork Arbor and
Jeweled Bearings

The steel casting rod is well liked by some casters, but only a
high-quality steel rod is worthy of consideration. It is generally not
classed as the equal of a well-made solid-wood or split-bamboo rod,
for casting. The better-grade steel rod is springy and flexible, and for
this reason is a good rod for casting for black bass. Still, most
anglers have a steel rod in their outfits, but it is not as generally used
at tournaments as the time-tried wooden rod.
The guide is an important detail of a casting rod. Inexpensive rods
are usually provided with large metal guides, while the better rods
are fitted with agate guides, or with agate tip and hand guides. A
guide, ¹⁄₂ in. in diameter, at the butt of the rod is large enough and
will handle the line smoothly.
The single grip is sometimes preferred by casters, but the
additional grip placed above the reel is desirable if much fishing is
done. Solid-cork hand grasps are most satisfactory, being less
slippery when wet and less tiring to the hand than handles made of
other materials. If the lower grip is properly shaped with a forward
edge, the forefinger will find a firm grip upon it, and the finger hook
will not be needed. However, some anglers prefer to use the hook,
and if the rod is not so fitted, a detachable hook may be fastened to
it. A reel band that locks securely is desirable, and all the better
grades of casting rods are so fitted.
 The quadruple reel is the logical choice of the bait caster, the
spool turning four times to one turn of the handle. As the work
demanded of a bait-casting reel is different and more exacting than
that of the click reel used in fly casting, its design is radically
different. The best type of click reel is of large spool diameter and
narrow between plates, while the typical casting reel is its direct
opposite—being long-barreled, with plates of comparatively small
diameter. For tournament casting the finest outfit will naturally be
selected, but for practical work it is unnecessary to pay $25 for a
finely wrought reel. Three typical reels are shown in the illustration.
The one at the left holds 80 yd. of line, can be easily taken apart for
cleaning, and costs $4. In the center is a level-winding reel with
thumb click and adjustable drag placed on the rims of the plates. It
holds 60 yd. of line, and costs $13.50. At the right is a very high-
grade reel for tournament casting. It has a cork arbor, jeweled
bearings holds 60 yd. of line, and costs $26.
Any of the better grades of braided silk are suitable for the bait-
casting line. The average price for a dependable line is about $1 for
50 yd. A small size line is best, and sizes G and E are preferred. A
brown and green, or other dark-colored, line is preferable for fishing.
The black and white, all white and other light-colored, lines are
satisfactory for tournament use. After a day’s fishing the line should
be unreeled and looped around the back of a chair in loose coils to
dry, or loosely wound upon a drying reel.
For practice and tournament casting one of the standard ¹⁄₄ and
¹⁄₂-oz. loaded wooden or metal plugs may be used. The popularity of
bait casting from the reel has brought forth a large variety of artificial
baits. These wooden and enameled lures are used by anglers
because they bring the catches.
The wooden minnows, and other forms of artificial bait, may be
roughly classed in three divisions: the surface lures for use on top of
the water; the diving and under-water baits; and the weedless
varieties, which are usually of the surface type. No bait caster’s outfit
is complete unless it includes one bait of each class, and to meet
fully the different and varying conditions of weather, season, and
water, it is a good plan to select one of each type in light and
medium-dark colors. A variety of typical baits are shown in the page
plate at A, B, C, etc. Many of the baits are provided with one or more
eyes for attaching the line, thus enabling the caster to spin his bait at
different depths. All of the wooden-body baits will float, and only dive
under the surface when reeled in; the quicker they are retrieved the
deeper they spin. While many of these baits have been designed
especially to entice the black bass, practically all of them are good
for pickerel and general fresh-water casting, particularly the minnow
baits.
The other items which complete the bait caster’s outfit are the
landing net and tackle box. The net should be well made, with a steel
hoop 13 in. or larger in diameter, and should have a handle about 3
ft. long. The folding type of net is the most convenient, and is
provided with a separate handle. The tackle box may be of sheet
metal or leather, with partitions for holding reels, baits, and the
angling “what not.” A convenient and serviceable landing net costs
$1 or more, and a tackle box will cost about as much.

Suggestions on Making the Cast

Although much practice is required before the caster can take his
place in a tournament, “getting the hang” of handling the short rod
and free-running reel comes quickly. As soon as the simple
principles are understood, length and accuracy will come only
through practice. Bait casting is easier than fly casting and the angler
need concern himself only with the length and accuracy of his cast,
since delicacy does not enter into this phase of fishing. The coarser
sweet-water fish, like black bass and pickerel, are not frightened
when the bait hits the water with a splash—indeed, the splash of the
bait attracts them.
Two methods of casts are used by bait casters, the overhead and
the underhand or side cast. The overhead cast is used largely by the
practical angler, as well as by those who take part in tournament
contests, because the bait is projected a longer distance and may be
more accurately placed. If the novice will thumb the line, as shown in
Fig. 1, with thumb bearing directly on the cross rod of the reel with
the ball of the thumb resting lightly on the line, one or two trials will
show the correct manner of spooling. The line is retrieved as shown
in Fig. 2, care being taken to form correct habits from the start. To
make the cast, reel in the bait until it is some 6 in. from the tip,
release the click or drag, and press the ball of the thumb firmly upon
the spooled line. Carry the rod over the shoulder until it is horizontal
or nearly so, as in Fig. 3, and then bring it smartly forward, checking
it quickly when it reaches an angle of some 45°, as shown in Fig. 4.
This projects the bait forward in a straight line. When the rod is
brought forward the tension of the thumb on the spool is released so
that the spool will revolve freely, but enough pressure must be
communicated by the thumb to prevent the reel from spinning faster
than the line is unreeled from the spool, otherwise backlash will
occur. The use of the whole arm and the weight of the body,
combined with the elasticity of the pliant rod, enables the caster to
cover a surprisingly long distance after a little practice. A distance of
150 ft. is not difficult to attain, but for fishing, a well-placed cast of 60
or 75 ft. is long enough to reach promising stretches of water.
 The Method of Thumbing the
A Line is Shown in Figure 1, and
the Correct form in Retrieving
the Line, in Figure 2. The
Overhead Cast is Illustrated in
Fig. 3 Figures 3 and 4, and the
B Underhand or Side Cast in
Figures 5 and 6
C
D
Fig. 4 Fig. 1
E
Fig. 2 Fig. 5
F
Typical Baits:
A, Under-Water Minnow; G
B, Wabbler Spoon;
C, Bass;
D, Fluted Wabbler;
E, Weedless; H Fig. 6
F, Diving;
G, Combination Color, Taste,
and Smell;
H, Spoon and Minnow Bass; I
I, Surface

The underhand or side cast is somewhat easier to master, and is

made much in the same way as the overhead cast, only the rod is
swung horizontally to the rear, the tip on a level with the angler’s
chest, as shown in Fig. 5. Bring the rod forward smartly in the
direction it is desired to shoot the bait, check the rod when about
horizontal, partly releasing the pressure of the thumb, just enough to
allow the spool to revolve freely, and check the line as the bait hits
the water. Figure 5 shows the beginning of the side cast and Fig. 6
the cast completed. This cast is useful when fishing from a shore
overhung with trees, and with practice the bait may be projected a
long distance.
 The chief factor of skill in bait casting from the reel lies in
controlling the line, and the thumb must be trained to do this through
practice. Some casters turn the reel sideways with the handle
downward while making the cast, but the majority keep the reel level
and cast with it in the same position as when reeling in. Thumbing
the line is easiest done when the spool is well filled, and if the
casting line is not long enough to do this, sufficient common line
should be first wound on to fill it.
The successful bait caster is one who possesses an intimate
knowledge of the fish he casts for, and while a lengthy exposition is
not to form a part of this article, reference to the fish commonly
sought by the caster will not be amiss. The fish generally of interest
to the bait caster are small and large-mouth black bass, pike,
muskellunge, and pickerel. The small-mouth black bass is classed
as the finest game fish, ounce for ounce, by many casters. The
muskellunge, often confused with the Great Lakes or English pike, is
more limited in range than the pike, being found in the Great Lakes
region and the waters of the St. Lawrence River.
Since the habits and general characteristics of the large-mouth
bass are much the same as its near relative, the small-mouth, the
following hints apply to both. When wading, it is well to cast
downstream. Cast in the shallows, near the weeds and snags, and
close to stumps. As little noise should be made as possible.
For early-spring casting, bass are likely to be found in the shallows
and sheltered places during the day, because the water is cool at
this season. The pickerel will also be found in the warmer and
shallower waters. The most successful baits at this time are the
diving minnows, and other under-water lures. As the warmer days of
late spring and early summer appear, the bass swim out into deeper
and cooler water, while the pickerel remain close to the lily pads. The
wooden minnows are excellent baits to use at this season. When the
weeds are very thick, one of the several kinds of weedless baits may
be used.
The early-morning and late-afternoon hours are the best for
midsummer casting, and the shallow places are the good spots to
work over. If there are any brooks or streams emptying into the lake
or pond, do not neglect to work this stretch well at the mouth. When
the sun is low, the white or light-colored surface baits may be
substituted for the medium-dark colored diving varieties, and for
casting after dark, a luminous or moonlight bait will often give good
results. During the fall months fishing conditions are similar to
casting in the spring, but the deep water is then the warmest, and
the under-water or diving baits are usually the best.
To catch bass, the angler must cast his bait where the fish are
feeding, and he will find it an advantage to match natural conditions,
so far as possible, in selecting his baits. For casting on bright days in
clear waters, the red, white and red, white, or green minnows are
good lures, as are also the silverplated spoon baits. Cast well ahead
of the boat, and reel in very slowly to keep the plug spinning well
below the surface. On cloudy days and in muddy waters, after a
heavy rain, for example, use red and yellow baits and goldplated or
burnished-copper spoons. The best all-around bait—equally good for
bass and pickerel—is one of the green-backed minnows, but no one
bait can be expected to meet all conditions, hence the caster should
have a fair assortment of baits.
A stiff casting rod with plenty of backbone is essential for
muskellunge. The angler is out after big game and must be prepared
for the sport. A hard, braided silk line, size E, and 3-0 or 4-0 hooks,
snelled and swiveled, are the best. Any of the wooden baits suitable
for pickerel—the minnows especially—are good for muskellunge. A
man to handle the oars is a handy companion, and after casting the
bait just outside the weed bed, make for deep water as soon as a
fish is hooked. The fish will be seen and felt quickly enough, and
when it breaks water and leaps in the air, give it a little slack line
quickly, otherwise the strain will snap the leader. A twisted, bronzed
trace is the best to use between line and bait, and this need not be
longer than a foot—6 in. is enough. When the fish is brought
alongside, shoot him through the head with a revolver which is
probably the most satisfactory manner of making an easy landing.
A Cat-Proof Bird Table
 Our bird table is a source of great enjoyment, particularly since the
birds feel secure from cats or other enemies because of the
construction of this ornament in our garden. The sketch shows the
arrangement of the table braced at the top of a 6-ft. post. Shrubbery
surrounds the table and a light evergreen climber clings to the post,
yet does not give the cats a good foothold.
Experience has taught us that birds in general prefer bread
crumbs to other varieties of food, and they are also fond of cracked
wheat. The linnets like oranges particularly. We cut an orange in two
and place the halves on the table. It is amusing to see the birds
balance on one side of the orange while they peck at the fruit. Soon
the orange peel is almost entirely emptied. We provide a small basin
of fresh water on the table, and the birds use it as a drinking cup as
well as a bathtub.—F. H. Sweet, Waynesboro, Va.