Download and Read online, DOWNLOAD EBOOK, [PDF EBOOK EPUB ], Ebooks

download, Read Ebook EPUB/KINDE, Download Book Format PDF

Ultrasonic Spectroscopy Applications in Condensed

Matter Physics and Materials Science 1st Edition
Robert G. Leisure

OR CLICK LINK
https://textbookfull.com/product/ultrasonic-
spectroscopy-applications-in-condensed-matter-
physics-and-materials-science-1st-edition-robert-
g-leisure/

Read with Our Free App Audiobook Free Format PFD EBook, Ebooks dowload PDF
with Andible trial, Real book, online, KINDLE , Download[PDF] and Read and Read
Read book Format PDF Ebook, Dowload online, Read book Format PDF Ebook,
[PDF] and Real ONLINE Dowload [PDF] and Real ONLINE
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Chemical Physics of Molecular Condensed Matter Kazuya

Saito

https://textbookfull.com/product/chemical-physics-of-molecular-
condensed-matter-kazuya-saito/

Topological Aspects of Condensed Matter Physics 1st

Edition Claudio Chamon

https://textbookfull.com/product/topological-aspects-of-
condensed-matter-physics-1st-edition-claudio-chamon/

Quantum Simulations with Photons and Polaritons Merging

Quantum Optics with Condensed Matter Physics 1st
Edition Dimitris G. Angelakis (Eds.)

https://textbookfull.com/product/quantum-simulations-with-
photons-and-polaritons-merging-quantum-optics-with-condensed-
matter-physics-1st-edition-dimitris-g-angelakis-eds/

String Theory Methods for Condensed Matter Physics 1st

Edition Horatiu Nastase

https://textbookfull.com/product/string-theory-methods-for-
condensed-matter-physics-1st-edition-horatiu-nastase/
Fundamentals of Charged Particle Transport in Gases and
Condensed Matter 1st Edition Robert Robson

https://textbookfull.com/product/fundamentals-of-charged-
particle-transport-in-gases-and-condensed-matter-1st-edition-
robert-robson/

Why More Is Different Philosophical Issues in Condensed

Matter Physics and Complex Systems Falkenburg

https://textbookfull.com/product/why-more-is-different-
philosophical-issues-in-condensed-matter-physics-and-complex-
systems-falkenburg/

Quantum Field Theory Approach to Condensed Matter

Physics 1st Edition Eduardo C. Marino

https://textbookfull.com/product/quantum-field-theory-approach-
to-condensed-matter-physics-1st-edition-eduardo-c-marino/

In situ Electron Microscopy Applications in Physics

Chemistry and Materials Science 1st Edition Gerhard
Dehm

https://textbookfull.com/product/in-situ-electron-microscopy-
applications-in-physics-chemistry-and-materials-science-1st-
edition-gerhard-dehm/

Advanced Quantum Condensed Matter Physics: One-Body,

Many-Body, and Topological Perspectives 1st Edition
Michael El-Batanouny

https://textbookfull.com/product/advanced-quantum-condensed-
matter-physics-one-body-many-body-and-topological-
perspectives-1st-edition-michael-el-batanouny/
 U LT R A S O N I C S P E C T RO S C O P Y

Ultrasonic spectroscopy is a technique widely used in solid-state physics, materials

science, and geology that utilizes acoustic waves to determine fundamental
physical properties of materials, such as their elasticity and mechanical energy
dissipation. This book provides complete coverage of the main issues relevant to
the design, analysis, and interpretation of ultrasonic experiments. Topics including
elasticity, acoustic waves in solids, ultrasonic loss, and the relation of elastic
constants to thermodynamic potentials are covered in depth. Modern techniques
and experimental methods including resonant ultrasound spectroscopy, digital
pulse-echo, and picosecond ultrasound are also introduced and reviewed. This
self-contained book includes extensive background theory and is accessible to
students new to the field of ultrasonic spectroscopy, as well as to graduate students
and researchers in physics, engineering, materials science, and geophysics.
r o b e r t g . l e i s u r e is Professor Emeritus in the Department of Physics at
Colorado State University, where he served as chair of the Physics Department
from 1984–1990. He is also Fellow of the Acoustical Society of America and the
Institute of Physics. His work focuses on ultrasonic studies of solids.
U LT R A S O N I C S P E C T RO S C O P Y
Applications in Condensed Matter Physics
and Materials Science

RO B E RT G . L E I S U R E
Colorado State University
 University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi – 110002, India
79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107154131
DOI: 10.1017/9781316658901
© Robert G. Leisure 2017
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2017
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Names: Leisure, Robert G., 1938– author.
Title: Ultrasonic spectroscopy : applications in condensed matter physics and
materials science / Robert G. Leisure, Colorado State University.
Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2017. |
Includes bibliographical references and index.
Identifiers: LCCN 2016054364 | ISBN 9781107154131 (Hardback ; alk. paper) |
ISBN 1107154138 (Hardback ; alk. paper)
Subjects: LCSH: Ultrasonic testing. | Materials–Testing. | Condensed matter.
Classification: LCC TA417.4 .L45 2017 | DDC 620.1/1274–dc23 LC record
available at https://lccn.loc.gov/2016054364
ISBN 978-1-107-15413-1 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party Internet Web sites referred to in this publication
and does not guarantee that any content on such Web sites is, or will remain,
accurate or appropriate.
 Contents

Preface page vii

1 Introduction 1
2 Elasticity 4
2.1 Strain 4
2.2 Stress 13
2.3 Elastic Constants 20
3 Acoustic Waves in Solids 56
3.1 Acoustic Waves in the Classical Elasticity Limit 56
3.2 Lattice Dynamics 66
3.3 Debye Theory of Solids 87
4 Experimental Methods 94
4.1 Plane-Wave Propagation Methods 94
4.2 Resonant Ultrasound Spectroscopy 107
4.3 Picosecond Ultrasonics 118
5 Elastic Constants 120
5.1 Introduction 120
5.2 Relevant Thermodynamics and Statistical Mechanics 120
5.3 Relation between Adiabatic and Isothermal Elastic Constants 122
5.4 Elastic Constants and the Helmholtz Free Energy 124
5.5 Ab Initio Computations 125
5.6 Analytical Methods 129
5.7 Phase Transitions 150
5.8 Simple Quantum Systems with a Small Number of Levels 163

v
vi Contents

6 Ultrasonic Loss 166

6.1 Introduction 166
6.2 Complex Elastic Constants 166
6.3 Measures of Ultrasonic Loss (Attenuation, Q, Internal Friction,
Loss Tangent, etc.) 168
6.4 Kramers-Kronig Relations 169
6.5 Response Functions, Fluctuations, and Dissipation 173
6.6 Relaxational Attenuation 177
6.7 Resonance Attenuation 181
6.8 Velocity-Dependent Damping 182
6.9 Qualitative Discussion of Various Sources of Loss 182
Appendix A Phase Shifts Due to Transducers and Bonds 210
Appendix B Diffraction 215
Appendix C Transducer Effects on Resonant Frequencies 217
Appendix D Damped, Driven Oscillator and Complex Force Constant 220
Appendix E Comparison of the Quasistatic and Experimental Temperature
Dependence for Specific Cases 222
E.1 Silver 222
E.2 Diamond 224
References 227
Index 238
 Preface

The use of ultrasonic methods for the study of materials continues to flourish and
evolve. These methods find uses in many areas including fundamental condensed
matter physics, materials science, various branches of engineering, geophysics, and
applied studies of device-related material parameters. Advancements in experimen-
tal methods, especially resonant ultrasound spectroscopy, have enabled quantitative
measurements on dramatically reduced specimen sizes, thereby vastly expanding
the possibilities for the study of novel materials. The title of the book “Ultrasonic
Spectroscopy” is taken here to mean simply the investigation of material properties
by the use of ultrasonic waves.
A major purpose of this book is to present an in-depth coverage of the main
issues underlying the planning and interpretation of ultrasonic investigations of
materials. It is intended that the level of the presentation be accessible to dedicated
upper-level undergraduate students, but at the same time achieve a depth of cover-
age useful to graduate students and other researchers. The approach is to present
in careful detail a number of topics, with two objectives in mind. One objective,
of course, is to educate the reader about basic concepts in the field – concepts
that should become familiar to any researcher in this area. A second objective,
perhaps more important, is to illustrate theoretical ideas that can be applied to a
wide variety of problems. The emphasis is on basic concepts, not specific materials.
The goal is to provide a fundamental background for beginning researchers so that –
with the help of a good scientific Internet search engine to obtain more focused
information – they are able to attack any of the gamut of interesting problems
amenable to ultrasonic methods.
The mathematical methods used should be familiar to upper-level undergraduate
students. Some knowledge of thermodynamics, statistical mechanics, and solid-
state physics is required, but an effort is made to present the key concepts from
these subjects as needed, and provide references to more detailed sources.

vii
viii Preface

The book includes one chapter on experimental methods. Both continuous wave
and pulse techniques are discussed.
I have benefited from interactions with many people over the years, too many to
list here. Special thanks go to Albert Migliori, Hassel Ledbetter, Ricardo Schwarz,
Paul Heyliger, David Hsu, Tatsuo Kanashiro, Alan Levelut, Jean-Yves Prieur, Ori
Yeheskel, many wonderful students, and my thesis advisor, the late Dan Bolef.
Special thanks also go to Dennis Agosta, Kate Ross, and Frank Willis for reading
large parts of the manuscript.
Most of all, I am especially thankful to my wife, Jeanine Smith Leisure, for
her contributions through steadfast support, encouragement, and thought-provoking
discussions.
 1
Introduction

The present chapter will serve as an overview of the material to be presented in

the rest of the book. While it is hoped that the material will prove useful to all
those involved in or interested in the use of ultrasound as a probe of condensed
matter, a special effort is made to present the material in sufficient detail so as to
be helpful to dedicated, upper-level undergraduate students and beginning graduate
students. Scientists from several different disciplines are nowadays finding ultra-
sonic spectroscopy a useful tool, thus a strong background in solid-state physics,
statistical physics, and quantum mechanics is not assumed of the readers. Brief
background material is presented as needed. Several monographs have contributed
to the advancement of ultrasonic spectroscopy, among them References [1, 2, 3].
The author is deeply indebted to those who have helped develop the field of ultra-
sonic studies of materials.
Chapter 2 deals with classical elasticity; the solid is treated as a continuum.
The continuum approximation is valid for virtually all ultrasonic experiments. The
present treatment of elasticity is more extensive than is usually found in books
on ultrasonic techniques, but this more extensive treatment seems important if the
researcher is to understand the widest implications of her/his ultrasonic research.
Basic physical parameters in this chapter are stress (a two-index tensor), strain
(a two-index tensor), and elastic constants (a four-index tensor, which by Hooke’s
Law connects stress and strain). Thus, many indices and sums over these indices
appear frequently. For pedagogical reasons, it was decided not to use the elegant
Einstein summation convention. For those new to the field, it seems better to
write out the sums explicitly. The relation of elastic constants to thermodynamic
potentials is derived. The condensed (Voigt) notation for stress, strain, and elastic
constants is explained in detail. Coordinate transformations are treated. The form of
the elastic constant matrix for each of the seven crystal systems is derived, as well as
the form for the icosahedral quasicrystal. Poisson’s ratio and the practical moduli –
bulk, Young’s, and torsion – are discussed for various crystal symmetries. It is

1
2 Introduction

difficult to visualize the directional dependence of elastic constants for anisotropic

materials; so, 3D representational surfaces are presented for several crystal
symmetries.
Chapter 3 treats acoustic waves in solids. The first part of the chapter deals with
waves in the continuum limit, the regime of ultrasonic experiments. The wave
equation is derived and the Christoffel equation for plane wave propagation is
found. It is shown how to calculate the wave velocities for any direction in a
crystal. Solutions are given for several crystalline directions in cubic and hexagonal
symmetries. Other symmetries are discussed. The Christoffel equation is valid in
the continuum limit and thus is restricted to wave vectors near the center of the
Brillouin zone. The second part of Chapter 3 treats lattice dynamics. To understand
many properties of solids it is necessary to go beyond the continuum limit; lattice
dynamics does just that. The usual 1D models for monatomic and diatomic cases
are treated and the dispersion relations obtained. Much insight is gained from these
models. However, it seems important to go beyond the 1D case, thus the full 3D
lattice dynamics model is treated. The relevant equations are derived and force
constant matrices are obtained for the face-centered cubic symmetry. A full 3D
calculation is performed, demonstrating the dispersion relations in various high-
symmetry directions. The last part of Chapter 3 discusses the highly simplified,
but highly successful, Debye model of solids. The thermal energy and the specific
heat are obtained, and it is shown how to calculate the Debye temperature from the
elastic constants.
Chapter 4 deals with common experimental methods for measuring ultrasonic
attenuation (or internal friction) and elastic constants (or ultrasound velocities).
The material in Chapters 2 and 3 was needed before the experimental methods
could be treated meaningfully. The well-known pulse-echo method is discussed.
For highly accurate results it is necessary to accurately determine the time delay
between echoes. This is not a trivial problem, and methods (pulse superposition,
pulse-echo overlap) have been developed to solve this problem. However, several
factors other than round-trip travel time usually contribute to the measured delay
time. These factors are discussed in detail as well as how to account for them. The
older methods were analog in nature. An important recent advance has been the
development of an all digital pulse-echo system. Resonant ultrasound spectroscopy
(RUS) is a resonance method dramatically different from the pulse-echo method.
RUS is also discussed at length in Chapter 4. Finally, picosecond ultrasound is
discussed in Chapter 4.
Chapter 5 treats the important subject of elastic constants. It starts with a
review of relevant thermodynamics and statistical mechanics, culminating in the
Helmholtz free energy for a harmonic oscillator. The individual lattice vibration
modes are treated as harmonic oscillators, thus the Helmholtz free energy for the
 Introduction 3

lattice is a sum of harmonic oscillators. A major part of Chapter 5 is concerned with

using the quasiharmonic approximation to find the temperature dependence of the
elastic constants. This rather lengthy calculation reveals the indirect effect of lattice
expansion on the temperature dependence of the elastic constants and shows that
it is of major importance. The effects of both first-order and second-order phase
transitions on the elastic constants are discussed in terms of the Landau theory.
Chapter 6 is concerned with ultrasonic loss. The chapter begins with several
general ideas related to ultrasonic loss: complex elastic constants and phase shifts;
units of measurement of loss; Kramers-Kronig relations; and fluctuations and dissi-
pation. Next, relaxational and resonance attenuation are discussed in general terms.
Many different specific loss mechanisms are known. In general, these have been
treated elsewhere. The present approach is to give a brief summary of ten of these
loss mechanisms, with references to the literature and previous monographs for
more detailed descriptions.
 2
Elasticity

Chapter 2 deals with the elastic response of materials, where elastic has the usual
meaning that a material returns to its original configuration after the removal of a
deformation-causing force. In fact, linear elasticity will be assumed, for which
the distortion from the equilibrium configuration is directly proportional to the
applied force. In the present chapter, classical elasticity will be discussed, in which
case the solid is treated as a continuum, and the discreteness of the underlying
lattice is ignored [4]. This approach is only valid when the length scale of the
spatial variations is much greater than the lattice constant, the usual situation for
ultrasonic vibrations in solids. Knowledge of classical elasticity is essential for an
understanding of ultrasonic vibrations and the propagation of ultrasonic waves in
materials.

2.1 Strain
2.1.1 The Strain Tensor
The problem at hand is to describe the distortions of a material from the equilibrium
configuration. Consider a point in a solid relative to an origin fixed in space. Before
deformation, the position of the point is given by r. After deformation, the position
is given by r = r + u. Thus, the displacement of a point from its equilibrium
position is
u = r − r. (2.1)
In the case of interest, u is a function of r, i.e u(r), otherwise the displacement just
corresponds to a uniform displacement of the material, and is not relevant to the
present discussion. A brief discussion of u(r) follows, leading to an expression for
the strain tensor.

4
 2.1 Strain 5
(a) (b)
dr B
du
A

uB uA
uB
uA

B
dr
A

Figure 2.1 (a) The two points A and B are initially separated by the vector dr.
After a distortion the points are separated by the vector dr . (b) The vector du is
shown.

Figure 2.1 shows two points in the material, A and B, which are originally
separated by dr. After a distortion the points move to A and B respectively and
are separated by dr . The various vectors are related by

dr = dr + (uB − uA ) = dr + du (2.2)

where du is the difference in the displacement vectors for the two original points,
A and B. An expression for the strain is obtained by considering the difference
in the square of the distances between the two points, and how it changes with
the distortion [5], [6]. This quantity will be unchanged for pure rotations, but will
change for distortions of the material. Squaring Equation 2.2 gives

(dr )2 = (dr)2 + 2dr · du + (du)2 . (2.3)

3
It is convenient to express the equations in component form, e.g. dr · du = i=1
dxi dui . In general dui will be a function of r, leading to
3 
 
∂ui
dui = dxj (2.4)
j=1
∂xj

with i = 1, 2, 3. Then, Equation 2.3 becomes

  ∂ui    ∂ui   ∂ui 
(dr )2 = (dr)2 + 2 dxj dxi + dxj dxl (2.5)
ij
∂xj ijl
∂xj ∂xl

where the sums over all indices range independently from 1 to 3. Rearranging the
summation indices yields the desired result. First, the second term of the right-hand
side of Equation 2.5 may be rewritten as
6 Elasticity
 ∂ui
   ∂ui    ∂uj 
2 dxj dxi = dxj dxi + dxi dxj (2.6)
ij
∂xj ij
∂xj ij
∂xi

where dummy summation labels have been interchanged in the last term of Equa-
tion 2.6. Similarly, i and l can be interchanged in the last term of Equation 2.5.
Finally, the result is

(dr )2 − (dr)2 = 2ij dxi dxj (2.7)
ij

where
 
1 ∂ui ∂uj  ∂ul ∂ul
ij = + + (2.8)
2 ∂xj ∂xi l
∂xj ∂xi

and ij is the Lagrangian [7] strain tensor. For small deformations the last term in
Equation 2.8 is neglected with the result that the elastic strain tensor is given by
 
1 ∂ui ∂uj
ij = + . (2.9)
2 ∂xj ∂xi

Notice that the strain tensor is symmetric, ij = ji .

Equation 2.9 is the basic definition of infinitesimal strain that will be used
throughout this work. In a few cases it will be necessary to use Equation 2.8
instead; and, Equation 2.8 will be referenced at those points.
Actually, Equation 2.9 could be taken as the starting point for the concept of
strain. In what follows, the concept will be developed further.

2.1.2 Geometrical Interpretation

One-Dimensional Example
A few simple examples will be given to illustrate the meaning of the strain tensor.
First, a one-dimensional distortion is discussed.
As shown in Figure 2.2, the points A and B are originally separated by a distance
x1 . After a stretch along the x1 -axis, the two points, A and B , are separated by
x1 + u1 . The strain in this 1D example is given by
 
1 ∂u1 ∂u1 ∂u1 u
11 = + =  . (2.10)
2 ∂x1 ∂x1 ∂x1 x

Thus, 11 gives the fraction change in the length separating two nearby points in the
x̂1 direction (where, as usual x̂i denotes a unit vector lying along the xi axis).
 2.1 Strain 7
x1 x1
Before
O A B

A B
After
O
x1+u1 x1+ u1

Figure 2.2 The two points A and B are initially separated by x1 . After a
distortion the points are separated by x1 + u1 .
B
x2 x2
After u2
θ1
Before
A x1+ u1
A x1 B

x1 x1

Figure 2.3 A distortion in two dimensions carries the points A and B to A and B
respectively.

Two-Dimensional Examples
The preceding discussion will now be extended to two dimensions. Figure 2.3
illustrates the situation. Points A and B are carried to A and B respectively by
a distortion.
The angle θ1 is exaggerated for clarity. For small strains, θ1 is correspondingly
small. Proceeding as for Equation 2.10 shows that 11 = 12 ∂u 1
∂x1
+ ∂u 1
∂x1
= ∂u 1
∂x1

u1
x1
, an extension per unit length in the x1 direction. Inspection of Figure 2.3 reveals
that
∂u2 u2
  θ1 . (2.11)
∂x1 x1
Thus, ∂u2 /∂x1 represents a counterclockwise rotation by an angle θ1 of a line
originally lying parallel to the x1 axis. A similar consideration of a line originally
lying parallel to the x2 axis shows that
∂u1 u1
  θ2 , (2.12)
∂x2 x2
represents a clockwise rotation of the line.
Still discussing 2D examples, if the angle between two lines is π/2 before a
homogeneous deformation, afterward it is
 
π π ∂u2 ∂u1 π
ψ = − (θ1 + θ2 ) = − + = − 212 (2.13)
2 2 ∂x1 ∂x2 2
8 Elasticity

x2 θ2

θ1

x1

Figure 2.4 The dashed lines indicate two lines that are perpendicular before a
distortion. After the distortion, the lines are rotated as shown.

Figure 2.5 The dashed lines indicate the initial position of a square sheet of the
material. The sides are parallel to the coordinate axes and one corner is located at
the origin. After a rigid-body rotation about the origin of an angle θ , the sheet is
oriented as shown.

as illustrated if Figure 2.4. Thus, 212 gives the change in the angle between two
lines that were perpendicular before the deformation. The illustration in Figure 2.4
is for ∂u1 /∂x2 and ∂u2 /∂x1 both > 0. Note that ∂u2 /∂x1 need not be equal to
∂u1 /∂x2 .

Rotations
Consider the situation in which a sheet of material undergoes a rigid-body rotation.
The situation is illustrated in Figure 2.5. By the arguments leading to Equations 2.11
and 2.12 the displacement of the material from the original position is described by
∂u2 /∂x1 = θ and ∂u1 /∂x2 = −θ. (The small-angle approximation is assumed.)
From these results it follows that the strain, as defined by Equation 2.9, is zero.
However, consider the quantity ωij defined as
 
1 ∂ui ∂uj
ωij = − (2.14)
2 ∂xj ∂xi
Applying Equation 2.14 to the present situation gives ω12 = −θ. Thus, ω12 repre-
sents a rotation; counterclockwise (clockwise) for ω12 negative (positive). It follows
 2.1 Strain 9

Figure 2.6 The dashed lines indicate the initial position of a square sheet of the
undeformed material. A simple shear deformation, characterized by the angle θ ,
may be regarded as a pure shear strain plus a rotation. (The angles are exaggerated
for clarity.)

Figure 2.7 A general shear deformation, characterized by the angles θ1 and θ2 ,

also may be regarded as a pure shear strain plus a rotation about an axis
perpendicular to the plane.

that a general distortion of the form ∂ui /∂xj can be written as the sum of a symmet-
ric and an antisymmetric part,
   
∂ui 1 ∂ui ∂uj 1 ∂ui ∂uj
= + + − (2.15)
∂xj 2 ∂xj ∂xi 2 ∂xj ∂xi
The first term on the right-hand sign of Equation 2.15 gives the tensor strain and
the second term represents the rotation. Figure 2.6 illustrates the relationship for
the case of ∂u2 /∂x1 = θ and ∂u1 /∂x2 = 0. Figure 2.6 shows how a simple shear
can be decomposed into a pure shear and a rotation. A transverse ultrasonic wave
propagating along a high-symmetry direction produces such a simple shear motion.
The rotation is commonly ignored, because in most cases there is no energy cost
for the pure rotation. However, such is not the case if an external torque acts on
the material (e.g. a magnetic material in a magnetic field [8]). In such cases, the
rotational term may become important.
Figure 2.7 illustrates a more general situation. For this case the general shear
distortion on the left side of Figure 2.7, ∂u2 /∂x1 = θ1 and ∂u1 /∂x2 = θ2 , can be
represented as a pure shear plus a rotation as shown.
10 Elasticity

2.1.3 Discussion of Strain in Three Dimensions

It is straightforward to generalize the interpretations of Section 2.1.2 to 3D. If u
represents the displacement of a point in the material from its equilibrium position
(relative to a coordinate system fixed in space), then the distortion of the material
is given by a strain ij , Equation 2.9, and a rotation ωij , Equation 2.14. These have
the following interpretations [4]:
ii is the extension per unit length of a line segment parallel to the xi axis,
ij , i = j equals 1/2 the change in the angle between two line elements originally
parallel to axes xi and xj respectively. (ij is the average of the angles of rotation for
the two line elements). The angle decreases for ij > 0.
ωij represents a rotation of the material about an axis along xk . The i, j, and k
directions are assumed to represent a right-handed Cartesian coordinate system, as
usual.

The Strain Tensor

For small deformations it is possible to express the local displacements from equi-
librium as
 ∂ui
ui = xj . (2.16)
j
∂xj

For the case of no rotations (∂ui /∂xj = ∂uj /∂xi ), application of Equations 2.9 and
2.16 give the local displacements from equilibrium as
u1 = 11 x1 + 12 x2 + 13 x3
u2 = 21 x1 + 22 x2 + 23 x3 (2.17)
u3 = 31 x1 + 32 x3 + 33 x3 .
where u is the displacement from equilibrium at the position x. Equation 2.17
can also be written as
⎛ ⎞ ⎛ ⎞⎛ ⎞
u1 11 12 13 x1
⎝ u2 ⎠ = ⎝ 21 22 23 ⎠ ⎝ x2 ⎠ . (2.18)
u3 31 32 33 x3
Equation 2.18 gives a linear relation between the vectors ui and xj , which
shows that ij is a tensor of the second rank, [4] the strain tensor. It is a symmetric
tensor because ij is symmetric by its definition (Equation 2.9). In general x is very
small (infinitesimal), but for homogeneous strains such is not necessarily the case.

Simple Examples of Pure Strain Deformations (No Rotations)

Equations 2.17 will now be used to illustrate the distortion produced by pure strains.
For simplicity in illustrations, 2D examples will be used. Consider the case with
 2.1 Strain 11

11 being the only strain. Equations 2.17 (or Equations 2.18) give the resulting
displacements, and the situation is illustrated in Figure 2.8b, i.e. a simple extension.
Next, consider the case of 12 = 21 being the only strain. If Equations 2.17 are
applied to each corner of the square indicated by the dashed lines in Figure 2.8c,
the result is the distortion illustrated by the solid line, a pure shear distortion.
Such considerations of distortions produced by different strains can be useful
in the interpretation of ultrasonic measurements. For example, in a cubic crys-
tal the [100]-type directions are equivalent. In general, applied strains – such as
those produced by ultrasonic waves – lower the crystalline symmetry. An 11 strain
renders the [100]-type directions inequivalent, while a 12 strain does not. With the
use of Equations 2.17, it is straightforward to determine the distortions produced
by other strains, e.g. 22 − 11 , a strain produced by certain ultrasonic waves. The
ultrasonic waves often consist of both a pure strain and a rotation, but as mentioned
earlier, the rotational part may be safely ignored in many cases.

Coordinate Transformations
Figure 2.8 can be understood from a different point of view. Because the strain, ,
is a second-rank tensor, it obeys the usual rule for coordinate transformations of
such tensors. Let xi represent a new coordinate system rotated with respect to the xi
system. The transformation of the strain tensor from the xi coordinate system to the
xi system is given by [4]

ij = aik ajl kl (2.19)
kl

where aik is a direction cosine; the first subscript refers to the prime system while
the second subscript refers to the original system. For example, a21 is the cosine of
the angle between x2 and x1 .

(a) (b) (c)

x2 x2 x2
11 12

x1 x1 x1

Figure 2.8 The dashed lines indicate the initial position of a square sheet of the
undeformed material. The solid lines show the shapes after deformation. (a) Initial
shape. (b) Deformed shape for a simple extension, 11 . (c) Deformed shape for a
pure shear, 12 .
12 Elasticity

Now consider the case of Figure 2.8c for which the strain matrix may be written
⎛ ⎞
0 o 0
 = ⎝ o 0 0 ⎠ (2.20)
0 0 0

in the unprimed system, where

 
1 ∂u1 ∂u2
o = + (2.21)
2 ∂x2 ∂x1

Next, a counterclockwise rotation about the x3 axis by π/4 is carried out by

applying Equation 2.19 to the matrix of Equation 2.20. The result is
⎛ ⎞
o 0 0
  = ⎝ 0 −o 0 ⎠ (2.22)
0 0 0

for the strain in the primed system (coordinate axes not shown). The interpretation
is simple; a pure shear in the x1 − x2 plane in the xi system becomes an extension
along x1 axis and a compression along the x2 axis in the xi coordinate system. An
examination of Figure 2.8c shows that this is exactly what happens.

Principal Axes
The preceding discussion is a special case of a general result. The general sym-
metrical matrix of Equation 2.18 can be diagonalized to find the eigenvalues and
the principal axes [9]. In the principal axes system the matrix of Equation 2.18, ij ,
becomes
⎛ ⎞
α 0 0
 = ⎝ 0 β 0 ⎠ (2.23)
0 0 γ
where α, β and γ refer to the principal axes. Thus, there are no shear strains in
the principal axes system, only extensions or contractions along a principal axes.
As a simple example, application of the diagonalization method to the matrix of
Equation 2.20 shows the principal axes to be rotated counterclockwise by π/4
about the x3 axes with the eigenvalues of Equation 2.22, the same result achieved
in the previous section by a different method. If the strain is homogeneous, then
the principal axes will be the same throughout the material; otherwise, the principal
axes may vary with position in the specimen.
 2.2 Stress 13

Engineering Strains
One sometimes encounters “engineering strains,” γij which are defined somewhat
differently from the tensor strains of Equation 2.9, the difference being that the 1/2
is omitted for i = j. The result is
∂ui
γii = = ii
∂xj
(2.24)
∂ui ∂uj
γij = + = 2ij i = j.
∂xj ∂xi
A cautionary note, an array formed from γij does not form a tensor, unlike the ij [4].
The consequence is that ij follows the tensor rules for transformation between
coordinate systems, whereas γij does not.

2.2 Stress
2.2.1 Definition of Stress
The deformations of a material are brought about by forces. Considering a partic-
ular small volume of the material, the forces acting on this volume are assumed
to arise from the material outside the volume, and to act at the surface of the
selected volume, i.e. the molecular forces are assumed to be of short range [5].
Long-range forces, (body forces) such as gravity which act throughout the material
are neglected in the present treatment. It is convenient to consider forces per unit
area, which are called stresses. The situation is illustrated in Figure 2.9 for the
situation of a small cube which is imagined to be embedded in the material [4].

Figure 2.9 The stresses acting on a small cube of material are illustrated. σij is the
force per unit area acting in the direction of the xi axis on a face perpendicular to
the xj direction.
14 Elasticity

In general, the force acting on a particular face of the cube will not be normal to
that face, hence there will be three components of stress acting on each face. The
stress, σij , is defined as the force per unit area acting in the direction of the xi axis
on a face perpendicular to the xj direction. For example, σ23 is the force per unit
area acting in the x2 direction on a face which is perpendicular to the x3 direction.
The stresses σ11 , σ22 , and σ33 are normal components of the stress, the others are
shear stresses. The stresses are exerted on the cube by the surrounding material.
By convention, positive normal components are taken as pointing outward. The
positive directions of the shear stresses then follow as shown in Figure 2.9.

2.2.2 Homogeneous Stress and Static Equilibrium

In the case that the stress is homogeneous throughout the body, the stresses on
opposite faces (not shown in Figure 2.9) are equal in magnitude and opposite in
direction, i.e. the net force acting on the cube is zero. Further, if static equilibrium
is required, then the net torque must be zero. An examination of Figure 2.9 shows,
e.g. that for the torque about the x3 axis to be zero, σ12 must equal σ21 . More
generally,
σij = σji . (2.25)

2.2.3 Inhomogeneous Stress and Equations of Motion

In general, the stress may vary throughout the material; in fact, this situation occurs
for many of the most interesting cases. Consideration of the stress variation across
a small cube of material leads to Newton’s Second Law for linear and rotational
motion of the cube. Referring to Figure 2.10 and considering first the forces in the
x1 direction, it is seen that the variation of σ11 leads to a force

Fx1 11 = (σ11 )(Area perpendicular to the x1 axis),

or
 
∂σ11
Fx1 11 = x1 (x2 x3 ). (2.26)
∂x1
In a similar manner, the variation of σ12 along the x2 axis leads to a net force in the
x1 direction. Reference to Figure 2.11 shows
 
∂σ12
Fx1 12 = x2 (x1 x3 ). (2.27)
∂x2

Of course there is a similar equation for the variation of σ13 along the x3 direction.
 2.2 Stress 15

Figure 2.10 The stress σ11 is assumed to vary in the x1 direction. The force, due
to σ11 , acting on a face perpendicular to the x1 axis is σ11 x2 x3 . Variation of
this force across the cube leads to a net force in the x1 direction.

Figure 2.11 The stress, σ12 is assumed to vary in the x2 direction. The force, due
to σ12 , acting on a face perpendicular to the x2 axis is σ12 x1 x3 . Variation of
this force across the cube leads to a net force in the x1 direction.

Totaling the three contributions to the force in the x1 direction, the equation of
motion, Fx1 = max1 becomes
       2 
∂σ11 ∂σ12 ∂σ13 ∂ u1
+ + = ρ 2 (2.28)
x1 x2 x3 ∂t
where m = ρ(x1 x2 x3 ), with ρ being the density. Also, ax1 = ∂ 2 u1 /∂t2 . A com-
mon factor of (x1 x2 x3 ) has been divided out of Equation 2.28. Equation 2.28
can be generalized to [4]
16 Elasticity


3
∂σij ∂ 2 ui
=ρ (2.29)
j=1
∂xj ∂t2

Equations 2.29 are basic equations of elasticity, equivalent to Newton’s second law,
and will be used for several problems later, including the development of the elastic
wave equation. The index i ranges from 1 to 3 to correspond to the three Cartesian
components of the motion.

2.2.4 The Stress Tensor

The stress will be shown later to be a second-rank tensor. Equation 2.25 indicates that
the tensor is symmetric for homogeneous stress. It will now be shown that Equation
2.25 holds rather generally, not just under the conditions of homogeneous stress.
Figure 2.12 illustrates the forces acting on a small volume element that produce
torque about the x1 axis [4]. The torque produced by the four forces illustrated is
       
∂σ32 x2 ∂σ32 x2 x2
σ32 + (x1 x3 ) + σ32 − (x1 x3 )
∂x2 2 ∂x2 2 2
       
∂σ23 x3 ∂σ23 x3 x3
− σ23 + (x1 x2 ) + σ23 − (x1 x2 )
∂x3 2 ∂x3 2 2
= (σ32 − σ23 )x1 x2 x3 . (2.30)

Figure 2.12 The stresses contributing to torque about the x1 axis are shown. The
x1 axis is taken to be at the center of the square. The stresses are considered to
vary spatially.
Another random document with
no related content on Scribd:
 Seeds numerous, angular.
Obs. In the greatest number of species the chives are bent downward, and
clothed with soft, coloured hairs on the lower part.

SPECIFIC CHARACTER.

Verbascum foliis subvillosis, rugosis; caulinis subsessilibus, æqualiter

crenatis, radicalibus oblongis, cordatis, duplicato crenatis.
Mullein with leaves a little hairy and rough: stem leaves almost without
foot stalks, equally scolloped; leaves from the root oblong, heart-shaped, and
doubly scolloped.

REFERENCE TO THE PLATE.

1. A Leaf of the plant, from the lower part.

2. The Empalement.
3. A Blossom, shewn from the front.
4. The same cut open from the side, to shew the insertion of the
chives.
5. One Chive magnified.
6. The Pointal and Seed-bud, magnified.
The rusty-flowered Mullein of the Kew Catalogue, collated from
Sutherland’s Hortus Medicus Edinburghensis, is undoubtedly our plant; but,
though it stands a name in the book, we much question its having graced the
garden, at the time of its publication. It certainly has been lost to us above a
century, and but recently introduced to the Oxford Botanic Garden by the
late Dr. Sibthorpe. Being a hardy biennial, it will perfect its seeds about July,
which may be sown in August the same year. It will grow in any soil, but a
sandy compost of light peat and loam suits it most. The flowers on the
spikes, which grow sometimes to the length of four feet, continue to expand,
upwards, for above two months; that is to say, from April, till June. Our
figure was taken from a specimen sent by the Hon. W. H. Irby, from his
garden at the Parsonage, Farnham Royal, Bucks.
 PLATE CLXIII.

AMARYLLIS FOTHERGILLIA.

Fothergillian Lily Daffodil.

CLASS VI. ORDER I.

HEXANDRIA MONOGYNIA. Six Chives. One Pointal.

ESSENTIAL GENERIC CHARACTER.

Corolla 6-petala, campanulata. Stigma trifidum.

Blossom 6-petalled, bell shaped. Summit three-cleft.
See Amaryllis radiata, Pl. XCV. Vol. II.

SPECIFIC CHARACTER.

Amaryllis spatha multiflora; petalis lanceolatis, apice revolutis;

genetalibus erectis; foliis linearibus, sub-canaliculatis, obtusis, glaucis.
Lily Daffodil with many flowers in the sheath; petals lance-shaped, rolled
back at the point; parts of fructification upright; leaves linear, rather
channelled, obtuse, and of a sea-green colour.

REFERENCE TO THE PLATE.

1. A Petal, with its Chive attached.

2. The Seed bud, and Pointal.
3. A ripe Seed-vessel, with the Seeds.
Even amidst this splendid family, Amaryllis Fothergillia stands
conspicuously preeminent; the flowers have the same vivid character of
refulgence, when exposed to the sun’s rays, which is perceived in A.
Sarniensis, or the Guernsey Lily, to which it is nigh affined. The late Dr.
Fothergill (to whom the botanical world is so much indebted, for the zeal
and extraordinary liberality he constantly manifested, in advancing the
science) received this plant from China at the same time with A. aurea, A.
radiata, &c. about the year 1777. It is rather a scarce and tender bulb, as it is
subject to rot by overmuch wet. It has hitherto been kept, as are most plants
on their first arrival from China, in the hot-house; but, from every
appearance, we should be led to suppose, it would not perish if kept in the
Green-house, as we question much, whether it is not equally hardy with the
Guernsey Lily, and a native, originally, as well as the latter, of Japan.
It should be planted in sandy loam, with a small quantity of rough peat,
above the tiles in the pot, to make it flower. The propagation is from offsets,
which are produced but rarely from the old bulbs. Our figure was taken from
a plant in the Hammersmith nursery this year, 1801, in the month of May.
 PLATE CLXIV.

GALAXIA GRANDIFLORA.

Large-flowered Galaxia.

CLASS XVI. ORDER I.

MONADELPHIA TRIANDRIA. Threads united. Three Chives.

ESSENTIAL GENERIC CHARACTER.

Spatha univalvis. Corolla monopetala, 6 fida, tubus filiformis. Stigma

multipartita.
Sheath of one valve. Blossom one petal, 6-cleft, tube thread-shaped.
Summit many divisions.
See Galaxia ovata, Pl. XCIV. Vol. II.

SPECIFIC CHARACTER.

Galaxia foliis canaliculatis, acuminatis, arcuatis; corolla magna, lutea,

folia æquantia.
Galaxia with channelled leaves, tapered to the point, and arched; blossom
large, yellow, and the length of the leaves.

REFERENCE TO THE PLATE.

1. The sheath of the blossom.

2. A Blossom spread open, with the Chives attached.
3. The Seed-bud, Shaft, and Summit, the summit detached,
magnified.
The figure of this delicate little bulb was taken from a plant in the
Hibbertian Collection, which had been imported in the last, and flowered in
February this year 1801. It is a tender plant, and the roots are very subject to
decay after flowering, for which reason, they should be removed from the
pot before the leaves are quite decayed. Mr. Allen informs us he has kept it,
in the greenhouse, with the other Cape bulbs, planted in light sandy peat
earth.
 PLATE CLXV.

ECHIUM GLAUCOPHYLLUM.

Sea-green-leaved Viper’s-Bugloss.

CLASS V. ORDER I.
PENTANDRIA MONOGYNIA. Five Chives. One Pointal.

ESSENTIAL GENERIC CHARACTER.

Corolla irregularis, fauce nuda.

Blossom irregular, mouth naked.
See Echium grandiflorum, Pl. XX. Vol. I.

SPECIFIC CHARACTER.

Echium caule fruticoso; ramis calycibusque glabris; foliis ovato-

lanceolatis, glaucis, glabris, margine serrulatis; corollis subæqualibus.
Viper’s-Bugloss with a shrubby stem; branches very smooth; leaves
between egg and lance-shaped, of a sea-green colour, smooth, and slightly
sawed at the margin; blossoms nearly equal.

REFERENCE TO THE PLATE.

1. The Empalement.
2. A Blossom cut open, to shew the insertion of the chives.
3. The Seed-bud, Shaft, and Summit, magnified.
This shrubby species of Echium was introduced, to the Royal Gardens at
Kew, by Mr. F. Masson, in the year 1792, from the Cape of Good Hope. It
grows to the height of three feet, or more, rather bushy at the top, flowers
about the month of May, and perfects its seeds with us; by which only
method, it is to be propagated. It thrives best in a light loamy soil, with a
small proportion of sandy peat, about one fourth.
 Professor Martyn, in his Miller’s Dic. has collated two descriptions of
Echiums under the titles of lævigatum, and glabrum; the first, No. 9, from
Lin. Sp. Plant. 199; the second, No 19, from Vahl. Symb. 3. 22. Thunberg
has likewise, in his Prodromus, 33. two names of plants, as E. lævigatum,
and E. glabrum; from whom Willdenow, in his new Edition of Sp. Plant, p.
785, has copied them under the same titles; adding, the various synonyms,
from Linnæus’s Sp. Plant, to the one; and to the latter, Prof. Jacquin’s
specific of glaucophyllum, taken from his Ic. rar. 2. t. 312, and his Collect. 2.
p. 325. Now, we have little hesitation in declaring our opinion, and we think,
those who choose to compare our figure with the different descriptions here
mentioned, will coincide with us, that this is the plant from which they must
all have originated, except the E. glabrum of Vahl, which we take to be a
different plant; as, the cup is described hairy, the length of the tube of the
blossom; the chives longer, and the flower smaller. Wherefore, we have
taken, for our plant, the specific title of Jacquin, as being, in our idea, the
most appropriate.
 PLATE CLXVI.

GLADIOLUS ABREVIATUS.

Shortened-petalled Gladiolus.

CLASS III. ORDER I.

TRIANDRIA MONOGYNIA. Three Chives. One Pointal.

ESSENTIAL GENERIC CHARACTER.

Corolla sexpartita, ringens. Stamina adscendentia.

Blossom six divisions, gaping. Chives ascending.
See Gladiolus roseus, Pl. XI. Vol. I.

SPECIFIC CHARACTER.

Gladiolus foliis linearibus, cruciatis, scapo longioribus; corolla tubulosa,

striata, ima angustata, lacinia summa magna, recta, ovata, reliquæ vero
parvæ, abreviatæ.
Gladiolus with linear leaves, cross shaped, longer than the flower-stem;
blossom tubular, streaked, and narrowed at the lower part, the upper petal is
large, grows straight out, and egg-shaped, the others are small, and appear as
if shortened.

REFERENCE TO THE PLATE.

1. The outer sheath of the Empalement.

2. The inner sheath of the Empalement.
3. A Blossom spread open, with the Chives attached.
4. The Seed-bud, Shaft, and Summits.
This most singular Gladiolus was received, amongst a variety of others, in
the year 1799, from the Cape of Good Hope, by Messrs. Lee and Kennedy,
Hammersmith; at whose nursery it flowered in March, this year 1801, for the
first time. It is a hardy bulb, and propagates freely from the root; grows
about two feet high, and, before flowering, has much the appearance of G.
tristis.
 PLATE CLXVII.

BRUNSFELSIA UNDULATA.

Waved-flowered Brunsfelsia.

CLASS XIV. ORDER II.

DIDYNAMIA ANGIOSPERMIA. Two Chives longer. Seeds covered.

GENERIC CHARACTER.

Calyx. Perianthium monophyllum, campanulatum, quinquedentatum,

obtusum, minimum, persitens.
Corolla monopetala, infundibuliformis; tubus longissimus, subincurvus;
limbus planus, quinquefidus, obtusus.
Stamina. Filamenta quatuor, brevissima. Antheræ oblongæ, erectæ,
filamentis dorso affixæ.
Pistillum. Germen subrotundum, parvum. Stylus filiformis, longitudine
tubi. Stigma crassiusculum.
Pericarpium. Capsula extus baccata, globosa, unilocularis, bivalvis.
Semina plurima, compressa, hinc convexa, inde angulata, punctato-
scabra.
Receptaculum fundo capsulæ adnatum, paleaceum; paleis coadunatis,
apice subulatis, semina distinguentibus.
Empalement. Cup one leaf, bell-shaped, five-toothed, blunt, very small,
remaining.
Blossom. One petal, funnel-shaped; tube very long, rather curved; border
flat, five-cleft, obtuse.
Chives. Four threads, very short. Tips oblong, upright, fixed by the back
to the threads.
Pointal. Seed-bud roundish, small. Shaft thread-shaped, the length of the
tube. Summit rather fleshy.
 Seed-vessel. Capsule on the outside a berry, globular, one cell, two
valves.
Seeds many, flat, convex on one side, angular on the other, roughly
punctured.
Receptacle fixed to the bottom of the capsule, chaffy; chaffs joining at
the base, awl shaped at the point, separating the seeds.

SPECIFIC CHARACTER.

Brunsfelsia foliis lanceolato-obovatis, utrinque acuminatis, petiolis

brevissimis; tubus parum incurvatus, laciniis limbi undulatis.
Brunsfelsia with leaves between lance and inversely egg-shaped, tapered
to both ends, footstalks very short; tube a little incurved, the segments of the
border waved.

REFERENCE TO THE PLATE.

1. The Empalement.
2. The tube of the blossom cut open, to shew the insertion and
character of the Chives.
3. The Pointal and Seed-bud.
The Genus Brunsfelsia (by Plumier so named, in honour of Otho Brunsfels a
monk, the first who edited in 1530 a Botanical work with good figures, it
contained 238 plates) has undergone a very necessary revision since the days
of Linnæus; indeed, it appears singular that Father Plumier, from whom
Linnæus quotes his Generic character, should have committed so great an
oversight, as to describe it with five chives; a feature scarcely to be
expected, from a sport of nature, in plants of this Class. So however it is on
record, and the fifth Class of the Gen. and Spe. Plant, and of the Syst. Nat.
tom. ii. each includes this Genus; from which, we should be induced to think
he had designed some other plant; if the figure, and the rest of the
description, did not confirm it. Schreber, in his Gen. Plant, has likewise an
observation, that the fruit should be named a capsule, rather than a berry; as
given by Linnæus, Miller, and Swartz; his reason, that it splits determinately,
by a suture, from the top to the base.
 Our species, is the third now in Britain; they are all natives of the West
India Islands. The 1st B. americana has been long an inhabitant of our
hothouses; the 2d B. maculata is but little known to most, though long
cultivated here; as it has been treated as a variety of the first species,
notwithstanding the strong specific difference, in the shape of the leaves and
blossoms; which in the B. maculata, are beautifully marked at the base, and
as large as the undulata; the leaves inversely egg-shaped and more downy.
The present plant seldom grows more than four feet high, flowers freely
about March, is very sweet scented, and is easily propagated by cuttings. It
was first sent to England in the year 1784 by Mr. Elcock from the Island of
Barbadoes, to Messrs. Lee and Kennedy, Hammersmith. Our drawing was
made from a plant in the Clapham Collection, in March 1800. It should be
planted in rich earth.
 PLATE CLXVIII.

GERANIUM PICTUM.

Painted-flowered Geranium.

CLASS XVI. ORDER IV.

MONADELPHIA DECANDRIA. Threads united. Ten Chives.

ESSENTIAL GENERIC CHARACTER.

Monogyna. Stigmata 5, Fructus rostratus, penta-coccus.

One Pointal. Five Summits. Fruit furnished with long awns, five dry
berries.
See Geranium Grandiflorum. Pl. XII. Vol. I.

SPECIFIC CHARACTER.

Geranium foliis cordato-oblongis, obtusis, inequaliter incisis, tomentosis,

humi adpressis; corolla alba, petalis superioribus profundè supra medium
rubro maculatis; staminibus septem fertilibus; radice tuberosa.
Geranium with oblong heart-shaped leaves, blunt, unequally gashed,
downy, and lying close to the ground; blossom white; the upper petals
deeply marked with red about the middle; seven fertile chives; root tuberous.

REFERENCE TO THE PLATE.

1. The Empalement.
2. The Chives and Pointal, magnified.
3. The Chives spread open, magnified.
4. The Seed-bud, Shaft and Summit, magnified.
No Genus of Plants claims our notice, for its beauty, more than
Geranium, and this species, certainly, ranks amongst the foremost. It is a
native of the Cape of Good Hope, and we believe only to be found in the
Clapham Collection; where, our figure was taken in April this year, from a
plant, the bulb or root of which had been received the preceding autumn. It is
nearly the only one, amongst twenty-two species of the tuberous kind, all
having irregular petalled blossoms and tubular cups, of which we possess
drawings, that has seven fertile chives; the greater number have two, four, or
five. It appears to flourish under the treatment given it by Mr. Allen, which
is, by keeping it in sandy peat, on a shelf, very dry, in the green-house. The
propagation appears to be the same for this, as the other tuberous kinds, that
is, by the root.