Statistical Thermodynamics Understanding the

Properties of Macroscopic Systems 1st Edition

Fai pdf download

https://ebookname.com/product/statistical-thermodynamics-
understanding-the-properties-of-macroscopic-systems-1st-edition-
fai/

Get the full ebook with Bonus Features for a Better Reading Experience on ebookname.com
Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Chemical thermodynamics of materials macroscopic and

microscopic aspects 1st Edition Svein Stølen

https://ebookname.com/product/chemical-thermodynamics-of-
materials-macroscopic-and-microscopic-aspects-1st-edition-svein-
stolen/

Elements of Statistical Thermodynamics Second Edition

Leonard K. Nash

https://ebookname.com/product/elements-of-statistical-
thermodynamics-second-edition-leonard-k-nash/

Entropy and the Time Evolution of Macroscopic Systems

Walter T. Grandy Jr.

https://ebookname.com/product/entropy-and-the-time-evolution-of-
macroscopic-systems-walter-t-grandy-jr/

The Premixing method prinicple design and construction

1st Edition Engan Kaihatsu Gijutsu Kenky■ Sent■ (Japan)

https://ebookname.com/product/the-premixing-method-prinicple-
design-and-construction-1st-edition-engan-kaihatsu-gijutsu-
kenkyu-senta-japan/
A HISTORY OF INDIA Second Edition Burton Stein

https://ebookname.com/product/a-history-of-india-second-edition-
burton-stein/

Fraught Intimacies Non Monogamy in the Public Sphere

Nathan Rambukkana

https://ebookname.com/product/fraught-intimacies-non-monogamy-in-
the-public-sphere-nathan-rambukkana/

Public Health 1st Edition Angela Scriven

https://ebookname.com/product/public-health-1st-edition-angela-
scriven/

The Volatility Smile 1st Edition Emanuel Derman

https://ebookname.com/product/the-volatility-smile-1st-edition-
emanuel-derman/

Fitness walking 2nd ed Edition Therese Iknoian

https://ebookname.com/product/fitness-walking-2nd-ed-edition-
therese-iknoian/
Getting into Films and Television How to Spot the
Opportunities and Find the Best Way in 9th Revised
Edition Robert Angell

https://ebookname.com/product/getting-into-films-and-television-
how-to-spot-the-opportunities-and-find-the-best-way-in-9th-
revised-edition-robert-angell/
 Statistical
Thermodynamics
Understanding the ProPerties of
MacroscoPic systeMs

Lukong Cornelius Fai

GARY Matthew Wysin
 Statistical
Thermodynamics
Understanding the ProPerties
of MacroscoPic systeMs
 Statistical
Thermodynamics
Understanding the ProPerties
of MacroscoPic systeMs

Lukong Cornelius Fai

GARY Matthew Wysin
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2013 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 2012918

International Standard Book Number-13: 978-1-4665-1068-5 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable
efforts have been made to publish reliable data and information, but the author and publisher cannot
assume responsibility for the validity of all materials or the consequences of their use. The authors and
publishers have attempted to trace the copyright holders of all material reproduced in this publication
and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any
future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information stor-
age or retrieval system, without written permission from the publishers.

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

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface.......................................................................................................................xi
Authors.................................................................................................................... xiii

Chapter 1 Basic Principles of Statistical Physics................................................... 1

1.1 Microscopic and Macroscopic Description of States................. 1
1.2 Basic Postulates.......................................................................... 2
1.3 Gibbs Ergodic Assumption.........................................................3
1.4 Gibbsian Ensembles................................................................... 4
1.5 Experimental Basis of Statistical Mechanics.............................5
1.6 Definition of Expectation Values................................................ 6
1.7 Ergodic Principle and Expectation Values................................. 9
1.8 Properties of Distribution Functions........................................ 14
1.8.1 About Probabilities...................................................... 14
1.8.2 Normalization Requirement of
Distribution Functions................................................. 15
1.8.3 Property of Multiplicity of Distribution
Functions..................................................................... 15
1.9 Relative Fluctuation of an Additive
Macroscopic Parameter............................................................ 16
1.9.1 Questions and Answers...............................................20
1.10 Liouville Theorem....................................................................26
1.10.1 Questions and Answers............................................... 30
1.11 Gibbs Microcanonical Ensemble.............................................. 39
1.12 Microcanonical Distribution in Quantum Mechanics..............44
1.13 Density Matrix.......................................................................... 48
1.14 Density Matrix in Energy Representation................................ 51
1.15 Entropy..................................................................................... 56
1.15.1 Entropy of Microcanonical Distribution..................... 58
1.15.2 Exact and “Inexact” Differentials...............................60
1.15.3 Properties of Entropy.................................................. 61

Chapter 2 Thermodynamic Functions................................................................. 67

2.1 Temperature.............................................................................. 67
2.2 Adiabatic Processes.................................................................. 72
2.3 Pressure.................................................................................... 73
2.3.1 Questions on Stationary Distributions Functions
and Ideal Gas Statistics............................................... 75
2.4 Thermodynamic Identity..........................................................84
2.5 Laws of Thermodynamics........................................................ 88

v
vi Contents

2.5.1 First Law of Thermodynamics.................................... 89

2.5.2 Second Law of Thermodynamics............................... 91
2.6 Thermodynamic Potentials, Maxwell Relations...................... 93
2.7 Heat Capacity and Equation of State........................................ 98
2.8 Jacobian Method..................................................................... 100
2.9 Joule–Thomson Process......................................................... 105
2.10 Maximum Work..................................................................... 109
2.11 Condition for Equilibrium and Stability in an
Isolated System....................................................................... 113
2.12 Thermodynamic Inequalities................................................. 118
2.13 Third Law of Thermodynamics............................................. 121
2.13.1 Nernst Theorem......................................................... 122
2.14 Dependence of Thermodynamic Functions
on Number of Particles........................................................... 124
2.15 Equilibrium in an External Force Field.................................. 129

Chapter 3 Canonical Distribution...................................................................... 135

3.1 Gibbs Canonical Distribution................................................. 135
3.2 Basic Formulas of Statistical Physics..................................... 139
3.3 Maxwell Distribution.............................................................. 160
3.4 Experimental Basis of Statistical Mechanics......................... 179
3.5 Grand Canonical Distribution................................................ 180
3.6 Extremum of Canonical Distribution Function...................... 185

Chapter 4 Ideal Gases........................................................................................ 189

4.1 Occupation Number................................................................ 189
4.2 Boltzmann Distribution.......................................................... 191
4.2.1 Distribution with Respect to Coordinates................. 195
4.3 Entropy of a Nonequilibrium Boltzmann Gas........................ 196
4.4 Free Energy of the Ideal Boltzmann Gas...............................200
4.5 Equipartition Theorem........................................................... 223
4.6 Monatomic Gas...................................................................... 227
4.7 Vibrations of Diatomic Molecules.......................................... 231
4.8 Rotation of Diatomic Molecules............................................. 235
4.9 Nuclear Spin Effects............................................................... 239
4.10 Electronic Angular Momentum Effect...................................244
4.11 Experiment and Statistical Ideas............................................ 245
4.11.1 Specific Heats............................................................246

Chapter 5 Quantum Statistics of Ideal Gases.................................................... 247

5.1 Maxwell–Boltzmann, Bose–Einstein, and
Fermi–Dirac Statistics............................................................ 247
5.2 Generalized Thermodynamic Potential
for a Quantum Ideal Gas........................................................248
5.3 Fermi–Dirac and Bose–Einstein Distributions...................... 249
Contents vii

5.4 Entropy of Nonequilibrium Fermi and Bose Gases............... 252

5.4.1 Fermi Gas.................................................................. 252
5.4.2 Bose Gas.................................................................... 256
5.5 Thermodynamic Functions for Quantum Gases.................... 258
5.6 Properties of Weakly Degenerate Quantum Gases................ 263
5.6.1 Fermi Energy............................................................. 263
5.7 Degenerate Electronic Gas at Temperature
Different from Zero................................................................266
5.8 Experimental Basis of Statistical Mechanics......................... 271
5.9 Application of Statistics to an Intrinsic Semiconductor......... 272
5.9.1 Concentration of Carriers.......................................... 273
5.10 Application of Statistics to Extrinsic Semiconductor............. 279
5.11 Degenerate Bose Gas..............................................................284
5.11.1 Condensation of Bose Gases.....................................284
5.12 Equilibrium or Black Body Radiation.................................... 288
5.12.1 Electromagnetic Eigenmodes of a Cavity................. 288
5.13 Application of Statistical Thermodynamics
to Electromagnetic Eigenmodes............................................. 294

Chapter 6 Electron Gas in a Magnetic Field...................................................... 305

6.1 Evaluation of Diamagnetism of a Free Electron Gas;
Density Matrix for a Free Electron Gas................................. 305
6.2 Evaluation of Free Energy...................................................... 316
6.3 Application to a Degenerate Gas............................................ 318
6.4 Evaluation of Contour Integrals............................................. 320
6.5 Diamagnetism of a Free Electron Gas; Oscillatory Effect..... 324

Chapter 7 Magnetic and Dielectric Materials.................................................... 329

7.1 Thermodynamics of Magnetic Materials in a
Magnetic Field........................................................................ 329
7.2 Thermodynamics of Dielectric Materials in an
Electric Field.......................................................................... 333
7.3 Magnetic Effects in Materials................................................ 336
7.4 Adiabatic Cooling by Demagnetization.................................340

Chapter 8 Lattice Dynamics.............................................................................. 343

8.1 Periodic Functions of a Reciprocal Lattice............................ 343
8.2 Reciprocal Lattice.................................................................. 343
8.3 Vibrational Modes of a Monatomic Lattice........................... 347
8.3.1 Linear Monatomic Chain..........................................348
8.3.2 Density of States........................................................ 358
8.4 Vibrational Modes of a Diatomic Linear Chain..................... 359
8.5 Vibrational Modes in a Three-Dimensional Crystal..............364
8.5.1 Properties of the Dynamical Matrix......................... 369
viii Contents

8.5.2 Cyclic Boundary for Three-Dimensional Cases....... 373

8.5.2.1 Born–Von Karman Cyclic Condition........ 373
8.6 Normal Vibration of a Three-Dimensional Crystal............... 375

Chapter 9 Condensed Bodies............................................................................. 389

9.1 Application of Statistical Thermodynamics to Phonons........ 389
9.2 Free Energy of Condensed Bodies in the
Harmonic Approximation...................................................... 391
9.3 Condensed Bodies at Low Temperatures............................... 394
9.4 Condensed Bodies at High Temperatures.............................. 397
9.5 Debye Temperature Approximation....................................... 398
9.6 Volume Coefficient of Expansion...........................................403
9.7 Experimental Basis of Statistical Mechanics.........................405

Chapter 10 Multiphase Systems...........................................................................407

10.1 Clausius–Clapeyron Formula.................................................407
10.2 Critical Point........................................................................... 413

Chapter 11 Macroscopic Quantum Effects: Superfluid Liquid Helium.............. 421

11.1 Nature of the Lambda Transition........................................... 421
11.2 Properties of Liquid Helium................................................... 424
11.3 Landau Theory of Liquid He II.............................................. 425
11.4 Superfluidity of Liquid Helium.............................................. 430

Chapter 12 Nonideal Classical Gases.................................................................. 435

12.1 Pair Interactions Approximation............................................ 435
12.2 Van Der Waals Equation........................................................ 441
12.3 Completely Ionized Gas......................................................... 442

Chapter 13 Functional Integration in Statistical Physics.....................................449

13.1 Feynman Path Integrals..........................................................449
13.2 Least Action Principle............................................................ 450
13.3 Representation of Transition Amplitude through
Functional Integration............................................................ 456
13.3.1 Transition Amplitude in Hamiltonian Form.............460
13.3.2 Transition Amplitude in Feynman Form................... 463
13.3.3 Example: A Free Particle.......................................... 472
13.4 Transition Amplitudes Using Stationary Phase Method........ 473
13.4.1 Motion in Potential Field........................................... 473
13.4.2 Harmonic Oscillator.................................................. 476
Contents ix

13.5 Representation of Matrix Element of Physical Operator

through Functional Integral.................................................... 479
13.6 Property of Path Integral Due to Events Occurring
in Succession.......................................................................... 481
13.7 Eigenvectors........................................................................... 482
13.8 Transition Amplitude for Time-Independent Hamiltonian...... 483
13.9 Eigenvectors and Energy Spectrum....................................... 485
13.9.1 Harmonic Oscillator Solved via
Transition Amplitude................................................ 485
13.9.2 Coordinate Representation of Transition
Amplitude of Forced Harmonic Oscillator............... 488
13.9.3 Matrix Representation of Transition Amplitude
of Forced Harmonic Oscillator................................. 490
13.10 Schrödinger Equation............................................................. 494
13.11 Green Function for Schrödinger Equation............................. 496
13.12 Functional Integration in Quantum Statistical Mechanics..... 497
13.13 Statistical Physics in Representation of Path Integrals.......... 497
13.14 Partition Function of Forced Harmonic Oscillator................ 503
13.15 Feynman Variational Method.................................................504
13.15.1 Proof of Feynman Inequality....................................506
13.15.2 Application of Feynman Inequality.......................... 507
13.16 Feynman Polaron Energy.......................................................509

References.............................................................................................................. 519
Preface
This book on statistical thermodynamics has been written with professional theoreti-
cal physicists in mind. The text is appropriate for training of graduate and postgradu-
ate students in theoretical physics. The authors present the modern level of evolution
of statistical thermodynamics. In presenting the material, the pivot is the application
of the general and fundamental principles of theoretical physics. The student should
be acquainted with the basics of statistical thermodynamics and in particular, its
experimental basis. The chapters go beyond classical thermodynamics and describe
how it relates to statistical mechanics. Knowledge of the microscopic constituents of
a system enables the evaluation of their macroscopic parameters, via the approach of
statistical thermodynamics.
Although statistical physics differs from thermodynamics, both complement each
other. Thermodynamics is the macroscopic phenomenological theory of heat. The
physical system in thermodynamics is mapped onto some large totality of macro-
scopic measurable quantities (temperature, pressure, heat capacity, etc.). The link
between them and the general laws they obey are developed from axioms in a logical
manner and postulated based on experimental facts.
Statistical physics deviates from thermodynamics in that it describes the atomic
system through a model or theory for the microscopic motions of matter in the pres-
ence of statistical fluctuations due generally to heat. The effects of heat are random
and must be calculated through a statistical approach, based on the ideas of probabil-
ity theory. All relations between macroscopic measurable quantities that arise from
thermodynamics are contained in the general statistical theory as relations between
corresponding statistical expectations. In this sense thermodynamics is a corollary
of statistical physics. However, thermodynamics represents just a general theory.
This is because on its basis the general axioms are inherent and do not contain con-
crete atomic representations and definite dynamic models of matter.
Both statistical physics and thermodynamics are partitioned into the theory of
equilibrium states and the theory of nonequilibrium processes. In the former, the
theory is based on the probabilities and expectation values of macroscopic param-
eters that are not dependent on time. In the latter, the probabilities and expectations
of measurable parameters depend on time, but generally have a tendency to move
toward the equilibrium conditions.
Statistical physics is partitioned into classical and quantum statistical physics,
dependent on the physical situation of the matter. Some physical phenomena may
be described correctly assuming that the atoms or molecules in the system move
according to the laws of classical mechanics. When statistical physics is constructed
on this model we call it classical statistics. The quantum model is required for the
motion of atoms, molecules, field oscillators, and other matter that when the discrete
energy states are perceptible, or when the wave functions of quantum particles over-
lap. In this case the needed theory is called quantum statistics. We establish relation-
ships between macroscopic parameters of the system when an equilibrium system

xi
xii Preface

is considered. This is done using classical thermodynamics. Valid conclusions are

drawn based on a minimum number of postulates.
Statistical thermodynamics and the related domains of statistical physics and
quantum mechanics nowadays are very important in many fields of research, such as
plasmas, rarefied gas dynamics, nuclear systems, lasers, semiconductors, supercon-
ductivity, ortho- and para-hydrogen, liquid helium, and so on. To study these systems
a statistical approach at microscopic as well as macroscopic scales is needed. Also,
in order to understand the foundations, laws, properties and applications of thermo-
dynamics, a study of statistical physics is extremely helpful.
This book starts with the fundamental principles of statistical physics, and then goes
to thermodynamics, where the criterion for equilibrium is that the entropy be a maxi-
mum. Statistical physics of equilibrium systems are constructed using the Gibbs formal-
ism that is more compatible with quantum mechanics and fits very well with ensemble
statistics. From the Gibbs formalism the Maxwell−Boltzmann, Bose−Einstein, and
Fermi−Dirac statistics follow in a more logical manner. The procedure in the book pres-
ents thermodynamics and statistical physics in a fundamentally cohesive and simple
manner. Lattice dynamics are treated in detail and clearly bring out understanding of
condensed bodies. The chapters aid very well in the understanding of condensed matter
and other areas of physics. The book contains applications on the polaron theory, the
electronic gas in a magnetic field, thermodynamics of dielectrics in an electric field, and
thermodynamics of magnetic materials in a magnetic field. The last chapter describes
statistical thermodynamics with the approach of functional integration and Feynman
path integrals. The book has a wide range of problems with solutions that explain the
theory very well. The authors do not claim originality neither of the problems nor of
their solutions. Their goal is that the problems may help in investigating, illustrating,
and concretizing various aspects of the theoretical course of statistical thermodynamics.
We would like to acknowledge those who have helped us at various stages of
the elaboration and writing, through discussions, criticism, and especially, encour-
agement and support. We single out the contributions of Professor Dr. Wolfgang
Dieterich of Universitaet Konstanz Fachbereich Physik and Professor Victor Fleurov
of Tel-Aviv University for the comments, suggestions, and corrections on the original
manuscript that have certainly contributed to improve the quality of this book. We
are grateful to Malcolm and Julie Eastoe of Norwich, UK for proofreading the first
manuscript in English. Fai Cornelius is very thankful to his wife, Dr. Fai Patricia Bi,
for all her support and encouragement and to his three sons (Fai Fanyuy Nyuydze,
Fai Fondzeyuv Nyuytari, and Fai Ntumfon Tiysiy) for their understanding and moral
support during the writing of this book. Gary Wysin expresses his appreciation for
the encouragement of his wife, Marcilene Sousa Wysin, and the understanding of his
daughter, Nayara Sousa Wysin, while working on the book.
Fai Cornelius wants to acknowledge with gratitude the financial and library sup-
port under the associate scheme received from the Abdus Salam International Centre
for Theoretical Physics, Trieste, Italy and the hospitality during his visits there.
While all the above persons have helped to improve the quality of the book, we
alone are responsible for all errors and inadequacies that remain. We will be grate-
ful to readers to bring errors to our attention so that corrections can be made in the
subsequent edition. The typing of the manuscript has been done by the authors.
Authors
Lukong Cornelius Fai is a professor of condensed matter physics, and head
of Condensed Matter Division as well as head of the Mesoscopic and Multilayer
Structures Laboratory in the Department of Physics, Faculty of Sciences, at the
University of Dschang, Cameroon. He also holds the position of chief of Division
for Cooperation at the same university, and delivers lectures on quantum mechan-
ics, statistical physics, and Feynman path integrals. Professor Fai performs research
in condensed matter and statistical physics, and also is a senior associate at the
Abdus Salam International Centre for Theoretical Physics, Trieste, Italy. Some of
his current research involves studies on the interactions between electrons and pho-
nons in metals, nanocrystals, and quantum dots—in which there are the so-called
polaron and bipolaron excitations. He was awarded a Doctor of Science in physics
and mathematics in 1997 from Moldova State University, based on studies of optical
absorption due to polarons. He obtained Master of Science from the same university
in 1991, which involved research on the quantum excitations in metal−dielectric−­
semiconductor structures.

Gary Wysin is a professor in the Physics Department at Kansas State University

(KSU) since 1988. He obtained a bachelor’s degree in electrical engineering in
1978 at University of Toledo, Ohio, and in 1980, a master’s degree in physics, for
research on optical bistability at the boundary of a metal and a dielectric, also at the
same institution. He was awarded a doctorate in theoretical physics from Cornell
University in 1985 while working on the theory and Monte Carlo and spin-dynamics
simulations for magnetic models with solitons and vortices. While at KSU, he has
lectured on classical electrodynamics, quantum mechanics, computational physics,
statistical physics, and introductory physics and astronomy. He carries out research
in condensed matter physics, including phase transitions, magnetization reversal,
spin waves, solitons and vortices in low-dimensional magnetism, and optical effects
in dielectrics and metals. Some of the research was carried out in collaborations
with scientists at Los Alamos National Laboratory, New Mexico, USA, University of
Bayreuth, Germany, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil,
Universidade Federal de Viçosa, Minas Gerais, Brazil, and Universidade Federal de
Santa Catarina, Florianópolis, Brazil.

xiii
 1 Basic Principles of
Statistical Physics

1.1  ICROSCOPIC AND MACROSCOPIC DESCRIPTION

M
OF STATES
Statistical thermodynamics is the science that describes the effect of heat interacting
with matter, on a microscopic and statistical basis. This heading constitutes the most
essential part of this course: to introduce the reader to the foundations of the subject
and prepare him/her for subsequent headings. In order to complete the structure of
statistical thermodynamics, we create a basis for relating macroscopic properties of
thermodynamic systems to the microscopic properties of their mechanical models at
the atomic level. The detailed studies of individual mechanical systems are carried
out by methods of classical and quantum mechanics [1–38], to obtain the physical
properties of their individual microstates.
Let us examine such microscopic and macroscopic systems. A system is micro-

scopic (small scale) if it is roughly of an atomic scale or of the order of 10 A or less. An
example of such a system is a molecule. Typically, a microscopic system has very few
degrees of freedom, in the mechanical sense, and so a description in terms of statistical
physics is not applicable. A system is macroscopic (large scale) when it is large enough
to be visible in the ordinary sense, say for example, it may be greater than 1 micron
such that it can at least be observed with a microscope with the use of ordinary light.
A macroscopic system usually possesses very many degrees of freedom; hence, a sta-
tistical description of its properties is not only applicable but also necessary. In dealing
with macroscopic systems, one is interested in macroscopic parameters of the system,
such as volume, pressure, magnetic moment, thermal conductivity, and so on, that can
be measured in experiments, and are representative of the average behavior of all its
molecules. For an isolated system, if the macroscopic parameters do not vary with
time, the system is in equilibrium. If the isolated system is not at equilibrium, the
parameters of the given system change with time so as to approach their equilibrium
values, possibly through some very complex processes. Generally, any system in equi-
librium will return to that state or condition even if affected by some temporary outside
force that is later removed. A great part of statistical physics considers only systems in
equilibrium. Unless otherwise stated further in the chapters, we deal with statistical
physics of equilibrium systems.
Macroscopic systems such as gases, liquids, or solids were first investigated from
a macroscopic phenomenological point of view in the nineteenth century and the
laws thereof formed the subject of thermodynamics. In the second half of the twenti-
eth century, the atomic theory of matter gained general acceptance and macroscopic
systems began to be analyzed from the fundamental microscopic point of view as

1
2 Statistical Thermodynamics

TABLE 1.1
Object, Goal, and Method of Thermodynamics and Statistical Physics
Object Goal Method
Thermodynamics Macroscopic body in the Link between macroscopic Phenomenological
thermodynamic parameters
equilibrium state
Statistical Macroscopic body in the From microscopic Probabilistic
thermodynamic description follows Microscopic
equilibrium state macroscopic properties

systems consisting of very many atoms or molecules. The development of quantum

mechanics (QM) after 1926 provided an adequate theory for the description of atoms
and thus opened the way for an analysis of systems composed from very many atoms
on the basis of realistic microscopic concepts. Thus, the basic concept applied in
statistical physics is the idea that macroscopically observed variables can be deter-
mined from the average behavior of many individual atoms. In this chapter, the main
objective is to derive the laws of thermodynamics from statistical physics. We will
derive statistical physics formulas for the evaluation of thermodynamic functions
of a system, giving their values in terms of various controlled variables such as the
system’s temperature.
The object of statistical physics is a study of particular types of laws that are
found in the behavior and properties of macroscopic states, that is, the states of a
large number of separate particles such as atoms and molecules. QM provides the
fundamental laws for evaluating the properties of individual microscopic particles
and their interactions. Statistical physics begin with these results and lays down the
statistical hypothesis about systems with a large number of particles. This hypothesis
includes the principle of equal a priori probabilities for a set of classical or quantum
states, that we see further in the chapters. In general, statistical physics is strongly
based on QM. Here, almost all the important results of statistical physics are devel-
oped using only certain basic concepts and results of QM.
Let us represent the above in Table 1.1.

1.2 BASIC POSTULATES

The goal of thermodynamics is an investigation of the link between macroscopic
parameters, and especially, the temperature dependence of measured quantities.
Statistical physics intends to find these links from a viewpoint of the microscopic
dynamics of the constituent atoms and molecules that make up the system. Although
a quantum mechanical description is the fundamental one, it is possible to describe
much of the problem in the language of classical mechanics and classical statis-
tics. It will become apparent that this point of view is only slightly modified when
the constraints of QM are included. Indeed, we can follow a system’s dynamics as
though it obeys classical mechanics, and then develop the corresponding formalism
Basic Principles of Statistical Physics 3

for quantum statistics based on that. In the end, it must be insured that the quantum
formalism satisfies the Heisenberg uncertainty principle, as well as fundamental
conservation laws. Further, the QM formalism must be able to describe the subtle
ways in which the presence of indistinguishable identical particles affects the QM
states and thus the thermodynamic properties.
In formulating the basic problem of classical statistics we first have to introduce
the notion of phase space that we will use frequently. We use the notion suggested
by Gibbs [38]. Suppose the given macroscopic mechanical system has s degrees of
freedom. This implies that there are s coordinates that we denote by qi, i = 1, 2, …,
s, which characterize the positions (or other generalized coordinates) of the micro-
scopic entities (atoms, molecules) that make up the system.
The state of a classical system at a given moment will be defined by the values of
s coordinates qi and s corresponding velocities qi. In statistics, we use for the charac-
teristics of the system its coordinates qi and canonical momenta pi and not the veloci-
ties, as the momenta are most closely connected to the coordinates, in both classical
and quantum mechanics. For one degree of freedom, the specification of the position
coordinate q and its corresponding momentum p is sufficient, as the laws of classical
mechanics are such that knowledge of q and p at any one time permits prediction of
the values of q and p at any other time. The specification of q and p is equivalent to
specifying a point in a two-dimensional space or phase space. The coordinate and
momentum of a particle change with time. This representative point (q, p) (phase
point) moves through this phase space, describing a curve called the phase trajectory.
Each point in the phase space corresponds to a microscopic state, and then the phase
trajectory shows exactly what states the dynamical system evolves through. As we
really have a thermodynamic system described by s coordinates qi, i = 1, 2, …, s and
s corresponding momenta pi, i = 1, 2, …, s, we then have a total of 2s parameters and
consequently a phase space of 2s dimensions. In general we will use the symbol (q,
p) to represent the full set of s coordinates and their corresponding momenta.
Thus, microscopic states in classical statistical physics make a continuous set of
points in phase space. In quantum statistical physics, the Heisenberg uncertainty
principle does not allow the coordinates and momenta to be known simultaneously
to such perfect precision. Instead, we might think of any point in the phase space
as being smeared out over an area ∆qi ∆pi ≈ h per degree of freedom, where h is
Planck’s constant. Classically or quantum mechanically, we cannot possibly follow
all the microscopic dynamics, especially of a macroscopic sample that might have
on the order of Avogadro’s number of particles. The basic idea of statistical physics
is the existence of a probability distribution for the relative frequency that the dif-
ferent microscopic states are visited by the system. If we only know the probability
that each state will be visited, then statistics can be used to predict real physical
measurements on the system.

1.3 GIBBS ERGODIC ASSUMPTION

It is impossible to calculate the full time evolution of a complex macroscopic system,
from which one could, in principle, calculate averages over long time. The founda-
tions of statistical physics and the success of its methods are based on the fact that
4 Statistical Thermodynamics

theoretical values for physical quantities may be found using statistical averages in
the phase space. This is to be contrasted with experimental measurements, which are
typically long-time averages. According to Boltzmann, a system left in free evolu-
tion for a sufficiently long time interval will pass through all of its accessible states,
namely the states with the given value of the total energy if the system is isolated.
This hypothesis is the so called the ergodic hypothesis. The ergodic hypothesis and
the use of the method of statistical methods assumed a central role indicated in the
work of J. Willard Gibbs of 1902 [38]. That approach is strongly based on using
phase space averages, with an appropriate type of probability distribution, to replace
long-time averages.
To make theoretical progress, it is necessary to introduce some postulate on the
relative probability of finding a system in any of its accessible states. For example, let
the system under consideration be isolated. Thus it cannot exchange energy with its
surroundings. From here, the laws of mechanics then imply that the total energy of
the system is conserved. With a given fixed energy, only a limited set of the possible
microscopic states can be visited; these are the accessible states referred to above.
Suppose the isolated system is in equilibrium, characterized by the fact that the
probability of finding the system in any one state is independent of time. All the mac-
roscopic parameters describing the isolated system are then also time independent.
From here, we may conclude that whenever the system is observed, it must be in one
of its accessible states consistent with the constant value of its energy. On the other
hand, this says nothing about exactly which of the accessible states it should be found
in, or more importantly, how often it would be found in a particular state. If the same
system is observed a short time later, it would be expected to be found in a different
one of the accessible states, as the phase point of the system moves with time.
An arbitrary system has complex dynamics in the phase space of 2s dimensions,
so that practically speaking, the future motion of the phase point cannot be obtained.
Instead, we use Boltzmann’s ergodic assumption: “An isolated system in equilib-
rium is equally likely to be in any of its accessible states.” There is no way to prove
this statement for arbitrary systems. In fact, one can find some simple mechanical
systems where it is not true. For very many realistic problems, however, it describes
the average long time behavior of the system exceedingly well, leading to the great
success of statistical physics in many areas.
What do we mean, most generally, by accessible states? There are generally some
specified physical conditions that a system is supposed to satisfy, even for a noniso-
lated system that may not have constant energy. These conditions act as constraints
that limit the number of states in which the system can be found. The accessible
states are the states consistent with these physical constraints.

1.4 GIBBSIAN ENSEMBLES

The ergodic hypothesis is necessary to justify and give physical meaning to this sta-
tistical ensemble method, which we describe in more detail. Consider a macroscopic
system of bodies. We suppose that this system is closed, that is, it does not inter-
act with any other body. Select a small macroscopic system from within the given
system. That selected small macroscopic system is called a subsystem of the given
Basic Principles of Statistical Physics 5

system. This subsystem is still mechanical but not closed as it is subjected to all pos-
sible interactions from the rest of the other particles. These interactions have a com-
plex and confusing character and as a consequence the state of a given subsystem
changes with time in a very complex and confusing manner. Due to the complicated
external influence of the rest of the particles for an exceedingly long interval of time,
the subsystem could be expected to attain all of its possible states.
Some physical quantity, say, the magnetization, could be measured by using
observations of a subsystem over a very long time. On the other hand, one could
imagine doing a different kind of measurement, by using observations of all of the
subsystems simultaneously, without a long-time average. If each subsystem is under
physical conditions the same as all the other subsystems, then they all move through
the same accessible region of phase space. At any given instant, however, they are
distributed with some likelihood or probability distribution over the accessible phase
space. We can use the basic methods of statistical averaging to calculate what should
be measured in the real system, especially, the quantities referred to as expectation
values. The idea is that averaging over the subsystems should give the same result
as a long-time averaging of one subsystem. Besides, one finds that this statistical
method can be carried out more readily than trying to calculate the time evolution of
a complex dynamical system.
Now, we can think about the terminology in a slightly different fashion.
Conceptually, the subsystems are themselves really nothing different than small
copies of the original macroscopic system. Instead of thinking of averages over sub-
systems, we can simply imagine doing averages over many identical copies of the
original system of interest. This is the idea of an ensemble and ensemble averaging.
The term ensemble was coined by Gibbs. A certain number N of copies of the system
are followed or averaged over, as a way to think about how to calculate expectation
values of macroscopic observables. We refer to these copies now not as subsystems
but as members of the ensemble. This is a useful concept especially because the aver-
ages we need are made over the phase space. As long as we know the probability for
the members of the ensemble to be in the different allowed parts of phase space, any
kind of averages can be predicted.

1.5 EXPERIMENTAL BASIS OF STATISTICAL MECHANICS

Statistical thermodynamics is a general theory designed to apply to a very broad
range of material systems and physical situations. It is based strongly on the idea
that all states of a system under given or desired physical conditions (e.g., constant
total energy as in the microcanonical ensemble that is discussed below) are equally
probable. Whether this idea is correct or not can only be verified by comparison of
the theoretical results with many experiments. The basic ideas, such as the ergodic
hypothesis, are not necessarily directly provable in general for all possible systems.
Indeed, even in a system that does nearly explore all of its available phase space,
it can only do so if observed over a sufficiently long time interval. This minimum
observation time depends on the fundamental physical time scales in the problem,
over which the system scans the phase space. As a simple example, an individual
harmonic oscillator must be observed, at the very least, over a time greater than its
6 Statistical Thermodynamics

period, in order that averages over time would be close to the averages obtained by
integration over the accessible phase space.
Of course, any physical system of interest is never truly isolated, and besides that,
we never know the true Hamiltonian. There are always additional interactions that
we cannot account for or even know about. Generally, they might be considered as
due to the “thermal reservoir” or “heat bath,” which in some vague sense can be
expected to mix the system more than we can possibly account for. Further, quan-
tum fluctuations will also be present due to interactions which we may not take into
account. We hope that these additional perturbations that we are not totally aware of
lead the system to behave as ergodic. But again, the best test of the validity of statisti-
cal mechanics is the entire realm of experiments for which it is found to apply, in a
wide variety of physical systems and physical measurements.

1.6 DEFINITION OF EXPECTATION VALUES

Any measurable physical quantity must be determined by the state of the system;
hence, it must be a function like F(q, p). As time passes, the observable F changes,
but when viewed over a long time, an average value will result, which is the expec-
tation value. We imagine first how it would be calculated based on an averaging
process over long time.
The state of any subsystem or system in an ensemble may be specified approxi-
mately by saying that its coordinate lies in some interval between q and q + dq and
its momentum lies in some interval between p and p + dp, that is, by stating that the
representative point (q, p) lies in a particular part of a phase space. We take q and p
to represent all s degrees of freedom, so that each of these has a certain small range.
The Gibbs space has 2s dimensions, where the number of degrees of freedom is s. The
coordinates are qi, i = 1, 2, …, s and the corresponding momenta are pi, i = 1, 2, …, s.
Thus, q ≡ {qi}, p ≡ {pi}, dq = dq1dq2 … dqs, dp = dp1dp2 … dps, and the infinitesimal
region surrounding a phase space point is a hypervolume dΓ = dpdq, with

dqdp = dq1dq2 … dqs dp1dp2 … dps (1.1)

Typically, we want to imagine observing when the system passes near a desired
point (q, p) in phase space, that is, in the small volume dΓ = dpdq. Suppose that for
an exceedingly long interval of time T, the confused phase trajectory passes through
an arbitrary part of phase space a very large number of times. Let dt = dt(q, p; dΓ)
be the part of the total time T for which the subsystem is found in that part of phase
space dΓ near (q, p). For an infinitely long time T, the ratio dt/T tends to a limit dW,
which is a number less than unity

dt (q, p; d Γ ) (1.2)
dW (q, p; d Γ ) = lim
T →∞ T

The value dW in Equation 1.2 is the probability that if we observe the subsystem at
an arbitrary moment in time, we find it in a desired infinitesimal volume dΓ = dpdq
Basic Principles of Statistical Physics 7

of phase space surrounding the point (q, p). This means that dW can be used to define
a probability density ρ(q, p) in phase space, which gives probability when multiplied
by a phase space volume:

dW (q, p; d Γ ) = ρ(q, p)d Γ (1.3)

Thus, ρ(q, p) is the probability per unit phase space volume to find the system in a
small region of phase space dΓ = dpdq surrounding the point (q, p). Further, being
a probability density means that a sum (or integral) over all the possible microstates
gives unity:

∫ ρ(q, p) dqdp = 1 (1.4)

Usually we would state this relationship by saying that the probability density is
normalized to unity. This relation just represents the fact that the sum of all the times
spent in different parts of phase space must add up to the total observation time, T.
Or, it can be interpreted to mean that the sum of probabilities for the system to be in
any state of phase space is unity.
Now suppose the observable F(q, p) is measured. The idea of measurement for a
macroscopic system, involves observing the system over long enough times that the
system evolves through a large part of phase space. During that time, the dynami-
cally measured quantity F(t) = F(q(t), p(t)) can be changing; however, the experi-
mental apparatus will only produce a time-averaged value because it cannot follow
the microscopically rapid motions. For an instantaneous measurement of F = F(t) to
give a desired value F(q, p) = F(t) requires that the system pass through the particu-
lar points (q, p)F=F(t) satisfying this constraint, F(q, p) = F(t). In fact, this constraint
could be visualized as contours in the phase space. Alternatively, we can look for
the observable to lie in some range, say, from F to F + dF. There will be some
limited region dΓ of phase space, bounded by contours at F and F + dF, possibly
made from disjoint sections, where F(q, p) lies in the desired range, with a width
dF. As time evolves, F(t) takes on values in some desired range from F to F + dF
only during some total time interval dt, out of a total observation time T, which is
a kind of mapping:

F ≤ F (t ) ≤ F + dF ⇒ t ∈ dt ( F ; dF ) (1.5)

Thought of in this way, Equation 1.2 can be interpreted as the probability for finding
F in a desired range:

dt ( F ; dF ) (1.6)
dW ( F ; dF ) = lim
T →∞ T

Here, of course, dt(F; dF) means the total time the system spent within the correct
region of phase space to give F(q, p) in the range from F to F + dF; the system may
be dynamically going into and out of that region.
8 Statistical Thermodynamics

The expectation value of the observable F(q, p), denoted as F or as 〈F〉, could be
found as its average using this probability:

F = ∫ F dW (F; dF ) (1.7)

In fact, this allows the definition of the corresponding unit-normalized probability

density for F, similar to the probability density per unit phase space ρ(q, p). We can
denote it as ρF(F), such that

dW ( F ; dF ) = ρF (F ) dF (1.8)

F = ∫ Fρ F ( F ) dF (1.9)

where ∫ρ F ( F ) dF = 1 (1.10)

These results show how the time averaging of F leads to a probability distribu-
tion for F. However, the expectation value of F can be viewed differently, using the
Gibbs phase space and the idea of ensemble averaging. Let us examine the quantity
F(t) of a dynamical state of a system. As time evolves, F(t) is changing ultimately
because the coordinates and momenta are evolving. Instead of thinking of a prob-
ability for F, it is better to think of the probability for a system to be in a desired
infinitesimal element of phase space, dΓ = dpdq, surrounding some point (q, p).
Thinking this way means that we imagine a very large number N of identical copies
of the system, all evolving together in phase space. This is the ensemble. When the
ensemble is observed at some instant of time, its coordinates and momenta will be
distributed in phase space, according to a probability distribution. At any instant,
there can be expected to be a number dN(q, p;dΓ) of members of the ensemble
within a desired volume element dΓ = dpdq. Indeed, that probability distribution
has the probability density ρ(q, p) introduced earlier. As time evolves, the “posi-
tions” (q, p)m of each member m of the ensemble change, following trajectories in
phase space. However, how they are distributed in phase space should remain the
same, if all are subjected to the same conditions and the system is in equilibrium.
Therefore, this probability distribution can be used to compute averages, that is,
expectation values. Further, one probability distribution will apply to averages for
all dynamical observables.
We already saw that the probability of observing any system in the volume
dΓ = dpdq is related to the time spent by a system within dΓ. Alternatively, we sim-
ply look at what number dN of the N systems in the ensemble are in the desired phase
space element to define the probability:

dt (q, p; d Γ ) dN (q, p; d Γ )
dW (q, p; d Γ ) = = (1.11)
T N
Basic Principles of Statistical Physics 9

The different systems occupy the different microstates with the probability dW.
Hence, this can be reexpressed in terms of the probability density in phase space:

dN (q, p) (1.12)
dW = = ρ(q, p) dqdp
N

where ρ(q, p) is the probability density (or distribution function) of phase points in
the given phase space. Here, the argument dΓ has been suppressed. The number
of macroscopically identical systems distributed among accessible microstates by
the distribution function ρ(q, p) defines a statistical ensemble. Thus, this statistical
ensemble is defined and characterized by ρ(q, p). The quantity dW or ρ(q, p) dqdp
is the probability that the dynamical state of any system in the ensemble is in dqdp
about the point (q, p).
With the probability of a given microstate being ρ(q, p) dqdp, the ensemble aver-
age (statistical average) of any function F(q, p) depending on the dynamical state of
a system is

F = ∫ F (q, p) ρ(q, p) dqdp (1.13)

The single integral sign means integration over all the p and q (the integral is taken
over all of phase space). Although we used the idea of an ensemble, there is no
explicit sum over its members. That has been accomplished by the use of the prob-
ability density. What has not been fully specified, however, is the actual probability
distribution ρ(q, p). Its definition depends on the conditions that determine the acces-
sible states.
We should also mention one further advantage of the ensemble method over the
concept of long-time averages. A physical system could be subjected to time-dependent
external forces. In that case, long-time averages would smear out possibly important
dynamical effects. The only way to make a sensible theoretical calculation of macro-
scopic averages that depend on time is to imagine using an ensemble average.

1.7 ERGODIC PRINCIPLE AND EXPECTATION VALUES

In this section, we discuss the correlation between the two types of expectation values
that are of interest, that is, statistical expectation values and chronological expecta-
tion values. The statistical expectation value is the ensemble average of a thermody-
namic quantity F(q, p), q = q(t), p = p(t), say, at a given time t over all systems of the
ensemble. The chronological expectation value is the time average of the thermody-
namic quantity F(q, p), say, for the given system of the ensemble over an exceedingly
long time interval 2T (i.e., where T → ∞). For simplicity, it is assumed that the only
time dependence of F is that implicitly due to the time variation in (q(t), p(t)).
The correlation between the statistical and the chronological expectations val-
ues will aid in choosing the specific form of the distribution function that will
be used in computing statistical expectations. In particular, for an equilibrium
10 Statistical Thermodynamics

situation, we will find that the distribution function ρ will be independent of

time t and dependent only on q and p.
The statistical expectation value of the quantity F(q, p) is denoted interchange-
ably by F or F and is defined by using an average over the ensemble:

1 1
F (t ) ≡ F (t ) ≡
N ∑ F (t ) = N ∑ F ({q(t )} ,{p(t )} ) → ∫ ρ(q, p; t ) F (q, p) dqdp
i i i
i i
(1.14)

or

∫
F (t ) = ρ(q, p; t ) F (q, p) dqdp (1.15)

Here, Fi(t) is the value assumed by F(q, p) in the ith system of the ensemble, N is
exceedingly large and is the total number of systems in the ensemble, and the coordi-
nates and momenta in the ith system, {q(t)}i, {p(t)}i, are evaluated at a particular time
t. The result could depend on the time t through ρ(q, p; t), if the number of members
of the ensemble within dΓ = dqdp depends on time, that is, dN(q, p; t).
The chronological expectation value of the quantity F(q, p) is denoted inter-
changeably by {F} or F . It is defined for the ith system (perfectly isolated) of the
ensemble by a symmetrical time average

T T
1 1
Fi (t ) = {Fi (t )} = ∫ Fi (t + t ′ ) dt ′ = ∫ F ({q(t + t ′)} ,{p(t + t ′)} ) dt ′
i i (1.16)
2T 2T
−T −T

The chronological and statistical expectations commute because the time and phase
space integration commute. To see this, consider the statistical expectation of the
time average, F (t ):

1 1  1 T
 1 T
 1 
F (t ) =
N ∑
i
Fi (t ) =
N ∑
i

 2T ∫ Fi (t + t ′ ) dt ′  =
 2T ∫  N ∑ i
Fi (t + t ′ )  dt ′

−T −T
T T
1 1 
→
2T ∫  ∫ ρ(q, p; t + t ′)F (q, p)dqdp dt ′ = 2T ∫ F (t + t ′) dt ′ = F (t )
−T −T
(1.17)

It follows that

F (t ) = F (t ) (1.18)

The quantity T is exceedingly long (“microscopically long”) so that F becomes inde-

pendent of T through the smoothing out of microscopic fluctuations.
Basic Principles of Statistical Physics 11

We consider now a stationary situation for the quantity F. This implies that there
exists no preferred origin in time for the statistical description of F. That is to say, we
ensure the same ensemble when all member functions F of the ensemble are shifted
by arbitrary amounts in time. This is true for all statistical quantities for an equilib-
rium situation. Here, there is an intimate union between the statistical and chrono-
logical expectations of the ensemble. This is on condition that the function F(t) for
an exceedingly long time will pass through all values accessible to it (the ergodic
assumption). Here, we do not consider some exceptional systems in the ensemble.
Consider again the function F(t) as observed in a system of the ensemble. We parti-
tion the time scale into exceedingly long intervals of magnitude 2T. As T is exceed-
ingly large, the behavior of F(t) in each partition is independent of its behavior in any
other partition. Consider some large number N for such partitions. Then, this is a
representation of the statistical behavior of F in the original ensemble. Let us view a
system in equilibrium. A way to define equilibrium is to say that chronological expec-
tation values do not depend on the choice of the origin of the long time interval. Then,
the chronological and statistical expectations should be equivalent. It follows that

F (t ) = F (1.19)

The chronological expectation is independent of the partition, say the ith. Similarly,
it must be true that for such a stationary ensemble the statistical expectation of F
must be independent of time, because ρ is independent of time:

F (t ) = F (1.20)

Consider again Equation 1.18. Examine the statistical expectation of Equation 1.19;
then

F (t ) = F (1.21)

Thus, if we consider Equation 1.18, then for a stationary ergodic ensemble, we have
the important result

F = F (1.22)

Let us again examine the phase space probability density ρ. We may get an exact
integral of motion for ρ, if a system is isolated. Gibbs proposed that the first integral
of motion of ρ is the energy E, and as a consequence E can be considered as the
independent variable:

ρ = ρ( E ), E = E (q, p) (1.23)

This form for ρ corresponds to what is called the microcanonical ensemble (system
at fixed energy). Let us examine a constant energy surface
E = E (q, p) = E0 = const (1.24)

in the 2s-dimensional phase space (q, p), represented abstractly in Figure 1.1.
12 Statistical Thermodynamics

(q1, p1)

q
(q2, p2)

FIGURE 1.1 An energy surface in a 2s-dimensional phase space.

On the given constant energy surface, consider the points (q1, p1) and (q2, p2), cor-
responding to two different accessible states, which must satisfy the requirements

E1 = E (q1 , p1 ) = E0 , E2 = E (q2 , p2 ) = E0 (1.25)

Then, under Gibbs’ assumption that the density can be determined from the energy
alone

ρ(q1 , p1 ) = ρ( E0 ), ρ(q2 , p2 ) = ρ( E0 ) (1.26)

It follows that on the constant energy surface, the densities are equal

ρ(q1 , p1 ) = ρ(q2 , p2 ) (1.27)

This is for an isolated equilibrium system.

From the definition

dW (q, p) = ρ(q, p)dqdp (1.28)

we conclude that the probabilities of all microstates with the given value of the
energy are the same. This is one of the most important fundamental principles in
statistical mechanics. It serves as a guiding principle for all ensemble averages: “The
probabilities of all accessible microstates are the same.”
In classical mechanics, q and p are continuous variables and define a microstate.
Mathematically, however, the probability is only defined for finding a system in a
microstate within some phase space region dqdp. In QM, the state of one degree
of freedom cannot be defined more precisely than ΔqΔp = 2π . It follows from the
uncertainty principle that one cannot completely localize a system in phase space. In
other words, each quantum state occupies a phase volume ΔqΔp = 2π  per degree of
Basic Principles of Statistical Physics 13

freedom, or a total hypervolume (2π)s for a system with s degrees of freedom. In

a quantum system, Gibbs’ assumption still holds: all quantum states consistent with
the conservation laws have equal probability.
Based on Gibbs’ assumption, the average (mean) statistical value F for the system
at chosen energy E 0 is

F = F = ∫ F (q, p) dW (q, p) = ∫ F (q, p) ρ(q, p) dqdp (1.29)

In the simplest case of an isolated system, the chronological expectation coincides

with the statistical expectation value

F = F (1.30)

We see that if the chronological expectation coincides with the expectation of the
statistical value, then statistical physics has a firm foundation. The ergodic problem
consists precisely of this justification and finds the correspondence between par-
ticular values of phase space functions and the experimental measurements for the
physical quantity. Relation (1.30) shows that time averaging is equivalent to averag-
ing over microcanonical ensembles and the property is often called ergodicity.
We see here that ρ really represents the probability density of points in phase
space and that is why the expectation value of the quantity F is properly computed
by relation (1.29).
Let us examine relation (1.18) from the light of the Schrödinger and the Heisenberg
pictures. We rewrite Equation 1.18 again in the form

F (t ) = ∫ dqdp F (q, p, t ) ρ(q, p, t ) = ∫ dqdp F (q, p, t ) exp{−iL t} ρ(q, p, 0)

0
s
(1.31)

The operator L̂s evolves the density or other physical quantities forward in
time. The quantity F at time t 0 corresponds to the Schrödinger picture in QM
where F(q, p, t 0) = F(q, p) is fixed in time and ρ(q, p, t) evolves with time. If we
perform integration by parts of the relation in Equation 1.31, then we have as
follows:

F (t ) = ∫ dqdp F (q, p, t ) ρ(q, p, 0) = ∫ dqdp ρ(q, p, 0) exp{−iLt} F (q, p, t )

s
(1.32)

This corresponds to the Heisenberg picture in QM where

∂F (q, p, t )  s (1.33)
i = L F (q, p, t )
∂t

We see that the probability density ρ(q, p) for any given configuration (q, p) may be
interpreted as representing a particular realization of a system of ensembles.
14 Statistical Thermodynamics

1.8 PROPERTIES OF DISTRIBUTION FUNCTIONS

A probability distribution function ρ(q, p) relates to the likelihood of finding a sys-
tem of the ensemble in a region of phase space surrounding (q, p). As a probability
distribution, it satisfies some basic properties.

1.8.1 About Probabilities

Let us review some principles of the probability, related to events or situations
denoted by symbols A and B, within some space of possible events:

1. Addition: If the two events can both occur but if they are mutually exclusive
(they occupy different nonoverlapping spaces in a Venn diagram), the prob-
ability that one or the other occurs is

P( A ∪ B) = P( A) + P( B) (1.34)

The symbol P(A ∪ B) is the probability of the union of A and B. This only
applies if the events A and B cannot take place together. Otherwise, they
have some overlapping part in the Venn diagram, which has the probability
of intersection, represented with the symbol P(A ∩ B). Then, the probabil-
ity that either A or B occurs, or both occur, is

P( A ∪ B) = P( A) + P( B) − P( A ∩ B) (1.35)

The last term corrects the overcounting of probability in the first two terms
by the intersection of the spaces of A and B.
2. Multiplication: If two events are mutually independent, the probability that
they both occur together is

P( A ∩ B) = P( A)P ( B) (1.36)

Further, if there are more than just two events, all mutually independent,
then the probability of all of them happening together is just the product of
all of their separate probabilities.
3. Conditional Probability: One may be interested in the probability that B
occurs, given that A is already present. This is the conditional probability,
denoted by the symbol, P ( B|A). It is equal to

P ( A ∩ B) (1.37)
P( B| A) =
P ( B)

If the events are independent, then Equation 1.36 can be applied and there
results only

P ( B| A) = P( A) (1.38)
Exploring the Variety of Random
Documents with Different Content
young man, before he had succeeded in making his own fortune,
and left a small family. Cornelis Van Heusen married Hetty Klink, of a
family which had been living in Albany some time. She was born
there, in 1782, and died at Albany, September 24. 1873. (VII)
Theodore Van Wyck Van Heusen, son of Cornelis and Hetty (Klink)
Van Heusen, was born in his father's home on the south side of Plain
street, Albany, New York, November 11, 1818, died at his residence.
No. 6 Madison Place, Albany, June 15, 1893. His father sent him to
the old Lancaster School, at the southwest corner of Eagle and
Lancaster streets, which was later converted into the Albany Medical
College, and he graduated there when thirteen years old. Shortly
afterwards, he entered the employ of Messrs. Mcintosh, being forced
by his father's early death to earn his living at an earlier age than
otherwise would have been the case, and in a few years he was
made the head clerk. In 1843, when twenty-five years old, he and a
fellow clerk, Daniel D. Tompkins Charles, formed the firm of Van
Heusen & Charles, to deal in crockery and a few household articles,
employing $1,000 they had saved as their cash basis. They then
occupied the building No. 66 State street, but in 1844 they removed
into larger quarters at Nos. 62-64 State street. By 1856 they had
succeeded so well that they purchased the famous old hostelry, Nos.
468470 Broadway, known all over the state as the Mansion House,
and upon its site erected their own large building, extending
westward through the block to include Nos. 9-1 1 James street. He
showed himself capable of rending the barriers of inauspicious
circumstances so as to win the victorj' of abundant success in the
business world, and before he died he was looked upon as the
leader among such business men in this country. The original name
of the firm of Van Heusen & Charles was changed in 1864 to Van
Heusen, Charles & Company, by the admission of George W. Pierce
as a partner. Mr. Charles, a muchrespected citizen of Albany, residing
opposite the Fort Orange Club on Washington avenue, died August i,
1892, and soon afterward the finn adopted its present style of the
Van Heusen-Charles Company. It is now the oldest and largest
enterprise of its kind not
 I038 SOUTHERN NEW YORK alone in Albany, but in eastern
New York, and vies with those in the metropolis. Mr. Van Heusen
was an original member of the Young Men's Association, organized
in Albany by Amos Dean, a philanthropist, for the promotion of
culture among the youth. At one time he was its president. He was a
charter member of the Fort Orange Club, the principal social
organization in the capitol city, and among those who started the
Albany Club in the heart of the business district. He was a member
of the Emmanuel Baptist Church from its foundation. He joined the
old and staid Albany Institute while a young man, in order to take
advantage of its lectures by scientists and learned men from the
world over. He joined the Holland Society, June 15, 1886, when it
was forming, and thoroughly delighted in participating at its
banquets, whereat he frequently spoke. When that society's first
president. Judge Hooper Gumming Van Vorst, died in New York City,
October 27, 1889, Mr. Van Heusen was chairman of the committee
providing for the public funeral which took place at Albany, for they
had grown up together, the judge having been born at Schenectady,
December 3, 1817, about a year previous to his own birthday. He
was able to enjoy the advantage of considerable travel, touring
Europe extensively, visiting Bermuda, Mexico, Canada and the West
Indies. Since 185 1, until the last years of his life, he was in the
habit of contributing letters on his travels to the newspapers. These
contained observations on the places and the people, and proved
not only of entertaining information for the public, but demonstrated
his kindly nature in his desire to have others participate in his
pleasure. He also wrote frequently and forcibly in discussing the
questions of the day, evincing literary ability and a taste which
secured him more than a mere local distinction, but which his limited
time prevented him from developing. Not only was he a noticeable
character upon the streets of his native city because of his
gentlemanly appearance and bearing and by reason of his large
enterprise as a merchant, but he also figured in the greater activities
of the city's life. In 1882 he was the Republican nominee for
congress, but the city vote at that period and for some time previous
had been overwhelmingly Democratic, so that to make any showing
the Republicans had to put forth their best representative, and
victory proved out of the question. He ever remained a staunch
Republican, but his business affairs never allowed him to enter the
political field at any other time. He was known widely as a man who
was fond of living and dressing well ; but beyond these material yet
consequential matters of a gentleman he had a still stronger liking
for art and history, especially that of his city, where he was regarded
as a fine type of a representative of the old Dutch stock of the days
of Fort Orange and Rensselaerwyck. He suft'ered a stroke of
paralysis, September 26, 1890, and after that was confined to his
residence. No. 6 Madison place, until he died. Theodore V. Van
Heusen married (first) at Amsterdam, New York, January 10, 1844,
Catherine Conover Vander Veer, of that city, by whom he had three
sons and one daughter, and she died in 1855. He married (second)
Helen M. Burroughs, of Medina, New York, who died when they had
been married four years, without issue. He married (third) at
Jamaica Plain, Boston, Massachusetts, June 23, 1863, Arabella
Manning. Her parents were \\''i!liam Manning, a highly respected
citizen of Boston, and his wife. She was descended from Joseph
Manning, in Captain Olney's company, Angell's Rhode Island
Regiment, in 1780-82. Mrs. Van Heusen was born at Boston, May 3,
1837, died at her home, No. 6 Madison place, Albany, January 18,
1912. She devoted nearly all of her time to philanthropy. In 1885
she was vice-president of the Women's Albany Indian Association,
whose purpose was to arouse sentiment to right the wrongs of the
Indians in order to have congress enact certain measures for their
upbuilding. Until her health failed, she was constant as a teacher of
a Bible class of the Emmanuel Baptist Church, and she took
particular interest in assisting the Chinese of her city, as well as in
visiting the sick, the poor and distressed. She and her husband were
buried in the Albany Rural Cemetery. (VIII) Alpha Tunis Van Heusen,
son of Theodore Van Wyck and Catherine Conover (Vander Veer)
Van Heusen, was born at /ilbany, January i, 1845, ^'^^ resides at
Amsterdam, New York. He was educated at the Albany Academy,
and in the fall of 1863 re 
 SOUTHERN NEW YORK 1039 moved to Amsterdam, settling
first upon a farm, where he Hved five years, when he moved into
the city, February, 1869, buying out the furniture and crockery store
of Kennicott & Wilde, but owing to ill health he was compelled to
give up business for three years. At the end of that time, he bought
out the oldest and leading insurance agency in Amsterdam, which,
with his son, Theodore, he conducts under the name of A. T. Van
Heusen & Son. In 1883 he was appointed justice of the peace, and
was re-elected three times, serving a total of sixteen years, the
longest period in which that office was held there by any one man.
The common council designated him associate recorder for eight
years. He formerly attended the Baptist church, holding every official
position except that of deacon, but now attends the Presbyterian
church. He is a Republican; was for thirtythree years a member of
Artisan Lodge of Masons, and for twelve years the king, or high
priest of Amsterdam Chapter, No. 81, Royal Arch Masons. Alpha
Tunis Van Heusen married, at Albany, March 24, 1864, Maria Vander
Veer, born at Amsterdam, March 3, 1845, died there, December 24,
1904, daughter of John Watts and Jane Eliza (Marsh) Vander Veer.
All their children were born at Amsterdam. Children: i. Frank \'ander
Veer, born May 13, 1865, died there, June 10, 1893. 2. Jennie
Marsh, born May 28, 1868, died there, October 24, 1882. 3. Howard,
born December 16, 1869, died there, July 24, 1870. 4. Edward
Alpha, born July 21, 1871, died there, August 19, 1872. 5. Frederick,
born November 19, 1874, died there, September 28, 1875. 6.
Catherine Conover, born February 23, 1876; married there, June,
1900, George L. Bennett. 7. Theodore Vander Veer, born November
4, 1884; married, at Amsterdam, June 5, 1907, Islay Wayne Brown.
(Vni) Friend Humphrey \'an Heusen, son of Theodore \'an Wyck and
Catherine Conover (Vander Veer) Van Heusen, was born at Albany,
June 9, 1847, and died there, in 1858. (VHI) Theodore Van Wyck
\'an Heusen, son of Theodore \'an Wyck and Catherine Conover
(Vander Veer) Van Heusen, was born at Albany, September 30,
1849. died at Washington, D. C, October 23, 1883; married, at
Detroit, Michigan, Margaret De Long. (VHI) Catherine Conover Van
Heusen, daughter of Theodore Van Wyck and Catherine Conover
(Vander Veer) Van Heusen, was born at Albany, December 3, 1851,
died at Cincinnati, Ohio ; married, at Albany, John A. Lamb. (VHI)
William Manning Van Heusen, son of Theodore Van Wyck and
Arabella (Manning) Van Heusen, was born at No. 6 Madison place,
Albany, New York, May 5, 1865. He received his early education at
the Albany Academy, where he was regarded by faculty and students
as an exceptionally bright student. He was one of the higher officers
of the cadet battalion, and figured brilliantly as the president of the
Beck Literary Society. \\'hen he graduated, in 1884, he was awarded
most of the honors, and thereafter entered Harvard University,
graduating in the class of 1888. He then took the course of the
Columbia Law School, and having qualified as an attorney, practiced
individually in NewYork City, opening his office in the Trinity Building,
No. 11 1 Broadway. Imbued with the same gentlemanly
characteristics for which his father had been noted, his courtly,
genial manner, coupled with an unusual brilliancy of intellect, made
many friends for him. Unfortunately, his career was all too soon
shortened by ill health, and he withdrew from activities to seek
strength by a sojourn in Cali/omia. He died at Pasadena, February 3,
1899, and on the loth was buried in the' Albany Rural Cemetery.
(VIII) Charles Manning \^an Heusen, son of Theodore Van Wyck
and Arabella (Manning) Van Heusen, was born at No. 6 Madison
Place, Albany, New York, June 27, 1866. He was educated at the
Albany Academy at the same time as his three brothers, while there
an officer of the cadet battalion and a member of the Beck Literary
Society. He then entered the large establishment conducted so long
by his father, the firm of the Van Heusen & Charles Company, at
Nos. 468-470 Broadway, Albany. When it was incorporated as the
Van Heusen-Charles Company, in 1893, he was made vice-
presid;ent, and not long afterward became its president. The
concern developed noticeably when under his control, enlarging its
quarters extensively both on the main floor and by converting the
several upper floors, which had been used as a warehouse, into
displays rivaling the best
 I040 SOUTHERN NEW YORK metropolitan establishments of
its kind. This allowed him to carry out his plan of adding fine
furniture, costly rugs, jewelry, statuary, etc. For many years he has
been accustomed to make extensive trips abroad during the
summer, combining duty and recreation. Like his father, he has been
a staunch Republican, and was an officer of several party
organizations, such as the Capital City Campaign Club. He belongs to
the Emmanuel Baptist Church and is a member of the Fort Orange,
the Albany Country, and the Racquet & Tennis clubs. He served
several years in the National Guard of New York State, in 1884,
joining Company A, of the Tenth Battalion, and afterwards entered
the Old Guard, known as the Albany Zouave Cadets, to become one
of its most active members. Mr. and Mrs. Van Heusen have a
summer place at Shohola, Pennsylvania, and another one at
Branford, Massachusetts, the latter, named Willbeth, is an estate of
some size containing a lake some three miles long within its
confines, stocked with fish and on which they have their launches.
She is the owner of the famous Greenacre Kennels, at Fairfield,
Connecticut, celebrated in dogdom for its prize chow, samoyede,
Blenheim and toy spaniels, the population of which keimel numljcrs
over one hundred and fifty thoroughbred canines. These have taken
highest prizes at all the great shows in this country, and in some
classes have never been outranked. Charles Manning Van Heusen
married, at New York City, January 21, 1895, Ada Olive (Proctor) ,
born at Elizabeth, New Jersey, September 25, 187 1, daughter of
William Fash and Vouletti Theresa (Singer) Proctor. Mrs. Proctor died
at her apartments in Hotel Lorraine, New York City, December 14,
1913, daughter of Isaac Merritt Singer, who was born in Oswego,
New York, October 17, 181 1, died at his home in Torquay, England,
July 23, 1875. To the credit of this world-famed man, he began life
as a young machinist; was fond of his calling, possessed an
abundance of ingenuity, and made a success of his life which will
never be forgotten. He turned his attention to the particular study of
the sewingmachine, determined to improve it so as to be practical,
simple in operation, and produced at a cost which would admit it to
many homes, so as to be a great helper of the household, which it
has proved. This required years of the closest application of one
skilled at inventions, in order to complete a single-thread, chain-
stitch machine, and his many trials finally terminated in
achievement. He obtained a patent in 1852, and thus insured the
result which amounted to millions of dollars. The agencies now run
into the thousands, and are to be found in India, Alaska, Egypt, and
in every country on the globe. In the early part of his career, he was
assisted by Edward Clark, a wealthy lawyer, and with his aid Singer
was able to establish a factory in New York. A competitor, the Howe
Sewing-Machine Company, sued him for alleged infringement on
their patents. The result was a compromise. He had a difficulty with
Mr. Clark, in consequence of which, while each retained an equal
interest in the machine, its manufacture was placed in the hands of
a company. On the accumulation of a fortune, he left this country
and for some time resided in Paris, but later removed to Torquay,
England, where he lived in a curiously constructed house. The Singer
Building at No. 149 Broadway, New York City, was named in his
honor. Children of Mr. and Mrs. Van Heusen: I. Elizabeth Proctor Van
Heusen (by a former marriage of Mrs. Van Heusen, but by law Van
Heusen), born in New York City, February 4, 1893 ; married, in New
York City, in 1914, Arthur Gould Hamilton; he was born in New York
City, August 17, 1892, son of William Arthur and Mary Lilian
(Brumell) Hamilton; he has allied himself with metropolitan
newspaper interests ; William A. Hamilton was born at Cookstown,
Ireland, September 27, i860, son of Rev. Robert and Jane (Corrigan)
Hamilton; resides in New York City; married, at the Church of the
Holy Trinity, Toronto, Canada, April 10, 1889, Mary Lilian Brumell,
born there, October i, 1866, daughter of Henry Peareth and Emily
(Carter) Brumell. 2. William Proctor Van Heusen, born in Albany,
January 23, 1898. 3. Olive Proctor Van Heusen, born in Albany,
September 6, 1907. (VIII) Dr. Richard Fletcher Van Heusen, son of
Theodore Van Wyck and Arabella (Manning) Van Heusen, was born
at No. 6 Madison place, Albany, New York, November 26. 1867, died
at his home, No. 611 West One Hundred and Tenth street. New York
City, June 16, 1913. He was buried in the
 SOUTHERN NEW YORK 1041 Van Heusen lot in the Albany
Rural Cemetery. His preparatory education was obtained at the
Albany Academy, and then, taking up the study of medicine, he
entered the Albany Medical College of Union University. Afterwards
he pursued his studies at the University of Michigan, and took a
post-graduate course in the Presbyterian Hospital in New York. He
practiced in that city, and while doing so made a number of
inventions. Dr. Richard Fletcher Van Heusen married Esther Nelson,
and by her had one child, Dorothea Ada Van Heusen. (VHI) John
Manning \^an Heusen, son of Theodore Van Wyck and Arabella
(Manning) \^an Heusen, was born at No. 6 Madison place, Albany,
New York, July 14, 1869, and now resides in Boston, Massachusetts.
He was educated at the Albany Academy, and for the next several
years was connected with the National Commercial and the First
National banks of Albany. He then turned his attention to inventions
of his own, the first of which was a sterilizer designed especially for
the medical fraternity, and it was acknowledged the best so far
produced. Other ideas have since been patented with success. He is
now engaged with the United Shoe. Innovation Shirt Company, and
others. Has been twice married and by first wife had children: I.
Dorothy Gray, married, at Montclair, New Jersey, May 10, 191 1.
Stephen Van Cullen White Connolly, born at Brooklyn, New York,
March 23, 1890, son of Patrick Henry and Jennie (Borland) Connolly,
of Montclair; issue : Stephen Van Cullen White Connolly, born at
Upper Montclair, New Jersey, August 29, 1913. 2. John Manning Jr.,
student at Upper Montclair, New Jersey. There is a good deal of
GREENOUGH doubt concerning the ] origin of the name Greenough.
The assumption usually is in a case of this kind that the name is of
English origin, and that it is a compound of the term or name Green
with some other substantive or qualifying affix. It is possible that the
name is a compound form like Greenwood, Greenend, Greenfield,
and so on, but the name, Greenough, is not found in most works
dealing with English surnames, and occurs in England, even in a
slightly differing form, so rarely as to suggest that it is of German
origin. It is quite possible that the name is an Americanized form of
the Teutonic name "Grunau," which is very similarly pronounced,
though again the termination "augh" or "ough'' would also suggest a
Gaelic origin. The name is a Massachusetts name of colonial origin,
and is now known in this or a slightly differing form in most of the
states. One form of the name that is not uncommon is Greenhow.
The Greenough arms are ; The sun in splendor, ppr. within the
circumference of a bugle horn sa., stringed gu. rimmed and mounted
or. (I) Robert Greenough, immigrant ancestor of the Greenough
family, died in Rowley, Massachusetts, March 30, 1718. Whence he
came and in what year can only be matters of conjecture. He was
recorder at Ipswich, Massachusetts, from 1690 to 1693. In 1691 he
was a selectman at Rowley, and paid a good tax that year. He
married (first) Martha Epes. He married (second) Sarah Mighill, in
1688 at Salem, Massachusetts. She was the widow of Stephen
Mighill, and daughter of the Rev. Samuel Phillips. He married (third)
at Rowley, in 1710, Mary Daniel. Children, by first marriage : Daniel,
born February 22, 1680; Robert, mentioned below. Children by
second marriage : Elizabeth, born December i, 1688, married
Thomas Kimball; Mary, September 17, 1696, married Enoch
Muttlebury. Child by third wife : John, born June 16, 1712. (II)
Robert (2), eldest son of Robert (i) and Martha (Epes) Greenough,
was born at Rowley, Massachusetts, October 28, 16S3, died
December 24, 171 7, aged about thirty-four years. He married, June
16, 1705, Hannah Dole. (III) Robert (3), son of Robert (2) and
Hannah (Dole) Greenough, was born at Rowley, Massachusetts,
November 21, 1712. died at Plaistow, New Hampshire. September 2,
1767. (IV) Richard, son of Robert (3) Greenough, was born at
Plaistow, New Hampshire. August 9, 1744, died at Salisbury,
November 17, 1834. He removed to Salisbury from Plaistow about
1780, clearing up the farm afterwards occupied by Reuben Greeley.
By trade he was a wheelwright and it is said that he built the first
thorough-brace wagons used in the town. He was a very ingenious
man, and made improvements upon many of the farming
implements. He married (first)
 1042 SOUTHERN NEW YORK Elizabeth Hoag or Hogg, who
died December 25, 1810, aged fifty-seven; (second) June 25, 181 1,
Mrs. Susan Shirley, who died December 7, 1848, at the age of
eighty-two. Richard Greenough had no issue by his second wife,
who, however, had borne five children to her first husband. Children
by first marriage: i. Betsey, born in 1771, died September, 1829;
married Thomas Baker, there being no children. 2. Charles, born
1773, died at Montreal. 3. Robert, born 1775, died March 13, 1827,
aged fifty-one or fifty-two; removed to South Road, built the E. P.
Eastman house, was one of the first to build coaches, the
construction of which was for a long time kept secret, having his
carriage shop near the house. 4. Brackett Leavitt, born April 22,
1777, died August 22, 1857, aged eighty years and four months. 5.
Hepzibah, born 1782, died at the age of eighty in 1862. 6. Moses,
born September, 1786, died December 16, 1828, aged forty-two. 7.
John, born May 18, 1790. 8. Ednah, born 1792, died 1821, aged
twentyeight of twenty-nine. 9. Agnes, born 1794. 10. Ezra,
mentioned below. 11. Hannah, born 1798, died at Castleton, 1831,
aged thirtythree. 12. James, born June 4, 1800. (V) Ezra, son of
Richard and Elizabeth (Hoag or Hogg) Greenough, was born April,
1796, died October 20, 1875, aged seventynine years and six
months. He married Lucy Caroline Ormes, born July 15, 1800.
Children: William Henry, died young; Charles Backhouse, mentioned
below. (\T) Charles Backhouse, son of Ezra and Lucy Caroline
(Ormes) Greenough, was born June 29, 1824. died at Paris, France,
January 3, 1880. At an early date he became interested in railroad
affairs. He had occupied a number of important positions in the
railroad service of the country, when his attention was first called to
the very profitable character of the street car enterprises in our large
cities. A franchise for the establishment of one of these roads in the
city of Rio de Janeiro had been granted by the government of Brazil,
but had proved a failure in the hands of the English managers, and
the owners of the franchise, after this failure, sought in vain for
years to find some one who would be willing to work it, even upon
his own terms. Through Mr. Camancho, a native of Venezuela, then
resident in New York City, the franchise was brought to the notice of
Mr. Greenough, who. associated with other gentlemen of his
acquaintance, agreed to purchase it, if, upon a visit to Rio de Janeiro
and a personal investigation, matters were found to be as
represented. As a result of that investigation the franchise was
acquired, the Botanical Garden Railroad Company was organized
under the laws of the State of New York, and Mr. Greenough was
chosen as its first president. There was much opposition at first from
certain local interests at Rio, and there were the usual protracted
delays incident to the establishment of new enterprises in South
American countries. But the company was finally permitted to begin
its work of construction, and October 9, 1808, the first part of the
line was completed and opened to traffic. The enterprise proved a
great success, and was so profitable to the founders that attempts
were made frequently to establish a rival road, but without success.
However, roads leading to other parts of the city were constructed,
and the developments of the suburbs were very rapid in
consequence. Mr. Greenough, having thus been the pioneer of the
successful establishment of street railroads at the Brazilian capital,
which quickly wrought so wonderful a change in the habits of the
people as well as greatly enhanced the value of the surrounding
suburban property, was highly regarded because of the extraordinary
services he had thus rendered to the community, and the press of
Rio still frequently speaks of him as one of Brazil's public
benefactors. The following references to Mr. Greenough and his
enterprises are taken from a work, entitled "Politics and Pen
Pictures," by the Hon. Henry W. Hilliard, who was the minister of the
United States to Brazil from 1879 to 1881: There are several street
car lines in Rio, which contributed much to the comfort of the people
of that great city. From the central part of the city the suburbs
extend for miles in several directions, and its five hundred thousand
inhabitants enjoy the increased facilities for travel. Of these the
Botanical Garden Railroad is by far the finest and most important.
Through the central part of the city, beginning at the Ouvidor, its
finest street, it e-xtends through the aristocratic quarter, Botafogo,
to the magnificent Botanical Gardens and to the suburbs beyond
tliem. This great tramway, one of the finest in the world, was
constructed by a company organized by Mr. Charles Backhouse
Greenough of the state of New York, w-ho possessed both the
capital and the enterprise. His plan, when first submitted to the
wealthy men of Rio, seemed to be imprac 
 SOUTHERN NEW YORK 1043 ticable, and he was able to
enlist but few capitalists in its support. But soon after the completion
of the road its success was such that as to place its stock higher in
the market; and in the course of a year or two it was quoted at such
a rate as to make its holders unwilling to part with it. It was a great
American enterprise, and its charter obtained from the imperial
government secured the stockholders against any trespass on their
right of way. Another charter for the similar road had been obtained
from the government, and its projectors from time to time seemed
determined to push this line into contact with the Botanical Garden
Line. I was frequently appealed to on behalf of those who held this
great American property to intervene for its protection, and I never
failed to do so successfully. The administration always vindicated the
good faith of the government. Mr. Greenough was an extraordinary
man; his person, manners, and intellect were all fine, and his
integrit)- was perfect. Unfortunately, the climate of Rio was not
favorable to his health and he said to me : "I must quit breathing
this hot air." Mrs. Greenough. a noble woman of engaging manners,
and full of character, cheered him, and shared all the dangers of a
residence at Rio with him to the last. !\Ir. Greenough returned to the
United States, leaving Colonel Shannon, in every way competent for
the place, in charge of the road, and he conducted its affairs with
ability and fidelity. Mr. Greenough resided for a time in Colorado, and
hoped that its fine climate would restore his health, but not
recovering his strength, he decided to go to Europe. Accompanied
by Mrs. Greenough he went to Paris and took a house in the
Boulevard Haussmann, fitting it up in accordance with their tastes.
He was for a time benefited by this agreeable residence, but did not
recover his failing health. Returning from an evening drive he was
fatigued, and reclining on a sofa, died suddenly and painlessly. It
was a peaceful close to a noble life. When the news of Mr.
Greenough's death reached Rio de Janeiro the offices and central
station of the company were draped in mourning and the
newspapers vied with each other in paying earnest tribute to the
inemory of the man who had done so much for the development and
progress of the city. Mr. Greenough married Martha Ann, daughter of
Silas and Rhoda (Scoville) Spaulding, of Clarkson, New York. The
widow married (second) Richard Cutts Shannon. Mr. and Mrs.
Charles Backhouse Greenough had one son, Charles Edward,
mentioned below. (VII) Charles Edward, son of Charles Backhouse
and Martha Ann (Spaulding) Greenough, was born in New York,
September 9, 1853, died in New York City, October 22, 1880. The
following note regarding hiin was published in one of the
newspapers of Brockport, New York, at the time of his death : This
young gentleman died at the Windsor Hotel in New York City last
week Friday. He was born in that city, September 9, 1853, and was
therefore a trifle over twenty-seven years of age. During his life he
spent a considerable portion of his time in this village in the home of
his grandparents where he attended the collegiate institute. His
bright and attractive manner always made him a welcome visitor and
surrounded him with many friends, who are now pained by the
announcement of his death. His education was completed in Geneva,
Switzerland. He did a great deal of travel both in North America,
South America, and Europe, having been just returned from across
the ocen, when his demise took place. At one time he was a
member of the Seventh Regiment of New York. He married, in New
York City, September, 1879, Frances Gray Dawson. There was one
child of the marriage, Charles Edward, mentioned below. (Vni)
Charles Edward (2), only son of Charles Edward (i) and Frances Gray
(Dawson) Greenough, was born in New York City, October 23, 1880.
He married, at Grace Church, New York City, November 16, 1909,
Eleanor Lucia, born in New York City, April 17, 1886, daughter of
Frederick Wallingford and Lucy (Arnold) Whitridge. Children: John
Whitridge, born New York City, October 5, 191 1 ; Richard Dawson,
London, England, September 4, 1912. The family name of Henry
HENRY would appear at first sight to be derived from the personal
name of the same form. Names applied as an hereditary possession
to a family and derived from a personal name without modification
are not very numerous. As a rule it will be found on tracing them
back that in former days they possessed some prefix such as Mac,
O, or Ap, or some suffix, such as son and sen, indicating an
ancestral connection. In almost all cases where the name Henry is
found in Britain or Scotland, its derivation will be found to be froin
MacEnri or MacHenry. Nor is the meaning of the name as obvious as
would appear froiu first sight. Its ancient form was Mac An Righ,
which is Gaelic for the "son of the king," the Celtic righ, being
related to the Latin rex. The origin and genealogy of the sept of
MacHenry is as princely as its name, and in genealogical
compilations like those of MacFirbis, O'Hart, Keating, Linea Antiqua,
and historical works like the "Annals of the Four Masters," the
 lo-pt SOUTHERN NEW YORK pedigree stem of the family is
found to cover about sixteen hundred years. (I) James Henry,
immediate ancestor of the Henry family, was born about I/69, and
was a farmer of western Pennsylvania. The records reveal very little
relating to him, and it is impossible to ascertain the name of his
wife, by whom he had nevertheless a large family, most of the
children growing to adult life. Children: John, Samuel S., James,
William, Joseph. Thomas. Robert, mentioned below; a daughter who
married a Vandiver; Jane, who married a MacClelland; Nancy. (li)
Rev. Robert Henry, son of James Henry, was born about 1800, and
died in 1837. He was a minister and was beginning a career of
promise when he died. He married a daughter of James Buchanan,
who came to this country from Ireland, and sister to James
Buchanan, President of the United States from the year 1857 to the
year 1861. She died in 1840. (HI) James Buchanan, son of Rev.
Robert and (Buchanan) Henry, was born in the year 1833. His
parents dying when he was very young, he was adopted into the
household of his uncle, James Buchanan, who afterwards became
president. He married Louisa, daughter of William A. C. (a physician)
and Louisa (Morgan) Anderson. (IV) Robert Edward, son of James
Buchanan and Louisa (Anderson) Henry, was born at Stapleton,
Staten Island, New York, June 12, 1877. He was educated at the
United States Military Academy, St. John's College, Annapolis,
Maryland, and Georgetown College, Washington, D. C. He is a
member of the firm of Hallowell & Henry, president of the Associated
Industrial Corporation, and has other business interests. He gives his
support to the Republican party in politics, and is an Episcopalian in
religion. He is fond of athetics and all outdoor recreations, his
favorite sports being riding, driving, and boating. He is a member of
several clubs, among them being the Crescent, Athletic. Riding and
Driving (Brooklyn). He married at Fort Hamilton, New York, October
26, 1904, Virginia Bell Tolar. There has been one son of the issue,
Robert Edward Jr., born in 1906. (The Buchanan Line.) This name is
stated to be derived from Anselan Buey O'Kyan or O'Bocainan,
anglicized Buchanan, the name or title of an Irish noble who
succeeded in the ninth century as provincial king of South Ulster,
Ireland. This monarch took part in the slaughter of Turgesius, the
Danish general, and his army at what is now called Limerick, and
was, with his followers, compelled to leave Ireland and flee to
Scotland, in 1016, and soon after entered the service of King
Malcolm II. against the Danes. He so signalized himself that he
obtained from this monarch many grants of land in the northern part
of Scotland, among which were the lands of Pitwhonidy and
Strathyre, and was recognized as the first Laird of Buchanan. He
married the heiress of Denniestown. Browning's "Americans of Royal
Descent" gives the Buchanans as descended from Fergus or
Fargallus, the 156th monarch of All Ireland, who was killed .'V.d. 718
by Moroch, King of Leinster, James Buchanan, president of the
United States, being a descendant in the thirty-second generation.
Anselan, seventh laird of Buchanan, and fourteenth in descent from
Fargallus, was chamberlain to Malduin, Earl of Lennox, in 1225, and
obtained a charter from him of an island in Loch Lomond, which he
called "Clareinch," the slughome, or warcry, proper to the family of
Buchanan. The first to assume the name of Buchanan was Gilbert,
eighth laird of Buchanan, fifteenth in descent from Fargallus. and
seventh in descent from Anselan Buey O'Kyan (O or Ua means
"descendant"— -"the descendant of Kyan") or O'Bocainan, the name
of the king, descent from whom Gilbert desired, according to the
usage among the Gaels, to perpetuate in the family surname. Sir
Walter Buchanan, another ancestor, married Lady Isabel Stewart,
daughter of Murdoch, second duke of Albany, and governor of
Scotland, a grandson of Robert the Second of Scotland, so that this
line of the Buchanans is descended from both Irish and Scottish
kings. George Buchanan, twenty-seventh in descent from Fargallus,
sold Blairlusk to his brother, William Buchanan, and removed to the
north of Ireland, the family having been then in Scotland seventeen
generations, counting from the last Irish ancestor's departure. (I)
James Buchanan, the first American ancestor of this line of the
Buchanans, was born at Romelton, county Donegal, Ireland, died at
Mercersburg, in Franklin county, Pennsylvania, in 1821. He removed
to
 SOUTHERN NEW YORK 1045 America about 1783. He was
the great-grandson of George Buchanan who left Scotland for
Ireland, grandson of Thomas, and son of John, who lived at
Romelton, county Donegal, Ireland. Soon after landing he went to
Pennsylvania and settled at Mercersburg, where he remained until
the time of his death. He married in Adams county, Pennsylvania,
Elizabeth Speer. Children : James, who became president, mentioned
below ; William Speer, died at the age of twenty-two ; George W.,
died at the age of twenty- five; Rev. Edward Young, D.D., of
Philadelphia, Pennsylvania, who married, in 1833, Elizabeth Ann
Foster ; Jane, who married in 1813, Elliott T. Lane; and , who
married the Rev. Robert Henry (see Henry). (II) James (2), son of
James (i) and Elizabeth (Speer) Buchanan, fifteenth president of the
ilnited States, was born at Mercersburg, Pennsylvania, April 23,
1791, died in Lancaster, Pennsylvania, June i, 1868. The days of his
youth were those of the nation's youth ; his public career of forty
years saw all our great extensions of boundary on the south and
west acquirements from foreign powers, the inclusion of thirteen
new states, the development of many important questions of
international policy, and the gradual rise and culmination of a great
insurrection. He was educated at a school in Mercersburg and at
Dickinson College, Pennsylvania, where he was graduated in 1809.
He began to practice law in Lancaster in 1812. His first public
address was made at the age of twenty-three on the occasion of a
popular meeting in Lancaster after the capture of Washington by the
British in 1814. In October of the same year he was elected to the
house of representatives in the legislature of Pennsylvania for
Lancaster county. He remained in the house of representatives ten
years in the second term. In 1834 the legislature of Pennsylvania
elected him to the United States senate to succeed Mr. Wilkins, who
had been appointed minister to Russia. This office was
acknowledged by Buchanan afterwards to be "the only public station
he desired to occupy." In 1843 h^ was elected to the senate for a
third term, and in 1844 his name was brought forward as the
Democratic candidate for Pennsylvania for the presidential
nomination, but before the national convention met he withdrew in
order that the whole strength of the party might be concentrated
upon one candidate. James K. Polk was elected; he asked Mr.
Buchanan to become his secretary of state and the invitation was
accepted. When in 1853 Franklin Pierce became president he
appointed Buchanan minister to England. He returned in 1856 and
met with a public reception from the authorities and people of New
York that evinced the interest that was already beginning to be
manifested in him as a possible future president. He was the
Democratic nominee in 1856 and was elected president after a
contest of nation-wide enthusiasm. He was inaugurated president,
March 4, 1857. The term of his administration was a memorable one,
preceding as it did the civil war. On March 9, 1861, Buchanan
returned to his home at Wheatland and there he lived quietly for the
remaining seven years of his life. His remains were interred in a
cemetery near Lancaster. (The Morgan Line.) The family name
Morgan is sometimes Welsh and sometimes Irish. The Welsh family
of the name is distinct from the Irish family, and goes back a long
way, being usually associated as regards territory with the county of
Glamorgan. The modern form Morgan is merely an anglicized form
of an older Celtic name. The old Irish or Gaelic form of the name is
O'Miurgain, the family being a branch of the Donnelan family, and
holding large and valuable landed properties in Ulster, which were
subsequently confiscated. Various genealogical compilations, such as
those of Dugald MacFirbis, and O'Hart, gave the genealogy of the
family through about sixty generations. (I) Thomas Morgan,
immediate ancestor of the family, was born in 1756, died in 1816.
He married Jane Jenner, and came to America in or around the year
1795. Children : Thomas, born 1790, died 1818, married Augusta
Wilmerding; Henry; Charles, 1794, died 1841, married, in 1832,
Adelia Walden, born 1814, died 1850; Joshua; Matthew, 1797, died
1862, married, 1846, Lucinda Lee Rogers, born 1807, died 1875;
Charlotte; Sidney; George, died 1850, married Paulina Druliard;
Edward, born 1806, died 1831 ; Louisa, mentioned below. (II)
Louisa Morgan, daughter of Thomas and Jane (Jenner) Morgan, was
born in 1810, died in 1873. She married, in 1838,
 1046 SOUTHERN NEW YORK William Astlcy Cooper
Anderson, M.D., born in 1814, died in 1882. Children: William
Edward, born 1839; Jane Morgan, 1841, died 1843; Charles Morgan,
1843, d'cd 1844; George Cooper, 1847, died 1847; Louisa,
mentioned below; Frank, 1852, medical director. United States navy,
married, 1886, Eleanor Coffin. (Ill) Louisa Anderson, daughter of
William Astley Cooper and Louisa (Morgan) Anderson, was born in
1848, died in 1886. She married James Buchanan Henry, secretary
to President Buchanan, an»5, father of Robert Edward Henry,
mentioned above. The origin of the Miller surMILLER name is
obvious, the name being manifestly derived from the occupation of a
miller. Families bearing the name were prominent in every state in
the early days of the colonies. They came from England, Ireland,
and Scotland, and they came from Germany and Holland. The form
of Mueller usual in these latter countries has in most cases in
America been modified to Miller. At almost every stage in American
history the name is met with, particularly in the south. In Virginia
and the states adjoining the name has been given to many localities
as Millers, Miller's Island, and Millersville in Maryland ' Miller's
Tavern, Miller's Fork (Creek), Miller's Camp Branch in Virginia ; Miller
county, Miller's Ferry and so on in other cases. There are
innumerable episodes connecting the name with the revolutionary
war. Robert Miller came from Ireland in the beginning of the
eighteenth century and founded a southern family of which there are
many branches He married Miss Ann Lynn in 173 1 and settled m
Goochland, Orange (or Albemarle) county where they reared their
family. They had six children, the boys being: Robert, Thomas and
Colonel John. Robert was born May 5 1714 married, about 1763,
Mary Maupin, a daughter of Daniel Maupin and Margaret (Via)
Maupin. his wife, of Albermarle county Virginia, and established his
home in oVange county. Thomas was born March 20 17^6 In the
family register of the late Colonei Thomas William Miller, of Stanford
Kentucky, ,s a note that he was born in Kentucky but the date and
place of his death and burial are not given. One Thomas Miller in
178^ was an attorney-at-law of the Albemarle bar at Charlottesville,
Virginia. Colonel John Miller was born January i, 1750, and married
Jane Dulaney. They immigrated into Kentucky and settled in Madison
county. It is probable that the line that follows belongs to some
branch of the foregoing. (I) Miles S. Miller married Fanny Virginia,
daughter of Mr. Harrison, and had issue' among whom was Hugh
Gordon, mentioned below. (II) Hugh Gordon, son of Miles S. and
Fanny Virginia (Harrison) Miller, was born in Virginia, March 2, 1875.
His maternal grandfather (Harrison) was clerk of the superior court
and probate judge of Camden county. North Carolina, for more than
a quarter of a century, so that it was quite in the nature of things
that the grandson should turn to the law. Mr. Miller was educated in
V irgima, for the most part in his native town of Norfolk, and was
admitted to the bar in Virginia on March 20, 1896, at once resigning
as deputy clerk of the courts of Norfolk to enter upon the active
practice of his profession, which has found him almost constantly
engaged in important litigations, first in Virginia and later in New
York City. He was admitted to practice before the United States
supreme court at the November term in 1898 and was probably the
youngest advocate that ever appeared to argue before that august
tribunal. Moreover, this highest court of the land by unanimous
consent on that occasion extended the usual time allowed an
advocate to address the court. Mr. Miller was one of the leading
promoters of the Jamestown Exposition of 1907, organized to
celebrate in Virginia the three-hundredth anniversary of the birth of
the nation. As far previous to the exposition of 1907 as in igos he
had conceived the benefits that would accrue to the American nation
by the observance of the ter-centenary, and in broaching the matter
to the Hon. Grover Cleveland, received an autograph letter endorsing
his proposition in the highest terms. It w-as this letter, when given
out to the pubhe through the press, which stirred the people the
congress and the state to action in making the exposition a reality,
and Governor Hughes used this letter as a text in his banquet
speech reading it aloud and eloquently commending the patriotic
terms at the banquet given in his honor on New York day at the
exposition.
 SOUTHERN NEW YORK 1047 Mr. Miller was likewise
instrumental in securing the interest of King Edward VII. and the co-
operation of the British government in advancing the exposition's
success and greatness through Rear-Admiral, Prince of Battenburg.
As a prominent Republican the forceful and direct delivery of Mr.
Miller has time and again been heard with enthusiasm as the chief
orator of great political gatherings throughout the country. It is thus
that he has come to be recognized as one of the most eloquent
speakers before the American public. In 1901 he delivered one of
the principal addresses at the annual Lincoln dinner and the
celebration of the Republican Club of the City of New York of which
he is a member. He delivered the Lincoln centennial address at
Convention Hall, Buffalo, at which Governor Hughes was the orator
and guest of honor, and later did the same at the Convention Hall,
Rochester. He is honorary vice-president of the Lincoln University
Endowment Association. Mr. Miller was selected by the commissioner
of the French government to be chairman of the American
committee of honor for the International Maritime Exposition at
Bordeaux, France, and in 1907 was chosen secretary and later vice-
president of the Robert Fulton Monument Association, of which he
was one of the founders. For four years before coming to New York,
Mr. Miller was chief assistant United States attorney for the eastern
district of Virginia, and at the direction of President Roosevelt was
commissioned special assistant attorney-general of the United
States, being placed in charge of important cases pending in the
supreme court of the United States. He is a member of the American
and other bar associations, and of the New York Southern Society.
He took an active part as a political speaker throughout the country,
but especially in the north and east in the campaigns resulting in the
election of President McKinley and President Roosevelt. Of his ability
as an advocate, the following tribute from the New York Financial
Times bears testimony : In legal circles the notable victory of Mr.
Hugh Gordon Miller in the Wilson case is a subject of highly
favorable comment, being accepted as additional evidence of Mr.
Miller's ability as a lawyer and an orator. With the details of the case
our readers are thoroughly familiar so that it is not necessary to
refer to them here. It is enough to state that Wilson, a temporary
messenger for the First National Bank, claimed that some one had
picked his pocket of $42,000 in checks and drafts and nearly $2,000
in cash. The bank officially refused to accept Wilson's story and he
was charged with the theft. Mr. Miller conducted the prisoner's case
with consummate tact and judgment and from the first succeeded in
creating a favorable impression for the defendant. His "summing up"
was a masterpiece — eloquent, impressive, logical and convincing
and unquestionably secured the acquittal of the prisoner. It was, in
every way, one of the ablest addresses ever heard in a New York
court room — and fully justified Mr. Miller's high reputation as a jury
pleader. For some time Mr. Miller, who is a native of the South, has
occupied a prominent position at the local bar and is known as a
lawyer of sound ability and experience. He is noted for the care with
which he prepares his cases and the fidelity with which he looks
after the interests entrusted to him. Mr. Miller married in New York
City, December 7, 1904, Edna M., daughter of Napoleon T. Allen, of
New York City. Mrs. Miller was the lady referred to on a prominent
occasion, when the president of the United States responded, as
"the last of the Martha Vineyard Aliens." The original grant of a large
portion of the present state of i\lassachusetts was made direct to
Governor Mayhue, a colonial great-grandfather of Mrs. Miller,
through a maternal line, and the colonial deed to a large subdivision
on this land on record in Massachusetts, reads from this colonial
governor jointly to Josias Standish (eldest son of Miles Standish),
William Pabodie (who through marriage to Elizabeth, their daughter,
was the son-in-law of Priscilla and John Alden), and James Allen, a
colonial great-grandfather of Mrs. Miller on the paternal side. Mrs.
Miller, as the child, the only one living, of Mr. Napoleon T. Allen, is a
descendant in the eighth generation of George Allen, who is stated
to have been born at Braintree, England, in 1567, and who was one
of a later party of Pilgrims that came over a few years after the
"Mayflower." His grandson was the James Allen above mentioned.
George Allen founded in America the famous family known as the
"Martha Vineyard Aliens." With Myles Standish, John Alden and
Governor Bradford, he permanently settled and established the
Plymouth colony. Mrs. Miller is also the granddaughter of three
colonial governors of Plymouth colony, several times removed, the
governors being Hickley and the two Mayhues ; and she is a
collateral descendant of Captain Myles
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

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

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com