Download pdf Graph Theory And Its Applications Third Edition Jonathan L Gross ebook full chapter

Graph Theory and Its Applications,
Third Edition Jonathan L. Gross

Visit to download the full and correct content document:
https://textbookfull.com/product/graph-theory-and-its-applications-third-edition-jonatha
n-l-gross/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
50 years of Combinatorics Graph Theory and Computing

Discrete Mathematics and Its Applications 1st Edition
Fan Chung (Editor)
https://textbookfull.com/product/50-years-of-combinatorics-graph-
theory-and-computing-discrete-mathematics-and-its-
applications-1st-edition-fan-chung-editor/
Biota Grow 2C gather 2C cook Loucas
https://textbookfull.com/product/biota-grow-2c-gather-2c-cook-
loucas/
Graph Theory Applications to Deregulated Power Systems

Ricardo Moreno Chuquen
https://textbookfull.com/product/graph-theory-applications-to-
deregulated-power-systems-ricardo-moreno-chuquen/
Graph Theory and Decomposition 1st Edition Kottarathil
https://textbookfull.com/product/graph-theory-and-
decomposition-1st-edition-kottarathil/
Game Theory and Its Applications to Takeovers 1st
Edition Suresh Deman
https://textbookfull.com/product/game-theory-and-its-
applications-to-takeovers-1st-edition-suresh-deman/
Fuzzy Graph Theory 1st Edition Sunil Mathew
https://textbookfull.com/product/fuzzy-graph-theory-1st-edition-
sunil-mathew/
Double Hierarchy Linguistic Term Set and Its

Extensions: Theory and Applications Xunjie Gou
https://textbookfull.com/product/double-hierarchy-linguistic-
term-set-and-its-extensions-theory-and-applications-xunjie-gou/
Basic Graph Theory 1st Edition Md. Saidur Rahman

(Auth.)
https://textbookfull.com/product/basic-graph-theory-1st-edition-
md-saidur-rahman-auth/
Introduction to Chemical Graph Theory 1st Edition

Stephan Wagner
https://textbookfull.com/product/introduction-to-chemical-graph-
theory-1st-edition-stephan-wagner/
i i
“K23723˙FM” — 2018/10/11 — 14:15 — page 2 — #2

i i
Graph Theory and Its

Applications, Third Edition
i i
i i
i i
“K23723˙FM” — 2018/10/11 — 14:15 — page 4 — #4

i i
Textbooks in Mathematics
Series editors:
Al Boggess and Ken Rosen
A CONCRETE INTRODUCTION TO REAL ANALYSIS, SECOND EDITION
Robert Carlson
MATHEMATICAL MODELING FOR BUSINESS ANALYTICS
William P. Fox
ELEMENTARY LINEAR ALGEBRA
James R. Kirkwood and Bessie H. Kirkwood
APPLIED FUNCTIONAL ANALYSIS, THIRD EDITION
J. Tinsley Oden and Leszek Demkowicz
AN INTRODUCTION TO NUMBER THEORY WITH CRYPTOGRAPHY, SECOND EDITION
James R. Kraft and Lawrence Washington
MATHEMATICAL MODELING: BRANCHING BEYOND CALCULUS
Crista Arangala, Nicolas S. Luke and Karen A. Yokley
ELEMENTARY DIFFERENTIAL EQUATIONS, SECOND EDITION
Charles Roberts
ELEMENTARY INTRODUCTION TO THE LEBESGUE INTEGRAL
Steven G. Krantz
LINEAR METHODS FOR THE LIBERAL ARTS
David Hecker and Stephen Andrilli
CRYPTOGRAPHY: THEORY AND PRACTICE, FOURTH EDITION
Douglas R. Stinson and Maura B. Paterson
DISCRETE MATHEMATICS WITH DUCKS, SECOND EDITION
Sarah-Marie Belcastro
BUSINESS PROCESS MODELING, SIMULATION AND DESIGN, THIRD EDITION
Manual Laguna and Johan Marklund
GRAPH THEORY AND ITS APPLICATIONS, THIRD EDITION
Jonathan L. Gross, Jay Yellen and Mark Anderson
i i
i i
i i
“K23723˙FM” — 2018/10/11 — 14:15 — page 6 — #6

i i
Graph Theory and Its

Applications, Third Edition
Jonathan L. Gross
Jay Yellen
Mark Anderson
i i
i i
i i
“K23723˙FM” — 2018/10/11 — 14:15 — page 8 — #8

i i
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2019 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
Printed on acid-free paper

Version Date: 20191011
International Standard Book Number-13: 978-1-4822-4948-4 (Hardback)
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 copy-
right 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 pho-
tocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission
from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (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 provides licenses and registration for a variety of users.
For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been
arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at

http://www.crcpress.com
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page v — #2

i i
Jonathan dedicates this book to Alisa.
Jay dedicates this book to the memory of his brother Marty.
Mark dedicates this book to Libertad.
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page vi — #3

i i
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page vii — #4

i i
CONTENTS
Preface xi
Authors xiii
1 INTRODUCTION TO GRAPH MODELS 1

1.1 Graphs and Digraphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Common Families of Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3 Graph Modeling Applications . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4 Walks and Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
1.5 Paths, Cycles, and Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
1.6 Vertex and Edge Attributes: More Applications . . . . . . . . . . . . . . . 49
1.7 Supplementary Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2 STRUCTURE AND REPRESENTATION 59

2.1 Graph Isomorphism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.2 Automorphisms and Symmetry . . . . . . . . . . . . . . . . . . . . . . . . 68
2.3 Subgraphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.4 Some Graph Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
2.5 Tests for Non-Isomorphism . . . . . . . . . . . . . . . . . . . . . . . . . . 93
2.6 Matrix Representations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
2.7 More Graph Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
3 TREES 121
3.1 Characterizations and Properties of Trees . . . . . . . . . . . . . . . . . . 122
3.2 Rooted Trees, Ordered Trees, and Binary Trees . . . . . . . . . . . . . . . 130
3.3 Binary-Tree Traversals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3.4 Binary-Search Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
3.5 Huffman Trees and Optimal Prefix Codes . . . . . . . . . . . . . . . . . . 148
3.6 Priority Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
3.7 Counting Labeled Trees: Prüfer Encoding . . . . . . . . . . . . . . . . . . 157
3.8 Counting Binary Trees: Catalan Recursion . . . . . . . . . . . . . . . . . . 163
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4 SPANNING TREES 173

4.1 Tree Growing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
4.2 Depth-First and Breadth-First Search . . . . . . . . . . . . . . . . . . . . 181
4.3 Minimum Spanning Trees and Shortest Paths . . . . . . . . . . . . . . . . 186
4.4 Applications of Depth-First Search . . . . . . . . . . . . . . . . . . . . . . 193
4.5 Cycles, Edge-Cuts, and Spanning Trees . . . . . . . . . . . . . . . . . . . 201
vii
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page viii — #5

i i
viii Contents
4.6 Graphs and Vector Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

4.7 Matroids and the Greedy Algorithm . . . . . . . . . . . . . . . . . . . . . 218
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
5 CONNECTIVITY 229
5.1 Vertex- and Edge-Connectivity . . . . . . . . . . . . . . . . . . . . . . . . 230
5.2 Constructing Reliable Networks . . . . . . . . . . . . . . . . . . . . . . . . 235
5.3 Max-Min Duality and Menger’s Theorems . . . . . . . . . . . . . . . . . . 243
5.4 Block Decompositions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
6 OPTIMAL GRAPH TRAVERSALS 259

6.1 Eulerian Trails and Tours . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
6.2 DeBruijn Sequences and Postman Problems . . . . . . . . . . . . . . . . . 264
6.3 Hamiltonian Paths and Cycles . . . . . . . . . . . . . . . . . . . . . . . . 279
6.4 Gray Codes and Traveling Salesman Problems . . . . . . . . . . . . . . . . 284
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
7 PLANARITY AND KURATOWSKI’S THEOREM 297

7.1 Planar Drawings and Some Basic Surfaces . . . . . . . . . . . . . . . . . . 298
7.2 Subdivision and Homeomorphism . . . . . . . . . . . . . . . . . . . . . . . 305
7.3 Extending Planar Drawings . . . . . . . . . . . . . . . . . . . . . . . . . . 309
7.4 Kuratowski’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
7.5 Algebraic Tests for Planarity . . . . . . . . . . . . . . . . . . . . . . . . . 324
7.6 Planarity Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
7.7 Crossing Numbers and Thickness . . . . . . . . . . . . . . . . . . . . . . . 340
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
8 GRAPH COLORINGS 351

8.1 Vertex-Colorings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
8.2 Map-Colorings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
8.3 Edge-Colorings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
8.4 Factorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
9 SPECIAL DIGRAPH MODELS 399

9.1 Directed Paths and Mutual Reachability . . . . . . . . . . . . . . . . . . . 400
9.2 Digraphs as Models for Relations . . . . . . . . . . . . . . . . . . . . . . . 411
9.3 Tournaments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
9.4 Project Scheduling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
9.5 Finding the Strong Components of a Digraph . . . . . . . . . . . . . . . . 429
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
10 NETWORK FLOWS AND APPLICATIONS 441

10.1 Flows and Cuts in Networks . . . . . . . . . . . . . . . . . . . . . . . . . . 442
10.2 Solving the Maximum-Flow Problem . . . . . . . . . . . . . . . . . . . . . 450
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page ix — #6

i i
Contents ix
10.3 Flows and Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460

10.4 Matchings, Transversals, and Vertex Covers . . . . . . . . . . . . . . . . . 469
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
11 GRAPH COLORINGS AND SYMMETRY 487

11.1 Automorphisms of Simple Graphs . . . . . . . . . . . . . . . . . . . . . . . 488
11.2 Equivalence Classes of Colorings . . . . . . . . . . . . . . . . . . . . . . . 493
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
A APPENDIX 501
A.1 Logic Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
A.2 Relations and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
A.3 Some Basic Combinatorics . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
A.4 Algebraic Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
A.5 Algorithmic Complexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511
A.6 Supplementary Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514
B BIBLIOGRAPHY 515
B.1 General Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515
B.2 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517
SOLUTIONS AND HINTS 523
INDEX OF APPLICATIONS 563
INDEX OF ALGORITHMS 565
GENERAL INDEX 567
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page x — #7

i i
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page xi — #8

i i
PREFACE
Graphs have proven to be useful models for a wide variety of problems, arising in
computer science, operations research, and in the natural and social sciences. This text
targets the need for a comprehensive approach to the theory, integrating a careful expo-
sition of classical developments with emerging methods, models, and practical needs. It
is suitable for classroom presentation at the introductory graduate or advanced under-
graduate level, or for self-study and reference by working professionals.
Graph theory has evolved as a collection of seemingly disparate topics. The intent
of the authors is to present this material in a more cohesive framework. In the process,
important techniques and analytic tools are transformed into a unified mathematical
methodology.
Emphasis throughout is conceptual, with more than 600 graph drawings included to
strengthen intuition and more than 1500 exercises ranging from routine drill to more chal-
lenging problem solving. Applications and concrete examples are designed to stimulate
interest and to demonstrate the relevance of new concepts.
Algorithms are presented in a concise format, and computer science students will find
numerous projects inviting them to convert algorithms to computer programs. Software
design issues are addressed throughout the book in the form of computational notes,
which can be skipped by those not interested in implementation. These design issues
include the efficient use of computational resources, software reusability, and the user
interface.
This book is a collection of topics drawn from the second edition of Graph Theory
and Its Applications, written by the first two authors of this book. The topics chosen
provide a strong foundation in graph theory.
Summary of Contents
Chapters 1 through 6 concentrate on graph representation, basic properties, modeling,
and applications. Graphical constructions here are concrete and primarily spatial. When
necessary, we introduce abstractions in a supportive role.
Chapter 7, which is devoted to planarity and Kuratowski’s theorem, presents some
of the underpinnings of topological graph theory. Chapter 8 is about graph colorings,
including vertex- and edge-colorings, map-colorings, and the related topics of cliques,
independence numbers, and graph factorization.
The material in Chapters 9 and 10, special digraph models and network flows, overlaps
with various areas in operations research and computer science.
Chapter 11 explores the connections between a graph’s automorphisms (its symme-
tries) and its colorings.
The appendix provides a review of a variety of topics which a student would typically
encounter in earlier math classes.
To the Instructor
The book is well-suited for a variety of one-semester courses. A general course in
graph theory might include many of the topics covered in the first eight chapters. A
xi
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page xii — #9

i i
xii Preface
course oriented toward operations research/optimization should include most or all of the
material in Chapters 4 (spanning trees), 5 (connectivity), 6 (traversability), 8 (colorings),
9 (digraph models), and 10 (flows). A course emphasizing the role of data structures and
algorithms might include more material from Chapter 3 (trees). A course designed to
have strong ties to geometry, linear algebra, and group theory, might include parts of
Chapters 4 (graphs and vector spaces), 7 (planarity), 8 (graph colorings), and 11 (graph
symmetry).
The definitions and results from all of Chapter 1, most of Chapter 2, and from §3.1
are used throughout the text, and we recommend that they be covered in any course
that uses the book. The remaining chapters are largely independent of each other, with
the following exceptions:
• §3.2 is used in §4.1.
• §4.1 and §4.2 are used in §5.4 and §9.5.

• §4.5 is referred to in §5.3 and §6.1.
• Parts of Chapter 5 are used in §10.3.
Features of This Book

• Applications. The Index of Applications lists 70 applications, distributed through-
out the text.
• Algorithms. The Index of Algorithms points to 51 algorithms, outlined in the text
with enough detail to make software implementation feasible.
• Foreshadowing. The first three chapters now preview a number of concepts, mostly
via the exercises, that are more fully developed in later chapters. This makes it
easier to encourage students to take earlier excursions in research areas that may
be of particular interest to the instructor.
• Solutions and Hints. Each exercise marked with a superscriptS has a solution or
hint appearing in the back of the book.
• Supplementary Exercises. In addition to the section exercises, each chapter con-
cludes with a section of supplementary exercises, which are intended to develop
the problem-solving skills of students and test whether they can go beyond what
has been explicitly taught. Most of these exercises were designed as examination
questions for students at Columbia University.
Websites
Suggestions and comments from readers are welcomed and may be sent to the
authors’ website at www.graphtheory.com. Thanks to our founding webmaster Aaron
Gross for his years of maintaining this website. The general website for CRC Press is
www.crcpress.com. As with our previous graph theory books, we will post the corrections
to all known errors on our website.
Jonathan L. Gross, Jay Yellen, and Mark Anderson
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page xiii — #10

i i
ABOUT THE AUTHORS

Jonathan L. Gross is Professor of Computer Science at
Columbia University. His research in topology, graph the-
ory, and cultural sociometry has earned him an Alfred P.
Sloan Fellowship, an IBM Postdoctoral Fellowship, and var-
ious research grants from the Office of Naval Research, the
National Science Foundation, and the Russell Sage Foun-
dation.
Professor Gross has created and delivered numerous soft-
ware development short courses for Bell Laboratories and
for IBM. These include mathematical methods for perfor-
mance evaluation at the advanced level and for develop-
ing reusable software at a basic level. He has received sev-
eral awards for outstanding teaching at Columbia University, including the career Great
Teacher Award from the Society of Columbia Graduates. His peak semester enrollment
in his graph theory course at Columbia was 101 students.
His previous books include Topological Graph Theory, coauthored with Thomas W.
Tucker; Measuring Culture, coauthored with Steve Rayner, constructs network-theoretic
tools for measuring sociological phenomena; and Combinatorial Methods with Computer
Applications.
Prior to Columbia University, Professor Gross was in the Mathematics Department at
Princeton University. His undergraduate work was at M.I.T., and he wrote his Ph.D.
thesis on 3-dimensional topology at Dartmouth College.
Jay Yellen is Professor of Mathematics at Rollins College.

He received his B.S. and M.S. in Mathematics at Polytech-
nic University of New York and did his doctoral work in
finite group theory at Colorado State University. Dr. Yellen
has had regular faculty appointments at Allegheny College,
State University of New York at Fredonia, and Florida In-
stitute of Technology, where he was Chair of Operations
Research from 1995 to 1999. He has had visiting appoint-
ments at Emory University, Georgia Institute of Technol-
ogy, and Columbia University, University of Nottingham,
UK, and KU Leuven, Belgium.
In addition to the Handbook of Graph Theory, which he co-edited with Professor
Gross, Professor Yellen has written manuscripts used at IBM for two courses in
discrete mathematics within the Principles of Computer Science Series and has con-
tributed two sections to the Handbook of Discrete and Combinatorial Mathematics. He
also has designed and conducted several summer workshops on creative problem solving
for secondary-school mathematics teachers, which were funded by the National Science
Foundation and New York State. He has been a recipient of a Student’s Choice Professor
Award at Rollins College.
Dr. Yellen has published research articles in character theory of finite groups, graph
theory, power-system scheduling, and timetabling. His current research interests include
graph theory, discrete optimization, and the development of course-scheduling software
based on graph coloring.
xiii
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page xiv — #11

i i
xiv Authors
Mark Anderson is Professor of Mathematics and Com-

puter Science at Rollins College. He received his doctor-
ate from the University of Virginia for work in algebraic
and differential topology. After arriving at Rollins College,
he switched his emphasis to graph theory and also earned
a masters degree in Computer Science from the Univer-
sity of Central Florida. He has publications in the areas of
algebraic topology (cobordism), topological graph theory
(crossing number and Cayley maps), graph coloring (using
voltage graphs), as well as graph security (eternal domina-
tion). He is currently working on problems in robust graph
coloring, which combines ideas from graph coloring with
those of graph security. Undergraduate students have done collaborative research with
Dr. Anderson in topological graph theory, graph coloring, and software testing.
Methods included in this book appear in a wide range of courses that Professor An-
derson has designed. In his Artificial Intelligence course, students explore graph search
algorithms; in Data Compression, they discover Huffman codes; in Coding Theory, they
work with the geometry of the n-dimensional hypercube. Courses that use algebra to
answer geometric questions include Algebraic Topology and Transformational Geometry.
The Geometry of Islamic Patterns involves imbeddings of graphs on surfaces as a way of
characterizing planar tilings. Courses he has developed for students with majors outside
math and computer science include The Mathematics of Games and Gaming and Zero,
A Revolutionary Idea.
i i
i i
i i
“Current˙Book” — 2018/10/16 — 17:36 — page 1 — #12

i i
Chapter 1
INTRODUCTION TO GRAPH MODELS
INTRODUCTION
Configurations of nodes and connections occur in a great diversity of applications.
They may represent physical networks, such as electrical circuits, roadways, or organic
molecules. And they are also used in representing less tangible interactions as might occur
in ecosystems, sociological relationships, databases, or the flow of control in a computer
program.
Formally, such configurations are modeled by combinatorial structures called graphs,
consisting of two sets called vertices and edges and an incidence relation between them.
The vertices and edges may have additional attributes, such as color or weight, or any-
thing else useful to a particular model. Graph models tend to fall into a handful of
categories. For instance, the network of one-way streets of a city requires a model in
which each edge is assigned a direction; two atoms in an organic molecule may have more
than one bond between them; and a computer program is likely to have loop structures.
These examples require graphs with directions on their edges, with multiple connections
between vertices, or with connections from a vertex to itself.
In the past these different types of graphs were often regarded as separate entities,
each with its own set of definitions and properties. We have adopted a unified approach by
introducing all of these various graphs at once. This allows us to establish properties that
are shared by several classes of graphs without having to repeat arguments. Moreover,
this broader perspective has an added bonus: it inspires computer representations that
lead to the design of fully reusable software for applications.
We begin by introducing the basic terminology needed to view graphs both as con-
figurations in space and as combinatorial objects that lend themselves to computer rep-
resentation. Pure and applied examples illustrate how different kinds of graphs arise as
models.
1
i i
i i
i i

i i
2 Chapter 1 INTRODUCTION TO GRAPH MODELS
1.1 GRAPHS AND DIGRAPHS

We think of a graph as a set of points in a plane or in 3-space and a set of line segments
(possibly curved), each of which either joins two points or joins a point to itself.
p k
q
v
A B c
g
u f
a b h
r s d
w
Figure 1.1.1 Line drawings of a graph A and a graph B.
Graphs are highly versatile models for analyzing a wide range of practical problems
in which points and connections between them have some physical or conceptual inter-
pretation. Placing such analysis on solid footing requires precise definitions, terminology,
and notation.
definition: A graph G = (V, E) is a mathematical structure consisting of two finite
sets V and E. The elements of V are called vertices (or nodes), and the elements of E
are called edges. Each edge has a set of one or two vertices associated to it, which are
called its endpoints.
terminology: An edge is said to join its endpoints. A vertex joined by an edge to a
vertex v is said to be a neighbor of v.
definition: The (open) neighborhood of a vertex v in a graph G, denoted N (v), is
the set of all the neighbors of v. The closed neighborhood of v is given by N [v] =
N (v) ∪ {v}.
notation: When G is not the only graph under consideration, the notations VG and
EG (or V (G) and E(G)) are used for the vertex- and edge-sets of G, and the notations
NG (v) and NG [v] are used for the neighborhoods of v.
Example 1.1.1: The vertex- and edge-sets of graph A in Figure 1.1.1 are given by
VA = {p, q, r, s} and EA = {pq, pr, ps, rs, qs}
and the vertex- and edge-sets of graph B are given by
VB = {u, v, w} and EB = {a, b, c, d, f, g, h, k}
Notice that in graph A, we are able to denote each edge simply by juxtaposing its
endpoints, because those endpoints are unique to that edge. On the other hand, in graph
B, where some edges have the same set of endpoints, we use explicit names to denote
the edges.
Simple Graphs and General Graphs

In certain applications of graph theory and in some theoretical contexts, there are
frequent instances in which an edge joins a vertex to itself or multiple edges have the
i i
i i
i i

i i
Section 1.1 Graphs and Digraphs 3
same set of endpoints. In other applications or theoretical contexts, such instances are
absent.
definition: A proper edge is an edge that joins two distinct vertices.
definition: A self-loop is an edge that joins a single endpoint to itself.†
definition: A multi-edge is a collection of two or more edges having identical end-
points. The edge multiplicity is the number of edges within the multi-edge.
definition: A simple graph has neither self-loops nor multi-edges.
definition: A loopless graph (or multi-graph) may have multi-edges but no self-
loops.
definition: A (general) graph may have self-loops and/or multi-edges.
Example 1.1.1 continued: Graph A in Figure 1.1.1 is simple. Graph B is not simple;
the edges a, b, and k are self-loops, and the edge-sets {f, g, h} and {a, b} are multi-edges.
terminology: When we use the term graph without a modifier, we mean a general
graph. An exception to this convention occurs when an entire section concerns simple
graphs only, in which case, we make an explicit declaration at the beginning of that
section.
terminology note: Some authors use the term graph without a modifier to mean
simple graph, and they use pseudograph to mean general graph.
Null and Trivial Graphs

definition: A null graph is a graph whose vertex- and edge-sets are empty.
definition: A trivial graph is a graph consisting of one vertex and no edges.
Edge Directions
An edge between two vertices creates a connection in two opposite senses at once.
Assigning a direction makes one of these senses forward and the other backward. In a line
drawing, the choice of forward direction is indicated by placing an arrow on an edge.
definition: A directed edge (or arc) is an edge, one of whose endpoints is designated
as the tail, and whose other endpoint is designated as the head.
terminology: An arc is said to be directed from its tail to its head.
notation: In a general digraph, the head and tail of an arc e may be denoted head(e)
and tail(e), respectively.
definition: Two arcs between a pair of vertices are said to be oppositely directed if
they do not have the same head and tail.
definition: A multi-arc is a set of two or more arcs having the same tail and same
head. The arc multiplicity is the number of arcs within the multi-arc.
definition: A directed graph (or digraph) is a graph each of whose edges is directed.
† We use the term “self-loop” instead of the more commonly used term “loop” because loop means
something else in many applications.
i i
i i
i i

i i
definition: A digraph is simple if it has neither self-loops nor multi-arcs.

notation: In a simple digraph, an arc from vertex u to vertex v may be denoted by uv
or by the ordered pair [u, v].
Example 1.1.2: The digraph in Figure 1.1.2 is simple. Its arcs are uv, vu, and vw.
u v w
Figure 1.1.2 A simple digraph with a pair of oppositely directed arcs.
definition: A mixed graph (or partially directed graph) is a graph that has both
undirected and directed edges.
definition: The underlying graph of a directed or mixed graph G is the graph that
results from removing all the designations of head and tail from the directed edges of G
(i.e., deleting all the edge-directions).
Example 1.1.3: The digraph D in Figure 1.1.3 has the graph G as its underlying graph.
D v G v
c c
u g u g
h h
d d
w w
Figure 1.1.3 A digraph and its underlying graph.
Simple and non-simple graphs and digraphs all commonly arise as models; the nu-
merous and varied examples in §1.3 illustrate the robustness of our comprehensive graph
model.
Formal Specification of Graphs and Digraphs

Except for the smallest graphs, line drawings are inadequate for describing a graph;
imagine trying to draw a graph to represent a telephone network for a small city. Since
many applications involve computations on graphs having hundreds, or even thousands,
of vertices, another, more formal kind of specification of a graph is often needed.
The specification must include (implicitly or explicitly) a function endpts that speci-
fies, for each edge, the subset of vertices on which that edge is incident (i.e., its endpoint
set). In a simple graph, the juxtaposition notation for each edge implicitly specifies its
endpoints, making formal specification simpler for simple graphs than for general graphs.
definition: A formal specification of a simple graph is given by an adjacency
table with a row for each vertex, containing the list of neighbors of that vertex.
i i
i i
i i

i i
Example 1.1.4: Figure 1.1.4 shows a line drawing and a formal specification for a simple
graph.
p q
p: q r s
q: p s
r: p s
s: p q r
r s
Figure 1.1.4 A simple graph and its formal specification.
definition: A formal specification of a general graph G = (V, E, endpts) consists

of a list of its vertices, a list of its edges, and a two-row incidence table (specifying
the endpts function) whose columns are indexed by the edges. The entries in the column
corresponding to edge e are the endpoints of e. The same endpoint appears twice if e is
a self-loop. (An isolated vertex will appear only in the vertex list.)
Example 1.1.5: Figure 1.1.5 shows a line drawing and a formal specification for a gen-
eral graph.
k
v V = {u, v, w} and E = {a, b, c, d, f, g, h, k}
c
g edge a b c d f g h k
u f
a b h endpts u u u w v v w v
u u v u w w v v
d
w
Figure 1.1.5 A general graph and its formal specification.
definition: A formal specification of a general digraph or a mixed graph

D = (V, E, endpts, head, tail) is obtained from the formal specification of the underly-
ing graph by adding the functions head : EG → VG and tail : EG → VG , which designate
the head vertex and tail vertex of each arc.
One way to specify these designations in the incidence table is to mark in each column
the endpoint that is the head of the corresponding arc.
Remark: A mixed graph is specified by simply restricting the functions head and tail
to a proper subset of EG . In this case, a column of the incidence table that has no mark
means that the corresponding edge is undirected.
Example 1.1.6: Figure 1.1.6 gives the formal specification for the digraph shown, in-
cluding the corresponding values of the functions head and tail. A superscript “h” is used
to indicate the head vertex.
Remark: Our approach treats a digraph as an augmented type of graph, where each
edge e of a digraph is still associated with a subset endpts(e), but which now also includes
a mark on one of the endpoints to specify the head of the directed edge.
This viewpoint is partly motivated by its impact on computer implementations of graph
algorithms (see the computational notes), but it has some advantages from a mathemat-
ical perspective as well. Regarding digraphs as augmented graphs makes it easier to view
certain results that tend to be established separately for graphs and for digraphs as a
single result that applies to both.
i i
i i
i i

i i
k
v
c
f g edge a b c d f g h k
u
a b h endpts u uh u w vh v w v
uh u vh uh w wh vh vh
d
w
head(a) = tail(a) = head(b) = tail(b) = head(d) = tail(c) = u;

head(c) = head(h) = head(f ) = tail(g) = head(k) = tail(k) = v;
head(g) = tail(d) = tail(h) = tail(f ) = w.
Figure 1.1.6 A general digraph and its formal specification.

Also, our formal incidence specification permits us to reverse the direction of an edge e at
any time, just by reversing the values of head(e) and tail(e). This amounts to switching
the h mark in the relevant column of the incidence table or reversing the arrowhead in
the digraph drawing.
computational note 1: These formal specifications for a graph and a digraph can
easily be implemented with a variety of programmer-defined data structures, whatever
is most appropriate to the application. A discussion of the comparative advantages and
disadvantages of a few of the most common computer information structures for graphs
and digraphs appears at the end of Chapter 2.
computational note 2: (A caution to software designers) From the perspective of
object-oriented software design, the ordered-pair representation of arcs in a digraph treats
digraphs as a different class of objects from graphs. This could seriously undermine
software reuse. Large portions of computer code might have to be rewritten in order to
adapt an algorithm that was originally designed for a digraph to work on an undirected
graph.
The ordered-pair representation could also prove awkward in implementing algorithms
for which the graphs or digraphs are dynamic structures (i.e., they change during the
algorithm). Whenever the direction on a particular edge must be reversed, the associated
ordered pair has to be deleted and replaced by its reverse. Even worse, if a directed edge
is to become undirected, then an ordered pair must be replaced with an unordered pair.
Similarly, the undirected and directed edges of a mixed graph would require two different
types of objects.
computational note 3: For some applications (network layouts on a surface, for in-
stance), the direction of flow around a self-loop has practical importance, and distin-
guishing between the ends of a self-loop becomes necessary. This distinction is made in
Chapters 8 and 16 but not elsewhere.
Mathematical Modeling with Graphs

To bring the power of mathematics to bear on real-world problems, one must first
model the problem mathematically. Graphs are remarkably versatile tools for modeling,
and their wide-ranging versatility is a central focus throughout the text.
i i
i i
i i

i i
Example 1.1.7: The mixed graph in Figure 1.1.7 is a model for a roadmap. The vertices
represent landmarks, and the directed and undirected edges represent the one-way and
two-way streets, respectively.
Gas Hardware Train

Station Store Station
Bank
Firehouse
Dentist Grocery
Town Hall Doctor
School
Pizzeria Library
Candy
Store Park
Figure 1.1.7 Road-map of landmarks in a small town.
Example 1.1.8: The digraph in Figure 1.1.8 represents the hierarchy within a company.
This illustrates how, beyond physical networks, graphs and digraphs are used to model
social relationships.
Department Head
Supervisors
Staff Members
Figure 1.1.8 A corporate hierarchy.
Degree of a Vertex
definition: Adjacent vertices are two vertices that are joined by an edge.
definition: Adjacent edges are two distinct edges that have an endpoint in common.
definition: If vertex v is an endpoint of edge e, then v is said to be incident on e, and
e is incident on v.
definition: The degree (or valence) of a vertex v in a graph G, denoted deg(v), is
the number of proper edges incident on v plus twice the number of self-loops.†
terminology: A vertex of degree d is also called a d-valent vertex.
notation: The smallest and largest degrees in a graph G are denoted δmin and δmax
(or δmin (G) and δmax (G) when there is more than one graph under discussion). Some
authors use δ instead of δmin and ∆ instead of δmax .
definition: The degree sequence of a graph is the sequence formed by arranging the
vertex degrees in non-increasing order.
† Applications of graph theory to physical chemistry motivate the use of the term valence.
i i
i i
i i

i i
Example 1.1.9: Figure 1.1.9 shows a graph and its degree sequence.
k
v
y x c h6, 6, 4, 1, 1, 0i
u f g v u wz y x
a b h
z d
w
Figure 1.1.9 A graph and its degree sequence.
Although each graph has a unique degree sequence, two structurally different graphs
can have identical degree sequences.
Example 1.1.10: Figure 1.1.10 shows two different graphs, G and H, with the same
degree sequence.
G H
Figure 1.1.10 Two graphs whose degree sequences are both h3, 3, 2, 2, 2, 2i.
The following result shows that the degree sequence of a simple graph must have at
least two equal terms. This has an interesting interpretation in a sociological model that
appears in Section 1.3 (see Application 1.3.2).
Proposition 1.1.1: A non-trivial simple graph G must have at least one pair of vertices
whose degrees are equal.
Proof: Suppose that the graph G has n vertices. Then there appear to be n possible
degree values, namely 0, . . . , n − 1. However, there cannot be both a vertex of degree
0 and a vertex of degree n − 1, since the presence of a vertex of degree 0 implies that
each of the remaining n − 1 vertices is adjacent to at most n − 2 other vertices. Hence,
the n vertices of G can realize at most n − 1 possible values for their degrees. Thus, the
pigeonhole principle implies that at least two of the n vertices have equal degree. ♦
The work of Leonhard Euler (1707–1783) is regarded as the beginning of graph theory
as a mathematical discipline. The following theorem of Euler establishes a fundamental
relationship between the vertices and edges of a graph.
Theorem 1.1.2: [Euler’s Degree-Sum Theorem] The sum of the degrees of the ver-
tices of a graph is twice the number of edges.
Proof: Each edge contributes two to the degree sum. ♦

Corollary 1.1.3: In a graph, there is an even number of vertices having odd degree.
Proof: Consider separately, the sum of the degrees that are odd and the sum of those
that are even. The combined sum is even by Theorem 1.1.2, and since the sum of the
even degrees is even, the sum of the odd degrees must also be even. Hence, there must
be an even number of vertices of odd degree. ♦
i i
i i
i i

i i
Remark: By Theorem 1.1.2, the sum of the terms of a degree sequence of a graph is
even. Theorem 1.1.4 establishes the following converse: any non-increasing, nonnegative
sequence of integers whose sum is even is the degree sequence of some graph. Example
1.1.11 illustrates the construction used in its proof.
Example 1.1.11: To construct a graph whose degree sequence is h5, 4, 3, 3, 2, 1, 0i, start
with seven isolated vertices v1 , v2 , . . . , v7 . For the even-valued terms of the sequence,
draw the appropriate number of self-loops on the corresponding vertices. Thus, v2 gets
two self-loops, v5 gets one self-loop, and v7 remains isolated. For the four remaining odd-
valued terms, group the corresponding vertices into any two pairs, for instance, v1 , v3
and v4 , v6 . Then join each pair by a single edge and add to each vertex the appropriate
number of self-loops. The resulting graph is shown in Figure 1.1.11.
v4 v6
v2
v1 v3 v5 v7
Figure 1.1.11 Constructing a graph with degree sequence h5, 4, 3, 3, 2, 1, 0i.
Theorem 1.1.4: Suppose that hd1 , d2 , . . . , dn i is a sequence of nonnegative integers

whose sum is even. Then there exists a graph with vertices v1 , v2 , . . . , vn such that
deg(vi ) = di , for i = 1, . . . , n.
Proof: Start with n isolated vertices v1 , v2 , . . . , vn . For each i, if di is even, draw d21
self-loops on vertex vi , and if di is odd, draw d12−1 self-loops. By Corollary 1.1.3, there
is an even number of odd di ’s. Thus, the construction can be completed by grouping the
vertices associated with the odd terms into pairs and then joining each pair by a single
edge. ♦
Graphic Sequences
The construction in Theorem 1.1.4 is straightforward but hinges on allowing the
graph to be non-simple. A more interesting problem is determining when a sequence is
the degree sequence of a simple graph.
definition: A non-increasing sequence hd1 , d2 , . . . , dn i is said to be graphic if it is the
degree sequence of some simple graph. That simple graph is said to realize the sequence.
Remark: If hd1 , d2 , . . . , dn i is the degree sequence of a simple graph, then, clearly, d1 ≤
n − 1.
Theorem 1.1.5: Let hd1 , d2 , . . . , dn i be a graphic sequence, with d1 ≥ d2 ≥ . . . ≥ dn .
Then there is a simple graph with vertex-set {v1 , . . . , vn } satisfying deg (vi ) = di for
i = 1, 2, . . . , n, such that v1 is adjacent to vertices v2 , . . . , vd1 +1 .
Proof: Among all simple graphs with vertex-set {v1 , v2 , . . . , vn } and deg(vi ) = di , i =
1, 2, . . . , n, let G be one for which r = |NG (v1 ) ∩ {v2 , . . . , vd1 +1 }| is maximum. If r = d1 ,
then the conclusion follows. If r < d1 , then there exists a vertex vs , 2 ≤ s ≤ d1 + 1,
such that v1 is not adjacent to vs , and there exists a vertex vt , t > d1 + 1 such that v1
is adjacent to vt (since deg(v1 ) = d1 ). Moreover, since deg(vs ) ≥ deg(vt ), there exists a
i i
i i
i i

i i
vertex vk such that vk is adjacent to vs but not to vt . Let G̃ be the graph obtained from
G by replacing the edges v1 vt and vs vk with the edges v1 vs and vt vk (as shown in Figure
1.1.12). Then the degrees are all preserved and vs ∈ NG̃ (v1 ) ∩ {v3 , . . . , vd1 +1 }. Thus,
|NG̃ (v1 ) ∩ {v2 , . . . , vd1 +1 }| = r + 1, which contradicts the choice of G and completes the
proof. ♦
v1 v1
vt vt
vs vs
vk vk
Figure 1.1.12 Switching adjacencies while preserving all degrees.
Corollary 1.1.6: [Havel (1955) and Hakimi (1961)] A sequence hd1 , d2 , . . . , dn i

of nonnegative integers such that d1 ≤ n − 1 and d1 ≥ d2 ≥ . . . ≥ dn is graphic if
and only if the sequence hd2 − 1, . . . , dd1 +1 − 1, dd1 +2 , . . . , dn i is graphic.
♦ (Exercises)
Remark: Corollary 1.1.6 yields a recursive algorithm that decides whether a non-
increasing sequence is graphic.
Algorithm 1.1.1: GraphicSequence (hd1 , d2 , . . . , dn i)

Input: a non-increasing sequence hd1 , d2 , . . . , dn i .
Output: TRUE if the sequence is graphic, FALSE if it is not.
If d1 ≥ n or dn < 0
Return FALSE
Else
If d1 = 0
Return TRUE
Else
Let ha1 , a2 , . . . , an−1 i be a non-increasing permutation
of hd2 − 1, . . . , dd1 +1 − 1, dd1 +2 , . . . , dn i .
Return GraphicSequence(ha1 , a2 , . . . , an−1 i)
Remark: An iterative version of the algorithm GraphicSequence based on repeated

application of Corollary 1.1.6 can also be written and is left as an exercise (see Exercises).
Also, given a graphic sequence, the steps of the iterative version can be reversed to
construct a graph realizing the sequence. However many zeroes you get at the end of
the forward pass, start with that many isolated vertices. Then backtrack the algorithm,
adding a vertex each time. The following example illustrates these ideas.
Example 1.1.12: We start with the sequence h3, 3, 2, 2, 1, 1i. Figure 1.1.13 illustrates an
iterative version of the algorithm GraphicSequence and then illustrates the backtracking
steps leading to a graph that realizes the original sequence. The hollow vertex shown in
each backtracking step is the new vertex added at that step.
i i
i i
i i

i i
< 3, 3, 2, 2, 1, 1 >
Cor. 1.1.7
< 2, 1, 1, 1, 1 >
Cor. 1.1.7
< 0, 0, 1, 1 >
permute
< 1, 1, 0, 0 >
Cor. 1.1.7
< 0, 0, 0 >
Figure 1.1.13 Testing and realizing the sequence h3, 3, 2, 2, 1, 1i.
Indegree and Outdegree in a Digraph

The definition of vertex degree is slightly refined for digraphs.
definition: The indegree of a vertex v in a digraph is the number of arcs directed to
v; the outdegree of vertex v is the number of arcs directed from v. Each self-loop at v
counts one toward the indegree of v and one toward the outdegree.
k
v
c vertex u v w
f g
a b
u indegree 3 4 1
h
outdegree 3 2 3
d
w
Figure 1.1.14 The indegrees and outdegrees of the vertices of a digraph.
The next theorem is the digraph version of Euler’s Degree-Sum Theorem 1.1.2.
Theorem 1.1.7: In a digraph, the sum of the indegrees and the sum the outdegrees
both equal the number of edges.
Proof: Each directed edge e contributes one to the indegree at head(e) and one to the
outdegree at tail(e). ♦
i i
i i
i i

i i
EXERCISES for Section 1.1

In Exercises 1.1.1 through 1.1.3, construct a line drawing, an incidence table, and the
degree sequence of the graph with the given formal specification.
1.1.1S V = {u, w, x, z} ; E = {e, f, g}

endpts(e) = {w}; endpts(f ) = {x, w}; endpts(g) = {x, z}
1.1.2 V = {u, v, x, y, z} ; E = {a, b, c, d}

endpts(e) = {u, v}; endpts(b) = {x, v}; endpts(c) = {u, v}; endpts(d) = {x}
1.1.3 V = {u, v, x, y, z} ; E = {e, f, g, h, k}

endpts(e) = endpts(f ) = {u, v}; endpts(g) = {x, z}; endpts(h) = endpts(k) = {y}
In Exercises 1.1.4 through 1.1.6, construct a line drawing for the digraph or mixed graph
with vertex-set V = {u, v, w, x, y}, edge-set E = {e, f, g, h}, and the given incidence table.
1.1.4S edges e f g h
endpts y xh vh vh
wh u x u
1.1.5 edges e f g h
endpts x vh v v
wh u uh uh
1.1.6 edges e f g h
endpts u xh v v
wh u y u
In Exercises 1.1.7 through 1.1.9, give a formal specification for the given digraph.
a
1.1.7S
x y
d b
z c
v b x
1.1.8
a d
e
u f g y
z
v x
1.1.9
h d a
c e
y
u f z g
b
In Exercises 1.1.10 through 1.1.12, give a formal specification for the underlying graph
of the digraph indicated.
1.1.10S The digraph of Exercise 1.1.7.
1.1.11 The digraph of Exercise 1.1.8.
1.1.12 The digraph of Exercise 1.1.9.
i i
i i
i i

i i
1.1.13 Draw a graph with the given degree sequence.

a. h8, 7, 3i b. h9, 8, 8, 6, 5, 3, 1i
1.1.14 Draw a simple graph with the given degree sequence.
a. h6, 4, 4, 3, 3, 2, 1, 1i b. h5, 5, 5, 3, 3, 3, 3, 3i
For each of the number sequences in Exercises 1.1.15 through 1.1.18, either draw a simple
graph that realizes it, or explain, without resorting to Corollary 1.1.6 or Algorithm 1.1.1,
why no such graph can exist.
1.1.15S a. h2, 2, 1, 0, 0i b. h4, 3, 2, 1, 0i

1.1.16 a. h4, 2, 2, 1, 1i b. h2, 2, 2, 2i
1.1.17 a. h4, 3, 2, 2, 1i b. h4, 3, 3, 3, 1i

1.1.18 a. h4, 4, 4, 4, 3, 3, 3, 3i b. h3, 2, 2, 1, 0i
1.1.19 Apply Algorithm 1.1.1 to each of the following sequences to determine whether
it is graphic. If the sequence is graphic, then draw a simple graph that realizes it.
a. h7, 6, 6, 5, 4, 3, 2, 1i b. h5, 5, 5, 4, 2, 1, 1, 1i
c. h7, 7, 6, 5, 4, 4, 3, 2i d. h5, 5, 4, 4, 2, 2, 1, 1i
1.1.20 Use Theorem 1.1.5 to prove Corollary 1.1.6.
1.1.21 Write an iterative version of Algorithm 1.1.1 that applies Corollary 1.1.6 repeat-
edly until a sequence of all zeros or a sequence with a negative term results.
1.1.22S Given a group of nine people, is it possible for each person to shake hands with
exactly three other people?
1.1.23 Draw a graph whose degree sequence has no duplicate terms.
1.1.24S What special property of a function must the endpts function have for a graph
to have no multi-edges?
1.1.25 Draw a digraph for each of the following indegree and outdegree sequences, such
that the indegree and outdegree of each vertex occupy the same position in both se-
quences.
a. in: h1, 1, 1i out: h1, 1, 1i
b. in: h2, 1i out: h3, 0i
definition: A pair of sequences ha1 , a2 , . . . , an i and hb1 , b2 , . . . , bn i is called digraphic if

there exists a simple digraph with vertex-set {v1 , v2 , . . . , vn } such that outdegree(vi ) = ai
and indegree(vi ) = bi for i = 1, 2, . . . , n.
1.1.26 Determine whether the pair of sequences h3, 1, 1, 0i and h1, 1, 1, 2i is digraphic.
1.1.27 Establish a result like Corollary 1.1.6 for a pair sequences to be digraphic.
1.1.28 How many different degree sequences can be realized for a graph having three
vertices and three edges?
i i
i i
i i

i i
1.1.29 Given a list of three vertices and a list of seven edges, show that 37 different
formal specifications for simple graphs are possible.
1.1.30 Given a list of four vertices and a list of seven edges, show that 72 56 210 different

formal specifications are possible if there are exactly two self-loops.

1.1.31 Given a list of three vertices and a list of seven edges, how many different formal
specifications are possible if exactly three of the edges are directed?
1.1.32S Does there exist a simple graph with five vertices, such that every vertex is
incident with at least one edge, but no two edges are adjacent?
1.1.33 Prove or disprove: There exists a simple graph with 13 vertices, 31 edges, three
1-valent vertices, and seven 4-valent vertices.
definition: Let G = (V, E) be a graph and let W ⊆ V. Then W is a vertex cover of

G if every edge is incident on at least one vertex in W. (See §10.4.)
1.1.34 Find upper and lower bounds for the size of a minimum (smallest) vertex cover
of an n-vertex connected simple graph G. Then draw three 8-vertex graphs, one that
achieves the lower bound, one that achieves the upper bound, and one that achieves
neither.
1.1.35 Find a minimum vertex cover for the underlying graph of the mixed graph shown
in Figure 1.1.7.
1.1.36S The graph shown below represents a network of tunnels, where the edges are
sections of tunnel and the vertices are junction points. Suppose that a guard placed at a
junction can monitor every section of tunnel leaving it. Determine the minimum number
of guards and a placement for them so that every section of tunnel is monitored.
a g
d
j
b e h
f
c k i
definition: An independent set of vertices in a graph G is a set of mutually non-

adjacent vertices. (See §2.3.)
1.1.37 Find upper and lower bounds for the size of a maximum (largest) independent
set of vertices in an n-vertex connected graph. Then draw three 8-vertex graphs, one
that achieves the lower bound, one that achieves the upper bound, and one that achieves
neither.
1.1.38 Find a maximum independent set of vertices in the underlying graph of the mixed
graph shown in Figure 1.1.7.
1.1.39S Find a maximum independent set of vertices in the graph of Exercise 1.1.36.
definition: A matching in a graph G is a set of mutually non-adjacent edges in G.

(See §8.3 and §10.4.)
i i
i i
Another random document with
no related content on Scribd:
compositors to put such texts correctly into type, the work of reviser,
press-corrector, or proof-reader, in the earlier days of printing,
demanded a very high standard of scholarship and a wide range of
knowledge. There was, therefore, no reason why Erasmus should
have been ashamed to admit that he had done work of this kind.
Some years later he gave to his friend Froben, the great publisher of
Basel, similar service and co-operation. The intimate relations of
Erasmus with Aldus and Froben, by far the greatest publishers of the
time, had no little influence in furthering the world-wide circulation
secured for his works.
While in Venice, Erasmus also supervised the printing of a revised
edition of his Adagia (Proverbs) which appeared in 1508. For this
work, Aldus obtained a privilege both in Venice and in Rome, and
there were printed in Venice alone eight editions. When, however, in
1520, Paul Manutius undertook again to reprint the Adagia, he found
that he had to contend with an increasing hostility on the part of the
Church against anything bearing the name of Erasmus. The book
was finally issued anonymously, and it was described in the
catalogue as the work of “Batavus quidam homo” (a certain
Hollander).
In 1512, Aldus printed, under the instructions of Erasmus, (who
was, however, at that time no longer in Italy) the Colloquies and the
Praise of Folly. There is unfortunately no record of the publishing
arrangement arrived at for these, but as Erasmus complained bitterly
of the loss and injury caused to the author through the wide sale of
the piracy issues, it is fair to assume that he had reserved an interest
in the authorised editions. In the introduction to his Adagia, Erasmus
writes as follows: “Formerly there was devoted to the correctness of
a literary manuscript as much care and attention as to the writing of
a notarial instrument. Such care and precision were held to be a
sacred duty. Later, the copying of manuscripts was entrusted to
ignorant monks and even to women. But how much more serious is
the evil that can be brought about by a careless printer, and yet to
this matter the law gives no heed. A dealer who sells English stuffs
under the guise of Venetian is punished, but the printer who in place
of correct texts, misleads and abuses the reader with pages the
contents of which are an actual trial and torment, escapes
unharmed. It is for this reason that Germany is plagued with so many
books that are deformed (i. e., untrustworthy). The authorities will
supervise with arbitrary regulations the proper methods for the
baking of bread, but concern themselves not at all as to the
correctness of the work of the printers, although the influence of bad
typography is far more injurious than that of bad bread.”
The relations of Aldus with Johann Reuchlin were longer and more
intimate than with Erasmus. It was natural enough that the scholar
who may properly be called the founder of Greek studies in
Germany, should have come into close relations with the publisher
who had undertaken to produce Greek texts for Europe and who had
founded a Greek academy in Venice. In 1498, Aldus printed the Latin
oration which Reuchlin had addressed to Pope Alexander VI., in
behalf of the Prince Palatine Philip, and from that date the two men
remained in regular correspondence with each other. In 1502, Aldus,
writing to Reuchlin (who was at that time in Pforzheim), gives, as to a
trusted friend upon whose sympathy and intelligent interest he could
depend, the details of his publishing undertakings and of his plans
and hopes for the future, and asks for counsel on various points. A
few months later, in another letter, Aldus writes:
“I am hardly able to express my gratification at your friendly words
concerning the importance and the value of my publishing
undertakings. It is no light thing to secure the commendation of one
of the greatest scholars of his time. If my life is spared to me, I hope
more fully to deserve the praise that you give to me for service
rendered to the scholarship and enlightenment of the age.”
Reuchlin was not only a friendly counsellor of the Venetian
publisher, but a valuable customer also for his books. In addition to
purchasing for his own library a full series of the Aldine editions,
Reuchlin appears to have interested himself keenly in commending
these to his scholarly acquaintances, not only, as he states, in order
to encourage a great undertaking, but for the purpose of doing
service to German students. In 1509, Reuchlin was appointed by the
Duke of Bavaria, Professor of Greek and Hebrew in the University of
Ingolstadt, the first professorship of Greek instituted in Germany.
Reuchlin said more than once that the work of his Chair had been
made possible only through the service rendered by Aldus in
providing the Greek texts.
The influence of Aldus not only on the publishing standards but on
the scholarly and literary conditions of Germany, was in fact
widespread and important. Kapp, the historian of the German book-
trade, speaks of it as more important than that of all the German
publishers of his generation. This influence was due not only to the
publishing undertakings of the Aldine Press, but to the intimate
relations maintained by its founder with many of the German
scholars, relations which helped to establish a community of
interests between the literary centres of Italy and Germany and to
direct German scholarship into new paths. The separation of political
boundaries had no significance for a man with the humanitarian
ideals of Aldus, while the fact that Latin was the universal language
of scholarship and of literature, helped not a little to bring about that
community of feeling among scholars which was the special aim of
the Venetian publisher. In 1502, Aldus writes to John Taberio, in
Brescia:
“I am delighted to learn that so many men of distinction in the
great city of Brescia are, under your guidance, devoting themselves
with ardour to Greek studies. The expectations with which I
undertook the publication of Greek texts are being more than
realised. I am, in fact, not a little astonished to find that even in these
sad times of war in which my undertakings have been begun, so
many are found ready to give the same ardour to scholarly pursuits
that they are giving to fighting against the infidel and to civil strife.
Thus it happens that even from the midst of war arises literature,
which has for so many years lain buried. And it is not only in Italy, but
also in Germany, in France, in Pannonia, in Spain, and in England,
and wherever the Latin language is known, that young and old are
devoting themselves to the study of Greek. The joy that this brings to
me causes me to forget my fatigues, and redoubles my zeal to do
what is in my power for the service of scholarship, and particularly
for the students who are growing up in this time of the renaissance of
letters.”
During the first years of the sixteenth century, the difficulties in the
transmission either of merchandise or of money were many. The
packages of books which Aldus had occasion to send to Reuchlin in
Stuttgart, for instance, came forward sometimes by way of Milan,
Vienna, or Basel, and later through Augsburg. The Augsburg
banking-house of Függer, founded about 1450, possessed in 1500
(and for half a century thereafter) connections which enabled them to
take charge not only of what we should call mercantile bills and
banking credits, but also of the forwarding and delivery of the goods
against which the bills were drawn. They carried on what to-day
would be called an express business, and in a majority of instances
the instructions were evidently to make collections on delivery.
During the first half of the sixteenth century, the Függers, with their
branch houses in Florence, Venice, and Genoa, supplied the most
valuable machinery for the transaction of business between Italy and
Germany. These communications, however, were of necessity very
frequently interrupted by the troubles of the times.
In 1510, Mutianus Rufus writes to Urban that “in connection with
the conflicts between the French and the Venetian soldiers, the
passes of the Alps have been blocked, so that literature from Venice
can no longer find its way into Germany. I had hoped with the next
Frankfort Fair, to be able to place in the hands of my students the
beautiful Aldine editions. But my hopes were in vain. When the Fair
was opened, there was not a single volume from Italy. We shall be
able this spring to do nothing in our classical schools. Oh, the
stupidities of war!”
In 1514, the Elector Frederic the Wise of Saxony applied to the
several powers interested for a safe conduct for his librarian,
Spalatin, whom he desired to send to Venice to purchase directly
from Aldus the Aldine classics for the library of Wittenberg. Some
difficulties intervened, however, as Spalatin appears never to have
reached Venice. It was doubtless due to the long-continued wars
between the Emperor and the States of Italy, that Aldus was unable,
during his own lifetime, to establish direct agencies in Germany for
his publications. We find record of such agencies in Frankfort, Basel,
Augsburg, and Nuremberg, first in the time of his son, agencies
which were extended by the grandson.
The active work of Aldus extended over a period of twenty years,
from 1495 to 1515. This time included the wars of 1500, 1506, 1510,
and 1511, in which Venice was directly engaged, wars which had of
necessity much to do with the interference with his business, and
with the difficulties, of which he makes continual complaint, in
securing returns for his sales. “For seven years,” writes Aldus in
1510, “books have had to contend against arms.” There appears to
have been no single year of the twenty in which he was free from
pressing financial cares, while from time to time the work of the
presses and in the composing room came to an actual standstill for
want of funds. During these twenty years he printed not less than
126 works which previously existed only in manuscript form, and the
manuscript copies of which had to be secured and carefully edited.
It is probable that Aldus, in his own enthusiasm concerning the
value and importance of the re-discovered classics, had
overestimated the extent of the interest that could be depended upon
for these classics throughout the world. It is evident, however, that
there were enough scholars in Italy, Germany, France, and the Low
Countries, to assure a widespread demand for the Aldine editions,
and that the larger part of the publisher’s difficulties consisted in the
lack of convenient machinery for making known to these scholars the
fact that such books had been prepared, for the delivery of such
copies as might be ordered, and for the collection of the payments
due.
Another serious difficulty with which Aldus had to contend was the
competition of the piratical copies of his editions which promptly
appeared in Cologne, Tübingen, Lyons, and even so close at home
as Florence. The most serious interference with his undertakings
appears to have come from the printers of Lyons, who in their
enterprising appropriations from Paris on the one hand and from
Nuremberg, Basel, and Venice on the other, speedily won for their
city notoriety as the centre of piratical publishing. The Lyons printers
printed editions of the Aldine Latin classics, making a very close
imitation of the cursive or italic type, and issued the volumes without
imprint, date, or place of publication.
The privileges secured from the government of Venice had effect,
of course, only in Venetian territory. Privileges were given by the
Pope for a number of the Aldine publications, and these covered, in
form, at least, not only the States of the Church but the territory of all
States recognising the papal authority, while the penalties for
infringing such papal privileges were not infrequently made to
include excommunication. There was, however, no machinery by
means of which the papal authority could be brought to bear upon
Catholics infringing or disregarding the privileges, and as a fact the
papal privileges proved of very little service in protecting the literary
property either of Aldus or of later literary workers. A further word
concerning the privileges issued in Venice and in the other States of
Italy will be given in a later division of this narrative.
Apart from this important work in the scholarly and editorial
divisions of publishing, Aldus made several distinctive contributions
to the art of book-making. He was, as before stated, the first printer
who founded complete and perfect fonts of Greek type, fonts which
for many years served as models for the printers of Europe. He
invented the type which was first called cursive, and which is known
to-day as italic, a type having the advantage of presenting the text in
a very compact form. (The cursive font was said to have been
modelled on the script of Petrarch.) And finally, he was the first
publisher who ventured upon the experiment of replacing the costly
and cumbersome folios and quartos, in which form alone all
important works had heretofore been issued, with convenient crown
octavo volumes, the moderate price of which brought them within the
reach of scholars of all classes and helped to popularise the
knowledge and the influence of classic literature. This constituted a
practical revolution in publishing methods.
Aldus had possibly read the remark of Callimachus, the librarian of
the Alexandrian library in 290 b.c., that “A big book is a big
nuisance.” These Aldine classics, while printed in octavo (i. e., upon
a sheet folded in eights), were of a size corresponding more nearly
to what would to-day be known as a sixteenmo, the size of the sheet
of paper being smaller than that used to-day. Aldus had no presses
which would print sheets large enough to fold in sixteen or even in
twelve. The price of these small octavos averaged three marcelli or
two francs, say forty cents. Making allowance for the difference in
the purchasing power of money between the year 1500 and the year
1895, I judge that this may represent about $2.00 of our currency.
For centuries the Aldine editions served as the authoritative texts
for the authors presented, and even to-day they stand as a
wonderful monument of the imagination, the learning, the courage,
and the persistency of their publisher. Good Italian though he were,
Aldus was by some of his countrymen charged with want of
patriotism on the ground that if he helped to make the study of the
classics easy for the Barbarians of the outer world, they would no
longer need to come for their learning to Italy, heretofore the centre
and source of all scholarly enlightenment. To this effect writes
Beatus Rhenanus in his introduction to the Works of Erasmus:
Quidam Venetiis olim Aldo Manutio commentarios Græcos in
Euripidem et Sophoclem edere paranti dixit: Cave, cave hoc facias,
ne barbari istis adjuti domi maneant et pauciores in Italiam ventilent.
Kapp is of opinion that the dread was well founded and that the
distribution throughout Germany and France of popular editions of
the classics, did have the result of keeping at home many students
who would otherwise have crossed the Alps. That they were now
able to secure, at moderate cost and in their own homes, learning for
which heretofore they had been obliged to make long and costly
journeys, was due to the unselfish and public-spirited labours of
Aldus. It was, therefore, with good reason that he was held in high
regard by the Humanists of Germany. They sought his friendship and
nearly overwhelmed him with correspondence. In 1498, Conrad
Celtes and Vincenzo Longinus commemorated his service in verse.
Aldus thanked them for their courtesy, and in sending them as an
acknowledgment copies of his Horace and Virgil, he asked them to
bring him into communication with any scholarly Germans who were
interested in the classics. Aldus did not, however, consider it wise to
print the ode of eulogy that Celtes had written upon the Emperor
Maximilian, because he was afraid of causing offence to the
Bohemians and Hungarians through whose scholars he had secured
not a few rare manuscripts.
Throughout Germany the productions of the Aldine presses were
received with enthusiasm. Mutianus Rufus speaks of himself as
weeping with joy when there came to him from a friend the precious
gift of the editions of Cicero, Lucretius, and other classics. He and
his friends Urban and Spalatin deprived themselves almost of the
necessaries of life, in order to save moneys with which to bring
across the Alps the other volumes of the series. Pirckheimer and
Reuchlin were among the first of the German buyers of the Aldine
classics. Hummelsburger writes in 1512 to Anselm in Tübingen, “I
shall buy my Hebrew books in Italy, where Aldus has printed them in
beautiful texts.... Germany no less than Latium owes a great debt to
Aldus.”
The political status of Italy and its division into a number of states
or principalities which carried on independent policies and which
were frequently in active warfare with each other, entailed serious
difficulties upon the new business of publishing, difficulties which,
while troublesome enough for Aldus in Venice, were still more
serious for his competitors in Florence and Milan. A privilege
secured for Venice was not binding even in times of peace outside of
Venetian territory, while in the frequently recurring times of war, any
privileges which a Venetian or a Milanese publisher had been
fortunate enough to secure in the Italian States were abrogated in
fact if not in form. In this respect, the early publishers of Paris, whose
privileges covered (nominally at least) the territory of the kingdom,
had a decided advantage over their rivals in the much divided
territory of Italy or of Germany.
Aldus had the feeling, for which in his case there appears to have
been sufficient ground, that his business undertakings, with which
were connected far-reaching plans for furthering scholarly
knowledge, were absolutely dependent upon his own continued and
persistent personal attention. While he had succeeded in securing
the services of scholarly associates to share with himself the editorial
responsibilities of his work, he does not appear to have been able,
with the material at his command, to train up any assistants
competent to take any important share in the business management.
One of his many complaints concerning the repeated interruptions
which interfere with his important daily labours, might have been
uttered by many a publisher of later times. He writes in 1514 (the
year before his death) to his friend Navagerus:
“I am hampered in my work by a thousand interruptions.... Nearly
every hour comes a letter from some scholar, and if I undertook to
reply to them all, I should be obliged to devote day and night to
scribbling. Then, through the day come calls from all kinds of visitors.
Some desire merely to give a word of greeting, others want to know
what there is new, while the greater number come to my office
because they happen to have nothing else to do. ‘Let us look in upon
Aldus,’ they say to each other. Then they loaf in and sit and chatter
to no purpose. Even these people with no business are not so bad
as those who have a poem to offer or something in prose (usually
very prosy indeed) which they wish to see printed with the name of
Aldus. These interruptions are now becoming too serious for me,
and I must take steps to lessen them. Many letters I simply leave
unanswered, while to others I send very brief replies; and as I do this
not from pride or from discourtesy, but simply in order to be able to
go on with my task of printing good books, it must not be taken
hardly.... As a warning to the heedless visitors who use up my office
hours to no purpose, I have now put up a big notice on the door of
my office to the following effect: ‘Whoever thou art, thou art earnestly
requested by Aldus, to state thy business briefly and to take thy
departure promptly. In this way thou mayst be of service even as
was Hercules to the weary Atlas. For this is a place of work for all
who may enter.’”
Aldus Manutius died January 25, 1515, (Venetian style,
corresponding to February 6, 1515, modern style) aged sixty-five
years. Until 1529, the business was carried on for the heirs by his
father-in-law, Torresano, and in that year was taken over by Paul
Manutius, the son of Aldus. In 1540, Paul took into partnership his
son, Aldus the younger, and the firm took the title of Aldi Filii. With
the death of Aldus the grandson, in 1597, the family, in its main line,
became extinct, and the work of the Aldine Press, which had
continued for a little more than a century, came to a close. To his
children, Aldus was able to bequeath little besides his fame and the
value of his name. The moneys that had been earned during his
work of twenty-five years from the successful undertakings had been
for the most part absorbed in other ventures which were either
unremunerative, or from which the returns came but slowly. The
carrying out of such great publishing plans required, in fact, business
connections and methods which did not yet exist, and was
dependent also upon the continuance of peace in Europe for a
quarter of a century, an impossible condition for the beginning of the
sixteenth century.
In entering upon business ventures under such difficult
circumstances, Aldus was doubtless, from a business point of view,
unwisely optimistic; but it is difficult not to admire the public spirit and
the pluck with which, in the face of all difficulties, he persisted till the
day of his death in the great schemes he had marked out for himself.
While his work had brought no wealth, his life had been rich in the
accomplishment of great things and in the appreciation given to his
labours. It was also his fortune to gather about him and to come into
relations with many noteworthy men, who as friends and co-workers
shared his enthusiasm, and who gave with him unselfish labour for a
scholarly ideal. Partly because the editors and the publishers were
working for results other than profits, partly because the books
published were (with a few noteworthy exceptions, like the writings of
Erasmus) not original works, but editions of old classics, and partly
because the whole business of publishing was still in its infancy, the
history of the Aldine Press does not present any important
precedents as to the compensation earned by authors for their
productions, or as to the protection of the author’s property rights in
these productions. The relations of Aldus with all the authors, editors,
and scholars with whom he had to do were however more than
satisfactory; they were cordial, resting in a number of cases on a
close personal friendship. The scholars regarded the publisher as
one of themselves, and, in fact, accepted him as a leader.
It is evident that Erasmus, whose writings formed an important
property, was satisfied with the returns secured for him by Aldus. He
speaks with cordial appreciation of the services rendered by his
“authorised publishers,” Aldus of Venice, and Froben of Basel, and
speaks further of the losses caused to himself by the competition of
the piracy reprints of Lyons and Paris. It appears, therefore, that he
retained a continued interest in the sale of his authorised editions,
but unfortunately no details of his publishing arrangements have
been preserved.
The history of the publishing work of Aldus, while not presenting
precedents for royalty or copyright arrangements, constitutes
nevertheless a very important chapter in the history of property in
literature. Aldus was able, by combining skilled editorial labour with
selected classics, to create a great literary property, which needed
only distributing machinery and a peaceable Europe to become
commercially valuable. He set the example also, for Italy at least, of
securing privileges in each of the Italian States possessing any
literary centres, and although he was not always able to prevent
piratical reprinting on the part of his competitors in Florence, or even
always to keep out of other cities in Italy the piracy editions from
Lyons, he accomplished something towards the ideal of a copyright
that should hold good for Italian territory. He even had hopes of
securing, through the authority of the Pope, a system of copyright
that should prove effective in all Catholic States, and it was not until
long after Aldus’s death that the attempts to establish a Catholic
copyright system were given up by publishers as practically futile.
His latest biographer, Didot, himself both a fine scholar and a great
publisher, contends that Aldus accomplished more than the greatest
scholars of his time for the spread of learning and the development
of literature; and the testimony of the three great scholars who were
contemporaries and near personal friends of the Venetian publisher,
Musurus, Reuchlin, and Erasmus, fully bears out M. Didot’s opinion.
It was the exceptional combination of a creative imagination and
scholarly knowledge with practical business ability and unfailing
pluck and persistency, that enabled the young tutor to create the
Aldine Press, the work of which will cause to be held in continued
honour, in the history alike of scholarship and of publishing, the
memory of Aldus Manutius.
The Successors of Aldus.—Paul Manutius, the son of
Aldus, continued for some years the business of the Aldine Press,
giving special attention to editions of the writings of Cicero. In 1561,
he accepted an invitation from Pope Pius IV. to come to Rome and
to take charge there of the publication of the writings of the Fathers
of the Church, and of such other works as might be selected. The
amount required for the organisation of an adequate printing-office
was to be supplied from the papal treasury. Paul was to receive an
annual stipend of 500 ducats, together with one half of the net profits
realised from the sales of the works published, and the contract was
to continue for twelve years.
An interesting series of letters has been preserved, written by Paul
to his brother Manutius in Asola, and to his son, Aldus the younger,
in Venice. These letters, which are quoted by Renouard, Frommann,
and Didot, contain a number of details and references which throw
light not only upon the personal relations of the writers, but upon the
business conditions of the time. We learn that Paul was a good deal
of an invalid throughout his working years, and we gather the
impression that his feeble health was an important ground for the
apparent lack of ambition which made him willing to give up his work
as an independent publisher in Venice and to accept the position of
Pope’s printer in Rome.
We also learn that his son Aldus, while bright-witted, was lacking
in persistency and in industry. The youngster never, in fact,
accomplished anything of importance. Paul had himself inherited the
scholarly tastes of his father, and had received a good classical
education, but he does not appear to have possessed very good
business faculty, and he made no distinctive mark as a publisher.
The Pope had, however, asked for his aid rather as a scholarly editor
than as an experienced man of business.
Pius appears to have been impressed with the belief that the
printing-press, under scholarly management, could be made of
service to the cause of the Church in withstanding the pernicious
influence of the increasing mass of the publications of the German
heretics. These Protestant pamphlets and books were not merely
undermining the authority of the Church in Germany, Switzerland,
and France, but were even making their way into Italy itself. The first
issues of the Aldine Press in Rome were the Decrees of the Council
of Trent, in a variety of editions, the writings of Cyprian, and the
letters of S. Jerome.
Pius V., who in 1565 succeeded Pius IV., was equally favourable
to the undertakings of the printing-office, and gave to Paul the
necessary support. The work was carried on in a building which was
the property of the municipality, and some issues arose with the
magistrates concerning its continued use as a printing-office. From a
letter dated September 27, 1567, it appears that the magistrates had
required that Paul should pay taxes or license-fees on his printing
business, which they classed as a trade. He took the ground that
printing was not a trade but an art, and that it was so defined in the
invitation given to him to come to Rome, and in the agreement
executed with him by the Pope. He contended, further, that, as the
Pope’s printer, whose work was devoted to the Church, he was in
any case entitled to exemption from the municipal taxes imposed on
traders. The Pope does not appear to have fully backed up his
printer in this contention, and a compromise was finally arrived at
under which a portion of the proceeds of the business was paid to
the magistracy. The precise terms of the arrangement are not clearly
stated, but it seems probable that the half share of the profits
previously payable to the papal treasury was divided into two
portions, one of which went to the municipality.
The profitable part of the business was in the printing of the official
editions of the Catechisms and Breviaries. Paul complains, in fact,
that the presses are so occupied with the work of the Breviaries, that
he is not able to make progress with the printing of his own
Commentaries on the Letters of Cicero. In June, 1568, Paul writes to
his son Aldus, who was now of age, expressing his regret that the
young man was not interested in devoting himself to carrying on the
printing-office in Venice. Aldus had, it seems, expressed a
preference for the study of law. The business in Venice was finally
turned over to Basa, who paid, for a term of five years, twenty scudi
gold a month for the use of the existing material and for the good-
will.
In July, 1569, difficulties began to accumulate about the printing-
office in Rome. The Pope was less interested and the magistrates
were troubling the office with what Paul calls unintelligent
interference. There were, in fact, too many parties interested in the
management of the business to enable its control to be easily or
consistently exercised. Paul’s health was also failing seriously and
he was longing for rest and for leisure to carry on his scholarly
undertakings. In 1570, the ownership of the receipts of the printing-
office was somewhat simplified, the change being probably due, in
part at least, to the representations of Paul that the many-headed
control was unworkable.
In May, 1570, Paul writes rather pathetically to Aldus: “In my case,
scholarship and industry have never brought rest or fortune.... I pray
God that you may be better favoured.... I must beseech you,
however, to put away childish things. It is full time that you recalled to
yourself the honourable traditions of our family.... My own active
work must be nearly over.”
In June, of the same year, he again counsels Aldus, who had for
some time been betrothed, to make a speedy marriage, and then to
concentrate himself upon the work of the printing-office in Venice. He
advises against a a plan that the young man had in view, of opening
a retail book-shop. He emphasises, however, that there is no chance
of success for a printer-publisher without the most persistent and
arduous labour.
In 1571, Paul’s failing strength compelled him to leave Rome,
resigning (as he hoped, for a time only) the income of the papal
printing-office. He devoted the winter months to the completion of his
Commentaries on the Orations of Cicero. The work was published in
1578-9 (after the author’s death) by his son Aldus in Venice, and,
under arrangement, by Plantin in Antwerp. The negotiations with
Plantin had been completed by Paul. He had specified the form and
style of the Antwerp edition, and had arranged to take his share of
the profits in the shape of a royalty on the sales.
In 1572, Paul being yet in Milan, one of his hopes was fulfilled in
the marriage of his son Aldus. “Now,” he wrote, “I can pass my days
in peace. I feel hopeful for your future and rejoice that our line is to
be continued.” Later in the year, with no little difficulty (partly on the
ground of his feeble health, and partly because of the floods and
wretched roads) he made his way to Venice for a brief visit. He
wanted to see his son’s wife, and he desired also to give personal
instructions for the printing of his Commentaries. “I feel very
hopeful,” he writes, “concerning the sale of my Cicero, and hopeful
also that it will not be reprinted (in piracy editions) during my
lifetime.”
Paul was obliged to leave Venice before the printing of his work
was begun, and the letter written after the receipt of the first sheets
expresses his bitter disappointment at the manner in which this all-
important commission had been attended to. “If you had had in your
hands some utterly contemptible scribble,” he writes, “you could
hardly have printed it in a more tasteless and slovenly style ... and
you knew I had this undertaking so much at heart!... I have instructed
Basa to burn all the sheets that have been printed, and to print these
signatures again, with a proper selection of type and on decent
paper.”
Aldus the younger seems never to have had his heart fairly in his
business, and under his management (or lack of management), the
prestige of the Aldine Press in Venice fell off sadly. He appears to
have been extravagant, or at least uncalculating, in his expenditures,
and was also spending moneys which he could ill afford, not like his
grandfather for manuscripts and type, but for clothes and artistic
curiosities.
Paul had accepted the pressing invitation of the new Pope,
Gregory XII.; to resume his place as manager of the printing-office in
Rome, but with less exacting duties, and with a fixed salary. A plan
was even talked over between the Pope and Paul for the
establishment of another printing-office, which should be devoted
entirely to the publication of classical works and of “expurgated”
editions of works, portions of which had been condemned in the
Index. Paul was to act as editor and supervisor of the series,
because his name was already recognised as that of a scholarly
authority. The scheme never, however, took shape. Paul’s strength
failed rapidly, and he died in the spring of 1574.
While he had devoted many years to his business as a printer-
publisher, and had maintained the reputation of his name for a high
standard as well of typography as of scholarly writing, his own
preference had been for a scholarly rather than a business career.
He went on with the work of his Press very largely because he felt
that it was a duty he owed to his father’s name and memory. His own
memory is, however, chiefly to be honoured for his scholarly edition
of Cicero, with its comprehensive and analytical commentaries, an
edition which long remained the accepted authority for Europe.
A few years after the death of Paul, his son Aldus gave up the
attempt to carry on the Press in Venice, a work for which he had
never been really fitted, and accepted a position in the University of
Bologna, as professor of archæology. The printing business was
sold, and the Aldine Press, after a century of work, came to an end.
Milan.—During the fifteenth century, Italy presents a curiously
complex and varied series of pictures and conditions. We find,
together with constantly recurring civil strife, successive wars of
invasion from the North and from the East, and in the train of the
frequent armies, those inevitable camp followers, pestilence, famine,
and misery. To the contests against the French and German
invaders and the strifes between states and cities, were added
schism and discord in the Church itself, and there were long periods
during which pope was contending against anti-pope for the right to
rule the world as the infallible head of an infallible church. Yet these
years, when the land was troubled by schism and devastated by
strife and pestilence, were years during which the cities of Italy were
becoming rich with an active and prosperous trade; while it was also
at this time that the art of Italy brought forth its greatest production
and that the development of its literature made most important
advances. The vitality of the people was so exuberant, its productive
force so enormous, that notwithstanding the frightful waste caused
by war and pestilence, its energies were still sufficient for some of
the greatest of artistic creations, for active and scholarly work in the
new learning and literature, and for a sharp competition for the
leadership of the world’s commerce and industries. A typical
example of the life and strife of the time is afforded by Milan, the
capital of Lombardy. Its position as the northernmost of the great
cities and in the centre of the open territory of the plains, exposed it
to the first attacks of invaders from across the Alps, while the
ambition of the rulers and of the people kept it in frequent strife with
its Italian rivals. Its trade seems to have continued active, however,
(except when armies were actually at its gates) and while in art more
important work was done in Florence, the first steps in the new
literature, that is, in the literature connected with printing, were taken
in Lombardy.
The first printing in Milan was done in 1469 by Philip of Lavagna,
who was followed in 1470 by Antonio Zarotus. In the printing of
books Milan holds precedence, therefore, over all the towns of Italy
except Subiaco and Rome, antedating Venice by about a year. The
publishing undertakings of the Lombardy capital never, however,
rivalled in importance those of Venice. In 1476, Paravisinus, printed
an edition of the Greek Grammar of Laskaris, the first volume printed
in Europe in Greek characters. In the previous volumes containing
Greek text, this had been printed in Latin characters. The editor of
the Grammar was Demetrius, a refugee from Crete. He was also the
editor of the first edition in Greek of Homer. The first Missal was
printed by Zarotus in 1475.
While in Rome the work of printing was begun by a German and in
Venice by a Frenchman, the first printers in Milan were native
Italians. Among the earlier of the Lombard printer-publishers, we find
the name of Alexander Minutianus, a learned professor, who devoted
himself to the editing of a valuable series of Latin classics, and
whose publishing activities extended over a term of twenty years.
Minutianus published in 1498-99, in four folio volumes, the first
complete edition of Cicero. The relations of Milan with the cities north
of the Alps were more intimate at this time than those of any other
Italian city, and it was natural, therefore, that as the printing business
in Lombardy increased in importance, German printers should begin
to seek employment there. The first whose name is recorded was
Waldorfer (or Valdarfer) from Regensburg, whose work began in
1474, and who brought with him fonts of Gothic type. Waldorfer
printed an edition of Pliny’s Letters and a selection of the Orations of
Cicero. These were followed by the Commentary of Servius on Virgil,
and by the first issue in print of the famous Decameron of Boccaccio.
The Decameron had been written in 1353, and had, therefore,
waited 120 years for a publisher. In 1493, Henricus Germanus and
Sebastian Pontremulo printed the first Greek edition of Isocrates. In
Milan, however, work in law, science, and medicine constituted a
more important proportion of the earlier publications than in Venice
or in Rome. The De Honate Brothers were printing as early as 1472,
works in jurisprudence, and Frommann is of opinion that before 1480
several firms were devoting their presses exclusively to the
departments of law and science. In 1472, a company was formed for
the printing and publishing of books, probably the first publishing
association in existence. There were at first five members or
associates, as follows:
Antonio Zarotus, a printer from Parma; Gabriel degli Orsoni, a
priest; Colla Montana, an instructor in the High School (he was
concerned some years later in the murder of the Duke Galeazzo
Maria); Pavero de’ Fontana, a professor of Latin, afterwards editor of
Horace; and Pedro Antonio de’ Burgo, of Castiglione, a lawyer.
Subsequently a sixth associate was added, Nicolao, a physician and
a brother of the last named.
The Association was organised for a term of three years and its
purpose was stated to be the instituting of a printing-office, with not
less than four presses, and the carrying on of a book-manufacturing
and publishing business. The capital was to be contributed in equal
shares by four of the associates, the printer, Zarotus, investing no
money, but contributing his knowledge of the business and
undertaking its general management. The printer was to receive one
third of the net proceeds, and the remaining two thirds were to be
divided equally among his four associates. From the printer’s share
were to be repaid the first expenditures contributed by the other four.
The subsequent expenditures were to be met by the sales of the
books. The person acting as corrector for the press, usually one of
the scholarly associates, secured as his compensation one or two
copies of the work corrected.
The selection of the books to be printed was to be made by the
unanimous decision of the whole board, and the selling price was
also to be fixed by the board. The organisation was to remain secret,
and all employees were to take an oath of secrecy and obedience.
Each member bound himself to give no council or aid to any other
publishing concern and to print no work with another printer except
under the permission of his associates. At the termination of the
agreement, the printer was to have a right to purchase at a valuation
the presses and the manuscripts.
The capitalist of the concern was the lawyer Antonio de’ Burgo,
and he found the funds (100 ducats) with which the first operations
were initiated. Under a supplementary agreement, the lawyer Burgo
and his brother the physician assumed for their individual account
one half of the rent of the premises and purchased three additional
presses. These presses were kept at work exclusively in the
production of a series of works in the departments of law and
medicine. The printer Zarotus took charge of the manufacture of
these books for the brothers Burgo, in addition to those printed for
the Association. The editorial work in selecting the material and in
preparing them for the press was cared for by the Burgos, who also
appear to have attended to the publishing details.
The brothers paid over to the treasury of the Association twenty-
five ducats for the use of the plant (type, etc.) outside of the presses,
and were to pay also one fourth of the proceeds of the sales of their
series. Each associate was also to receive a copy of each book
printed.
The brothers agreed to print no books excepting in the
departments of canon and civil law and of medicine, and the
Association was to include in its list no works in these departments.
The penalty for infringing this provision was fixed at 200 ducats.
The brothers were not at liberty to dispose of their portion of the
printing-office to any other parties. At the end of three years, the
presses and publications belonging to the two Burgos were
transferred, on an appraisal, to Zarotus.
No records have been preserved of the results of their
undertakings, or of those of the Association as a whole. The fact,
however, that as early as 1472, only eight years after the introduction
of printing into Italy, there should have been sufficient business, or
even expectation of business, to warrant the organisation of such a
publishing company, is certainly noteworthy, if only as evidence of
the intellectual activity and business enterprise of the Italy of the
fifteenth century. It is curious also that special provision should have
been made for legal and medical publications, as the literary
interests of the period of the Renaissance, which had so much
influence in furthering the activities of the earlier Italian printers, were
so largely classical.
It was necessary for the first publishers to be both printers and
scholars, and this necessary condition of early publishing
undertakings, the association of adequate scholarship with technical
knowledge required for the making of books, was fully provided for in
the Milan company, which included, as we have seen, two classical
professors, one theologian, one jurist, and one physician.
More than a century later, in 1589, was organised the Guild of the
Printers, Publishers, and Booksellers of Milan. During the hundred
years that had passed since the printing-press began its work in
Lombardy, the city had known various rulers, and had, for a brief
term, enjoyed independence. By far the larger portion of the century
had been for Lombardy periods of turmoil, and the years of
uninterrupted peace had been few. It was, therefore, not surprising
that the business of the production of books had developed more
rapidly and more prosperously in Venice, Rome, and Bologna, which
were from their position better protected against the mischances of
war.
In 1589, Lombardy was a portion of the great Spanish Empire, and
(as it contained few heretics) it was enjoying under the rule of Philip
II., a period of peace and of comparative prosperity. The charter of
the Guild or Corporation of the Printers and Publishers was

Download pdf Graph Theory And Its Applications Third Edition Jonathan L Gross ebook full chapter

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Download pdf Graph Theory And Its Applications Third Edition Jonathan L Gross ebook full chapter

Uploaded by

Copyright:

Available Formats

Graph Theory and Its Applications,

Third Edition Jonathan L. Gross

50 years of Combinatorics Graph Theory and Computing

Biota Grow 2C gather 2C cook Loucas

Graph Theory Applications to Deregulated Power Systems

Graph Theory and Decomposition 1st Edition Kottarathil

Fuzzy Graph Theory 1st Edition Sunil Mathew

Double Hierarchy Linguistic Term Set and Its

Basic Graph Theory 1st Edition Md. Saidur Rahman

Introduction to Chemical Graph Theory 1st Edition

“K23723˙FM” — 2018/10/11 — 14:15 — page 2 — #2

Graph Theory and Its

“K23723˙FM” — 2018/10/11 — 14:15 — page 4 — #4

A CONCRETE INTRODUCTION TO REAL ANALYSIS, SECOND EDITION

MATHEMATICAL MODELING FOR BUSINESS ANALYTICS

ELEMENTARY LINEAR ALGEBRA

James R. Kirkwood and Bessie H. Kirkwood

APPLIED FUNCTIONAL ANALYSIS, THIRD EDITION

J. Tinsley Oden and Leszek Demkowicz

AN INTRODUCTION TO NUMBER THEORY WITH CRYPTOGRAPHY, SECOND EDITION

James R. Kraft and Lawrence Washington

MATHEMATICAL MODELING: BRANCHING BEYOND CALCULUS

Crista Arangala, Nicolas S. Luke and Karen A. Yokley

ELEMENTARY DIFFERENTIAL EQUATIONS, SECOND EDITION

ELEMENTARY INTRODUCTION TO THE LEBESGUE INTEGRAL

LINEAR METHODS FOR THE LIBERAL ARTS

David Hecker and Stephen Andrilli

CRYPTOGRAPHY: THEORY AND PRACTICE, FOURTH EDITION

Douglas R. Stinson and Maura B. Paterson

DISCRETE MATHEMATICS WITH DUCKS, SECOND EDITION

BUSINESS PROCESS MODELING, SIMULATION AND DESIGN, THIRD EDITION

Manual Laguna and Johan Marklund

GRAPH THEORY AND ITS APPLICATIONS, THIRD EDITION

Jonathan L. Gross, Jay Yellen and Mark Anderson

“K23723˙FM” — 2018/10/11 — 14:15 — page 6 — #6

Graph Theory and Its

“K23723˙FM” — 2018/10/11 — 14:15 — page 8 — #8

© 2019 by Taylor & Francis Group, LLC

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-4822-4948-4 (Hardback)

Visit the Taylor & Francis Web site at

and the CRC Press Web site at

“Current˙Book” — 2018/10/16 — 17:36 — page v — #2

Jonathan dedicates this book to Alisa.

Jay dedicates this book to the memory of his brother Marty.

Mark dedicates this book to Libertad.

“Current˙Book” — 2018/10/16 — 17:36 — page vi — #3

“Current˙Book” — 2018/10/16 — 17:36 — page vii — #4

1 INTRODUCTION TO GRAPH MODELS 1

2 STRUCTURE AND REPRESENTATION 59

4 SPANNING TREES 173

“Current˙Book” — 2018/10/16 — 17:36 — page viii — #5

4.6 Graphs and Vector Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

6 OPTIMAL GRAPH TRAVERSALS 259

7 PLANARITY AND KURATOWSKI’S THEOREM 297

8 GRAPH COLORINGS 351

9 SPECIAL DIGRAPH MODELS 399

10 NETWORK FLOWS AND APPLICATIONS 441

“Current˙Book” — 2018/10/16 — 17:36 — page ix — #6

10.3 Flows and Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460

11 GRAPH COLORINGS AND SYMMETRY 487

SOLUTIONS AND HINTS 523

INDEX OF APPLICATIONS 563