Integer Sequences Divisibility Lucas

and Lehmer Sequences Masum Billal

Samin Riasat
Visit to download the full and correct content document:
https://ebookmeta.com/product/integer-sequences-divisibility-lucas-and-lehmer-seque
nces-masum-billal-samin-riasat/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Integer Sequences Divisibility Lucas and Lehmer

Sequences 1st Edition Masum Billal

https://ebookmeta.com/product/integer-sequences-divisibility-
lucas-and-lehmer-sequences-1st-edition-masum-billal/

Theory of Infinite Sequences and Series 1st Edition

Andrei Bourchtein

https://ebookmeta.com/product/theory-of-infinite-sequences-and-
series-1st-edition-andrei-bourchtein/

Function and Evolution of Repeated DNA Sequences 1st

Edition Guy-Franck Richard

https://ebookmeta.com/product/function-and-evolution-of-repeated-
dna-sequences-1st-edition-guy-franck-richard/

Bioinformatic and Statistical Analysis of Microbiome

Data: From Raw Sequences to Advanced Modeling with
QIIME 2 and R Yinglin Xia

https://ebookmeta.com/product/bioinformatic-and-statistical-
analysis-of-microbiome-data-from-raw-sequences-to-advanced-
modeling-with-qiime-2-and-r-yinglin-xia/
Bioinformatic and Statistical Analysis of Microbiome
Data From Raw Sequences to Advanced Modeling with QIIME
2 and R Yinglin Xia Jun Sun

https://ebookmeta.com/product/bioinformatic-and-statistical-
analysis-of-microbiome-data-from-raw-sequences-to-advanced-
modeling-with-qiime-2-and-r-yinglin-xia-jun-sun/

Behavior and Culture in One Dimension Sequences

Affordances and the Evolution of Complexity Resources
for Ecological Psychology Series 1st Edition Dennis P.
Waters
https://ebookmeta.com/product/behavior-and-culture-in-one-
dimension-sequences-affordances-and-the-evolution-of-complexity-
resources-for-ecological-psychology-series-1st-edition-dennis-p-
waters/

The Prevention and Management of Violence (2nd Edition)

Masum Khwaja

https://ebookmeta.com/product/the-prevention-and-management-of-
violence-2nd-edition-masum-khwaja/

Integer Optimization and its Computation in Emergency

Management 1st Edition Zhengtian Wu

https://ebookmeta.com/product/integer-optimization-and-its-
computation-in-emergency-management-1st-edition-zhengtian-wu/

Fuzzy Data Matching with SQL: Enhancing Data Quality

and Query Performance 1st Edition Jim Lehmer

https://ebookmeta.com/product/fuzzy-data-matching-with-sql-
enhancing-data-quality-and-query-performance-1st-edition-jim-
lehmer/
Masum Billal
Samin Riasat

Integer
Sequences
Divisibility, Lucas and Lehmer Sequences
Integer Sequences
Masum Billal · Samin Riasat

Integer Sequences
Divisibility, Lucas and Lehmer Sequences
Masum Billal Samin Riasat
Dhaka, Bangladesh Toronto, ON, Canada

ISBN 978-981-16-0569-7 ISBN 978-981-16-0570-3 (eBook)

https://doi.org/10.1007/978-981-16-0570-3

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Singapore Pte Ltd. 2021
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Dedicated to Morgan Ward
Preface

It gives us great pleasure that Springer Nature is publishing this book. Our initial
purpose was to popularize the theories in this text within the mathematical olympiad
community seeing how useful they are for solving many olympiad problems.
However, we soon realized that this alone cannot be the sole purpose of the book as
it covers areas that can also be classified as research material. We hope that the book
will be able to help people on both fronts.
The Fibonacci sequence is one of the most popular topics of study in math-
ematics. Lucas sequences are a generalization of the Fibonacci sequence. In this
book, we discuss Lucas sequences and their generalizations: Lehmer sequences,
divisibility sequences, generalized binomial coefficients, and integer sequences with
exponent lifting property. Our goal is to discuss theories that can be derived using
mostly elementary means. Some notions have been borrowed from linear and abstract
algebra, but they have been kept in check. We will not be using any particular idea that
screams abstract algebra. Rather, we will use some basic notions to help develop our
theories. Most results in this book are not usually found in common number theory
texts. The theories discussed in this book are very interesting yet not as popular as
one would hope. Considering the impact and usefulness of these theories, it seems
unfortunate that theories like divisibility sequences are not part of regular studies
and olympiad mathematics.
We had a very hard time finding proper references for some topics. For example,
double modulus was found only in Venkov [2]. Fortunately, an English translation
of the book was available, see Venkov [3]. Considering that the theory of double
modulus has been known since the time of Dedekind, it is a shame that almost no
common elementary text today covers this topic. Another gem we want to mention is
the Schönemann theorems. We have yet to find proper references for Schönemann’s
theorems, especially his second theorem. The only references we could find of these
theorems are in Ward’s papers, e.g., Ward [4]. Apparently, Ward was a big fan of
these theorems and often applied them in his work. Moreover, judging by papers
of Engstrom [1] and others we are convinced they too knew these results well. For
example, Engstrom [1] uses a factorization similar to that of Schönemann’s in his
paper. But today there is no record of these results either online or in any common
books that we know of.
vii
viii Preface

This book is dedicated to Morgan Ward due to his enormous contribution to

the subject. Ward wrote 33 papers on recurring series, which is closely related to
both divisibility sequences and Lucas sequences. We do not know for sure what
the motivation was for Ward to study these sequences in such detail. For the first
author, the initial motivation to study these topics was to find out when an integer
sequence has integer binomial coefficients. For the second author, they were the many
interesting divisibility properties of Fibonacci-like sequences. Ward’s motivation
might have been similar. A lot of the results and papers referenced are available only
because of Ward. Although many original proofs have been replaced by modern
notions and notations, it seems quite fitting that this book be dedicated to him.
Finally, we would like to express our gratitude to the kind referees whose opinions
helped improve the original manuscript significantly. Nevertheless, in the case of any
typo or error in results, the authors bear full responsibility and request to be contacted
via email at billalmasum93@gmail.com or sriasat@uwaterloo.ca.

Dhaka, Bangladesh Masum Billal

Toronto, Canada Samin Riasat
April 2021

References

1. H.T. Engstrom. Periodicity in Sequences Defined by Linear Recurrence Relations. in Proceed-

ings of the National Academy of Sciences 16.10 (1930), pp. 663–665. https://doi.org/10.1073/
pnas.16.10.663.
2. B.A. Venkov. Elementarnaia Teoriia Chisel. The main edition of general technical and techno-
theoretical literary, 1937.
3. B.A. Venkov. Elementary Number Theory. Wolters-Noordhoff, 1970.
4. Morgan Ward. The Arithmetical Theory of Linear Recurring Series. Trans. Amer. Math. Soc.
35.3 (1933), pp. 600–628. ISSN: 00029947. http://www.jstor.org/stable/1989851.
Who This Book Is For

This book is written for pretty much anybody interested in Fibonacci-like sequences.
We have tried to write the book in a way that not only undergraduates but people
with a focus on mathematical olympiads can also read the book with little effort.
As for pre-requisites, we have tried our best to keep the book self-contained and
as elementary as possible. We have even made an extra effort to use things like the
double modulus which one could argue that should be replaced by ideals. Although
we do discuss ideals as part of the prologue to the actual discussion, we have kept
the usage of anything abstract to a bare minimum and used them only if they were
absolutely necessary. The thought of writing the whole book based on elementary
methods only did cross our minds. However, writing everything from scratch and
redeveloping known theory does not really favor the subject at hand in the long run.
That being said, we used little abstract algebra most of which is only for the first
two chapters. The rest of the chapters are almost free from abstract algebra concepts.
Our purpose is to not only let an undergraduate student study these topics but also
show an olympiad-level student how useful such theory can be in many cases. For
this purpose, we have also added some exercises that appeared in various popular
mathematical contests.

ix
Contents

1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Groups, Rings, Fields, and Vector Spaces . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Rings and Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.3 Vector Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Cyclotomic Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2 Linear Recurrent Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2 Periodicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.3 Fundamental Theorems on Periodicity . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.4 Finding Pre-Period and Least Period . . . . . . . . . . . . . . . . . . . . . . . . . . 43
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3 Divisibility Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.2 Characterization of Divisibility Sequences . . . . . . . . . . . . . . . . . . . . . 61
3.3 Binomial Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4 Lucas Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.2 Divisibility of Second-Order Lucas Sequences . . . . . . . . . . . . . . . . . . 74
4.3 Lucas Sequence of Arbitrary Order . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5 Lehmer Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.2 Divisibility of Lehmer Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.3 Periodicity of Lehmer Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.4 Extending Wilson’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
xi
xii Contents

5.5 LCM Sequence of Lehmer Sequences . . . . . . . . . . . . . . . . . . . . . . . . . 111

5.6 Subsequence of Lehmer Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
6 On Primitive Divisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.1 The History of Primitive Divisors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.2 Primitive Divisors of a n − bn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.3 Primitive Divisors of Real Lucas Sequences . . . . . . . . . . . . . . . . . . . . 128
6.4 Primitive Divisors of Real Lehmer Sequences . . . . . . . . . . . . . . . . . . 133
6.5 Primitive Divisors of Complex Lehmer Sequences . . . . . . . . . . . . . . 138
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
About the Authors

Masum Billal is working as a senior data scientist in Dhaka, Bangladesh. He has

authored a book on number theory and he has been a trainer at math camps of
Bangladesh for 5/6 years. His research interests include mathematics (specially
number theory), machine learning, cryptography, etc.

Samin Riasat has completed his Master of Mathematics in Computer Science at the
University of Waterloo, Canada. He has authored a book on mathematical olympiads
and his research articles have been published in journals of repute. He is also a
reviewer for Mathematical Reviews.

xiii
Chapter 1
Preliminaries

In this chapter, we discuss some topics from algebra that are prerequisites to the theory
we will develop. First, we discuss groups, rings, fields, vector spaces, and matrices
very briefly in Sects. 1.1 and 1.2. Then, we discuss properties of polynomials in
Sect. 1.3 and cyclotomic polynomials in particular in Sect. 1.4. Later, we will talk
about recurrent sequences (mostly linear recurrent sequences over fields) in Chap. 2.
We will talk about divisibility sequences and explore their connections to binomial
and related sequences in Chap. 3. We will use the theory developed in these chapters
to characterize Lucas sequences in Chap. 4 and Lehmer sequences in Chap. 5. Finally,
we will discuss special divisors of such sequences, particularly primitive divisors, in
Chap. 6.
We will be using mostly elementary methods in this book. Occasionally we will
require some advanced theory. However, our use of abstract theory will be minimal,
to the point where we can even claim that it is still elementary. So a reader completely
unfamiliar with abstract algebra should not feel overwhelmed . We could probably
avoid using abstract theory altogether. For example, it is possible to only develop the
theory of cyclotomic polynomials. However, doing so would make our discussion
too restrictive and very case specific when we can obtain far more general results
with the introduction of little theory. We initially wanted to avoid going into abstract
theory altogether but soon realized that the amount of effort this will require will not
be worth it.
To be more specific, we will need some basic ideas from linear and abstract
algebra. In order to keep the book self-contained, we will discuss these topics from
a beginner’s perspective. We will be using a standard notation common to most
texts for your convenience. You can consult texts like [Lan86, LN97] but we do not
require such heavy machinery. Therefore, if you are unfamiliar with these subjects,
you may find it easier to focus on what is presented in this book alone. We suggest
you to use examples to make sense of the definitions and ideas. Then, if you feel like
you want to dig deeper you can consult other texts. For example, linear algebra will
be required mostly for the notion of a basis. And abstract algebra will be required

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 1
M. Billal and S. Riasat, Integer Sequences,
https://doi.org/10.1007/978-981-16-0570-3_1
2 1 Preliminaries

for some properties of polynomials and matrices that would otherwise be difficult to
describe. You may be wondering why we might need these things in a number theory
text at all. To your surprise, you will see that recurrent sequences, polynomials, and
matrices are very much connected to one another. This will make more sense once
we develop the theories and show the connections. To give you an idea, most of
the results regarding primitive divisors in Chap. 6 will be derived using cyclotomic
polynomials.

1.1 Groups, Rings, Fields, and Vector Spaces

If you are completely new to this topic, do not be afraid. As long as you are familiar
with sets, you should be able to grasp the ideas in this section. We also urge you to
try to prove as many of the facts that we state as you can.
Vector spaces are just sets with some interesting properties. But before getting
into vector space, we will discuss some more basic algebraic structures. These will
help us define and develop a lot of properties with ease. First, we will talk about
groups, rings, and fields. Then we will discuss vector spaces.
We mentioned that groups and vector spaces are just sets with interesting proper-
ties. One property that is common in all such algebraic structures is closure. We will
be talking about binary operations defined on these sets. Such an operation takes two
elements of the set as input. Now, it makes sense to be interested in the case when
the element produced is again an element of the original set. This is why we call this
property closure.

1.1.1 Groups

A group G is a set equipped with a binary operation ◦ satisfying the following

properties. These properties are called group axioms.
(1) (Closure) For any x, y ∈ G, we have x ◦ y ∈ G.
(2) (Associativity) For any x, y, z ∈ G,

x ◦ (y ◦ z) = (x ◦ y) ◦ z

(3) (Identity) There is an element e ∈ G such that for any x ∈ G, we have x ◦ e =

e ◦ x = x. This element e is called the identity of G.
(4) (Inverses) For any x ∈ G, there is an element y ∈ G such that x ◦ y = e. This
element y is the inverse of x, usually denoted x −1 or −x.
Some consequences of these axioms are that the identity and the inverse of a group
element must be unique, which justifies the above notation.
1.1 Groups, Rings, Fields, and Vector Spaces 3

For example, the set Z of integers equipped with addition is a group. Observe
that 0 is the identity of Z, and −x is the inverse of x for any x ∈ Z. Note that
the set N of natural numbers is not a group under addition. Also, Z is not a group
under multiplication. The set {1, ω, ω 2 } where ω is an imaginary cubic root of unity
(ω 3 = 1) is a group under multiplication.
Note that a familiar property is missing from the list of properties of a group,
which is commutativity. If

x◦y = y◦x

for any x, y ∈ G, then G is called an abelian group. It is however true in general that

(x ◦ y)−1 = y −1 ◦ x −1 (1.1)

for any group G and x, y ∈ G. All the examples of groups we have mentioned so far
are abelian. An example of a group that is not abelian is the set of invertible matrices
under matrix multiplication. We will talk about this group near the end of Sect. 1.2.
Consider a group G with binary operation ◦. We write

x n = x ◦ ·
· · ◦ x, x 0 = e
n times

for a positive integer n, and define x −n = (x −1 )n . However, when ◦ represents addi-

tion (+) , it is more convenient to write x n as nx and to use 0 for the identity element,
so that 0x = 0.
A subset of a group G that is also a group under the operation of G is a subgroup
of G. The order of a group G is the number of elements |G| of G. For an element
a ∈ G, we write ord(a) for the smallest positive integer k such that a k = e. Observe
that the set a = {a i : 0 ≤ i < |G|} is a subgroup of G with ord(a) elements. More
generally, we define

a1 , . . . , an  = {a1i1 · · · anin : 0 ≤ i j < |G|, 1 ≤ j ≤ n}

If G = a1 , . . . , an , we call each ai a generator of G. We call G cyclic if G = a

for some a.
We give the following theorems without proof.
Theorem 1.1 Every subgroup of a cyclic group is cyclic.
Theorem 1.2 Let G = a be a cyclic group of order n. The order of the subgroup
a k  of G is gcd(n,k)
n
for any positive integer k.
Theorem 1.3 Let G = a be a cyclic group of order n. If d is a divisor of n, then
G has exactly ϕ(d) elements of order d, where ϕ is the Euler Totient Function.
Theorem 1.4 Let G = a be a cyclic group of order n. Then G = a i  for each i
with 1 ≤ i ≤ n and gcd(i, n) = 1.
4 1 Preliminaries

The following theorem of Lagrange is a fundamental result in group theory and

has a wide range of applications in all of mathematics. For instance, Fermat’s little
theorem and Euler’s theorem are immediate consequences of Lagrange’s theorem.
Theorem 1.5 (Lagrange’s Theorem) If H is a subgroup of G, then |H | divides |G|.
We will now generalize the notion of a group.

1.1.2 Rings and Fields

Consider a set S equipped with two operations + (called addition) and · (called
multiplication). We call S a ring if it satisfies the following properties.
(1) (Abelian Additive Group) S is an abelian group under + with identity 0.
(2) (Associativity of Multiplication) For any x, y, z ∈ S,

x · (y · z) = (x · y) · z

(3) (Identity of Multiplication) There is an element 1 ∈ S such that x · 1 = 1 · x = x

for any x ∈ S. We call 1 the multiplicative identity of S.
(4) (Distributivity of Multiplication over Addition) For any x, y, z ∈ S,

x · (y + z) = x · y + x · z
(y + z) · x = y · x + z · x

Note that two familiar properties are missing for multiplication: commutativity
and inverses. If S also has these properties, then S is called a field. That is, a field is
a ring with the following additional properties.
i. (Commutativity of Multiplication) For any x, y ∈ S,

x·y = y·x

ii. (Inverses for Multiplication) For any x ∈ S \ {0}, there is an element x −1 ∈ S

such that x · x −1 = x −1 · x = 1.
If S is a field satisfying the properties above, we call S a field under the operations
+ and ·. Notice that these are all familiar properties of real numbers. In fact, the
set R of real numbers and the set C of complex numbers are both fields under their
regular operations. The set Q of rational numbers likewise is also a field. However,
the set Z of integers and the set N of natural numbers are not fields under their regular
operations. If x is an element of Z, then −x is in Z but x1 is not in Z, since x1 is not
integer unless |x| = 1. And if x is an element of N, then −x is not always an element
of N.
1.1 Groups, Rings, Fields, and Vector Spaces 5

As we can see, a field is simply a set with some desired properties. More generally,
the objects studied in abstract algebra are√sets with special properties. For instance,
the set of all numbers of the form a + b 2, where a, b are rational √ numbers, is a
field. Similarly, the set of numbers of the form a + bi, where i = −1 and a, b are
rational numbers, is also a field.
The non-zero elements of a field form an abelian group under multiplication. We
write F ∗ for the set of non-zero elements of the field F and call F ∗ the multiplicative
group of F. We also write x y instead of x · y for convenience.
Theorem 1.6 If F is a field, then every finite subgroup of F ∗ is cyclic.
If F, K are fields under the same operations with F a subset of K , we call F a
subfield of K . If, in addition, F = K , then we call F a proper subfield of K . A field
that has no proper subfield is called a prime field. Observe that R and Q are subfields
of C.
For a positive integer n, let Fn be the set {0, 1, 2, . . . , n − 1} of residues modulo
n. Then Fn is a ring under addition modulo n and multiplication modulo n, for any
n. Moreover, F p is a prime field for any prime p.
Let R be a ring. If there is a positive integer n such that nx = 0 for any x ∈ R,
then the least such n is called the characteristic of the ring R. If no such n exists,
then we say R has characteristic 0. For example, F p has characteristic p, while R
and C both have characteristic 0. A domain is a ring with no zero divisor; that is, if
x y = 0 then either x = 0 or y = 0. If R is a domain with a non-zero characteristic,
then the characteristic of R must be a prime number. As a result, all finite fields have
a prime characteristic.
Theorem 1.7 Let R be a commutative ring of prime characteristic p. Then for any
a, b ∈ R and integer n ≥ 0,
n n n
(a + b) p = a p + b p

Proof We prove the result for n = 1 and leave the general case as an exercise. Recall
the binomial theorem (see Binomial Theorem in the glossary),
   
n n−1 n
(a + b)n = a n + a b + ··· + a 1 bn−1 + bn
1 n−1
p p  p−1 p
Now, since k
= k k−1
, it follows that p divides k
for 0 < k < p. Hence
   
p p−1 p
(a + b) = a +
p p
a b + ··· + ab p−1 + b p
1 1
= ap + bp

as desired. 
Let F be a field, K a subfield of F and S a subset of F. Consider all the subfields
of F that contain both K and S. The intersection of these subfields is also a field.
6 1 Preliminaries

We call this field the extension field of K adjoining S and write it as K (S). We
call K (S) an extension of K and if S is finite, a finite extension of K . Observe that
K (S) is the smallest subfield of F that contains both K and S. For a finite S, if
S = {s1 , s2 , . . . , sk }, then we write K (S) = K (s1 , s2 , . . . , sk ). If S = {s}, then we
call K (s) a simple extension.
A set I is called an ideal of a ring R if I ⊆ R and for all x ∈ R, a ∈ I , we have
xa ∈ I . For instance, note that Z is not an ideal of Q . For the ring Z, the subset I
which is the set of all multiples of some fixed integer n is an ideal of Z. In fact, the
smallest ideal containing a particular element x ∈ R is the set {xa : a ∈ R}. We call
this the ideal generated by a. An ideal I of R is called a principal ideal if is generated
by some a ∈ I . We call I the principal ideal generated by a, and write I = (a). A
domain R is called a principal ideal domain if every ideal of R is a principal ideal.
An ideal I of a ring R is called a prime ideal if x y ∈ I only if x ∈ I or y ∈ I for all
x, y ∈ R.

1.1.3 Vector Spaces

We now define a vector space V over a field F, which is a set with the following
properties.

(1) (Abelian Additive Group) V is an abelian group under + with identity 0.

(2) (Closure under Scalar Multiplication) If x is an element of V , then ax is an
element of V for any a in F.
(3) (Properties of Scalar Multiplication) If a, b are elements of F and x, y are ele-
ments of V , then

a(x + y) = ax + ay
(a + b)x = ax + bx
a(bx) = (ab)x
0x = 0
1x = x

We call a subset U of V a subspace of V if it is a subgroup of V and is closed

under scalar multiplication.
If F is a field, we denote by F n the set of all n-tuples (x1 , x2 , . . . , xn ) where
x1 , x2 , . . . , xn are elements of F. Moreover, if x = (x1 , x2 , . . . , xn ) and
y = (y1 , y2 , . . . , yn ), then we define

x + y = (x1 + y1 , x2 + y2 , . . . , xn + yn )
ax = (ax1 , ax2 , . . . , axn )
1.1 Groups, Rings, Fields, and Vector Spaces 7

for a in F. Then F n satisfies the properties of a vector space over F. In this case we
just say F n is a vector space and omit the mention of the base field. For example,
Rn , Cn , Qn are all vector spaces.
Theorem 1.8 If F is an extension of K , then F is a vector space over K .
If V is a vector space over a field F, then for x1 , x2 , . . . , xn ∈ F and vec-
tors v1 , v2 , . . . , vn ∈ V , we call v1 x1 + v2 x2 + · · · + vn xn a linear combination of
v1 , v2 , . . . , vn .
Theorem 1.9 For a vector space V over F and vectors v1 , v2 , . . . , vn ∈ V , the set
of all linear combinations of v1 , v2 , . . . , vn is a subspace of V .
We call the subspace of V in Theorem 1.9 the subspace generated by v1 , v2 , . . . , vn .
Consider a vector space V over F and a set B = {v1 , v2 , . . . , vn } of vectors in V .
We say that B is linearly dependent if

a1 v1 + a2 v2 + · · · + an vn = 0

for some a1 , a2 , . . . , an ∈ F not all 0. If such elements do not exist, then we say that
B is linearly independent. So, B is linearly independent if

a1 v1 + a2 v2 + · · · + an vn = 0

implies a1 = a2 = · · · = an = 0. On the other hand, if every v ∈ V can be written as

a linear combination of v1 , v2 , . . . , vn , we say B spans V . If B is linearly independent
and also spans V , then we call B a basis of V .
Theorem 1.10 Let F be a field. The set of vectors

v1 = (1, 0, 0, . . . , 0)
v2 = (0, 1, 0, . . . , 0)
..
.
vn = (0, 0, 0, . . . , 1)

in F n is a basis of F n .
A subset B of V is called maximally linearly independent if B is linearly inde-
pendent, but for any v ∈ V \ B, the set B ∪ {v} is linearly dependent. Likewise, a
subset B of V is called minimally spanning if B spans V , but for any b ∈ B, the set
B \ {b} does not span V .
Theorem 1.11 Let V be a vector space and B ⊆ V . Then the following are equiv-
alent.
1. B is a basis of V .
2. B is maximally linearly independent.
3. B is minimally spanning.
The dimension of a vector space V over a field F is the cardinality of a basis of
V over F, and this is well-defined.
8 1 Preliminaries

1.2 Matrices

Matrices are ubiquitous and an important object of study in mathematics. A matrix

A over a field F is a rectangular array of elements from F.
We write
⎛ ⎞
a1,1 a1,2 · · · a1,n
⎜ a2,1 a2,2 · · · a2,n ⎟
⎜ ⎟
A=⎜ . .. . . .. ⎟ ∈ F
m×n
⎝ . . . . . ⎠
am,1 am,2 · · · am,n

and call A an m × n matrix over F, where m and n are the number of rows and
columns, respectively, and are called the dimensions of A. The entry in the i-th row
and the j-th column is ai, j . We can also use the shorthand A = (ai, j ), 1 ≤ i ≤ m, 1 ≤
j ≤ n to represent A.
Observe that each row and each column of A represents a vector over F. For
instance, we can express the i-th row of A as the 1 × n matrix (or row vector)

Ai, = ai,1 ai,2 · · · ai,n

which represents (ai,1 , ai,2 , . . . , ai,n ) ∈ F n . Similarly, we can express the j-th col-
umn of A as the m × 1 matrix (or column vector)
⎛ ⎞
a1, j
⎜ a2, j ⎟
⎜ ⎟
A, j = ⎜ . ⎟
⎝ .. ⎠
am, j

which represents (a1, j , a2, j , . . . , am, j ) ∈ F m .

We can also talk about operations on matrices. Naturally, two matrices of the same
size can be added or subtracted coordinate-wise. That is, if A, B ∈ F m×n , we define
A ± B = (ai, j ± bi, j ). For example, if
⎛ ⎞
123
A = ⎝4 5 6⎠
789
⎛ ⎞
10 11 12
B = ⎝13 14 15⎠
16 17 18

then
1.2 Matrices 9
⎛ ⎞
11 13 15
A + B = ⎝17 19 21⎠
23 25 27
2A = A + A
⎛ ⎞
2 4 6
= ⎝ 8 10 12⎠
14 16 18
⎛ ⎞
876
B − 2 A = ⎝5 4 3⎠
210

etc. Notice that this makes F m×n into a group under +. The identity is the zero matrix
0, defined to be the matrix with all entries equal to 0. The (additive) inverse of A
is 0 − A, which we denote simply by −A. This also allows us to define equality of
A, B ∈ F m×n as

A = B if and only if A − B = 0 (1.2)

In general, we say two matrices are equal if they have the same dimensions and
satisfy 1.2.
For c ∈ F, we define c A = (cai, j ). This defines scalar multiplication on F n×n
over F and allows F m×n to be viewed as a vector space over F.
It is also possible in some sense to define multiplication and division for matri-
ces. Traditional matrix multiplication is defined as follows. The product AB of two
matrices A and B is well-defined if and only if the number of columns of A is equal
to the number of rows of B. In particular, if A is m × n and B is p × q, then AB is
well-defined if and only if n = p. In this case, the product C = AB = (ci, j ) is the
m × q matrix given by


n
ci, j = ai,k bk, j
k=1

It is important here to note that AB = B A in general. In other words, matrix multi-

plication in general is not commutative.
For instance, with A and B in the example above,
⎛ ⎞
1 · 10 + 2 · 13 + 3 · 16 1 · 11 + 2 · 14 + 3 · 17 1 · 12 + 2 · 15 + 3 · 18
AB = ⎝4 · 10 + 5 · 13 + 6 · 16 4 · 11 + 5 · 14 + 6 · 17 4 · 12 + 5 · 15 + 6 · 18⎠
7 · 10 + 8 · 13 + 7 · 10 7 · 11 + 8 · 14 + 9 · 17 7 · 12 + 8 · 15 + 9 · 18
⎛ ⎞
84 90 96
= ⎝201 216 231⎠
318 342 366
10 1 Preliminaries

In order to understand the motivation behind this mechanism, we look at systems

of linear equations. The simplest equations we deal with are usually linear equations
of the form

2x = 6 (1.3)

To solve (1.3), we divide both sides by 2 to obtain x = 3. Now, consider the system
of linear equations

2x + 3y = 5
x+y=2

It is routine to solve the system by hand. However, representing the system using
matrices as follows can give us new insight.
    
23 x 2x + 3y
=
11 y x+y

So the system of equations can be represented using matrices as

   
2x + 3y 5
=
x+y 2
    
23 x 5
=
11 y 2
    
A X B
AX = B (1.4)

We can see that (1.4) resembles (1.3). In order to solve (1.4) like (1.3) we wish to
“divide” both sides of (1.4) by A. Note that this is equivalent to multiplying both
sides of (1.4) on the left by a matrix A−1 that satisfies A−1 A = I . If successful, this
would then give X = A−1 B, as desired. Such a matrix A−1 would be called the (left)
inverse of A. Unfortunately, it does not always exist.

AX = B
(A )AX = A−1 B
−1

(A−1 A)X = A−1 B

I X = A−1 B
X = A−1 B

Also note that it was important to multiply on the left because, as noted earlier, matrix
multiplication is not commutative in general.
1.2 Matrices 11

Ideally, we would like to view F m×n as a ring. For this purpose it makes sense to
consider the case m = n in particular. We call an element of F n×n a square matrix.
So a matrix is square if it has an equal number of rows and columns. Let A = (ai, j ) ∈
F n×n be a square matrix. A diagonal matrix is a square matrix with all off-diagonal
entries equal to 0. Formally, A is diagonal if ai, j = 0 whenever i = j. If A is diagonal
with ai,i = 1 for all i, then A is called the identity matrix and is denoted In , or simply
I when the context is clear. So
⎛ ⎞
1 0 0 ··· 0
⎜0 1 0 · · · 0 ⎟
⎜ ⎟
⎜ ⎟
I = ⎜0 0 1 · · · 0 ⎟
⎜ .. .. .. . . .. ⎟
⎝. . . . .⎠
0 0 0 ··· 1

With addition, multiplication, the zero matrix and the identity matrix defined as
above, F n×n becomes a ring.
It is natural to ask when the multiplication operation defined above is reversible.
A square matrix A is said to be invertible or non-singular if there is a matrix A−1 ,
called the inverse of A, such that A−1 A = A A−1 = I . Otherwise we call A singular.
An example of a singular matrix is
 
12
B=
24

Note that B corresponds, for example, to the following system of equations:

x + 2y = 4
2x + 4y = 9

which can be seen to have no solution, in accordance with the fact that B is a singular.
We give a few more useful definitions.
Let A = (ai, j ) ∈ F m×n . The transpose of A is defined to be the matrix A T =
(a j,i ). When A = A T , we call A a symmetric matrix. When m = n, the determinant
of A is defined recursively as det(A) = a1,1 if n = 1, and if n > 1, then


n
det(A) = (−1) j−1 ai, j det(Mi, j )
j=1

for any i, where Mi, j is the (n − 1) × (n − 1) matrix obtained from A by deleting

its i-th row and j-th column. For example, if
 
ab
A=
cd
12 1 Preliminaries

then det(A) = ad − bc, and if

⎛ ⎞
a bc
B = ⎝p q r⎠
x yz

then
     
qr pr pq
det(B) = a det − b det + c det
yz x z x y
= a(qz − r y) − b( pz − r x) + c( py − q x)

etc. Observe also that det(I ) = 1.

We now state the following theorems without proof.

Theorem 1.12 If A, B ∈ F n×n , then (AB)T = B T A T .

Theorem 1.13 If A, B ∈ F n×n , then det(AB) = det(A) det(B).

Theorem 1.14 A square matrix is invertible if and only if its determinant is non-zero.

Theorem 1.15 The determinant of a diagonal matrix is equal to the product of its
diagonal entries. In particular, a diagonal matrix with non-zero diagonal entries is
invertible.

It follows from Theorems 1.13 and 1.14 that the product of two invertible matrices
A, B ∈ F n×n is again invertible, and by (1.1) that (AB)−1 = B −1 A−1 . In particular,
the invertible matrices in F n×n form a group under matrix multiplication, called the
general linear group GLn (F). It also follows from Theorem 1.13 that the matrices in
F n×n with determinant 1 form a group under matrix multiplication, called the special
linear group SLn (F). In fact, SLn (F) is a subgroup of GLn (F). Note that neither
GLn (F) nor SLn (F) is abelian for n > 1 if F contains more than one element.

1.3 Polynomials

Let R be a ring. A polynomial f over R in the indeterminate x is an expression of

the form

f (x) = ak x k + ak−1 x k−1 + · · · + a1 x + a0

where k ≥ 0 and, for each i, ai ∈ R is the coefficient of x i in f . If ak = 0, we call k

the degree of f and write deg( f ) = k. If deg( f ) > 0, we call f non-constant, and
call f constant otherwise. We say f is monic if ak = 1. We denote the set of all
polynomials over R as R[x].
1.3 Polynomials 13

As before,
m we wish to define
addition and multiplication for polynomials. Write
m
f (x) = i=0 ai x i and g(x) = i=0 bi x i with f, g ∈ R[x]. Then we define f + g
and f g via


m
f (x) + g(x) = (ai + bi )x i
i=0

m+n 
f (x)g(x) = ck x k where ck = ai b j
k=0 i+ j=k
0≤i≤m
0≤ j≤n

These make R[x] into a ring with 0 and 1 coming from R. Observe that if f g = 0,
then deg( f + g) ≤ max{deg( f ), deg(g)} and deg( f g) = deg( f ) + deg(g).
We will mostly consider polynomials over fields. The reason being if F is a field,
then F[x] is a principal ideal domain. Consequently, we have a division algorithm for
polynomials over a field. Henceforth, we write a | b (resp. a  b) to mean a divides
(resp. does not divide) b.

Theorem 1.16 Let F be a field and f, g ∈ F[x] with g non-constant. Then f =

gq + r for unique polynomials q, r ∈ F[x] with r = 0 or deg(r ) < deg(g).

A non-constant polynomial f ∈ F[x] is reducible if f = gh for some non-

constant polynomials g, h ∈ F[x]. Otherwise we call f irreducible. For example,
any polynomial f ∈ F[x] with deg( f ) = 1 is irreducible. On the other hand, x 2 + 1
is irreducible in Z[x] but is reducible in C[x] as x 2 + 1 = (x + i)(x − i). Like prime
numbers, if an irreducible polynomial f ∈ F[x] divides a product f 1 f 2 · · · f n with
each f i ∈ F[x], then f divides f i for some 1 ≤ i ≤ n.
As a consequence of Theorem 1.16, we also obtain the following result.
Theorem 1.17 If F is a field, then F[x] is a principal ideal domain, and every ideal
of F[x] is of the form ( f ) for some monic polynomial f ∈ F[x]. In particular, every
f ∈ F[x] has a unique factorization of the form

f = c f 1e1 f 2e2 · · · f mem

where c ∈ F and the f i are distinct monic irreducible polynomials in F[x].

Theorem 1.17 allows us to view polynomials over a field just like integers with
unique factorization, enabling us to talk about their greatest common divisor (gcd).

Theorem 1.18 Let f ∈ F p [x]. Then f (x) p = f (x p ).

Proof Apply Theorem 1.7. 

An element α ∈ F is called a root (sometimes called a zero) of f ∈ F[x] if

f (α) = 0. Here f (α) is calculated by substituting α for x in f (x).
14 1 Preliminaries

Theorem 1.19 (Remainder Theorem) Let F be a field with f ∈ F[x] and α ∈ F.

The remainder when f is divided by x − α is f (α).

Proof By Theorem 1.16, there exist q ∈ F[x] and r ∈ F such that f (x) = (x −
α)q(x) + r . Substituting α for x gives f (α) = r , as desired. 

Theorem 1.20 (Factor Theorem) Let F be a field with f ∈ F[x] and α ∈ F. Then
x − α divides f if and only if f (α) = 0.

Proof This follows immediately from Theorem 1.19. 

If α is a root of f , then the largest positive integer k for which (x − α)k divides
f is called the multiplicity of α. If k > 1, then we say that α is a multiple root. If
f does not have a multiple root, then f is said to be square-free. Observe that a
polynomial f in F[x] can have at most deg( f ) roots in F. Observe also that x − α
for any α ∈ F is irreducible and square-free.
Let f (x) = ak x k + · · · + a1 x + a0 ∈ F[x]. The derivative of f is the polynomial

f (x) = kak x k−1 + · · · + 2a2 x + a1 ∈ F[x]

Theorem 1.21 Let F be a field and f ∈ F[x]. If f = c f 1e1 f 2e2 · · · f mem with c ∈ F
and the f i distinct monic irreducible polynomials in F[x], then


m
f = a f 1e1 −1 f 2e2 −1 · · · f mem −1 ei f 1 f 2 · · · f i−1 f i+1 · · · f m
i=1

Since f 1e1 −1 f 2e2 −1 · · · f mem −1 divides both f and f , it follows immediately from
Theorem 1.21 that α is a multiple root of f if and only if α is also a root of f . In
particular, we have the following.

Theorem 1.22 Let F be a field and f ∈ F[x]. Then f is square-free if and only if
gcd( f, f ) = 1.

An immediate consequence is the following interesting result.

Theorem 1.23 Let p be a prime number. If there is an integer α such that x n − 1 =

(x − α)2 g(x) in F p [x], then p | n. That is, x n − 1 has a multiple root modulo p if
and only if p | n.

Proof Since f (x) = x n − 1 ∈ F p [x] is not square-free, it follows that p | gcd(x n −

1, nx n−1 ). Therefore p | n. 

For a polynomial f ∈ F[x], the set F[x]/( f ) consists of the residue classes
modulo f , written g + ( f ) or g (mod f ) or [g], for g ∈ F[x]. Two polynomials
g, h ∈ F[x] are said to be in the same residue class if and only if g − h ∈ ( f ). In
this case we write g + ( f ) = h + ( f ) or g ≡ h (mod f ) or [g] = [h].
We can define operations on the residue classes modulo f as follows.
1.3 Polynomials 15

[g] ± [h] = [g ± h]
[g][h] = [gh]

These operations are well-defined on F[x]/( f ) and make F[x]/( f ) into a ring,
called the residue class ring or quotient ring of F[x] by ( f ). The additive identity
is [0] and the multiplicative identity is [1].

Theorem 1.24 Let F be a field and f ∈ F[x]. The quotient ring F[x]/( f ) is a field
if and only if f is irreducible over F.

When F = F p in Theorem 1.24, we can identify F[x]/( f ) with F pd , where d =

deg( f ). So, F pk is a field for any prime p and positive integer k. These are in fact
the only finite fields. Observe that F pk is an extension of F p .
If a polynomial f of degree n has roots α1 , α2 , . . . , αn , then its discriminant is
n(n−1) 
D( f ) = (−1) 2 an2n−2 (αi − α j )
1≤i< j≤n

where an is the coefficient of x n in f (x). For example, the discriminant of the

quadratic polynomial ax 2 + bx + c is b2 − 4ac. Observe that D( f ) = 0 if and only
if f has a multiple root.
Note that D( f ) = an2n−2 det(V ( f )), where V ( f ) is the Vandermonde matrix of
f defined as
⎛ ⎞
1 α1 · · · α1n−1
⎜1 α2 · · · α2n−1 ⎟
⎜ ⎟
V ( f ) = ⎜. .... .. ⎟
⎝ .. . . . ⎠
1 αn · · · αnn−1

We can similarly define the resultant or eliminant of two polynomials

f (x) = am x m + · · · + a1 x + a0
g(x) = bn x n + · · · + b1 x + b0

as


m 
n
R( f, g) = amn bnm (αi − β j )
i=1 j=1

where α1 , α2 , . . . , αm are the roots of f and β1 , β2 , . . . , βn are the roots of g. Note

that R( f, g) = det(S( f, g)), where S( f, g) is the Sylvester matrix of f, g defined as
16 1 Preliminaries
⎛ ⎞
am am−1 · · · a0 ⎫
⎜ am am−1 · · · a0 ⎟ ⎬
⎜ ⎟ n
⎜ .. .. .. .. ⎟ ⎭
⎜ . . . .⎟
S( f, g) = ⎜
⎜ bn
⎟
⎟
⎫
⎜ bn−1 · · · b0 ⎟ ⎬
⎜ bn bn−1 · · · b0 ⎟ m
⎝ ⎠ ⎭
.. .. .. ..
. . . .

with the blank entries representing 0.

Let F be a field and K be a subfield of F. If α ∈ F is a root of a non-zero
polynomial f (x) ∈ K [x], then α is said to be algebraic root over K . If every α ∈ F
is algebraic over K , then F is said to be an algebraic extension of K .
Let F be a field and K be a subfield of F. Let α ∈ F be algebraic over K . Observe
that the set of polynomials f ∈ K [x] for which f (α) = 0 is an ideal of K [x]. By
Theorem 1.17, this ideal is of the form (m) for some monic polynomial m ∈ K [x].
We call m the minimal polynomial of α over K . Observe that m is irreducible in
K [x]. Furthermore, α is a root of some f ∈ K [x] if and only if m | f . So, we have
the following theorem.
Theorem 1.25 Let F be a field and f ∈ F[x] be an irreducible polynomial . If α is
a root of f , then α is a root of g ∈ F[x] if and only if f | g in F[x].
If α is algebraic over Q with minimal polynomial m, we call α an algebraic
number. If in addition m ∈ Z[x], we call α an algebraic integer. It is not hard to see
that α is rational if and only if deg(m) = 1.
Let K be a field and F be a finite extension of K . Then F is algebraic over K . If
a polynomial f ∈ K [x] has roots α1 , α2 , . . . , αn ∈ F, then F is a splitting field of
f.
Theorem 1.26 Let p be a prime number and f ∈ F p [x] be a polynomial of degree
d ≥ 1 such that f (0) = 0. Then there is a positive integer k ≤ p d − 1 such that
f | x k − 1.
Proof The residue class ring F p [x]/( f ) contains p d − 1 non-zero residue classes.
Consider the elements x i + ( f ) for 0 ≤ i ≤ p d − 1 in F p [x]/( f ). Since f (0) = 0,
it follows that x  f , so x i ∈
/ ( f ) for 0 ≤ i ≤ p d − 1. Hence there must be i, j with
0 ≤ i < j ≤ p − 1 such that x i and x j belong to the same residue class. Then
d

x i ≡ x j (mod f ), so x j−i ≡ 1 (mod f ). Thus f | x j−i − 1, as desired. 

Theorem 1.26 enables us to define the order ord( f ) of f ∈ F p [x] to be the small-
est positive integer d such that f | x d − 1, which is well-defined if f (0) = 0. For
an arbitrary f , write f (x) = x r g(x), where gcd(x, g(x)) = 1. Then we can define
ord( f ) to be ord(g).
2 n−1
Consider an extension F pn of F p and let α ∈ F pn . We call α, α p , α p , . . . , α p
the conjugates of α over F p . Observe that these are roots of the minimal polynomial
m of α over F p . Moreover, they are distinct if and only if deg(m) ≥ n. Then we have
the following.
1.3 Polynomials 17

Theorem 1.27 The conjugates of α ∈ F∗pn over F p have the same order.

Theorem 1.28 Let f ∈ F p [x] be an irreducible polynomial of degree d ≥ 1 such

that f (0) = 0 . If α ∈ F∗p is a root of f , then ord( f ) = ord(α).

Proof Observe that F pd is a splitting field of f over F p . By Theorem 1.27, the roots
of f have the same order in F∗pn . Let n = ord(α). Then α is a root of x n − 1. Thus
f | x n − 1 and so ord( f ) ≤ n. If m = ord( f ) < n then f | x m − 1, whence αm = 1,
which is impossible. Thus ord( f ) = n. 

Theorem 1.29 Let f ∈ F p [x] be an irreducible polynomial such that f (0) = 0.

Then ord( f ) | p deg( f ) − 1.

Proof Note that F∗pn is a group of order p n − 1. For α a root of f , if α has order
k, then ord( f ) = ord(α) = k by Theorem 1.28. So, k | p n − 1 by Theorem 1.5 . 

Theorem 1.30 Let k be a positive integer and f ∈ F p [x] . Then f | x k − 1 if and

only if ord( f ) | k.

Proof If f (0) = 0 then we are done. Otherwise, let d = ord( f ). Then f | x d − 1.

If d | k, then x d − 1 | x k − 1. Hence f | x k − 1.
Conversely, if f | x k − 1, write k = dq + r for integers q, r with 0 ≤ r < d.
Then

x k − 1 = x dq+r − 1
= x dq x r − 1
= (x dq − 1)x r + x r − 1
= (x d − 1)g(x)x r + x r − 1

for some polynomial g ∈ F p [x]. Since f | x k − 1 and f | x d − 1, we have f | x r −

1. Since r < d, it follows that r = 0. Thus d | k. 

The purpose of discussing all this background from abstract algebra was to intro-
duce the next theorem. The theorems below will be used to develop the theory of
linear recurrent sequences in Chap. 2.

Theorem 1.31 Let f ∈ F p [x] be an irreducible polynomial . Suppose that ord( f ) =

d and let r be the smallest positive integer such that pr ≥ k. Then ord( f k ) = dpr .
r r
Proof Since f | x d − 1, it follows that f p | (x d − 1) p . By Theorem 1.7,
r r
(x d − 1) p = (x d ) p − 1
r
= x dp − 1
r r
Thus f k | f p | x dp − 1. Hence ord( f k ) | dpr by Theorem 1.30.
18 1 Preliminaries

On the other hand, since f k | x ord( f ) − 1, we have f | x ord( f ) − 1. Therefore

k k

d | ord( f k ) by Theorem 1.30.

Hence ord( f k ) = dp s for some 0 ≤ s ≤ r .
s
By Theorem 1.23, x d − 1 has no multiple root. Hence every root of x dp − 1 =
ps dps ps
(x − 1) has multiplicity p . Since f | x − 1 = (x − 1) , it follows that p s ≥
d s k d

k, whence s ≥ r . Therefore s = r and ord( f k ) = dpr . 

Theorem 1.32 Let f 1 , f 2 , . . . , f n ∈ F p [x] be irreducible pairwise relatively prime

polynomials and put f = f 1 f 2 · · · f n . Then

ord( f ) = lcm(ord( f 1 ), ord( f 2 ), . . . , ord( f n ))

Using the theorems above we obtain the following general result.

Theorem 1.33 Let f ∈ F p [x] such that f (0) = 0. Suppose that f = c f 1e1 f 2e2 · · ·
f kek , where c ∈ F p and each f i is a monic irreducible polynomial in F p [x]. Then

ord( f ) = lcm(ord( f 1 ), ord( f 2 ), . . . , ord( f k )) pr

where r is the smallest positive integer such that pr ≥ max(e1 , e2 , . . . , ek ).

[War37, P. 279, Lemma 4.1] states the following two theorems. An integer poly-
nomial here means a (possibly multi-variable) polynomial with integer coefficients.
The first one is known as Schatanovski’s principle.

Theorem 1.34 Let F(x1 , x2 , . . . , xk ) be an integer polynomial symmetric in

x1 , x2 , . . . , xk . Then for a positive integer m and an integer polynomial f (x), if

F(x) ≡ (x − α1 )(x − α2 ) · · · (x − αk )
≡ (x − β1 )(x − β2 ) · · · (x − βk ) (mod m)

then

F(α1 , α2 , . . . , αk ) ≡ F(β1 , β2 , . . . , βk ) (mod m)

Theorem 1.35 Let f (x) and F(x) be integer polynomials such that

f (x) = x k − ak−1 x k−1 − · · · − a1 − a0

F(x) = x k − bk−1 x k−1 − · · · − b1 x − b0

and F(x) has roots α1n , α2n , . . . , αkn , where α1 , α2 , . . . , αk are the roots of f (x).
Then, for any positive integer l ≤ k and prime p, p | gcd(a0 , a1 , . . . , al−1 ) if and
only if p | gcd(b0 , b1 , . . . , bl−1 ).

Proof First we show that if p | gcd(a0 , a1 , . . . , al−1 ) then p | gcd(b0 , b1 , . . . , bl−1 ).

W rite
1.3 Polynomials 19

f (x) ≡ x k−l (x l − al−1 x l−1 − · · · − ak−1 ) (mod p)

Let g(x) = x k−l (x l − al x l−1 − · · · − ak−1 ) with roots γ1 , γ2 , . . . , γl and γi = 0 for

l < i ≤ k. Then

(x − α1 )(x − α2 ) · · · (x − αk ) ≡ (x − γ1 )(x − γ2 ) · · · (x − γk ) (mod p)

Using Vieta’s formulas (see Vieta’s Formulas in the glossary), the coefficients of a
polynomial can be expressed as symmetric polynomials in the roots of the polyno-
mial. Let φi ( f ) be the elementary symmetric polynomial of degree i in the roots of
f . By Theorem 1.34,

bk−i+1 ≡ φi ( f ) (mod p)
bk−i+1 ≡ φi (g) (mod p)

Now, consider the polynomial G(x) with roots γ1n , γ2n , . . . , γkn . Again, using
Vieta’s Formulas, if ci is the coefficient of x i in G(x), then

φk−i+1 (G) = γ1n γ2n · · · γk−i+1
n

= ci

But if i < l, then at least one of the roots will be 0 in this product, hence ci ≡ 0
(mod p). Similarly, bi ≡ 0 (mod p) for i < l and p | bi for 0 ≤ i < l.
We now prove that if p | gcd(b0 , b1 , . . . , bl−1 ), then p | gcd(a0 , a1 , . . . , al−1 ).
Note that

bl ≡ φk−l+1 (G) (mod p)


≡ γ1n γ2n · · · γk−l+1
n
(mod p)
≡ γ1n γ2n · · · γk−l+1
n
(mod p)
≡ (γ1 γ2 · · · γk−l+1 ) (mod p)
n

≡ cln (mod p)

Thus, if p  cl , then p  bl . Therefore, if p  bi for i < l, then p  ai , as desired. 

1.4 Cyclotomic Polynomials

Cyclotomic polynomials have been extensively studied over the past few centuries.
[Gau01, §§341, P. 599] proved that the cyclotomic polynomial  p (x) is irreducible.
[Sch46] and [Kro45] also proved the same result. [Kro54] and [Ded57] proved the
general case that n (x) is irreducible. And mathematicians have kept working on
20 1 Preliminaries

this, for example, the method shown by [Syl79] was used by [War55] to prove that
real Lehmer numbers have primitive divisors. However, in old times, notations were
not as consistent and convenient as they are now. Different authors such as [Car29],
[War55], [BV04], [Syl79] have used different notations throughout history. Ward
followed in Carmichael’s footprints, while other authors used whatever notation
they deemed fit. Nowadays it is a different story. Even though sometimes Cn (x) is
used, most authors, e.g., [Lan12, Section 3, Chapter 6, Part 2, Roots of unity] use
n (x). Cyclotomic polynomials have wide applications in number theory. However,
most common elementary texts do not cover this topic in detail even though one can
argue that the subject is elementary. We review cyclotomic polynomials briefly in
this section for our purposes in this book.
It is known that every positive integer greater than 1 can be written as a product of
primes. Here, primes are irreducible in the sense that they cannot be divided by any
other positive integer greater than 1. Consider a similar factorization for polynomials.
To be more specific, consider the factorization of x n − 1 into irreducible polynomials.
A general question is, how far can we keep factoring polynomials ? In the case of
x n − 1, simple algebra gives

x 2 − 1 = (x + 1)(x − 1)
x 3 − 1 = (x − 1)(x 2 + x + 1)
x 4 − 1 = (x − 1)(x + 1)(x 2 + 1)

Here we cannot further factorize any of the factors present into two or more non-
constant polynomials with integer coefficients. Without such a restriction, however,
one can obtain factorizations such as x 2 + x + 1 = (x − ω)(x − ω 2 ), where ω is a
complex cubic root of unity. The natural question that arises then is how we can find
a general procedure of factoring x n − 1 into irreducible polynomials. Fortunately,
cyclotomic polynomials are exactly these irreducible polynomials. The complex fac-
torization of x 2 + x + 1 should shed some light on how to obtain such a factorization
in the general case.
We can define the cyclotomic polynomial of order n as

n (x) = (x − ζ i ) (1.5)
1≤i≤n
gcd(i,n)=1

where ζ is a primitive n-th root of unity. The product is over all the positive integers
less than n which are relatively prime to n. Recall that ζ is an n-th root of unity if
ζ n = 1. At first, the definition (1.5) may seem intimidating. However, if you look
at it closely and compare it with the cubic case, you will see that it is simply a
generalization of that factorization.
The idea of factoring x n − 1 with roots of unity is a lot more powerful than one
would deem at first sight. It can even be connected with the history of Fermat’s Last
Theorem. In case you are not familiar with Fermat’s Last Theorem , it states that
1.4 Cyclotomic Polynomials 21

x n + y n = z n is not true for any positive integers x, y, z and n > 2. It has eluded
mathematicians for over three centuries. Finally, Andrew Wiles and Richard Taylor
proved it in around 1993−1994. Many prominent mathematicians failed to prove
this theorem despite significant efforts. In the process, a lot of theories have been
developed. Prior to the proof of the general case, mathematicians used to try to prove
the theorem in special cases. [WE78] is a good read for a historical perspective
on this. Euler proved the case n = 3. Fermat himself prove the case n = 4 using
Pythagorean triplets. Legendre and Dirichlet proved the case n = 5 in 1823. Then
Dirichlet proved the case n = 14 in 1832. Gabriel Lamé proved the case n = 7 using
roots of unity in 1839. Note that n = 7 covers the case n = 14, but the converse is
false. So, even Dirichlet failed to prove the case n = 7. Lamé believed that he could
generalize the idea. The main idea behind the proof was to write

x p + y p = (x + y)(x + ζ y)(x + ζ 2 y) · · · (x + ζ p−1 y)

and if, in some sense, the factors can be shown to be pairwise relatively prime, then
x p + y p = z p would imply that each factor above must be a p-th power. Then one
could use an infinite descent argument. Lamé’s work was based mostly on Liouville’s.
However, when Lamé presented this idea for the general case, Liouville himself
was not very enthusiastic. This idea indeed seems to be too cheap for so many
prominent mathematicians to miss for so long. For example, Lagrange showed the
factorization above as well. Lamé forgot to take into account the complexities the
introduction of complex numbers could present. So the concern was raised whether
prime factorization would remain unique if complex numbers were introduced. Going
forward, another mathematician Wantzel claimed he could prove unique factorization
for n ≤ 4. Then he said that it was easy to see the same argument applies for n > 4,
which was wrong of course. One does not easily see that the same argument applies
in general. Later, Kummer went on to introduce ideal complex numbers which, we
can say, laid the foundations of modern algebraic number theory. Anyway, as you
can see, the history of cyclotomic integers and polynomials is quite rich, dramatic,
and influential. Let us now continue our original discussion.
The smallest positive integer k for which ζ k = 1 is called the order of ζ. We
denote the order of ζ by ord(ζ). And ζ is a primitive n-th root of unity if ord(ζ) = n;
that is, ζ k = 1 does not hold for any positive integer k < n. If ζ is a primitive n-th
root of unity of order k, letting n = kq + r with 0 ≤ r < k, we see that k must divide
n. Otherwise, we would have ζ r = 1 with r smaller than k which would contradict
the minimality of k.
Fix a primitive n-th root of unity ζ and consider the powers of ζ modulo n. We are
interested in finding k such that ζ k is a primitive n-th root of unity; that is, ord(ζ k ) = n.
Letting ord(ζ k ) = d, we have n | dk, since ord(ζ) = n. If g = gcd(n, k) with n =
gm, k = gl, and gcd(m, l) = 1, then m | d. Using ζ dk = 1 along with d = mr for
some positive integer r , we get ζ nlr = 1. In order to make ord(ζ k ) minimal, we need
to have r = 1. This gives ord(ζ k ) = n/ gcd(n, k). If gcd(n, k) > 1, then ord(ζ k ) =
n/ gcd(n, k) < n. In this case, ζ k is not primitive. Therefore, ζ k is primitive only if
gcd(k, n) = 1. This shows that n (x) is a polynomial of degree ϕ(n).
22 1 Preliminaries

Primitive roots of unity are interesting and worth studying because they generate
all roots of unity.
We will now formally prove that n defined in (1.5) is indeed a polynomial with
integer coefficients, that is, a polynomial in Q[x]. We first prove the following result.

Theorem 1.36 For any positive integer n,


xn − 1 = d (x) (1.6)
d|n

Proof It suffices to check that both sides have the same roots. The roots of the
polynomial on the left are exactly the n-th roots of unity. The roots of the polynomial
on the right are the primitive d-th roots of unity for each divisor d of n. Since each
n-th root of unity must be a primitive dth root of unity for a unique divisor d of n
and vice-versa, the conclusion follows. 

This theorem also gives another proof of


n= ϕ(d) (1.7)
d|n

Applying the Möbius Inversion (see Möbius Function and Möbius Inversion) to (1.6)
gives
 n
μ(d)
n (x) = xd −1 (1.8)
d|n

Theorem 1.37 n is monic in Q[x].

The proof of this theorem relies on the fact that if f, g ∈ Q[x] are monic and f g ∈
Q[x], then f, g ∈ Q[x]. Then we can simply induct on n to prove that n ∈ Q[x].
Proof The claim is trivially true for n = 1. Assume that m is monic in Q[x] for all
m < n. Then

xn − 1 = d (x)
d|n

= n (x) d (x)
d|n
d<n
= n (x) f (x)

where 
f (x) = d (x)
d|n
d<n
1.4 Cyclotomic Polynomials 23

By the induction hypothesis, d is monic in Q[x] for d < n, so f is monic in

Q[x]. Hence n is the quotient of two monic polynomials in Q[x], so is monic in
Q[x]. Finally, since x n − 1 = n (x) f (x) ∈ Q[x], we conclude by the above fact
that n ∈ Q[x]. 
We now give a few results that can be used to compute n for different values of n.

Theorem 1.38 For a positive integer n and a prime number p,

⎧
⎨n (x p ) if p | n
np (x) = n (x p )
⎩ otherwise
n (x)

Proof Using (1.8),

 np
μ(d)
np (x) = x d −1
d|np
 np
μ(d)   np μ(d)
= x d −1 x d −1
d|n d|np
dn
 n
μ(d)   np μ(d)
= (x p ) d − 1 x d −1
d|n d|np
dn
 np
μ(d)
= n (x p ) x d −1
d|np
dn

= n (x p )P(x)

where we split the product into two parts based on whether d | n.

First, consider the case p | n. To evaluate P, we need to determine when d | np
but d  n. Let n = p α m with α ≥ 1 and p  m. Then d | p α+1 m but d  p α m. Hence
p 2 | d, so that μ(d) = 0. This gives P(x) = 1.
Next, assume that p  n. To evaluate P, we note that if d | np but d  n, then
d = ap with a | n and gcd(a, p) = 1. Then

μ(d) = μ(ap) = μ(a)μ( p) = −μ(a)

using the multiplicative property of μ. Therefore,

24 1 Preliminaries

 np
μ(ap)
P(x) = x ap − 1
a|n
 n
−μ(a)
= xa −1
a|n

= n (x)−1

as desired. 
This result extends to prime powers as follows.
Theorem 1.39 Let p be a prime number and n, k be positive integers. Then
⎧
⎪ k
⎨n (x p ) if p | n
 pk n (x) = n (x )pk
⎪
⎩ otherwise
n (x pk−1 )

Theorem 1.40 If m, n are positive integers having the same prime factors such that
m | n, then
n
n (x) = m (x m )

Proof Let ν p (n) denote the largest integer a such that pa | n. Observe that ν p (m) ≥ 1
for all p | n. In a similar fashion as before, if d | n but d  m, then ν p (d) ≥ ν p (m) +
1 ≥ 2 for some prime p. Then μ(d) = 0. Assuming n = mk, we have
 n
μ(d)
n (x) = xd −1
d|n
 n
μ(d)   n μ(d)
= xd −1 xd −1
d|m d|n
dm
 mk
μ(d)
= x d −1 ·1
d|m
 m
μ(d)
= (x k ) d − 1
d|m

= m (x k )
n
= m (x m )

as desired. 
Theorem 1.41 If n > 1 is odd, then 2n (x) = n (−x).
Theorem 1.42 If n = p1e1 p2e2 · · · pkek with the pi distinct primes, then
1.4 Cyclotomic Polynomials 25
 e1 −1 e2 −1 ek −1 
n (x) =  p1 p2 ··· pk x p1 p2 ··· pk

Theorem 1.43 Let k, n be positive integers. Then k (x) | k (x n ) if and only if

gcd(k, n) = 1.

We now look at the prime divisors of cyclotomic polynomials evaluated at the positive
integers. They allow us to establish some important results. Let a denote an arbitrary
positive integer.

Theorem 1.44 For a positive integer n and a prime divisor p of n (a), if p | d (a)
for some d | n with d < n, then p | n.

Proof Since a n − 1 = d|n d (a), if p is a prime divisor of d (a) and n (a) for
some d | n with d < n, then a is a multiple root of n (x) modulo p. Hence p | n by
Theorem 1.23. 

Theorem 1.45 For a positive integer n and a prime divisor p of n (a), either p | n
or p ≡ 1 (mod n).

Proof If p | d (a) for some d | n with d < n, then p | n by Theorem 1.44. Other-
wise, let k = ord(a) in F∗p . We claim that k = n. This will imply n | |F∗p | = p − 1
by Theorem 1.5, and consequently, p ≡ 1 (mod n).
To prove the claim, observe that p | n (a) | a n − 1, so that k | n. If k < n, then
p | a k − 1 implies p | d (a) for some d | k by Theorem 1.36. Then d ≤ k < n,
contradicting the minimality of n. 

An important special case is the following.

Theorem 1.46 Every prime divisor of 1 + a + a 2 + · · · + a p−1 is either p or con-

gruent to 1 (mod p).

Proof Let q be a prime divisor of

ap − 1
1 + a + a 2 + · · · + a p−1 =
a−1

d| p d (a)
=
a−1
1 (a) p (a)
=
1 (a)
=  p (a)

By Theorem 1.45, either q = p or q ≡ 1 (mod p). 

Theorem 1.46 can be used to prove a special case of Dirichlet’s theorem on

arithmetic progressions, which states that if gcd(a, b) = 1, then the sequence (x)
with xn = an + b contains infinitely many prime numbers. Although not particularly
relevant to our discussion, we present the result as a nice corollary to Theorem 1.46.
26 1 Preliminaries

Theorem 1.47 There are infinitely many primes congruent to 1 (mod n) for any
positive integer n.

Proof It suffices to consider the case n > 1. If the number of such primes is finite,
let P be the product of all such primes and the prime divisors of n. Pick a positive
integer k such that n (P k ) > 1, so that it has at least one prime divisor q. Such a
k must exist since n (x) is a non-constant monic polynomial. Now q | P nk − 1, so
q  P. On the other hand, by Theorem 1.45, either q | n or q ≡ 1 (mod n), hence
q | P. This is a contradiction. 

Theorem 1.48 If m, n are positive integers such that gcd(m (a), n (a)) > 1, then
m/n is a prime power.

Proof Let p be a prime divisor of gcd(m (a), n (a)) and write m = p u b and n =
p v c for integers u, v ≥ 0 such that p  bc. It suffices to prove that b = c.
Since p | m (a) and p | n (a), we have p | a m − 1 and p | a n − 1 by Theorem
1.36. Then p  a. So, if u = 0, then p | b (a). Otherwise,
u
b (a p )
m (a) =
b (a pu−1 )
u u
Since p | m (a), we have p | b (a p ). By Fermat’s little theorem, a p ≡ a (mod p)
u
and so b (a p ) ≡ b (a) (mod p). Thus, in every case, p | b (a) and similarly,
p | c (a). Then p | a b − 1 and p | a c − 1 by Theorem 1.36, so that p | a h − 1,
where h = gcd(b, c). So by Theorem 1.36, p | d (a) for some d | h.
If b = c, we may assume that b > c. Then h < b. Hence p | d (a) and p | b (a)
with d | b and d < b. Therefore p | b by Theorem 1.44. This contradicts p  bc. Thus
b = c. 

The next theorem is from [Ser85, P. 133−134, Lemma 347]. It follows from
Theorem 1.18.

Theorem 1.49 Let p be a prime number and s, m ≥ 0 be integers such that p  m.

Then
s s
m (x p ) ≡ m (x) p (mod p)

References

[BV04] Geo. D. Birkhoff and H. S. Vandiver. “On the Integral Divisors of a n − bn ”. In: Ann.
Mat. 5.4 (1904), pp. 173-180. issn: 0003486X. http://www.jstor.org/stable/2007263
[Car29] R. D. Carmichael. “A Simple Principle of Unification in the Elementary Theory of Num-
bers”. In: Amer. Math. Monthly 36.3 (1929), pp. 132-143. issn: 00029890, 19300972.
http://www.jstor.org/stable/2299678
References 27

[Ded57] R. Dedekind, Beweis für die Irreductibilität der Kreistheilungs-Gleichungen. Journal für
die reine und angewandte Mathematik 54, 27–30 (1857)
[Gau01] Carl Friedrich Gauss. Disquisitiones arithmeticae. 1801
[Kro45] L. Kronecker, Beweis dass für jede Primzahl p die Gleichung 1 + x + x 2 + +x p−1 =
0 irreductibel ist. Journal für die reine und angewandte Mathematik (Crelles Journal)
1845(29), 280–280 (1845). https://doi.org/10.1515/crll.1845.29
[Kro54] L. Kronecker, Memoire sur les facteures irréductibles de l’expression x n − 1. Journal
de mathématiques pures et appliquées : ou recueil mensuel de mémoires sur les diverses
parties des mathématiques 19, 177–192 (1854)
[Lan86] Serge Lang, Introduction to Linear Algebra (Springer, New York, 1986)
[Lan12] Serge Lang. Algebra. Springer-verlag New York Inc., 2012
[LN97] Rudolf Lidl and Harald Niederreiter. Finite Fields: Encyclopedia of Mathematics and Its
Applications. Vol. 7. 1997, p. 136
[Sch46] L. Schönemann. “Von denjenigen Moduln, welche Potenzen von Primzahlen sind.” In:
Journal für die reine und angewandte Mathematik 32 (1846), pp. 93-105. http://eudml.
org/doc/147334
[Ser85] Joseph Alfred Serret. Cours d’Algèbre supérieure Tome 1. fre. Paris: Gauthier-Villars,
1885. <error l="292" c="Undefined command " />http://eudml.org/doc/202932
[Syl79] J. J. Sylvester. “On Certain Ternary Cubic-Form Equations”. In: Amer. J. Math. 2.4
(1879), pp. 357-393. issn: 00029327, 10806377. http://www.jstor.org/stable/2369490
[War37] Morgan Ward. “Linear Divisibility Sequences”. In: Trans. Amer. Math. Soc. 41.2 (1937),
pp. 276-286. issn: 00029947. http://www.jstor.org/stable/1989623
[War55] Morgan Ward. “The Intrinsic Divisors of Lehmer Numbers”. In: Ann. Mat. 62.2 (1955),
pp. 230-236. issn: 0003486X. http://www.jstor.org/stable/1969677
[WE78] Lawrence Washington and Harold M. Edwards. “Fermat’s Last Theorem: A Genetic
Introduction to Algebraic Number Theory.” In: Mathematics of Computation 32.143
(1978), p. 943. doi: https://doi.org/10.2307/2006502.
Chapter 2
Linear Recurrent Sequences

In this chapter, we discuss linear recurrent sequences over a field. We give results
on when such a sequence is periodic and obtain an upper bound on the length of the
period, as well as show how to produce this bound. Finally, we discuss the theory
developed by Morgan Ward on the periodicity of such sequences with the help of the
double modulus. We will see a lot of results that are of fundamental importance in
this theory.

2.1 Introduction

A sequence (a) of integers defined by

an+k = ck−1 an+k−1 + · · · + c1 an+1 + c0 an + α (2.1)

where ai is the i-th term of the sequence and the ci are constants, is called a (linear)
recurrent sequence of order k. Note that like monic polynomials, ck = 1. We call
(a) homogeneous if α = 0, otherwise call (a) non-homogeneous. The order here is k
and not k + 1 because an+k is represented as a linear combination of the k preceding
terms of the sequence. This will make more sense when we define the characteristic
polynomial of (a) below. Recurrent sequences have also been called solutions to
linear difference equations [Bel30, War33a].
We will mostly focus on homogeneous sequences. In this chapter we discuss some
properties of recurrent sequences, especially periodicity. Considering (2.1) modulo
a positive integer m, we get a new sequence (r ) of residues where ri is the least non-
negative residue of ai upon division by m. We call (r ) the reduction of (a) modulo
m.
Consider the recurrent sequence (a) defined in (2.1) over the finite field Fm ,
so that the coefficients ck−1 , . . . , c0 and the terms an+k−1 , . . . , an are elements of

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 29
M. Billal and S. Riasat, Integer Sequences,
https://doi.org/10.1007/978-981-16-0570-3_2
30 2 Linear Recurrent Sequences

Fm . Here we assume that m = p s for a prime p and a positive integer s. We call

si = (ai , ai+1 , . . . , ai+k−1 ) the i-th state vector of (a). State vectors will be very
useful for characterizing periodicity of recurrent sequences as we will see later.
We now define the characteristic polynomial of a recurrent sequence. The moti-
vation behind this comes from assuming that the terms of a recurrent sequence (a)
can be written as the powers of some constant c; that is, an = cn for all n, so that (a)
is in fact just a geometric sequence. This may seem like a wild guess, but it is not
completely unreasonable. For example, the recurrent sequence an = 2an−1 can be
obtained this way. Another non-trivial example is the Fibonacci Sequence (F) given
by

Fn+1 = Fn + Fn−1 (2.2)

Although this is not exactly a geometric sequence, it has an interesting property.

Looking at the ratio of consecutive Fibonacci numbers we observe that

Fn+1
lim
n→inf Fn

exists. This suggests that (F) behaves in some sense like a geometric sequence. To
find the common ratio, we let

Fn+1
φ = lim
n→∞ Fn
Fn−1
= lim 1 +
n→∞ Fn
1
=1+
φ
φ =1+φ
2
(2.3)
√
1± 5
φ=
2
As a matter of fact, we have
 √ n  √ n 
1 1+ 5 1− 5
Fn = √ −
5 2 2

So Fn is a linear combination of the powers of the roots of (2.3). To understand this

behavior, say it possible to write Fn = r n by wishful thinking. Then r n = r n−1 +
r n−2 , so that r 2 = r + 1, which is exactly (2.3). If the roots of this equation are r1
and r2 , then this means that Fn = r1n and Fn = r2n satisfy (2.2). Just like how the set
of solutions to a linear system of equations is a vector space, it turns out that the set
of solutions to a linear recurrent sequence is also a vector space. In particular, the
general solution to (2.2) is of the form
2.1 Introduction 31

Fn = b1r1n + b2 r2n

for some constants b1 , b2 . We can solve for b1 , b2 using the initial values F0 =
0, F1 = 1 of the Fibonacci Sequence and obtain

r1n − r2n
Fn =
r1 − r2

The above analysis worked because the roots of the characteristic polynomial of
(F) were distinct. If they were not distinct, then we would expect the solution to be

r1n − r2n
lim = nr2n−1
r1 →r2 r1 − r2

We can extend this idea to the general sequence (2.1). Assuming α = 0, so that
the sequence (a) is homogeneous, the characteristic polynomial of (a) is

f (x) = x k − ck−1 x k−1 − · · · − c1 x − c0

Theorem 2.1 Given the initial state vector (a0 , a1 , . . . , ak−1 ) of the homogeneous
recurrent sequence (a) of order k, if the characteristic polynomial f (x) of (a) has k
distinct roots α1 , α2 , . . . , αk , then for any n ∈ N,

an = b1 α1n + b2 α2n + · · · + bk αkn

for some constants b1 , b2 , . . . , bk that are uniquely determined by the initial state
vector.

As in the case of the Fibonacci Sequence, the assumption that the roots of f (x)
are distinct is very important here. We will simply mention when this theorem is
applicable.
The characteristic polynomial f (x) of (2.1) can be obtained in another way.
Consider the k × k matrix C defined as follows. If k > 1, then
⎛ ⎞
0 0 ···0 c0
⎜1 0 ···0 c1 ⎟
⎜ ⎟
⎜ ···0 c2 ⎟
C = ⎜0 1 ⎟
⎜ .. .. .. .. .. ⎟
⎝. . .. . ⎠
0 0 · · · 1 ck−1

and otherwise, C = (c0 ). The matrix C is called the companion matrix of (a). As
we will soon see, this matrix will play a big role in our theory. For now, just notice
that the characteristic polynomial can be written as

f (x) = det(x I − C)
32 2 Linear Recurrent Sequences

A matrix M is called diagonalizable if there exists an invertible matrix S such

that S −1 M S is a diagonal matrix . It turns out that C is diagonalizable if the roots of
f (x) are distinct. For example, the companion matrix of an = 2an−1 − an−2 is

0 −1
C=
1 2

which cannot be diagonalized.

A very important property of the companion matrix is its association with the
state vectors. One can easily check by induction that

sn = s0 C n

Theorem 2.1 can be generalized to non-homogeneous recurrences as follows. For a

non-homogeneous linear recurrence (a) of order k given by (2.1), if the characteristic
polynomial f (x) of (a) has k distinct roots α1 , α2 , . . . , αk , then
α
an = + b1 α1n + b2 α2n + · · · + bk αkn
f (1)

As in Theorem 2.1, we obtain the unique constants b1 , b2 , . . . , bk from the initial

state vector (a0 , a1 , . . . , ak−1 ).

2.2 Periodicity

Consider a sequence of integers (henceforth referred to as an integer sequence)

(a). We are interested in the periodicity of this sequence. It is possible to interpret
periodicity in multiple ways. As usual, when we say a sequence is periodic, we mean
the sequence repeats from some point. In particular, if there exist positive integers
n 0 , ρ such that an+ρ = an for all n ≥ n 0 , then we say (a) is ultimately periodic.
If n 0 = 0, then we simply call (a) periodic. Here we call n 0 the pre-period and
ρ the period. Note, however, that these values may not be unique; we are mostly
interested their smallest values. Carmichael [Car29] and Ward called this smallest ρ
the characteristic of (a). However, this term can be easily confused with other things
termed characteristic so we will simply call it the least period (or just period when
there is no ambiguity) of (a).
For example, if (r ) is the reduction of the recurrent sequence (a) modulo m, then
we will show later that (r ) is ultimately periodic. In particular, if every term of (a)
is divisible by m, then we call (a) a null sequence modulo m.
Recall our definition of ord( f ) for a polynomial f . If f (x) = x h g(x) for some
h ≥ 0 such that gcd(g(x), x) = 1, equivalently g(0) = 0, then we defined ord( f ) as
the smallest positive integer d for which g(x) | x d − 1. This was also a pre-condition
for some theorems in Chapter 1. This will now make better sense as follows. If the
2.2 Periodicity 33

characteristic polynomial f (x) of a linear recurrent sequence (a) is of the form

x h g(x), where gcd(g(x), x) = 1, then the first h terms of the initial state vector s0
can be ignored in the sense that they will have no impact on the periodicity of (a).
Thus, we have the following theorem.

Theorem 2.2 Let (a) be a linear recurrent sequence of order k with characteristic
polynomial f (x) = x h g(x), where gcd(g(x), x) = 1. Then the sequence (c) with
ci = ai+h has characteristic polynomial g(x).

We first give some results on the period of a recurrent sequence.

Theorem 2.3 Let m be a positive integer and (a) be a recurrent sequence of order
k over Fm . Then (a) is ultimately periodic with period

mk if (a) is non-homogeneous
ρ≤
m − 1 otherwise
k

Proof The number of different possible state vectors of (a) is m k . So, among the state
vectors si for 0 ≤ i ≤ m k , there must be two that are equal. In particular, there exists
0 ≤ i < j ≤ m k such that si = s j . Incorporating this into (2.1) gives sn+ j−i = sn
for all n ≥ i. Hence (a) is ultimately periodic with period ρ = j − i ≤ m k . If (a) is
homogeneous, then excluding the zero vector from our count gives ρ ≤ m k − 1. 

So any recurrent sequence over a finite field is ultimately periodic, although not
necessarily periodic . It is easy to construct an ultimately periodic sequence that is
not periodic.

Theorem 2.4 Let F be a finite field and (a) be a recurrent sequence over F given
by

an+k = ck−1 an+k−1 + · · · + c1 an+1 + c0 an + α

If c0 = 0 , then (a) is periodic.

Proof We want to show that (a) has pre-period 0. Let ρ and n 0 be the least period and
pre-period of (a) respectively. Then an+ρ = an for all n ≥ n 0 . Since c0 = 0, there is
an inverse c0−1 ∈ F. Then if an+ρ = an for some n ≥ n 0 , we can obtain an+ρ−1 = an−1
as follows. If k ≥ 1, then an+k+ρ−1 = an+k−1 , so

an+k+ρ−1 = ck−1 an+k+ρ−1 + · · · + c0 an+ρ−1 + α


an+ρ−1 = c0−1 an+k+ρ−1 − ck−1 an+ρ−2+k − · · · − c1 an+ρ − α
= c0−1 (an+k−1 − ck−1 an−2+k − · · · − c1 an − α)

and
34 2 Linear Recurrent Sequences

an+k−1 = ck−1 an+k−2 + · · · + c0 an−1 + α

an−1 = c0−1 (an+k−1 − ck−1 an−2+k − · · · − c1 an − α)

Thus, an+ρ−1 = an−1 and by induction, an+ρ−i = an−i for 0 ≤ i ≤ n. This shows that
n 0 = 0 and an+ρ = an for all n ≥ 0, as desired. 

Theorem 2.5 Let F be a finite field. If ρ is the least period of the recurrent sequence
(a) over F and d is a period of (a), then ρ | d.

Proof By definition, we have an+ρ = an and an+d = an for all n. Let d = qρ + r

with 0 ≤ r < ρ.

an = an+d
= an+qρ+r
= an+(q−1)ρ+r
= ···
= an+r

If r > 0, then we have a period r with 0 < r < ρ , which is a contradiction. Thus
ρ | d. 

Bell [Bel30] gave the next theorem that connects the characteristic polynomial to
the least period of a recurrent sequence. This is a very nice result and while we could
use the theory we developed for fields, we present his original proof that uses only
the ideas developed in this section.

Theorem 2.6 Let (a) be a recurrent sequence of order k with an irreducible char-
acteristic polynomial f . Then (a) has a unique least period ρ if and only if k = ϕ(ρ)
and the roots of f are the primitive ρ-th roots of unity.

Proof Let α1 , α2 , . . . , αk be the roots of f . Since f is irreducible, the roots are

distinct and D( f ) = 0. Then we have

an = bk αkn + · · · + b1 α1n

for some b1 , b2 , . . . , bk are some constant real numbers. These constants b1 , b2 , . . . , bk

are uniquely determined by the initial state vector (a0 , a1 , . . . , ak−1 ).
Since an+ρ = an for all n, we have
n+ρ n+ρ
bk αk + · · · + b1 α1 = bk αkn + · · · + b1 α1n
ρ ρ
bk αkn (αk − 1) + · · · + b1 α1n (α1 − 1) = 0

Observe that none of the roots are 0. Setting n = 0, 1, . . . , k − 1 we obtain the

following system of equations.
2.2 Periodicity 35

ρ ρ
bk (αk − 1) + · · · + b1 (α1 − 1) = 0
ρ ρ
bk αk (αk − 1) + · · · + b1 α1 (α1 − 1) = 0
..
.
ρ ρ
bk αkk−1 (αk − 1) + · · · + b1 α1k−1 (α1 − 1) = 0
ρ
We can see that the system has a solution only if bi (αk − 1) = 0 for 1 ≤ i ≤ k. So,
first we show that bi = 0 for 1 ≤ i ≤ k. Let h be the number of i such that bi = 0.
Then (a) can be written as a recurrent sequence of order r − h. If the characteristic
polynomial in this case is g, then g shares those r − h roots with f . So g is an
integer polynomial that divides f . This forces g to be constant, whence h = 0. Thus,
ρ
αi − 1 = 0 and αi is a ρ-th root of unity. Furthermore, αi must be a primitive ρ-th
ρ
root of unity as ρ is the smallest positive integer that satisfies αi = 1. Finally, the
number of primitive ρ-th roots is ϕ(ρ). So, we must have k = ϕ(ρ). 

So we can express the characteristic polynomial f (x) as


(x d − 1)μ( d )
n

d|ρ

which is the cyclotomic polynomial ρ (x). [War33a] extended this theorem as fol-
lows.

Theorem 2.7 Let (u) be a recurrent sequence of order k such that for fixed positive
integers a, b we have

u a = u a+b = u a+2b = · · · = u a+kb = 0 (2.4)

If the characteristic polynomial f of (u) is irreducible in Z[x], then (u) is periodic

and the roots of f are roots of unity.

Proof Let α1 , α2 , . . . , αk be the roots of f . Since f is irreducible, the roots are

distinct and we can write u n as

u n = b1 α1n + b2 α2n + · · · + bk αkn

for some non-zero constants bi . Let c = u a+ib for 0 ≤ i ≤ k. We get the following
system of equations from (2.4)

(b1 α1 )αib
1 + (b2 α2 )α2 + · · · + (bk αk )αk − c = 0
ib ib

for 0 ≤ i ≤ k. By assumption, c = 0 and we have k + 1 homogeneous linear equa-

tions in b1 α1 , b2 α2 , . . . , bk αk , and c. So the determinant of the coefficient matrix of
the system must be 0. We can write this determinant as the Vandermonde determinant
V (α1b , α2b , . . . , αkb , 1).
Another random document with
no related content on Scribd:
 87. It is a mistaken notion, and injurious to health, for a young
wife, or for any one else, to eat, just before retiring to rest, a hearty
meat supper:
“Oppress, not nature sinking down to rest
With feasts too late, too solid, or too full.”

88. How often we hear a delicate lady declare that she can only eat
one meal a day, and that is a hearty meat supper the last thing at
night; and who, moreover, affirms that she can neither sleep at night,
nor can she have the slightest appetite for any other meal but her
supper, and that she should really starve if she could not have food
when she could eat it! The fact is, the oppressed stomach oppresses
the brain, and drives away sleep, and appetite, and health. The habit
is utterly wrong, and oftentimes demands professional means to
correct it.
89. How is it that sometimes a lady who has an excellent appetite
is, notwithstanding, almost as thin as a rake? It is not what she eats,
but what she digests, that makes her fat. Some people would fatten
on bread and water, while others would, on the fat of the land, be as
thin as Pharaoh’s lean kine. Our happiness and our longevity much
depend on the weakness or on the soundness of our stomachs: it is
the stomach, as a rule, that both gauges our happiness and that
determines the span of the life of both men and women. How
necessary it is, then, that due regard should be paid to such an
important organ, and that everything should be done to conduce to
the stomach’s welfare,—not by overloading the stomach with rich
food; not by a scanty and meager diet; but by adopting a middle
course, betwixt and between high living and low living—the juste
milieu. We should all of us remember that glorious saying—those
immortal words of St. Paul—“Be temperate in all things.”
90. Where a lady is very thin, good fresh milk (if it agree) should
form an important item of her diet. Milk is both fattening and
nourishing, more so than any other article of food known; but it
should never be taken at the same meal (except it be in the form of
pudding) with either beer or wine: they are incompatibles, and may
cause disarrangement of the stomach and bowels. Milk would often
agree with an adult, where it now disagrees, if the admixture of milk
with either beer or wine were never allowed.
 91. Let me advise my fair reader to take plenty of time over her
meals, and to chew her food well; as nothing is more conducive to
digestion than thoroughly masticated food. No interruptions should
be allowed to interfere with the meals; the mind, at such times,
should be kept calm, cheerful, and unruffled, for “unquiet meals
make ill digestions.”
92. Many persons bolt their food! When they do, they are drawing
bills on their constitutions which must inevitably be paid! The teeth
act as a mill to grind and prepare the food for the stomach; if they do
not do their proper work, the stomach has double labor to perform,
and being unable to do it efficiently, the stomach and the whole body
in consequence suffer.
93. The teeth being so essential to health, the greatest care should
be taken of them: they should be esteemed among one’s most
precious possessions.[19]
94. With regard to beverage, there is, as a rule, nothing better for
dinner than either toast and water, or, if it be preferred, plain spring
water—
“Naught like the simple element dilutes;”[20]

and after dinner, one or two glasses of sherry. A lady sometimes,

until she has had a glass of wine, cannot eat her dinner; when such
be the case, by all means let a glass of wine be taken,—that is to say,
let her have it either just before or during dinner, instead of after
dinner; or let her have one glass of sherry before or during dinner,
and one glass after dinner.
95. A young wife sometimes has a languid circulation, a weak
digestion, and constipated bowels; then, a glass of sherry during
dinner and another glass after dinner is beneficial; and however
much she might dislike wine, she should be induced to take it, as the
wine will improve her circulation, will strengthen her digestion, and
will tend to open her bowels. But let me urge her never, unless
ordered by a medical man, to exceed the two glasses of wine daily.
96. If wine does not agree, and if she require a stimulant, a
tumblerful either of home-brewed ale or of Burton bitter ale ought,
instead of water, to be taken at dinner. But remember, if she drink
either beer or porter, she must take a great deal of out-door exercise;
otherwise it will probably make her bilious. If she be inclined to be
bilious, wine is superior to either beer or porter.
91. Brandy ought never to be taken by a young wife but as a
medicine, and then but rarely, and only in cases of extreme
exhaustion. It would be a melancholy and gloomy prospect for her to
drink brandy daily; she would, in all probability, in a short time
become a confirmed drunkard. There is nothing, when once
regularly taken, more fascinating and more desperately dangerous
than brandy-drinking. It has caused the destruction of tens of
thousands both of men and of women!
98. A wife ought not, if she feel low, to fly on every occasion to
wine to raise her spirits, but should try the effects of a walk in the
country, and
“Draw physic from the fields in draughts of vital air.”[21]

99. An excitable wife is a weakly wife: “excitement is the effect of

weakness, not of strength.” Wine in large quantities will not
strengthen, but, on the contrary, will decidedly weaken; the more the
wine, the greater the debility and the greater the excitement—one
follows the other as the night the day. A person who drinks much
wine is always in a state of excitement, and is invariably weak, low,
and nervous, and frequently barren. Alcoholic stimulants in excess
are “a delusion and a snare,” and are one of the most frequent causes
of excitement, and therefore both of weakness and of barrenness.
Alcohol, pure and undiluted, and in excess, is a poison, and is ranked
among the deadly poisons; if a person were to drink at one draught
half a pint of undiluted alcohol it would be the last draught he or she
would ever, in this world, drink,—it would be as surely fatal as a large
dose of either arsenic or strychnine! Brandy, whisky, gin, and wine
are composed of alcohol as the principal ingredient; indeed, each and
all of them entirely owe their strength to the quantity of alcohol
contained therein. Brandy, whisky, gin, and wine, without the
alcohol, would, each one of them, be as chip in porridge—perfectly
inert. Brandy and wine, the former especially, contain large
proportions of alcohol, and both the one and the other, in excess,
either prevents a woman from conceiving, and thus makes her
barren, or if she do conceive, it poisons the unborn babe within her;
and it either makes him puny and delicate, or it downright kills him
in the womb, and thus causes a miscarriage. If he survive the poison,
and he be born alive, he is usually, when born, delicate and
undersized; if such a one be suckled by such a mother, he is
subjected, if the mother can nurse him, which in such cases she
rarely can, to a second course of poisoning; the mother’s milk is
poisoned with the alcohol, and the poor unfortunate little wretch,
having to run the gantlet in the womb and out of the womb, pines
and dwindles away, until at length he finds a resting-place in the
grave! If you wish to make a dog small, give him, when he is a puppy,
gin; the alcohol of the gin will readily do it: this is a well-known fact,
and is, by dog-fanciers, constantly practiced. If you desire, in like
manner, to make a Tom Thumb of a baby, give him the milk of a
mother or of a wet-nurse who imbibes, in the form of wine or of
brandy or of gin, alcohol in quantities, and the deed is done! Gin-
drinking nursing mothers, it is well known, have usually puny
children; indeed, the mother drinking the gin is only another way of
giving gin to the babe—an indirect instead of a direct route, both
leading to the same terminus. Brandy was formerly sold only by the
apothecary; brandy is a medicine—a powerful medicine—and ought
only to be prescribed as a medicine; that is to say, but seldom, in
small and in measured quantities at a time, and only when absolutely
necessary: now it is resorted to on every occasion as a panacea for
every ill! If taken regularly, and in quantities, as unfortunately it
frequently now is, it becomes a desperate poison—a pathway leading
to the grave! It is utterly impossible for any person to hold in the
mouth, for five minutes at a time, a mouthful of neat brandy without
experiencing intense suffering: if it has this fearful effect on the
mouth, what effect must this burning fluid, when taken in quantities,
have upon the stomach? Injury, most decided injury to the stomach,
and, through the stomach, disease and weakness to the remainder of
the body! Brandy is a wonderful and powerful agent: brandy has the
effect, if taken in excess and for a length of time, of making the liver
as hard as a board. Brandy in large quantities, and in the course of
time, has the power of making the body marvelously big—as big
again; but not with firm muscle and strong sinew, not with good
blood and wholesome juices—nothing of the kind; but of filling it full,
even to bursting, with water! Brandy has the power of taking away a
giant’s strength, and of making him as helpless as a little child!
Habitual brandy-drinking poisons the very streams of life! It would
take more time and space than I have to spare to tell of the wonderful
powers of brandy; but unfortunately, as a rule, its powers are more
those of an angel of darkness than those of an angel of light! If the
above statements be true (and they cannot be contravened), they
show the folly, the utter imbecility, and the danger, both to mother
and to babe, of dosing a wife, be she strong or be she delicate, and
more especially if she be delicate, with large quantities either of wine
or of brandy. Brandy, gin, and whisky act on the human economy
very much alike; for, after all, it is the quantity of alcohol contained
in each of them that gives them their real strength and danger. I have
selected brandy as the type of all of them, as brandy is now the
fashionable remedy for all complaints, and, unfortunately, in too
many instances the habit of drinking it imperceptibly but rapidly
increases, until at length, in many cases, that which was formerly a
teaspoonful becomes a tablespoonful, and eventually a wineglassful,
with what result I have earnestly endeavored faithfully to portray.
Avoid, then, the first step in regular brandy-drinking: it is the first
step that ofttimes leads to danger, and eventually to destruction!
100. I am quite convinced that one cause of barrenness among
ladies of the present day is excessive wine-drinking. This is an age of
stimulants, and the practice is daily increasing. A delicate lady is
recommended to take three or four glasses of wine daily. It seems for
the moment to do her good, and whenever she feels low she flies to it
again. The consequence is, that she almost lives upon wine, and takes
but little else besides! Who are the fruitful women? Poor women who
cannot afford to drink stimulants; for instance, poor Irish women
and poor curates’ wives, who have only, principally, water and milk
and buttermilk to drink.
101. There is decidedly, among the higher ranks, more barrenness
than formerly, and one cause of it, in my opinion, is the much larger
quantity of wine now consumed than in the olden times. Many ladies
now drink as many glasses of wine in one day as their grandmothers
drank in a week; moreover, the wineglasses of the present day are
twice the size of old-fashioned wineglasses; so that half a dozen
glasses of wine will almost empty a bottle; and many ladies now
actually drink, in the day, half a dozen glasses of wine!
102. In the wine-growing and wine-drinking country of France,
barrenness prevails to a fearful extent; it has become there a serious
consideration and a State question. Wine is largely consumed in
France by ladies as well as by gentlemen. The usual and everyday
quantity of wine allowed at dinner at the restaurants of Paris, for
each lady, is half a wine quart bottle-full—a similar quantity to that
allowed for each gentleman. Where a gentleman and a lady are
dining together, and have a bottle of wine between them, it is
probable that the former might consume more than his own share of
the wine; but whether he does or not, the quantity the lady herself
drinks is sadly too much either for her health or for her fruitfulness. I
am, moreover, quite convinced that the quantity of wine—sour wine
—consumed by French wives is not only very antagonistic to their
fertility, but likewise to their complexions.
103. Wine was formerly a luxury, it is now made a necessary of life.
Fruitful women, in olden times, were more common than they are
now. Riches, and consequently wine, did not then so much abound,
but children did much more abound. The richer the person, the
fewer the children.
104. Wine is now oftentimes sucked in with a mother’s milk! Do
not let me be misunderstood; wine and brandy, in certain cases of
extreme exhaustion, are, even for very young children, most valuable
remedies; but I will maintain that both wine and brandy require the
greatest judgment and skill in administering, and do irreparable
mischief unless they are most carefully and judiciously prescribed.
Wine ought to be very rarely given to the young; indeed, it should be
administered to them with as much care and as seldom as any other
dangerous or potent medicine.
105. Statistics prove that wine-bibbing in England is greatly on the
increase, and so is barrenness. You might say there is no connection
between the two. I maintain that there is a connection, and that, the
alcohol contained in the wine (if wine be taken to excess, which
unfortunately it now frequently is) is most antagonistic to
fruitfulness.
106. It is surprising, nowadays, the quantity of wine some few
young single ladies, at parties, can imbibe without being intoxicated;
but whether, if such ladies marry, they will make fruitful vines is
quite another matter; but of this I am quite sure, that such girls will,
as a rule, make delicate, hysterical, and unhealthy wives. The young
are peculiarly sensitive to the evil effects of overstimulation.
Excessive wine-drinking with them is a canker eating into their very
lives. Time it is that these facts were proclaimed through the length
and breadth of our land, before mischief be done past remedy.
107. Champagne is a fashionable and favorite beverage at parties,
especially at dances. It is a marvel to note how girls will, in
quantities, imbibe the dangerous liquid. How cheerful they are after
it; how bright their colors; how sparkling their eyes; how voluble
their tongues; how brilliant their ideas! But, alas! the effects are very
evanescent—dark clouds soon o’ershadow the horizon, and all is
changed! How pale, after it, they become; how sallow their
complexions; how dim their eyes; how silent their tongues; how
depressed their spirits—depression following in an inverse ratio to
overstimulation; and if depression, as a matter of course, weakness
and disease! Champagne is one of the most fascinating but most
desperately dangerous and deceptive drinks a young girl can imbibe,
and should be shunned as the plague! Young men who witness their
proceedings admire them vastly as partners for the evening, but
neither covet nor secure them as partners for life. Can they be
blamed? Certainly not! They well know that girls who, at a dance,
imbibe freely of the champagne-cup, and who at a dinner party
drink, as some few are in the habit of drinking, four or five, or even
six, glasses of wine,—that such wives as these, if ever they do become
mothers (which is very doubtful), will be mothers of a degenerate
race. It is folly blinking the question; it is absolutely necessary that it
be looked boldly in the face, and that the evil be remedied before it
be too late.
108. There is an immense deal of drinking in England, which, I am
quite convinced, is one reason of so few children in families, and of
so many women being altogether barren. It is high time that these
subjects were looked into, and that the torrent be stemmed, ere it
o’erflow its banks, and carry with it a still greater amount of
barrenness, of misery, and of destruction.
109. It might be said that the light wines contain but little alcohol,
and therefore can cause, even if taken to excess, but slight injurious
effects on the constitution. I reply, that even light wines, taken in
quantities, conduce to barrenness, and that, as a rule, if a lady once
unfortunately takes to drinking too much wine, she is not satisfied
with the light wines, but at length flies to stronger wines—to wines
usually fortified with brandy, such as either to sherry or to port wine,
or even, at last, to brandy itself! I know that I am treading on tender
ground, but my duty as a medical man, and as a faithful chronicler of
these matters, obliges me to speak out plainly, without fear or
without favor, and to point out the deplorable consequences of such
practices. I am quite aware that many ladies have great temptations
and great inducements to resort to wine to cheer them in their hours
of depression and of loneliness; but unless the danger be clearly
pointed out and defined, it is utterly impossible to suggest a remedy,
and to snatch such patients from certain destruction.
110. I am quite convinced of one thing, namely, that the drinking
of much wine—be it light as claret, or be it heavy as port—sadly
injures the complexion, and makes it muddy, speckled, broken-out,
and toad-like.
111. It is high time that medical men should speak out on the
subject, and that with no “uncertain sound,” before mischief be done
past remedy, and before our island become as barren of children as
France unfortunately now is.
112. If a lady be laboring under debility, she is generally dosed with
quantities of wine—the greater the debility the more wine she is
made to take, until at length the poor unfortunate creature almost
lives upon wine. Her appetite for food is by such means utterly
destroyed, and she is for a time kept alive by stimulants; her stomach
will at length take nothing else, and she becomes a confirmed invalid,
soon dropping into an untimely grave! This is a most grievous, and,
unfortunately, in this country, not an uncommon occurrence. Much
wine will never make a delicate lady strong—it will increase her
weakness, not her strength. Wine in excess does not strengthen, but,
on the contrary, produces extreme debility. Let this be borne in
mind, and much misery might then be averted.
113. Remember I am not objecting to a lady taking wine in
moderation—certainly not; a couple of glasses, for instance, in the
day, of either sherry or claret, might do her great good; but I do
strongly object to her drinking, as many ladies do, five or six glasses
of wine during that time. I will maintain that such a quantity is most
detrimental both to her health and to her fecundity.
114. The effect of the use of wine is beneficial; but the effect of the
abuse of it is deplorable in the extreme. Wine is an edge-tool, and
will, if not carefully handled, assuredly wound most unmercifully. I
have not the slightest doubt that the quantity of wine consumed by
many ladies is one cause, in this our day, of so much delicacy of
constitution. It is a crying evil, and demands speedy redress; and as
no more worthy medical champion has appeared in the field to fight
the battle of moderate wine-drinking, I myself have boldly come
forward to commence the affray, fervently trusting that some earnest
men may join me in the conflict. I consider that the advocates for a
plentiful supply of alcoholic stimulants are wrong, and that the
upholders of total abstinence principles are equally wrong; and that
the only path of health and of safety lies between them both—in
moderation. A teetotaller and an advocate for a plentiful supply of
alcoholic drinks are both very difficult to please; indeed, the one and
the other are most intemperate. I am aware that what I have written
will be caviled at, and will give great offense to both extreme parties;
but I am quite prepared and willing to abide the consequences, and
sincerely hope that what I have said will be the means of ventilating
the subject, which is sadly needed. It is the violence and obstinacy of
the contending parties, each of whom is partly right and partly
wrong, that have long ago prevented a settlement of the question at
issue, and have consequently been the means of causing much heart-
burning, misery, and suffering. The Times once pithily remarked that
it would be well if the two combatants were “to mix their liquors.”
115. A young wife ought to rise betimes in the morning, and after
she be once awake should never doze. Dozing is both weakening to
the body and enervating to the mind. It is a species of dram-
drinking; let my fair reader, therefore, shun it with all her might. Let
her imitate the example of the Duke of Wellington, who, whenever he
turned in bed, made a point of turning out of it; indeed, so
determined was that illustrious man not to allow himself to doze
after he was once awake, that he had his bed made so small that he
could not conveniently turn in it without first of all turning out of it.
Let her, as soon as she is married, commence early rising; let her
establish the habit, and it will for life cling to her:
“Awake! the morning shines, and the fresh field
Calls us; we lose the prime, to mark how spring
Our tender plants; how blows the citron grove,
What drops the myrrh, and what the balmy reed;
 How Nature paints her colors; how the bee
Sits on the bloom.”[22]

116. It is wonderful how much may be done betimes in the

morning. There is nothing like a good start. It makes for the
remainder of the day the occupation easy and pleasant—
“Happy, thrice happy, every one
Who sees his labor well begun,
And not perplexed and multiplied
By idly waiting for time and tide.”[23]

117. How glorious, and balmy, and health-giving, is the first breath
of the morning, more especially to those living in the country! It is
more exhilarating, invigorating, and refreshing than it is all the rest
of the day. If you wish to be strong, if you desire to retain your good
looks and your youthful appearance, if you are desirous of having a
family, rise betimes in the morning; if you are anxious to lay the
foundation of a long life, jump out of bed the moment you are awake.
Let there be no dallying, no parleying with the enemy, or the battle is
lost, and you will never after become an early riser; you will then lose
one of the greatest charms and blessings of life, and will, probably,
not have the felicity of ever becoming a mother; if you do become
one, it will most likely be of puny children. The early risers make the
healthy, bright, long-lived wives and mothers. But if a wife is to be an
early riser, she must have a little courage and determination; great
advantages in this world are never gained without; but what is either
man or woman good for if they have not those qualities?
118. An early riser ought always to have something to eat and
drink, such as a little bread and butter, and either a cup of tea or a
draught of new milk, before she goes out of a morning; this need not
interfere with, at the usual hour, her regular breakfast. If she were to
take a long walk on an empty stomach, she would for the remainder
of the day feel tired and exhausted, and she would then, but most
unfairly, fancy that early rising did not agree with her.
119. The early morning is one of the best and most enjoyable
portions of the day. There is a perfect charm in nature which early
risers alone can appreciate. It is only the early riser that ever sees
“the rosy morn,” the blushing of the sky, which is gloriously
beautiful! Nature, in the early morning, seems to rejoice and be glad,
and to pour out her richest treasures: the birds vie with each other in
their sweetest carols; the dew on the grass, like unto myriads of
diamonds, glittering and glistening, and glinting in the rays of the
sun; occasionally the cobwebs on the shrubs and bushes, like
exquisite lace sparkling with gems; the fresh and matchless perfume
and fragrance of the earth and flowers;—these, one and all, are
gloriously beautiful to behold, and can only be enjoyed to perfection
in the early morning, while the majority of people, during the
choicest periods of their existence, are sweltering, and dozing, and
deteriorating both in body and mind, on beds of down, when they
ought to be up, out, and about! Can it be wondered at, when such
weakening and enervating practices are so much in vogue—for
luxury is the curse of the day—that there are so many barren wives in
England? It looks, on the first blush, that many of the customs and
practices of the present day were to cause barrenness; for, assuredly,
if they had been instituted on purpose, they could not have
performed their task more surely and successfully.
120. It might be said that the dews of the morning are dangerous!
The dews of the early morning are beneficial to health, while the
dews of the evening are detrimental. How truly the poet sings—
“Dew-drops are the gems of morning,
But the tears of mournful eve!”[24]

121. Early rising imparts health to the frame, strength to the

muscles, and comeliness to the countenance; it clears the brain, and
thus brightens the intellect; it is a panacea for many of the ills of life,
and, unlike many panaceas, it is both simple and pleasant in its
operation; it calms the troubled breast; it gives a zest to the after-
employments and pleasures of the day; and makes both man and
woman look up from “nature’s works to nature’s God!”
122. Early rising rejuvenizes the constitution: it makes the middle-
aged look young, and the old look middle-aged; it is the finest
cosmetic in the world, and tints the cheeks with a bloom the painter
emulates, but in vain! On the other hand, late rising adds years to the
looks, fills the body with aches and pains, and the countenance with
crow-feet and wrinkles; gives a yellowness and pimples to the face,
and depression to the spirits. Aged looks and ill health invariably
follow in the wake of late rising.
 123. If a mistress rise early the servants are likely to follow suit: a
lazy mistress is almost sure to have lazy servants; the house becomes
a sluggard’s dwelling! Do not let me be misunderstood; I do not
recommend any unreasonable hours for rising in the morning; I do
not advise a wife to rise early for the sake of rising early: there would
be neither merit nor sense in it; I wish her to have her full
complement of sleep—seven or eight hours; but I do advise her to go
to bed early, in order that she might be up every morning at six
o’clock in the summer, and at seven o’clock in the winter. I maintain
that it is the duty of every wife, unless prevented by illness, to be an
early riser. This last reason should have greater weight with her than
any other that can possibly be brought forward! All things in this
world ought to be done from a sense of duty; duty ought to be a
wife’s and every other person’s pole-star!
124. There is a wonderful and glorious object in creation which
few, very few, ladies, passing strange though it be, have ever seen—
the rising of the sun! The few who have seen it are, probably, those
who have turned night into day, who are returning home in the early
morning, jaded and tired, after dancing the whole of the previous
night. These, of course, cannot enjoy, and most likely do not even
see, the magnificent spectacle!
125. I am not advising my fair reader to rise every morning with
the rising of the sun—certainly not; but if she be an early riser, she
might occasionally indulge herself in beholding the glorious sight!
126. “The top of the morning to you” is a favorite Irish salutation,
and is very expressive and complimentary. “The top of the
morning”—the early morning, the time when the sun first rises in his
majesty and splendor—is the most glorious, and health-giving, and
best part of the whole day; when nature and all created beings rejoice
and are glad:
“But mighty Nature bounds as from her birth,
The sun is in the heavens, and life on earth;
Flowers in the valley, splendor in the beam,
Health in the gale, and freshness in the stream.”[25]

127. Let a young wife, if she be anxious to have a family and

healthy progeny, be in bed betimes. It is impossible that she can rise
early in the morning unless she retire early at night. “One hour’s
sleep before midnight is worth three after.” Sleep before midnight is
most essential to health, and if to health, to beauty; hence, sleep
before midnight is called beauty-sleep. The finest cosmetic is health!
128. She ought to pay particular attention to the ventilation of her
sleeping apartment, and she herself, before leaving the chamber in
the morning, ought never to omit to open the windows; and in the
summer, if the room be large, she should during the night leave, for
about six or eight inches, the window-sash open. If the room be small
it will be desirable to have, instead of the window, the door (secured
from intrusion by a door-chain) unclosed; and to have, as well, either
the skylight or the landing window open. There ought by some
means or other, if the inmates of the room are to have sweet and
refreshing sleep, to be thorough ventilation of the sleeping
apartment. I have no patience to hear some men assert that it is
better to sleep in a close room—in a foul room! They might, with
equal truth, declare that it is desirable for a healthy person to
swallow every night a dose of arsenic in order to prolong his life!
Carbonic acid gas is as truly a poison as arsenic! If there be a
dressing-room next to the bedroom, it will be well to have the
dressing-room window, instead of the bedroom window, open at
night. The dressing-room door will regulate the quantity of air to be
admitted into the bedroom, opening it either little or much as the
weather might be cold or otherwise.[26] The idea that it will give cold
is erroneous; it will be more likely, by strengthening the system and
by carrying off the impurities of the lungs and skin, to prevent cold.
129. Some persons, accustomed all their lives to sleep in a close,
foul room—in a room contaminated with carbonic acid gas—cannot
sleep in a fresh, well-ventilated chamber, in a chamber with either
door or window open: they seem to require the stupefying effects of
the carbonic acid gas, and cannot sleep without it! If such be the
case, and as sleep is of such vital importance to the human economy,
let both window and door be closed, but do not, on any account, let
the chimney be stopped, as there must be, in a bedroom, ventilation
of some kind or another, or ill health will inevitably ensue.
130. It is madness to sleep in a room without ventilation—it is
inhaling poison; for the carbonic acid gas, the refuse of respiration,
which the lungs are constantly throwing off, is a poison—a deadly
poison—and, of course, if there be no ventilation, a person must
breathe this carbonic acid gas mixed with the atmospheric air. Hence
the importance, the vital importance, of either an open chimney or of
an open window, or of both. The chimney, then, even if the window
be closed, ought never to be stopped; and the window, either of the
bedroom or of the dressing-room, should not be closed, even in the
night, unless the weather be either very wet or bitterly cold. I should
strongly recommend my fair reader, and, indeed, every one else, to
peruse the good and talented Florence Nightingale’s Notes on
Nursing. They ought to be written in letters of gold, and should be
indelibly impressed on the memory of every one who has the interest
of human life and happiness at heart. Florence Nightingale declares
that no one, while in bed, ever catches cold from proper ventilation.
I believe her; and I need not say that no one has had more experience
and better opportunities of judging about what she writes than this
accomplished authoress.
131. I fearlessly assert that no one can sleep sweetly and
refreshingly unless there be thorough ventilation of the chamber.
She may have, in an unventilated apartment, heavy, drowsy,
deathlike sleep, and well she might! She is under the stupefying
effects of poison; the carbonic acid gas, which is constantly being
evolved from the lungs, and which wants a vent, but cannot obtain it,
is, as I have before remarked, a deadly poison! She may as well take
every night a stupefying opiate, as breathe nightly a bedroom
charged with carbonic acid gas; the one would, in the long run, be as
pernicious as the other. To show the power of carbonic acid gas in
sending people to sleep, we have only to notice a crowded church of
an evening; when, even if the preacher be an eloquent man, the
majority of the congregation is fast asleep,—is, in point of fact, under
the soporific influence of the carbonic acid gas, the church being at
the time full of it. Carbonic acid gas is as certain, if not more certain,
to produce a heavy deathlike slumber as either numbing opium or
drowsy poppy.
132. I moreover declare that she cannot have sweet refreshing
sleep at night unless during the day she take plenty of exercise, and
unless she has an abundance of active, useful occupation.
133. Occupation—active, useful occupation—is the best composing
medicine in the world; and the misfortune of it is that the wealthy
have little or no occupation to cause them to sleep. Pleasure they
have in abundance, but little or no real occupation. “The sleep of a
laboring man is sweet, whether he eat little or much: but the
abundance of the rich will not suffer them to sleep.”[27]
134. Sleep is of more consequence to the human economy than
food. Nothing should therefore be allowed by a young wife to
interfere with sleep. And as the attendance on large assemblies, balls,
and concerts sadly, in every way, interfere with sleep, they ought, one
and all, to be sedulously avoided.
135. As exercise is very conducive and provocative of sleep—sound,
sweet, childlike sleep—exercise must be practiced, and that not by
fits and starts, but regularly and systematically. She ought, then,
during the day, with exercise and with occupation, to tire herself, and
she will then have sweet and refreshing sleep. But some ladies never
do tire themselves except with excitement; they do not know what it
is to be tired either by a long walk or by household work. They can
tire themselves with dancing at a ball; poor fragile creatures can
remain up the whole night waltzing, quadrilling, and galloping, but
would be shocked at the idea and at the vulgarity of walking a mile at
a stretch! Poor creatures, they are to be pitied; and, if they ever
marry, so are their husbands. Are such wives as these likely to be
mothers, and if they are, are their offspring likely to be strong? Are
such wives as these likely to be the mothers of our future warriors, of
our future statesmen, and of our other worthies—men of mark, who,
“Departing, leave behind them
Footprints on the sands of time!”

136. Sleep is the choicest gift of God. Sleep is a comforter, a solace,

a boon, a nourisher, a friend. Happy, thrice happy, is a wife who can
sleep like unto a little child! When she is well, what a comfort is
sleep; when she is ill, what a soother of pain is sleep; when she is in
trouble, what a precious balm is sleep!
137. Hear what our noblest poet, Shakspeare, says of sleep:
“Sleep that knits up the raveled sleeve of care,
The death of each day’s life, sore labor’s bath,
Balm of hurt minds, great nature’s second course
Chief nourisher in life’s feast.”
 138. A luxurious, idle wife cannot sleep; she, night after night,
tumbles and tosses on her bed of down. What has she done during
the day to tire herself, and thus to induce sleep? Alas, nothing! She in
consequence never experiences
“Tired nature’s sweet restorer, balmy sleep.”

For, after all, out-door exercise and useful occupation are the best
composing medicines in the world! Encompassed as she is with every
luxury—partaking of all the delicacies of the season, of the richest
viands, and of the choicest wines—decked out in costly apparel—
reclining on the softest cushions—surrounded with exquisite scenery,
with troops of friends, and with bevies of servants;—yet,
notwithstanding all these apparent advantages, she is oftentimes one
of the most debilitated, complaining, “nervous,” hysterical and
miserable of mortals. The causes of all these afflictions are—she has
nothing to do; she is overwhelmed with prosperity; she is like a fire
that is being extinguished in consequence of being overloaded with
fuel; she is being killed with overmuch kindness; she is a drone in a
hive where all must work if they are to be strong and well, and bright
and cheerful; for labor is the lot of all and the law for all, for “God is
no respecter of persons.” The remedies for a lady affected as above
described are simple and yet efficacious—namely, simplicity of living,
and an abundance of out-door exercise and of useful occupation. It
would have been to the manifest advantage of many a fair dame if
she were obliged to put down her close carriage, and were compelled
to walk instead of drive. Riding in close carriages nurse many
ailments which walking would banish; a brisk walk is the best tonic
and the most reviving medicine in the world, and would prevent the
necessity of her swallowing so much nauseous physic. Nature’s
simple remedies are oftentimes far superior and far more agreeable
than any to be found in the Pharmacopœia. It would have been a
blessing to many a rich, indolent, and luxurious lady if she had been
born in a lower rank—in one in which she had been compelled to
work for her daily bread; if she had been, she would, in many
instances, have been far happier and healthier than she now is.
Indolence and luxury kill more than hard work and hard fare ever
did or ever will kill. Indolence and luxury are slow poisons; they
destroy by degrees, but are in the end as certain in their deleterious
effects as either arsenic or deadly nightshade—
“Come hither, ye that press your beds of down,
And sleep not; see him sweating o’er his bread
Before he eats it. ’Tis the primal curse,
But softened into mercy—made the pledge
Of cheerful days, and nights without a groan.”

139. I must not forget to speak of the paramount importance in a

dwelling of an abundance of light—of daylight. Light is life, light is
health, light is a physician! Light is life: the sun gives life as well as
light; if it were not for the sun, all creation would wither and die.
There is “no vitality or healthful structure without light.”[28] Light is
health: it strengthens the frame, it cheers the heart, and tints the
cheeks with a roseate hue! Light is a physician: it drives away many
diseases, as the mists vanish at the approach of the sun; and it cures
numerous ailments which drugs alone are unable to relieve.
140. Look at the bloom on the face of a milkmaid! What is it that
tints her cheeks? An abundance of light. Behold the pallid, corpselike
countenance of a factory girl! What blanches her cheek? The want of
light, of air, and of sunshine.
141. A room, then, ought to have large windows in order that the
sun might penetrate into every nook and corner of the apartment. A
gardener thoroughly appreciates the importance of light to his
flowers; he knows, also, that if he wishes to blanch some kinds of
vegetables—such as celery and sea-kale—he must keep the light from
them; and if my fair reader desires to blanch her own cheeks, she
ought to keep the light from them; but, on the other hand, if she be
anxious to be healthy and rosy, she must have plenty of light in her
dwelling.
142. The want of light stunts the growth, dims the sight, and
damps the spirits. Colliers, who a great part of their lives live in the
bowels of the earth, are generally stunted; prisoners, confined for
years in a dark dungeon, frequently become blind; people who live in
dark houses are usually melancholic.
143. Light banishes from rooms foulness, fustiness, mustiness, and
smells. Light ought therefore to be freely allowed to enter every
house, and be esteemed as the most welcome of visitors. Let me then
advise every young wife to admit into her dwelling an abundance of
light, of air, and of sunshine.
144. Some ladies, to keep off the sun, to prevent it from fading the
furniture, have, in the summer time, all the blinds of the windows of
the house down. Hence they save the fading of their furniture, and,
instead of which, they fade their own and their children’s cheeks.
Many houses, with all their blinds down, look like so many prisons,
or as if the inmates were in deep affliction, or as if they were
performing penance; for is it not a penance to be deprived of the
glorious light of day, which is as exhilarating to the spirits as, and
much more beneficial than, a glass of champagne?
145. It is a grievous sin to keep out from a dwelling the glorious
sunshine. We have heard of “a trap to catch a sunbeam:” let the open
windows be a trap, and a more desirable prize cannot be caught than
a sunbeam. Sunbeams, both physical and metaphorical, make a
house a paradise upon earth!
146. Let me strongly caution the newly made wife against the evil
effects of tight lacing. The waist ought, as a rule, to be from twenty-
seven to twenty-nine inches in circumference; if, therefore, she bind
and gird herself in until she be only twenty-three inches, and, in
some cases, until she be only twenty-one inches, it must be done at
the expense of comfort, of health, and happiness. If stays be worn
tightly, they press down the contents of the lower part of the belly,
which might either prevent a lady from having a family, or might
produce a miscarriage.[29]
147. Let her dress be loose, and be adapted to the season. She
ought not to adopt the fashion of wearing in the morning warm
clothes with long sleeves, and in the evening thin dresses with short
sleeves. “It is hopeless to battle with fashion in matters of dress;
women will never believe that their bonnets, neck-wrappers, or huge
petticoats (until they go out of fashion) can have anything to do with
headaches, sore throats, or rheumatism; but they ought to know that
the more they swathe themselves, the more tender and delicate they
are likely to be. If they wish to withstand cold, they should accustom
themselves to bear it.”[30]
148. If a young wife be delicate, and if her circulation be languid, a
flannel vest next the skin, and in the daytime, should, winter and
summer, be worn. Scarlet is, in such a case, a favorite color, and may
be selected for the purpose.
149. It is important that it should be borne in mind that the
wearing of flannel next the skin is more necessary in the summer
than in the winter time. A lady in the summer is apt, when hot, either
from the weather or from exertion, to get into a draught to cool
herself, and not wearing flannel next the skin, she is almost sure at
such times to catch a cold. Now, flannel being a bad conductor of
heat, keeps the body at a tolerably equal temperature, and thus
materially lessens the risk. When it is considered that many of the
diseases afflicting humanity arise from colds, the value of wearing
flannel next the skin as a preventive is at once apparent.
150. Never was there such a time as the present when dress was so
much thought of. Grand dresses now sweep our dirty streets and
thoroughfares; rich velvets, silks, and satins are as plentiful as dead
leaves in autumn. “There is so much to gaze and stare at in the dress,
one’s eyes are quite dazzled and weary, and can hardly pierce
through to that which is clothed upon.” Dress is become a crying evil;
many ladies clothe themselves in gorgeous apparel at the expense of
household comforts, and even of household necessaries:
“We sacrifice to dress, till household joys
And comforts cease. Dress drains our cellars dry,
And keeps our larder lean—puts out our fires,
And introduces hunger, frost, and woe,
Where peace and hospitality might reign.”[31]

151. It might be said, What has all this to do with the health of a
wife? I reply, much. The customs, habits, and luxuries of the present
day are very antagonistic both to health and fecundity.
152. She must not coddle, nor should she muffle up her throat with
furs. Boas are the most frequent cause of sore throats and quinsies,
and therefore the sooner they are discarded the better. “And this is
perfectly true, though few seem to be aware of the fact. Relaxed
throats would be rare if cold water was more plentifully used, both
externally and internally, and mufflers were laid aside.”[32]
153. If my gentle reader will freely use cold water ablutions, she
will find that she will not require nearly so much clothing and
muffling up. It is those who use so little water who have to wear so
much clothing; and the misfortune of it is, the more clothes they
wear the more they require. Many young people are wrapped and
muffled up in the winter time like old folks, and by coddling they
become prematurely old—frightened at a breath of air and at a
shower of rain, and shaking in their shoes at an easterly wind!
Should such things be?
154. Pleasure, to a certain degree, is as necessary to the health of a
young wife, and every one else, as the sun is to the earth—to warm, to
cheer, and to invigorate it, and to bring out its verdure. Pleasure, in
moderation, rejuvenizes, humanizes, and improves the character,
and expands and exercises the good qualities of the mind; but, like
the sun, in its intensity it oppresseth drieth up, and withereth.
Pleasures kept within due bounds are good, but in excess are utterly
subversive of health and happiness. A wife who lives in a whirl of
pleasure and excitement is always weakly and “nervous,” and utterly
unfitted for her duties and responsibilities.
155. Let the pleasures of a newly-married wife, then, be dictated by
reason, and not by fashion. She ought to avoid all recreations of an
exciting kind, as depression always follows excitement. I would have
her prefer the amusements of the country to those of the town, such
as a flower-garden, botany, archery, croquet, bowls,—everything, in
fact, that will take her into the open air, and will cause her to
appreciate the pure, simple, and exquisite beauties of nature.
Croquet I consider to be one of the best games ever invented: it
induces a lady to take exercise which perhaps she would not
otherwise do; it takes her into the open air, it strengthens her
muscles, it expands her chest, it promotes digestion, it circulates her
blood, and it gives her an interest in the game which is most
beneficial both to mind and body.
156. Oh, that my countrywomen should prefer the contaminated
and foul air of ball and of concert-rooms, to the fresh, sweet, and
health-giving air of the country!
157. Let me in this place enter my strong protest against a young
wife dancing, more especially if she be enceinte. If she be anxious to
have a family, it is a most dangerous amusement, as it is a fruitful
source of miscarriage; and the misfortune is, that if she once have a
miscarriage, she might go on again and again, until her constitution