(eBook PDF) Knowing the Odds An Introduction to

Probability pdf download

https://ebooksecure.com/product/ebook-pdf-knowing-the-odds-an-
introduction-to-probability/

Download more ebook from https://ebooksecure.com

We believe these products will be a great fit for you. Click
the link to download now, or visit ebooksecure.com
to discover even more!

(eBook PDF) An Introduction to Probability and

Statistical Inference 2nd Edition

http://ebooksecure.com/product/ebook-pdf-an-introduction-to-
probability-and-statistical-inference-2nd-edition/

(eBook PDF) Introduction to Probability, Second Edition

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-second-edition/

(eBook PDF) Introduction to Probability Models 12th

Edition

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-models-12th-edition/

(eBook PDF) Introduction to Probability by Mark Ward

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-by-mark-ward/
(eBook PDF) Introduction to Probability and Statistics
15th Edition

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-and-statistics-15th-edition/

Introduction to Probability and Statistics 14th Edition

(eBook PDF)

http://ebooksecure.com/product/introduction-to-probability-and-
statistics-14th-edition-ebook-pdf/

(eBook PDF) Introduction to Probability and Statistics

4rd Canadian Edition

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-and-statistics-4rd-canadian-edition/

(eBook PDF) Introduction to Probability and Statistics

3rd by William Mendenhall

http://ebooksecure.com/product/ebook-pdf-introduction-to-
probability-and-statistics-3rd-by-william-mendenhall/

(eBook PDF) The Atmosphere An Introduction to

Meteorology 13th

http://ebooksecure.com/product/ebook-pdf-the-atmosphere-an-
introduction-to-meteorology-13th/
 Knowing the Odds
An Introduction to
Probability

John B. Walsh

Graduate Studies
in Mathematics
Volume 139

American Mathematical Society

Contents

Preface xi
Introduction xiii

Chapter 1. Probability Spaces 1

§1.1. Sets and Sigma-Fields 1
§1.2. Elementary Properties of Probability Spaces 6
§1.3. The Intuition 8
§1.4. Conditional Probability 15
§1.5. Independence 18
§1.6. Counting: Permutations and Combinations 22
§1.7. The Gambler’s Ruin 30

Chapter 2. Random Variables 39

§2.1. Random Variables and Distributions 39
§2.2. Existence of Random Variables 45
§2.3. Independence of Random Variables 48
§2.4. Types of Distributions 49
§2.5. Expectations I: Discrete Random Variables 54
§2.6. Moments, Means and Variances 60
§2.7. Mean, Median, and Mode 63
§2.8. Special Discrete Distributions 65
Chapter 3. Expectations II: The General Case 75
§3.1. From Discrete to Continuous 75

vii
viii Contents

§3.2. The Expectation as an Integral 81

§3.3. Some Moment Inequalities 85
§3.4. Convex Functions and Jensen’s Inequality 86
§3.5. Special Continuous Distributions 89
§3.6. Joint Distributions and Joint Densities 96
§3.7. Conditional Distributions, Densities, and Expectations 103
Chapter 4. Convergence 117
§4.1. Convergence of Random Variables 117
§4.2. Convergence Theorems for Expectations 122
§4.3. Applications 127
Chapter 5. Laws of Large Numbers 133
§5.1. The Weak and Strong Laws 134
§5.2. Normal Numbers 137
§5.3. Sequences of Random Variables: Existence 140
§5.4. Sigma Fields as Information 142
§5.5. Another Look at Independence 144
§5.6. Zero-one Laws 145
Chapter 6. Convergence in Distribution and the CLT 151
§6.1. Characteristic Functions 151
§6.2. Convergence in Distribution 162
§6.3. Lévy’s Continuity Theorem 170
§6.4. The Central Limit Theorem 176
§6.5. Stable Laws 182
Chapter 7. Markov Chains and Random Walks 191
§7.1. Stochastic Processes 191
§7.2. Markov Chains 192
§7.3. Classification of States 201
§7.4. Stopping Times 204
§7.5. The Strong Markov Property 208
§7.6. Recurrence and Transience 211
§7.7. Equilibrium and the Ergodic Theorem for Markov Chains 218
§7.8. Finite State Markov Chains 226
§7.9. Branching Processes 234
§7.10. The Poisson Process 242
Contents ix

§7.11. Birth and Death Processes 250

Chapter 8. Conditional Expectations 265
§8.1. Conditional Expectations 265
§8.2. Elementary Properties 268
§8.3. Approximations and Projections 272
Chapter 9. Discrete-Parameter Martingales 275
§9.1. Martingales 275
§9.2. System Theorems 282
§9.3. Convergence 290
§9.4. Uniform Integrability 295
§9.5. Applications 304
§9.6. Financial Mathematics I: The Martingale Connection 315
Chapter 10. Brownian Motion 335
§10.1. Standard Brownian Motion 336
§10.2. Stopping Times and the Strong Markov Property 344
§10.3. The Zero Set of Brownian Motion 348
§10.4. The Reflection Principle 351
§10.5. Recurrence and Hitting Properties 352
§10.6. Path Irregularity 354
§10.7. The Brownian Infinitesimal Generator 359
§10.8. Related Processes 363
§10.9. Higher Dimensional Brownian Motion 368
§10.10. Financial Mathematics II: The Black-Scholes Model 374
§10.11. Skorokhod Embedding 377
§10.12. Lévy’s Construction of Brownian Motion 388
§10.13. The Ornstein-Uhlenbeck Process 390
§10.14. White Noise and the Wiener Integral 394
§10.15. Physical Brownian Motion 404
§10.16. What Brownian Motion Really Does 410
Bibliography 413
Index 415
Preface

In the long-forgotten days of pre-history, people would color peach pits dif-
ferently on the two sides, toss them in the air, and bet on the color that
came up. We, with a more advanced technology, toss coins. We flip a coin
into the air. There are only two possible outcomes, heads or tails, but until
the coin falls, we have no way of knowing which. The result of the flip may
decide a bet, it may decide which football team kicks off, which tennis player
serves, who does the dishes, or it may decide a hero’s fate.
The coin flip may be the most basic of all random experiments. If the
coin is reasonably well-made, heads is as likely as tails to occur. But. . . what
does that mean?
Suppose we flip a coin, and call “Heads” or “Tails” while it is in the
air. Coins are subject to the laws of physics. If we could measure the exact
position, velocity, and angular velocity of the coin as it left the hand—
its initial conditions—we could use Newton’s laws to predict exactly how it
would land. Of course, that measurement is impractical, but not impossible.
The point is that the result is actually determined as soon as the coin is in
the air and, in particular, it is already determined when we call it; the result
is (theoretically) known, but not to us. As far as we are concerned, it is just
as unpredictable as it was before the flip. Let us look at the physics to see
why.
The outcome is determined by the exact position, angular position, ve-
locity, and angular velocity at the time of the flip. Physicists represent these
all together as a point in what they call phase space. We can picture it as
follows.

xi
xii Preface

T
H
T

H H
T T T T

H H H H

T T T

H H H

Figure 1. Phase space

This represents the initial condition of the coin in phase space. Some
points lead to heads, some to tails. But a small difference in initial conditions
completely changes the result. The conditions leading to heads are a union
of very small regions, which are evenly mixed up with those leading to tails.
This means that no matter how we try to toss the coin, we cannot zero
in on a particular result—our toss will be smeared out, so to speak, over the
“Heads” and “Tails” regions, and this will happen no matter how carefully
we toss it. This leads us to say things like: “Heads and tails are equally
likely,” or “Heads and tails each have probability one-half.”
Philosophers ask deep questions about the meaning of randomness and
probability. Is randomness something fundamental? Or is it just a measure
of our ignorance? Gamblers just want to know the odds.
Mathematicians by and large prefer to duck the question. If pressed,
they will admit that most probability deals with chaotic situations, like the
flip of a coin, where the seeming randomness comes from our ignorance of the
true situation. But they will then tell you that the really important thing
about randomness is that it can be measured—for probabilities measure
likelihood—and that we can construct a mathematical model which enables
us to compute all of the probabilities, and that, finally, this model is the
proper subject of study.
So you see, mathematicians side with the gamblers: they just want to
know the odds.
From now on, probability is mathematics. We will be content just to
note that it works—which is why so few casino owners go broke—and we
will leave the deeper meanings of randomness to the philosophers.
Introduction

There is an order to chaos. Unpredictability is predictable. In fact, random-

ness itself is so regular that we can assign a number to a random occurrence
which tells us in a precise way how likely it is. The number is called its
probability.
That is not to say that we can predict the result of a single toss of a fair
coin. We cannot. But we can predict that between forty and sixty out of a
hundred tosses will be heads. We might—rarely—be wrong about that, but
only once or twice in a hundred tries, and if we continue to toss: a thousand
times, a million times, and so on, we can be sure that the proportion of
heads will approach 1/2.
So randomness has its own patterns. Our aim is to understand them.
Probability is a rather unusual part of mathematics. While its full birth
as a mathematical subject can be traced to the correspondence between
Fermat and Pascal1 in the summer of 1654, the subject wasn’t put on a
rigorous footing until 1934, 270 years later, when A. N. Kolmogorov showed
it was properly a part of measure theory2. But probability had been around
for several centuries before measure theory existed, and it is quite possible to
study the subject without it. In fact, probability is taught at many different

1
Pascal and Fermat were by no means the first to study probabiity, but their work on the
“problem of points” was so much deeper than what had gone before that it is properly considered
the true beginning of the subject. See Keith Devlin’s “The Unfinished Game” [13] for an account.
2
See [22] for an English translation of Kolmogorov’s landmark paper. It showed that all of
probability theory could be regarded as a part measure theory, giving a general existence theorem
for stochastic processes (not present, alas, in this book, but see [12] or [9]) and a rigorous definition
of conditional expectations (see Chapter 8), which had previously been confined to special cases.
This was quite a change from the more intuitive approach, and it took some time to replace “could
be taken” by “is.” That was completed by Doob, culminating in his seminal book Stochastic
Processes [12].

xiii
xiv Introduction

levels, according to the mathematics the students know: in elementary and

high school, first year college, third or fourth year college, as well as in
graduate school. Certain things are common to all of these courses, but the
the more mathematics the student knows, the deeper he or she can go. This
particular text is drawn from a two-semester course taught over the years
at the University of British Columbia, mainly to fourth-year mathematics
honors students. It assumes the student is familiar with calculus and knows
some analysis, but not measure theory. Many of the students, but by no
means all, take a concurrent course in Lebesgue measure. It is not necessary,
but it adds depth, and gives the student some “Aha!” moments, such as
the sudden realization: “Aha! The expectation is nothing but a Lebesgue
integral3!”
We begin with the basic axioms of probability, and the all-important
ideas of conditional probability and independence. Then we quickly develop
enough machinery to allow the students to solve some interesting problems
and to analyze card games and lotteries. Just to show how quickly one can
get into non-trivial questions, we work out the problem of the gambler’s
ruin.
The systematic study of classical probability begins in Chapter Two. Its
aim is to prove two of the basic classical theorems of the subject: the law
of large numbers and the central limit theorem. Far from being recondite,
these theorems are practically part of Western folklore. Who has not heard
of the law of averages? That is another name for the law of large numbers.
What student has not been subject to “grading on a curve”, a direct (and
often mistaken) application of the central limit theorem? It is surprising how
much of the curriculum is determined by the modest aim of understanding
those two results: random variables, their expectations and variances, their
distributions, the idea of independence, and the ideas of convergence are
needed merely to state the theorems. A number of inequalities, the theory
of convergence in distribution, and the machinery of characteristic functions,
are necessary to prove them. This, along with enough examples to supply
the intuition necessary to understanding, determines the first six chapters.
The second part of the book introduces stochastic processes, and changes
the viewpoint. Stochastic processes evolve randomly in time. Instead of
limit theorems at infinity, the emphasis is on what the processes actually
do; we look at their sample paths, study their dynamics, and see that many
interesting things happen between zero and infinity. There is a large se-
lection of stochastic processes to study, and too little time to study them.

3
On the other hand, students who take probability before measure theory have their “Aha!”
moment later, when they realize that the Lebesgue integral is nothing but an expectation.
Introduction xv

We want to introduce processes which are major building blocks of the the-
ory, and we aim the course towards Brownian motion and some of its weird
and wonderful sample path properties. Once more, this determines much
of the curriculum. We introduce the Markov property and stopping times
with a study of discrete-parameter Markov chains and random walks, in-
cluding special cases such as branching processes. Poisson and birth and
death processes introduce continuous parameter processes, which prepares
for Brownian motion and several related processes.
The one non-obvious choice is martingales. This deserves some expla-
nation. The subject was once considered esoteric, but has since shown itself
to be so useful4 that it deserves inclusion early in the curriculum. There are
two obstructions. The first is that its whole setting appears abstract, since
it uses sigma-fields to describe information. Experience has shown that it
is a mistake to try to work around this; it is better to spend the necessary
time to make the abstract concrete by showing how sigma-fields encode in-
formation, and, hopefully, make them intuitive. The second obstruction is
the lack of a general existence theorem for conditional expectations: that
requires mathematics the students will not have seen, so that the only case
in which we can actually construct conditional expectations is for discrete
sigma-fields, where we can do it by hand. It would be a pity to restrict
ourselves to this case, so we do some unashamed bootstrapping. Once we
show that our hand-constructed version satisfies the defining properties of
the general conditional expectation, we use only these properties to develop
the theory. When we have proved the necessary martingale theorems, we
can construct the conditional expectation with respect to a general sigma
field as the limit of conditional expectations on discrete sigma fields. This
gives us the desired existence theorem . . . and shows that what we did was
valid for general sigma-fields all along. We make free use of martingales in
the sequel. In particular, we show how martingale theory connects with a
certain part of mathematical finance, the option pricing, or Black-Scholes
theory.
The final chapter on Brownian motion uses most of what we have learned
to date, and could pull everything together, both mathematically and artis-
tically. It would have done so, had we been able to resist the temptation
to spoil any possible finality by showing—or at least hinting at—some of

4
The tipping point was when engineers started using martingales to solve applied problems,
and, in so doing, beat the mathematicians to some very nice theorems. The coup de grâce was
struck by the surprising realization that the celebrated Black-Scholes theory of finance, used by all
serious option-traders in financial markets was, deeply, martingale theory in disguise. See sections
9.6 and 10.10
xvi Introduction

the large mathematical territory it opens up: white noise, stochastic inte-
grals, diffusions, financial mathematics, and probabilistic potential theory,
for example.
A last word. To teach a course with pleasure, one should learn at the
same time. Fair is fair: the students should not be the only learners. This
is automatic the first time one teaches a course, less so the third or fourth
time. So we tried to include enough sidelights and interesting byways to
allow the instructor some choice, a few topics which might be substituted at
each repetition. Most of these are starred:  . In fact, we indulged ourselves
somewhat, and included personal favorites that we seldom have time to cover
in the course, such as the Wiener stochastic integral, the Langevin equation,
and the physical model of Brownian motion.
 Chapter 1

Probability Spaces

It is said that the best way to enter a cold swimming pool is to dive in head
first, not to inch in one toe at a time. Let us take that advice and begin
with a splash: the basic mathematical model of probability. We will explain
the intuition behind it afterwards.

1.1. Sets and Sigma-Fields

Let Ω be a set.
Definition 1.1. A class G of subsets of Ω is a field if
(i) ∅ ∈ G,
(ii) A ∈ G =⇒ Ac ∈ G,
(iii) A, B ∈ G =⇒ A ∪ B ∈ G;
it is a σ-field if
(i) ∅ ∈ G,
(ii) A ∈ G =⇒ Ac ∈ G,
∞
(iii) A1 , A2 , · · · ∈ G =⇒ i=1 Ai ∈ G;
it is a monotone class if
∞
(i) A1 ⊂ A2 ⊂ · · · ∈ G =⇒ Ai ∈ G,
i=1
∞
(ii) A1 ⊃ A2 ⊃ · · · ∈ G =⇒ i=1 Ai ∈ G.

Thus a field1 is closed under complementation and finite unions, a σ-

field is closed under complementation and countable unions, and a monotone
1
You might encounter these under other names. A field has also been called an algebra, a
Boolean algebra, and a finitely additive class. A σ-field—pronounced sigma-field—has been called
a σ-algebra, a Borel field, and a countably additive class.

1
2 1. Probability Spaces

class is closed under countable increasing unions and countable decreasing

intersections.
For our purposes, the σ-field is by far the most useful. This is because
it is closed under countable set operations. Finite unions and intersections
are not enough. We will be dealing with limits, and limits involve infinite
operations. However, in practice, the field is the easiest to come by. Once
we have a field, we often ask, “Is it actually a σ-field? If not, can we extend
it to be one?” The monotone class is an important tool for this: it is easier
to show that something is a monotone class than to show it is a σ-field.
Under certain conditions, this implies that it is a σ-field.
Remark 1.2. (i) Ac is the complement of A relative to Ω: Ac = Ω − A.
(ii) Note that a field or σ-field contains Ω as well as ∅, since Ω = ∅c .
(iii) A field (resp. σ-field) F is closed under  finite (resp.
 c countable)
c
intersections, too. For if A1 , A2 , · · · ∈ F , then n An = n A n ∈ F by
(ii) and (iii).

There are some obvious relations between the three:

(i) σ-field =⇒ field and monotone class.
(ii) Field and monotone class =⇒ σ-field.
 (iii) Let {G α , α ∈ I} be fields (resp. σ-fields, monotone classes). Then
α G α is a field (resp. σ-field, monotone class.)

(iv) If C is a collection of sets, there exists a minimal field (resp σ-

field, monotone class) containing C. This is called the field (resp. σ-field,
monotone class) generated by C.

Proof. (i) is obvious. For (ii), if G is a field, it is closed under complemen-

tation and finite unions, and if A1 , A2 , · · · ∈ G, then
   n

An = Aj ∈ G .
n n j=1

Indeed, the finite unions are in G since G is a field, so a countable union

can be written as a union of increasing sets, and hence is also in G because G
is a monotone class, proving (ii). And (iii) is easy! Just verify the defining
conditions.
To see (iv), note that if C is any class of subsets of Ω, there is at least
one field (resp. σ-field, monotone class) containing it, namely the class of
all subsets of Ω. Thus the intersection of all fields (resp. σ-field, monotone
classes) containing C makes sense, and, by (iii), is itself a field (resp. σ-field,
monotone class). It is necessarily minimal. 
1.1. Sets and Sigma-Fields 3

Example 1.2.1. Here are some σ-fields.

• All subsets of Ω.
• {Ω, ∅}. (This is called the trivial σ-field.)
• All countable subsets of R together with their complements.
• The Lebesgue-measurable subsets of R.
The first two are clearly σ-fields, for (i), (ii) and (iii) are immediate.
Note that every σ-field contains both ∅ and Ω, so the trivial σ-field is also
the smallest possible σ-field.
The third example contains the empty set and is clearly closed under
complements. Inaddition, it contains countable unions: if A1 , A2 , . . . are all
countable, so is n An since a countable union of countable sets is countable.

If at least one of the An is the complement of a countable set, so is n An .
In either case, it is closed under countable unions, and therefore a σ-field.
The last example—the Lebesgue-measurable sets—is for the benefit of
those who are familiar with Lebesgue measure, where they are defined and
shown to form a σ-field. It is also proved there that there are non-Lebesgue-
measurable sets, so that the first and fourth examples are actually different2.
Exercise 1.1. Show that the following two classes are fields, but not σ-fields.
(a) All finite subsets of R together with their complements.
(b) All finite unions of intervals in R of the form (a, b], (−∞, a], and (b, ∞).

Let us take a close look at one σ-field.

Definition 1.3. The class B of Borel sets in R is the smallest σ-field which
contains all open sets.

This is a natural class of subsets of R. While it is smaller than the class

of Lebesgue sets, it contains nearly any set we can think of.
We will only be concerned with the Borel sets of the line for the time
being, but let us note for future reference that they can be defined for any
topological space: if Z is a topological space, the class of Borel sets B(Z)
of Z is the smallest σ-field of subsets of Z which contain all open sets. In
particular, the Borel sets B(Rn ) can be defined for n-dimensional Euclidean
space.
They can be generated by far smaller classes than the open sets. For
instance, an open set is a countable union of open intervals, so that any σ-
field containing the open intervals also contains all open sets, and therefore
contains B, which is the smallest σ-field containing the open sets. So the
2
That is, they are different if you believe in the Axiom of Choice: the existence of non-
Lebesgue measurable sets is equivalent to the Axiom of Choice.
4 1. Probability Spaces

open intervals generate the Borel sets. Here are a few more possibilities.
Note that to prove that a class of sets generates the Borel sets, we need only
show they generate the open intervals.
Exercise 1.2. Show that the Borel sets are generated by any one of the following
classes:
(a) The closed sets.
(b) All closed intervals.
(c) All closed intervals with rational end points.
(d) All intervals of the form (a, b].
(e) All intervals of the form (−∞, x].
(f) All intervals of the form (−∞, x], where x is rational.

We say two sets A and B are disjoint if A ∩ B = ∅.

Definition 1.4. Let F be a σ-field. A probability measure P on F is a
real-valued function defined on F such that
(i) if A ∈ F , P {A} ≥ 0;
(ii) P {Ω} = 1;
(iii) if A1 , A2 , . . . is a finite or countably infinite sequence of disjoint
elements of F (i.e., i = j =⇒ Ai ∩ Aj = ∅), then
 
P An = P {An }.
n n
P {A} is called the probability of A.

Property (iii) is called countable additivity. Note that it holds for

infinite unions, not just finite ones. We will see shortly that this is really a
continuity property in disguise.

The Monotone Class Theorem . This section states and proves the
Monotone Class Theorem, and can be safely skipped for the moment. We
will not use it until Section 2.3. It is one of those theorems which seems very
technical. . . until it’s needed to make some otherwise-painful proof easy.
Theorem 1.5 (Monotone Class Theorem). Let F 0 be a field and G a mono-
tone class. Suppose F 0 ⊂ G. Then σ{F 0 } ⊂ G. In particular, the monotone
class and the σ-field generated by F are the same.

Proof. This proof makes extensive use of minimality. Let G  be the smallest
monotone class containing F 0 . Then G  ⊂ G. Since σ(F 0 ) is also a monotone
class containing F 0 , G  ⊂ σ(F 0 ). We will show σ(F 0 ) = G  .
Define two classes of subsets of Ω:
C 1 = {A ∈ G  : A ∩ B ∈ G  , ∀B ∈ F 0 } ,
1.1. Sets and Sigma-Fields 5

C 2 = {A ∈ G  : A ∩ B ∈ G  , ∀B ∈ G  } .
Let A1 , A2 , · · · ∈ C 1 , (resp. C 2 ) and B ∈ F 0 (resp. B ∈ G  ). If A1 ⊃
A2 ⊃ . . . , then
A1 ∩ B ⊃ A2 ∩ B ⊃ . . . ↓ An ∩ B ∈ G 
n
since An ∩ B ∈ G  and G  is a monotone class. Similarly, if A1 ⊂ A2 ⊂ . . . ,
then 
A1 ∩ B ⊂ A2 ∩ B ⊂ . . . ↑ An ∩ B ∈ G 
n
for the same reason. Thus C 1 (resp. C 2 ) is a monotone class. Moreover, C 1
contains F 0 , since F 0 is already closed under intersections, so that F 0 ⊂
C 1 ⊂ G  . But G  is the minimal monotone class containing F 0 , hence C 1 = G  .
Now C 2 ⊂ G  by definition, and F 0 ⊂ C 2 . Indeed, if B ∈ F 0 and
A ∈ G  = C 1 , then A ∩ B ∈ G  =⇒ B ∈ C 2 . By minimality again, C 2 = G  .
Therefore, G  contains F 0 and is closed under finite intersections.
Let us check complementation. Let C 3 = {A ∈ G  : Ac ∈ G  }. Then
F 0 ⊂ C 3 . If A1 ⊂ A2 ⊂ · · · ∈ C 3 , then
∞
c
Aj = Acj .
j=1 j
∞
But Ac1 ⊃ Ac2 ⊃ · · · ∈ G  , and G  is a monotone class, so that j=1 Aj ∈

G  =⇒ ∞ j=1 Aj ∈ C 3 . Similarly, if A1 ⊃ A2 ⊃ · · · ∈ C 3 , then
∞ 
c
Aj = Acj .
j=1 j

But Ac1 ⊂ Ac2 ⊂ · · · ∈ G  and G  is a monotone class, so that ∞ j=1 Aj ∈

G  =⇒ ∞ j=1 Aj ∈ C 3 . Therefore C 3 is a monotone class containing F 0 .
Since C 3 ⊂ G  , and G  is minimal, C 3 = G  . Therefore, G  is closed under
complementation.
Thus G  is closed under finite intersections and complementation and
therefore it is also closed under finite unions, so it is a field. But it is a
monotone class, so it is also a σ-field. This proves the first conclusion of the
theorem.
The second conclusion is almost immediate. Since σ(F 0 ) is the smallest
σ-field containing F 0 , σ(F 0 ) ⊂ G  . But σ(F 0 ) is also a monotone class
containing F 0 , and G  is minimal, G  ⊂ σ(F 0 ). Thus σ(F 0 ) = G  ⊂ G. 
Here is a typical application of the monotone class theorem. Let (Ω, F, P )
be a probability space. Suppose that F 0 is a field, but not a σ-field. Then
it generates a σ-field, σ(F 0 ). The question is, what kind of sets do we add
6 1. Probability Spaces

to make it a σ-field? Are the new sets close to the old? In fact they are, in
the sense that every set in σ(F 0 ) can be approximated arbitrarily well by a
set in F 0 . (We could borrow some terminology from topology and say that
in a sense, F 0 is dense in σ(F 0 ).)
Definition 1.6. The symmetric difference A Δ B of two sets A and B is
A Δ B = (A − B) ∪ (B − A) .
  ∞ 
Note that P { ∞ j=1 Aj  j=1 Bj } ≤
∞
j=1 P {Aj  Bj }.

Theorem 1.7. Let F 0 ⊂ F be a field and Λ ∈ σ(F 0 ). For any ε > 0, there
exists Γε ∈ F 0 such that P {Λ  Γε } < ε.

Proof. This is typical of many proofs in measure theory: it defines a class

of sets with the desired property, and then shows it is a monotone class.
Let G = {Λ ∈ F : ∀ ε > 0 ∃ Γε ∈ F 0 such that P {Λ  Γε } < ε}.
Clearly F 0 ⊂ G.
n Let ε > 0 and let Λ1 ⊂ Λ2 ⊂ · · · ∈ G. Choose n such that P {Λ −
j=1 Λj } < ε/2.
= 1, . . . , n choose Γj ∈ F 0 such that P {Λj  Γj } < ε/2j+1 . Then
For j 
set Γε = nj=1 Γj . Note that Γε ∈ F 0 since F 0 is a field, and
 
n   
n 
n 
P {Γε  Λ} ≤ P Λ − Λj + P Λj  Γj
j=1 j=1 j=1
n
≤ ε/2 + P {Λj  Γj }
j=1
n
ε
≤ ε/2 + ≤ ε.
2j+1
j=1

Thus Λ ∈ G. A similar argument shows that if Λ1 ⊃ Λ2 ⊃ · · · ∈ G, then


j Λj ∈ G. Therefore, G is a monotone class. It contains the field F 0 , so by
the monotone class theorem, G contains σ(F 0 ). 
This corollary follows immediately:
Corollary 1.8. Let P and Q be probability measures on a sample space
(Ω, F). Let F 0 ⊂ F be a field. Suppose Q{A} = P {A} for A ∈ F 0 . Then
Q{A} = P {A} for A ∈ σ(F 0 ).

1.2. Elementary Properties of Probability Spaces

Let (Ω, F , P ) be a probability space.
1.2. Elementary Properties of Probability Spaces 7

Proposition 1.9. Let A and B be events. Then

(i) P {∅} = 0;
(ii) P {Ac } = 1 − P {A};
(iii) A ⊂ B =⇒ P {A} ≤ P {B};
(iv) P {A ∪ B} = P {A} + P {B} − P {A ∩ B}.

Proof. (i) ∅ ∈ F and ∅ ∩ ∅ = ∅—the empty set is disjoint from itself (!)—so
that P {∅} = P {∅} + P {∅} = 2P {∅}. Thus P {∅} = 0.
(ii) A ∩ Ac = ∅ and A ∪ Ac = Ω, so P {A} + P {Ac } = P {Ω} = 1.
(iii) Since A and B − A are disjoint, and B = A ∪ (B − A), P {B} =
P {A}+P {B −A}. This is greater than or equal to P {A}, for P {B −A} ≥ 0.
(iv) Write A, B, and A ∪ B as disjoint unions: P {A} = P {A ∩ B} +
P
 {A − B} and P {B} = P {A ∩ B}+ P {B − A}. Thus P {A} + P {B} =
P {A − B} + P {B − A} + P {A ∩ B} + P {A ∩ B} = P {A ∪ B} + P {A ∩ B},
and (iv) follows. 
Countable additivity ((iii) in the definition) is important. It is not
enough to have additivity just for finite unions. In fact, countable ad-
ditivity is a continuity property in disguise. The following proposition
shows that probability measures are continuous from above and from be-
low: limn P {An } = P {limn An } if the sequence (An ) is either increasing or
decreasing.
Proposition
  1.10. (i) If A1 ⊂ A2 ⊂ A3 ⊂ . . . are events, then limj→∞ P {Aj }
=P A
j j .
   
(ii) If B1 ⊃ B2 ⊃ B3 ⊃ . . . are events, then limj→∞ P Bj = P j Bj .

Proof. (i) j Aj = A1 ∪ (A2 − A1 ) ∪ (A3 − A2 ) ∪ . . . , which is a disjoint
union. Thus, by countable additivity and the definition of the sum of a
series:
  ∞
P Aj = P {A1 } + P {Aj − Aj−1 }
j j=2
N
= P {A1 } + lim P {Aj − Aj−1 } .
N →∞
j=2

But Aj−1 ⊂ Aj , so P {Aj − Aj−1 } = P {Aj } − P {Aj−1 }, and the series is

N
 
= P {A1 } + lim P {Aj } − P {Aj−1 }
N →∞
j=2
= lim P {AN } .
N →∞
8 1. Probability Spaces

(ii) follows from (i) by taking complements. We leave the details as an

exercise. 
Definition 1.11. A probability space is a triple (Ω, F, P ), where Ω is
a set, called the sample space, F is a σ-field of subsets of Ω, and P is a
probability measure on F .
Definition 1.12. Elements of F are called events.
Remark 1.13. (i) In many cases, F is the class of all subsets of Ω. But
sometimes we do not know the probabilities of all subsets, and sometimes—
for technical reasons we need not go into here—it is simply not possible to
define the probability of all sets. For those who have seen it, the primary
example of this is Lebesgue measure. It cannot be defined on all subsets of
the line, but only on a class of sets called the Lebesgue-measurable sets.
(ii) In any case, the class of sets on which we can define a probability
is a σ-field. This is implicit in the very definition of a probability measure,
since, for example, (iii) implies that the probability of a countable union of
events can be defined, so the union must itself be an event.
(iii) There is another important reason for using σ-fields. It will not
come up until Chapter 5.4, but it is worth mentioning here. It turns out
that σ-fields are a very flexible way to represent information. One could say
that they are information in the raw.
Problems 1.2
1.3. Let A, B, and C be sets. Prove or disprove:
(a) (A − B) − C = A ∩ B c ∩ C c .
(b) (A ∩ B) − C = (A − C) ∩ (B − C).
(c) A ∪ B = B ⇐⇒ A ⊂ B.
1.4. Let n be an integer, let Ω = R, let F n be the smallest σ-field containing all
n ], and let P be any probability measure on R. Describe the
the intervals ( ni , i+1
random variables on (Ω, F n , P ).
1.5. Show that the following two classes are fields, but not sigma-fields.
(a) All finite subsets of R together with their complements.
(b) All finite unions of intervals in R of the form (a, b], (−∞, a], and (b, ∞).
1.6. Let C be the class of all subsets of R which are either countable or are the
complement of a countable set. Show that C is a σ-field.
1.7. Finish the proof of Proposition 1.10 by showing that if B and Bn are events,
then Bn ↓ B =⇒ P (Bn ) ↓ P (B).

1.3. The Intuition

Think of a probability space as the mathematical description of an exper-
iment. The experiment could be, for example, tossing a coin, rolling dice,
Another Random Scribd Document
with Unrelated Content
butler, and she did not like to hear his discretion aspersed.

"It looks as if he was right this time, however," she replied;

"though it wasn't Tim the tinker as stole Sir Thomas's spoons, which
Mr. Downing never had a good opinion of him; but when there ain't
nothing clearer than the person who was seen at the eating-house
with the victim" (Martha "took in" the Hatchet of Horror every week,
and framed her language on that delightful model) "had on a coat as
Evans made, it looks as if he wasn't altogether in the wrong, now
don't it, Mrs. Brookes?"

Mrs. Brookes could not deny that it looked very like that
complimentary conclusion, and her brave old heart almost died
within her. But she kept down her fear and horror, and dismissed
Martha, telling her to bring her the paper as soon as she could. The
woman returned in a few moments, laid the newspaper beside Mrs.
Brookes, and then went off to enjoy a continuation of the gossip of
the servants' hall. Very exciting and delightful that gossip was, for
though the servants had no inkling of the terribly strong interest, the
awfully near connection, which existed for Poynings in the matter, it
was still a great privilege to be "in" so important an affair by even
the slender link formed by the probable purchase of a coat at
Amherst by the murderer. They enjoyed it mightily; they discussed it
over and over again, assigning to the murdered man every grade of
rank short of royalty, and all the virtues possible to human nature.
The women were particularly eloquent and sympathizing, and
Martha "quite cried," as she speculated on the great probability of
there being a broken-hearted sweetheart in the case.

In the housekeeper's room, Mrs. Brookes sat poring over the

terrible story, to which she had listened carelessly on the previous
day, as the servants talked it vaguely over. From the first words
Martha had spoken, her fears had arisen, and now they were
growing every instant to the terrible certainty of conviction. What if
the wretched young man, who had already been the cause of so
much misery, had added this fearful crime to the long catalogue of
his follies and sins?

All the household sleeps, and the silence of the night is in every
room but one. There Mrs. Brookes still sits by the table with the
newspaper spread before her, lost in a labyrinth of fear and anguish;
and from time to time her grief finds words, such as:

"How shall I tell her? How shall I warn her? O George, George! O
my boy! my boy!"

CHAPTER XIII.

THE SHADOW OF DEATH.

Mr. Carruthers was an early man; no danger of any skulking

among the numerous hands which found employment on the
Poynings estate. If the eye of the master be indeed the spur of the
servant, Mr. Carruthers's dependents had quite enough of that
stimulant. He made his rounds every morning at an hour which the
in-door servants, who were obliged to have breakfast ready on his
return, considered heathenish, and the out-door servants declared
savoured of slave-driving. Mrs. Brookes knew that she should have
no difficulty in procuring a private interview with her mistress on the
morning following Mr. Dalrymple's visit, as an hour and a half always
elapsed between Mr. Carruthers's leaving the house and his wife's
ringing for her maid. The old woman looked worn and weary and
very old, as she peered from behind a red-cloth door, which shut off
the corridor on which Mr. Carruthers's dressing-room opened from
the grand gallery, and watched her master take his creaking way
down the staircase, looking as he went more full of self-importance
than usual, and treading more heavily, as if the weight of the Home
Office communication had got into his boots.

When he had disappeared, and she had heard the click of the lock
as he opened the great door and went out into the pure fresh
morning air, Mrs. Brookes emerged from behind the partition-door,
and softly took the way to Mrs. Carruthers's bedroom. The outer
door was slightly open, the heavy silken curtain within hung closely
over the aperture. The old woman pushed it gently aside, and,
noiselessly crossing the room, drew the window curtain, and let in
sufficient light to allow her to see that Mrs. Carruthers was still
sleeping. Her face, pale, and even in repose bearing a troubled
expression, was turned towards the old woman, who seated herself
in an arm-chair beside the bed, and looked silently and sadly on the
features, whose richest bloom and earliest sign of fading she had so
faithfully watched.

"How am I to tell her?" she thought. "How am I to make her see

what I see, suspect what I suspect? and yet she must know all, for
the least imprudence, a moment's forgetfulness, would ruin him.
How am I to tell her?"

The-silver bell of a little French clock on the chimney-piece rang

out the hour melodiously, but its warning struck upon the old
woman's ear menacingly. There was much to do, and little time to
do it in; she must not hesitate longer. So she laid her withered,
blanched old hand upon the polished, ivory-white fingers of the
sleeper, lying with the purposelessness of deep sleep upon the
coverlet, and addressed her as she had been used to do in her
girlhood, and her early desolate widowhood, when her humble friend
had been well-nigh her only one.

"My dear," she said, "my dear." Mrs. Carruthers's hand twitched in
her light grasp; she turned her head away with a troubled sigh, but
yet did not wake. The old woman spoke again: "My dear, I have
something to say to you."

Then Mrs. Carruthers awoke fully, and to an instantaneous

comprehension that something was wrong. All her fears, all her
suspicions of the day before, returned to her mind in one flash of
apprehension, and she sat up white and breathless.

"What is it, Ellen? Has he found out? Does he know?"

"Who? What do you mean?"

"Mr. Carruthers. Does he know George was here?"

"God forbid!" said the old woman, in a trembling tone.

She felt the task she had before her almost beyond her power of
execution. But her mistress's question, her instinctive fear, had given
her a little help.

"No," she said, "he knows nothing, and God send he may neither
know nor suspect anything about our dear boy! but you must be
quiet now and listen to me, for I must have said my say before
Dixon comes--she must not find me here."

"Why are you here?" asked Mrs. Carruthers, who had sat up in
bed, and was now looking at the old woman, with a face which had
no more trace of colour than the pillow from which it had just been
raised. "Tell me, Ellen; do not keep me in suspense. Is anything
wrong about George? It must concern him, whatever it is."

"My dear," began Mrs. Brookes--and now she held the slender
fingers tightly in her withered palm--"I fear there is something very
wrong with George."

"Is he--is he dead?" asked the mother, in a faint voice.

 "No, no; he is well and safe, and far away from this, I hope and
trust."

Mrs. Carruthers made no answer, but she gazed at her old friend
with irresistible, pitiful entreaty. Mrs. Brookes answered the dumb
appeal.

"Yes, my dear, I'll tell you all. I must, for his sake. Do you know
what was the business that brought that strange gentleman here, he
that went out with master, and dined here last night? No, you don't.
I thought not. Thank God, you have got no hint of it from any one
but me."

"Go on, go on," said Mrs. Carruthers, in a yet fainter voice.

"Do you remember, when George was here in February, you gave
him money to buy a coat?"

"Yes," Mrs. Carruthers rather sighed than said.

"He bought one at Evans's, and he was remarked by the old man,
who would know him again if he saw him. The business on which
the strange gentleman came to master was to get him to help, as a
magistrate, in finding the person who bought that coat at Evans's,
Amherst."

"But why? What had he done? How was the coat known?"

"My dear," said Mrs. Brookes--and now she laid one arm gently
round her mistress's shoulder as she leaned against the pillows--"the
wearer of that coat is suspected of having murdered a man, whose
body was found by the river-side in London the other day."

"My God!" moaned the mother, and a hue as of death overspread

her features.
 "My dear, he didn't do it. I'm sure he didn't do it. I would stake my
soul upon it. It is some dreadful mistake. Keep up until I have done,
for God's sake, and George's sake, keep up--remember there is no
danger unless you lose courage and give them a hint of anything. Be
sure we shall find he has sold the coat to some one else, and that
some one has done this dreadful thing. But you must keep up--here,
let me bathe your face and hands while I am talking, and then I'll go
away, and, when Dixon comes, you must just say you are not well,
and don't mean to get up to breakfast, and then I shall have an
excuse for coming to you. There! you are better now, I am sure.
Yes, yes; don't try to speak; I'll tell you without asking," she went
on, in a rapid whisper. "The strange gentleman and master saw
Evans, and he told them when he sold the coat, and the sort of
person he sold it to; but Gibson and Thomas say he could not have
told them distinct, for they heard the strange gentleman saying to
master, in the carriage, that the description was of no use. And I am
certain sure that there is not the least suspicion that he has ever
been in Amherst since he bought the coat."

"I don't understand," stammered Mrs. Carruthers. "When--when

did this happen?"

"A few days ago: it's all in the papers."

Mrs. Carruthers groaned.

"Nothing about George, but about finding the body and the coat.
It is all here." The old woman took a tightly folded newspaper from
her pocket. The light was too dim for her to read its contents to her
mistress, who was wholly incapable of reading them herself. Mrs.
Brookes, paper in hand, was going to the window, to withdraw the
curtain completely, when she paused.

"No," she said; "Dixon will be here too soon. Better that you
should ring for her at once, and send her for me. Can you do this,
my dear? keeping yourself up by remembering that this is only some
dreadful mistake, and that George never did it--no, no more than
you did. Can you let me go away for a few minutes, and then come
back to you? Remember, we cannot be too careful, for his sake; and
if Dixon found me here at an unusual hour, the servants would know
there is some secret or another between us."

"I can bear anything--I can do anything you tell me," was Mrs.
Carruthers's answer, in a whisper.

"Well then, first lie down, and I will close the curtains and leave
you. When I have had time to get to my room, ring for Dixon. Tell
her you are ill. When she lets the light in she will see that for herself,
and desire her to send me to you."

In another minute the room was once more in darkness, and Mrs.
Brookes went down the grand staircase, in order to avoid meeting
any of the servants, crossed the hall, and gained her own apartment
without being observed. A short time, but long to her impatience,
had elapsed, when Mrs. Carruthers's maid knocked at the door, and
having received permission to enter, came in with an important face.
She delivered the message which Mrs. Brookes was expecting, and
added that she had never seen her lady look so ill in all her born
days.

"Looks more like a corpse, I do assure you, than like the lady I
undressed last night, and circles under her eyes, dreadful. I only
hope it ain't typus, for I'm dreadful nervous, not being used to
sickness, which indeed I never engaged for. But, if you please, Mrs.
Brookes, you was to go to her immediate, and I'm to let Miss
Carruthers know as she's to make tea this morning for master, all to
their two selves, which he won't like it, I dare say."

Then the talkative damsel went her way to Miss Carruthers's

room, and Mrs. Brookes hurried to that of her unhappy mistress. She
had again raised herself in the bed, and was looking eagerly towards
the door, with hollow haggard eyes, and lips ashy pale, whose
trembling she in vain tried to control.

"Lock both doors, Ellen," she said, "and tell me all. Give me the
paper; I can read it--I can indeed."

She took it and read it steadily through--read it with the same

horrible emotion, a thousand times intensified, which had agitated
the faithful servant a few hours previously. Standing by the bedside,
Mrs. Brookes gazed upon her pale, convulsed features, as she read,
and ever, as she saw the increasing agony which they betrayed, she
murmured in accents of earnest entreaty:

"Don't, my dear, for God's sake, don't, not for a moment, don't
you believe it. He sold the coat, depend upon it. It looks very bad,
very black and bad, but you may be sure there's no truth in it. He
sold the coat."

She spoke to deaf ears. When Mrs. Carruthers had read the last
line of the account of the inquest on the body of the unknown man,
the paper dropped from her hand; she turned upon the old nurse a
face which, from that moment, she never had the power to forget,
and said:

"He wore it--I saw it on him on Friday," and the next moment
slipped down among the pillows, and lay as insensible as a stone.

The old woman gave no alarm, called for no assistance, but

silently and steadily applied herself to recalling Mrs. Carruthers to
consciousness. She had no fear of interruption. Mr. Carruthers
invariably went direct to the breakfast-room on returning from his
morning tour of inspection, and Clare would not visit Mrs. Carruthers
in her own apartment unasked. So Mrs. Brookes set the windows
and doors wide open, and let the sweet morning air fan the
insensible face, while she applied all the remedies at hand. At length
Mrs. Carruthers sighed deeply, opened her eyes, and raised her hand
to her forehead, where it came in contact with the wet hair.
 "Hush, my dear," said Mrs. Brookes, as she made an almost
inarticulate attempt to speak. "Do not try to say anything yet. Lie
quite still, until you are better."

Mrs. Carruthers closed her eyes again and kept silent. When, after
an interval, she began to look more life-like, the old woman said,
softly:

"You must not give way again like this, for George's sake. I don't
care about his wearing the coat. I know it looks bad, but it is a
mistake, I am quite sure. Don't I know the boy as well as you do,
and maybe better, and don't I know his tender heart, with all his
wildness, and that he never shed a fellow-creature's blood in anger,
or for any other reason. But it's plain he is suspected--not he, for
they don't know him, thank God, but the man that wore the coat,
and we must warn him, and keep it from master. Master would go
mad, I think, if anything like suspicion or disgrace came of Master
George, more than the disgrace he thinks the poor boy's goings on
already. You must keep steady and composed, my dear, and you
must write to him. Are you listening to me? Do you understand me?"
asked the old woman, anxiously, for Mrs. Carruthers's eyes were wild
and wandering, and her hand twitched convulsively in her grasp.

"Yes, yes," she murmured, "but I tell you, Ellen, he wore the coat-
-my boy wore the coat."

"And I tell you, I don't care whether he wore the coat or not,"
repeated Mrs. Brookes, emphatically. "He can explain that, no doubt
of it; but he must be kept out of trouble, and you must be kept out
of trouble, and the only way to do that, is to let him know what
brought the strange gentleman to Poynings, and what he and
master found out. Remember, he never did this thing, but, my dear,
he has been in bad hands lately, you know that; for haven't you
suffered in getting him out of them, and I don't say but that he may
be mixed up with them that did. I'm afraid there can't be any doubt
of that, and he must be warned. Try and think of what he told you
about himself, not only just now, but when he came here before,
and you will see some light, I am sure."

But Mrs. Carruthers could not think of anything, could not

remember anything, could see no light. A deadly horrible conviction
had seized upon her, iron fingers clutched her heart, a faint
sickening terror held her captive, in body and spirit; and as the old
woman gazed at her, and found her incapable of answering, the fear
that her mistress was dying then and there before her eyes took
possession of her. She folded up the newspaper which had fallen
from Mrs. Carruthers's hand upon the bed, replaced it in her pocket,
and rang the bell for Dixon.

"My mistress is very ill," she said, when Dixon entered the room.
"You had better go and find master, and send him here. Tell him to
send Dr. Munns at once."

Dixon gave a frightened, sympathizing glance at the figure on the

bed, over which the old woman was bending with such kindly
solicitude, and then departed on her errand. She found Mr.
Carruthers still in the breakfast-room. He was seated at the table,
and held in his hand a newspaper, from which he had evidently been
reading, when Dixon knocked at the door; for he was holding it
slightly aside, and poising his gold eye-glass in the other hand, when
the woman entered. Mr. Carruthers was unaccustomed to being
disturbed, and he did not like it, so that it was in a tone of some
impatience that he said:

"Well, Dixon, what do you want?"

"If you please, sir," replied Dixon hesitatingly, "my mistress is not
well."

"So I hear," returned her master; "she sent word she did not mean
to appear at breakfast. He said it rather huffily, for not to appear at
breakfast was, in Mr. Carruthers's eyes, not to have a well-regulated
mind, and not to have a well-regulated mind was very lamentable
and shocking indeed.

"Yes, sir," Dixon went on, "but I'm afraid she's very ill indeed. She
has been fainting this long time, sir, and Mrs. Brookes can't bring her
to at all. She sent me to ask you to send for Dr. Munns at once, and
will you have the goodness to step up and see my mistress, sir?"

"God bless my soul," said Mr. Carruthers, pettishly, but rising as he

spoke, and pushing his chair away. "This is very strange; she has
been exposing herself to cold, I suppose. Yes, yes, go on and tell
Mrs. Brookes I am coming, as soon as I send Gibson for Dr. Munns."

Dixon left the room, and Mr. Carruthers rang the bell, and desired
that the coachman should attend him immediately. When Dixon had
entered the breakfast-room, Clare Carruthers had been standing by
the window, looking out on the garden, her back turned towards her
uncle. She had not looked round once during the colloquy between
her uncle and his wife's maid, but had remained quite motionless.
Now Mr. Carruthers addressed her.

"Clare," he said, "you had better go to Mrs. Carruthers." But his

niece was no longer in the room; she had softly opened the French
window, and passed into the flower-garden, carrying among the
sweet, opening flowers of the early summer, and into the serene air,
a face which might have vied in its rigid terror with the face upstairs.
When Mr. Carruthers had come in that morning, and joined Clare in
the pretty breakfast-room, he was in an unusually pleasant mood,
and had greeted his niece with uncommon kindness. He had found
everything in good order out of doors. No advantage had been taken
of his absence to neglect the inexorable sweepings and rollings, the
clippings and trimmings, the gardening and grooming. So Mr.
Carruthers was in good humour in consequence, and also because
he was still nourishing the secret sense of his own importance,
which had sprung up in his magisterial breast under the flattering
influence of Mr. Dalrymple's visit. So when he saw Clare seated
before the breakfast equipage, looking in her simple, pretty morning
dress as fair and bright as the morning itself, and when he received
an intimation that he was not to expect to see his wife at breakfast,
he recalled the resolution he had made last night, and determined to
broach the subject of Mr. Dalrymple's visit to his niece without delay.

A pile of letters and newspapers lay on a salver beside Mr.

Carruthers's plate, but he did not attend to them until he had made
a very respectable beginning in the way of breakfast. He talked to
Clare in a pleasant tone, and presently asked her if she had been
looking at the London papers during the last few days. Clare replied
that she seldom read anything beyond the deaths, births, and
marriages, and an occasional leader, and had not read even so much
while she had been at the Sycamores.

"Why do you ask, uncle?" she said. "Is there any particular news?"

"Why, yes, there is," replied Mr. Carruthers, pompously.

"There is a matter attracting public attention just now in which I

am, strange to say, a good deal interested--in which responsibility
has been laid on me, indeed, in a way which, though flattering--very
flattering indeed--is, at the same time, embarrassing."

Mr. Carruthers became more and more pompous with every word
he spoke. Clare could not repress a disrespectful notion that he bore
an absurd resemblance to the turkey-cock, whose struttings and
gobblings had often amused her in the poultry-yard, as he mouthed
his words and moved his chin about in his stiff and spotless cravat.
His niece was rather surprised by the matter of his discourse, as she
was not accustomed to associate the idea of importance to society
at large with Mr. Carruthers of Poynings, and cherished a rather
settled conviction that, mighty potentate as he was within the
handsome gates of Poynings, the world outside wagged very
independently of him. She looked up at him with an expression of
interest and also of surprise, but fortunately she did not give
utterance to the latter and certainly predominant sentiment.

"The fact is," said Mr. Carruthers, "a murder has been committed
in London under very peculiar circumstances. It is a most mysterious
affair, and the only solution of the mystery hitherto suggested is that
the motive is political."

He paused, cleared his throat, once more settled his chin

comfortably, and went on while Clare listened, wondering more and
more how such a matter could affect her uncle. She was a gentle-
hearted girl; but not in the least silly, and quite free from any sort of
affectation; so she expressed no horror or emotion at the mere
abstract idea of the murder, as a more young-ladyish young lady
would have done.

"Yes, uncle?" she said, simply, as he paused.

Mr. Carruthers continued:

"The murdered man was found by the river-side, stabbed, and

robbed of whatever money and jewelry he had possessed. He was a
good-looking man, young, and evidently a foreigner; but there were
no means of identifying the body, and the inquest was adjourned--in
fact, is still adjourned."

"What an awful death to come by, in a strange country!" said

Clare, solemnly. "How dreadful to think that his friends and relatives
will perhaps never know his fate! But how did they know the poor
creature was a foreigner, uncle?"

"By his dress, my dear. It appears he had on a fur-lined coat, with

a hood--quite a foreign article of dress; and the only person at the
inquest able to throw any light on the crime was a waiter at an
eating-house in the Strand, who said that the murdered man had
dined there on a certain evening--last Thursday, I believe--and had
worn the fur coat, and spoken in a peculiar squeaky voice. The
waiter felt sure he was not an Englishman, though he spoke good
English. So the inquest was adjourned in order to get more
evidence, if possible, as to the identity of the murdered man and
also that of the last person who had been seen in his company. And
this brings me to the matter in which I am interested."

Clare watched her uncle with astonishment as he rose from his

chair and planted himself upon the hearth-rug before the fireplace,
now adorned with its summer ornaments of plants and flowers, and
draped in muslin. Taking up the familiar British attitude, and looking,
if possible, more than ever pompous, Mr. Carruthers proceeded:

"You will be surprised to learn, Clare, that the visit of the

gentleman who came here yesterday, and with whom I went out,
had reference to this murder."

"How, uncle?" exclaimed Clare. "What on earth have you, or has

any one here, to do with it?"

"Wait until I have done, and you will see," said Mr. Carruthers in a
tone of stately rebuke. "The last person seen in the company of the
man afterwards found murdered, and who dined with him at the
tavern, wore a coat which the waiter who recognized the body had
chanced to notice particularly. The appearance of this person the
man failed in describing with much distinctness; but he was quite
positive about the coat, which he had taken from the man and hung
up on a peg with his own hands. And now, Clare, I am coming to the
strangest part of this strange story."

The girl listened with interest indeed, and with attention, but still
wondering how her uncle could be involved in the matter, and
perhaps feeling a little impatient at the slowness with which, in his
self-importance, he told the story.

"I was much surprised," continued Mr. Carruthers, "to find in the
gentleman who came here yesterday, and whose name was
Dalrymple, an emissary from the Home Office, intrusted by Lord
Wolstenholme with a special mission to me"--impossible to describe
the pomposity of Mr. Carruthers's expression and utterance at this
point--"to me. He came to request me to assist him in investigating
this most intricate and important case. It is not a mere police case,
you must understand, my dear. The probability is that the murdered
man is a political refugee, and that the crime has been perpetrated"-
-Mr. Carruthers brought out the word with indescribable relish--"by a
member of one of the secret societies, in revenge for the defection
of the victim, or in apprehension of his betrayal of the cause."

"What cause, uncle?" asked Clare innocently. She was not of a

sensational turn of mind, had no fancy for horrors as horrors, and
was getting a little tired of her uncle's story.

"God knows, my dear--some of their liberty, fraternity, and

equality nonsense, I suppose. At all events, this is the supposition;
and to ask my aid in investigating the only clue in the possession of
the government was the object of Mr. Dalrymple's visit yesterday.
The man who was seen in the company of the murdered man by the
waiter at the tavern, and who went away with him, wore a coat
made by Evans of Amherst. You know him, Clare--the old man who
does so much of our work here. I went to his shop with Mr.
Dalrymple, and we found out all about the coat. He remembered it
exactly, by the description; and told us when he had made it (two
years ago), and when he had sold it (six weeks ago), to a person
who paid for it with a ten-pound note with the Post-office stamp
upon it. The old man is not very bright, however; for though he
remembered the circumstance, and found the date in his day-book,
he could not give anything like a clear description of the man who
had bought the coat. He could only tell us, in general terms, that he
would certainly know him again if he should see him; but he talked
about a rather tall young man, neither stout nor thin, neither ugly
nor handsome, dark-eyed and dark-haired,--in short, the kind of
description which describes nothing. We came away as wise as we
went, except in the matter of the date of the purchase of the coat.
That does not help much towards the detection of the murderer, as a
coat may change hands many times in six weeks, if it has been
originally bought by a dubious person. The thing would have been to
establish a likeness between the man described by Evans as the
purchaser of the coat, and the man described by the waiter as the
wearer of the coat at the tavern. But both descriptions are very
vague."

"What was the coat like?" asked Clare in a strange, deliberate

tone.

"It was a blue Witney overcoat, with a label inside the collar
bearing Evans's name. The waiter at the tavern where the murdered
man dined had read the name, and remembered it. This led to their
sending to me; and my being known to the authorities as a very
active magistrate"--here Mr. Carruthers swelled and pouted with
importance--"they naturally communicated with me. The question is
now, how I am to justify the very flattering confidence which Lord
Wolstenholme has placed in me? It is a difficult question, and I have
been considering it maturely. Mr. Dalrymple seems to think the clue
quite lost. But I am not disposed to let it rest; I am determined to
set every possible engine at work to discover whether the
description given by the waiter and that given by Evans tally with
one another."

"You said the inquest was adjourned, I think," said Clare.

"Yes, until to-day; but Mr. Dalrymple will not have learned
anything. There will be an open verdict"--here Mr. Carruthers
condescendingly explained to his niece the meaning of the term--
"and the affair will be left to be unravelled in time. I am anxious to
do all I can towards that end; it is a duty I owe to society, to Lord
Wolstenholme, and to myself."

Clare had risen from her chair, and approached the window. Her
uncle could not see her face, as he resumed his seat at the
breakfast-table, and opened his letters in his usual deliberate and
dignified manner. Being letters addressed to Mr. Carruthers of
Poynings, they were, of course, important; but if they had not had
that paramount claim to consideration, the communication in
question might have been deemed dull and trivial. Whatever their
nature, Clare Carruthers turned her head from the window and
furtively watched her uncle during their perusal. He read them with
uplifted eyebrows and much use of his gold-rimmed eye-glasses, as
his habit was, but then laid them down without comment, and took
up a newspaper.

"I dare say we shall find something about the business in this," he
said, addressing his niece, but without turning his head in her
direction. "Ah, I thought so; here it is: 'Mysterious circumstance;
extraordinary supineness and stupidity of the police; no one arrested
on suspicion; better arrest the wrong man, and tranquillize the
public mind, than arrest no one at all.' I'm not convinced by that
reasoning, I must say. What!--no reason for regarding the murder as
a political assassination? Listen to this, Clare;" and he read aloud,
while she stood by the window, her back turned towards him, and
listened intently, greedily, with a terrible fear and sickness at her
heart:

"'The supposition that this atrocious crime has been committed

from political motives has, in our opinion, no foundation in
probability, and derives very little support from common sense. The
appearance of the body, the fineness of the linen, the expensive
quality of the attire, the torn condition of the breast and sleeves of
the shirt, which seems plainly to indicate that studs, probably of
value, had been wrenched violently out; the extreme improbability
that an individual, so handsomely dressed as the murdered man,
would have been out without money in his pocket,--all indicate
robbery, at least; and if perhaps more than robbery, certainly not
less, to have been the motive of the crime. An absurd theory has
been founded upon the peculiarity in the dress of the victim, and
upon the remark made by the only witness at the inquest about his
tone of voice. Nothing is more likely, in our opinion, than a complete
miscarriage of justice in this atrocious case. Suspicion has been
arbitrarily directed in one channel, and the result will be, probably,
the total neglect of other and more likely ones. While the political
murderer is being theorized about and "wanted," the more ordinary
criminal--the ruffian who kills for gain, and, not for patriotism or
principle--is as likely as not to escape comfortably, and enjoy his
sway in some pleasant, unsuspected, and undisturbed retreat.'

"Now, I call this most unjustifiable," said Mr. Carruthers in a tone

of dignified remonstrance and indignation. "Really, the liberty of the
press is going quite too far. The Government are convinced that the
murder is political, and I can't see--"

It was at this point of Mr. Carruthers's harangue that he was

interrupted by his wife's maid. When he again looked for Clare she
had disappeared, nor did he or any of the frightened and agitated
household at Poynings see the young lady again for many hours. Dr.
Munns arrived, and found Mr. Carruthers considerably distressed at
the condition in which Mrs. Carruthers was, also a little annoyed at
that lady's want of consideration in being ill, and unable to refrain
from hinting, with much reserve and dignity of manner, that he was
at present more than usually engaged in business of the last
importance, which rendered it peculiarly unfortunate that he should
have an additional care imposed on him--public importance, he took
care to explain, and no less onerous than mysterious. But the worthy
gentleman's pride and pompousness were soon snubbed by the
extreme gravity of Dr. Munns's manner, as he answered his inquiries
and put questions in his turn relative to his patient. The doctor was
both alarmed and puzzled by Mrs. Carruthers's state. He told her
husband she was very seriously ill: he feared brain-fever had already
set in. Could Mr. Carruthers account for the seizure in any way? No,
Mr. Carruthers could not; neither could the housekeeper, nor Mrs.
Carruthers's maid, both of whom were closely questioned, as having
more and more frequent access to that lady's presence than any
other members of the household.
 Had Mrs. Carruthers heard any distressing intelligence? had she
received a shock of any kind? the doctor inquired. Mr. Carruthers
appeared to sustain one from the question. Of course not; certainly
not; nothing of the kind, he replied, with some unrepressed irritation
of manner, and secretly regarded the bare suggestion of such a
possibility as almost indecent. Mrs. Carruthers of Poynings receive
shocks indeed! The doctor, who knew and disregarded his
peculiarities, calmly pursued his inquiries undeterred by Mr.
Carruthers's demeanour; and finding that nothing particular had
happened, acknowledged that, there being no apparent cause to
which so sudden and serious an illness could be attributed, he was
the more uneasy as to its probable result. Then Mr. Carruthers
caught the infection of his alarm, and all the best side of his
character, all the real love and appreciation of his wife, ordinarily
overlaid by his egotism, came out in full force, and the staunchest
stickler for domestic fealty could not have demanded greater
solicitude than the frightened husband exhibited.

In a wonderfully short of space of time the house assumed the

appearance which illness always gives. The servants went about
their work whispering, and the sitting-rooms were silent and
deserted. No one bestowed a thought on Clare. The attendants on
the suffering woman, busily engaged in carrying out the orders given
them by Dr. Munns, who remained for several hours with his patient;
the alarmed husband, who wandered about disconsolately between
his own library and his wife's room--all forgot the girl's existence. It
was very late--within a few minutes of the usual dinner-hour (an
inflexible period at Poynings)--when Clare Carruthers crossed the
flower-garden, entered the house by the window through which she
had left it, and stole gently upstairs to her own room. She threw her
hat and shawl upon her bed and went to her dressing-table. There
she stood for some minutes before the glass, holding her disordered
hair back with her hands--there were bits of grass and fragments of
leaves in it, as though she had been lying with her fair head prone
upon the ground--and gazing upon her young misery-stricken face.
White about the full pure lips, where the rich blood ordinarily
glowed; purple about the long fair eyelids and the blushing cheeks,
heavy-eyed,--the girl was piteous to see, and she knew it. The hours
that had passed over since she left her uncle's presence in the
morning had been laden with horror, with dread, with such anguish
as had never in its lightest form touched her young spirit before; and
she trembled as she marked the ravages they had made in her face.

"What shall I do?" she murmured, as though questioning her own

forlorn image in the glass. "What shall I do? I dare not stay away
from dinner, and what will they say when they see my face?"

She fastened up her hair, and bathed her face with cold water;
then returned to the glass to look at it again; but the pallor was still
upon the lips, the discoloration was still about the heavy eyelids. As
she stood despairingly before the dressing-table, her maid came to
her.

"The dinner-bell will not ring, ma'am," said the girl. "Mr.
Carruthers is afraid of the noise for Mrs. Carruthers."

"Ay," said Clare, listlessly, still looking at the disfigured image in

the glass. "How is she?"

"No better, ma'am; very bad indeed, I believe. But don't take on
so, Miss Clare," her maid went on, affectionately. "She is not so bad
as they say, perhaps; and, at all events, you'll knock yourself up, and
be no comfort to Mr. Carruthers."

A light flashed upon Clare. She had only to keep silence, and no
one would find her out; her tears, her anguish, would be imputed to
her share of the family trouble. Her maid, who would naturally have
noticed her appearance immediately, expressed no surprise. Mrs.
Carruthers was very ill, then. Something new had occurred since the
morning, when there had been no hint of anything serious in her
indisposition. The maid evidently believed her mistress acquainted
with all that had occurred. She had only to keep quiet, and nothing
would betray her ignorance. So she allowed the girl to talk, while
she made some trifling change in her dress, and soon learned all the
particulars of Mrs. Carruthers's illness, and the doctor's visit, of her
uncle's alarm, and Mrs. Brookes's devoted attendance on her
mistress. Then Clare, trembling though relieved of her immediate
apprehension of discovery, went down-stairs to join her uncle at
their dreary dinner. He made no comment upon the girl's
appearance, and, indeed, hardly spoke. The few words of sympathy
which Clare ventured to say were briefly answered, and as soon as
possible he left the dining-room. Clare sat by the table for a while,
with her face buried in her hands, thinking, suffering, but not
weeping. She had no more tears to-day to shed.

Presently she went to Mrs. Carruthers's room, and sat down on a

chair behind the door, abstracted and silent. In the large dimly-
lighted room she was hardly seen by the watchers. She saw her
uncle come in, and stand forlornly by the bed; then the doctor came,
and several figures moved about silently and went away, and then
there was no one but Mrs. Brookes sitting still as a statue beside the
sufferer, who lay in a state of stupor. How long she had been in the
room before the old woman perceived her Clare did not know; but
she felt Mrs. Brookes bending over her, and taking her hand, before
she knew she had moved from the bedside.

"Pray go away and lie down, Miss Carruthers," the old woman
said, half tenderly, half severely. "You can do no good here--no one
can do any good here yet--and you will be ill yourself. We can't do
with more trouble in the house, and crying your eyes out of your
head, as you've been doing, won't help any one, my dear. I will send
you word how she is the first thing in the morning."

The old woman raised the girl by a gentle impulse, as she spoke,
and she went meekly away, Mrs. Brookes closing the door behind
her with an unspoken reflection on the uselessness of girls, who,
whenever anything is the matter, can do nothing but cry.
 The night gradually fell upon Poynings--the soft, sweet, early
summer night. It crept into the sick-room, and overshadowed the
still form upon the bed--the form whose stillness was to be
succeeded by the fierce unrest, the torturing vague effort of fever; it
closed over the stern pompous master of Poynings, wakeful and
sorely troubled. It darkened the pretty chamber, decorated with a
thousand girlish treasures and simple adornments, in which Clare
Carruthers was striving sorely with the first fierce trial of her
prosperous young life. When it was at its darkest and deepest, the
girl's swollen weary eyelids closed, conquered by the irresistible
mighty benefactor of the young who suffer. Then, if any eye could
have pierced the darkness and looked at her as she lay sleeping, the
stamp of a great fear upon her face even in her slumber, and her
breast shaken by frequent heavy sighs, it would have been seen that
one hand was hidden under the pillow, and the fair cheek pressed
tightly down upon it, for better security. That hand was closed upon
three letters, severally addressed to the advertising department of
three of the daily newspapers. The contents, which were uniform,
had cost the girl hours of anxious and agonizing thoughts. They
were very simple, and were as follows, accompanied by the sum
which she supposed their insertion would cost, very liberally
estimated:

"The gentleman who showed a lady a sprig of myrtle on last

Saturday is earnestly entreated by her not to revisit the place where
he met her. He will inevitably be recognized."

"God forgive me if I am doing wrong in this!" Clare Carruthers had

said with her last waking consciousness. "God forgive me, but I must
save him if I can!"
 CHAPTER XIV

THE SHADOW LIGHTENED.

Long before Mr. Carruthers, impelled by the irresistible force of

routine, which not all the concern, and even alarm, occasioned him
by Mrs. Carruthers's condition could subdue, had issued forth upon
his daily tour of inspection, Clare's letters had been safely posted by
her own hand at the village. She had slept but little on the night
which had fallen on her first experience of fear and grief; and
waking at dawn, oppressed by a heavy sense of some dimly-
understood calamity, she had recalled it all in a moment; and having
hurriedly dressed herself, she went down to the breakfast-room, and
let herself out through the window, accompanied by her dog, whose
joyous gambols in the bright morning air she did not notice. That
morning air struck chill to the weary limbs and aching head of the
sad, bewildered girl as she pursued her rapid way through the
shrubbery, brushing the dew from the branches of the trees as she
passed hurriedly along heart-sick, and yet wandering and confused
in her thoughts.

Her walk was quite solitary and uninterrupted. She slid the letters
into a convenient slit of a window-shutter of the general-shop, to
which the dignity and emoluments of a post-office were attached;
glanced up and down the little street, listened to certain desultory
sounds which spoke of the commencement of activity in adjacent
stable-yards, and to the barking with which some vagabond dogs of
her acquaintance greeted her and Cæsar; satisfied herself that she
was unobserved, and then retraced her steps as rapidly as possible.
The large white-faced clock over the stables at Poynings--an
unimpeachable instrument, never known to gain or lose within the
memory of man--was striking six as Clare Carruthers carefully
replaced the bolt of the breakfast-room window, and crept upstairs
again, with a faint flutter of satisfaction that her errand had been
safely accomplished contending with the dreariness and dread which
filled her heart. She put away her hat and cloak, changed her dress,
which was wet with the dew, and sat down by the door of the room
to listen for the first stir of life in the house.

Soon she heard her uncle's step, lighter, less creaky than usual,
and went out to meet him. He did not show any surprise on seeing
her so early, and the expression of his face told her in a moment
that he had no good news of the invalid to communicate.

"Brookes says she has had a very bad night," he said gravely. "I
am going to send for Munns at once, and to telegraph to London for
more advice." Then he went on in a state of subdued creak; and
Clare, in increased bewilderment and misery, went to Mrs.
Carruthers's room, where she found the reign of dangerous illness
seriously inaugurated.

Doctor Munns came, and early in the afternoon a grave and polite
gentleman arrived from London, who was very affable, but rather
reserved, and who was also guilty of the unaccountable bad taste of
suggesting a shock in connection with Mrs. Carruthers's illness. He
also was emphatically corrected by Mr. Carruthers, but not with the
same harshness which had marked that gentleman's reception of Dr.
Munns's suggestion. The grave gentleman from London made but
little addition to Dr. Munns's treatment, declined to commit himself
to any decided opinion on the case, and went away, leaving Mr.
Carruthers with a sensation of helplessness and vague injury, to say
nothing of downright misery and alarm, to which the Grand Lama
was entirely unaccustomed.

Before the London physician made his appearance Clare and her
uncle had met at breakfast, and she had learned all there was to be
known on the subject which had taken entire and terrible possession
of her mind: It seemed to Clare now that she had no power of
thinking of anything else, that it was quite impossible that only
yesterday morning she was a careless unconscious girl musing over
a romantic incident in her life, speculating vaguely upon the
possibility of any result accruing from it in the future, and feeling as
far removed from the crimes and dangers of life as if they had no
existence. Now she took her place opposite her uncle with a face
whose pallor and expression of deep-seated trouble even that
unobservant and self-engrossed potentate could not fail to notice.
He did observe the alteration in Clare's looks, and was not altogether
displeased by it. It argued deep solicitude for Mrs. Carruthers of
Poynings--an extremely proper sentiment; so Mr. Carruthers
consoled his niece after his stately fashion, acknowledging, at the
same time, the unaccountable vagaries of fever, and assuring Clare
that there was nothing infectious in the case--a subject on which it
had never occurred to the girl to feel any uneasiness. Not so with Mr.
Carruthers, who had a very great dread of illness of every kind, and
a superstitious reverence for the medical art. The conversation was
interrupted by the arrival of the post, and Mr. Carruthers's attention
was again drawn to the subject of the murder and the possibility of
promoting his own importance in connection with it. Clare's pale face
turned paler as her uncle took up the first letter of the number
presented to him by Thomas (footman), that official looking
peculiarly intelligent on the occasion; for the letter bore the magic
inscription, "On Her Majesty's Service," and the seal of the Home
Office.

Mr. Carruthers took some time to read the letter, even with the aid
of the gold eye-glasses. It came from Mr. Dalrymple, who wrote an
abnormally bad hand even for a government official--a circumstance
which Mr. Carruthers mentally combined with the beard, of which he
retained an indignant remembrance as a sign of the degeneracy of
the age. The irrepressible pompousness of the man showed itself
even in this crisis of affairs, as he perused the document, and laid it
down upon the table under the hand armed with the eye-glasses.

Clare waited breathless.

 "Hem! my dear," he began; "this letter is connected with the
matter I mentioned to you yesterday. You remember, I daresay,
about the murder, and the inquiry I was requested by the
government to make at Amherst."

O yes, Clare remembered; she had been very much interested.

Had anything since transpired?

"Nothing of any moment. This letter is from Mr. Dalrymple--the

gentleman who came here, as I told you, from Lord Wolstenholme."

Clare, still breathless, bowed. There was no use in trying to

accelerate Mr. Carruthers's speech. He was not to be hurried.

"He writes to me that the Home Secretary regrets very much the
failure of our inquiries at Amherst, in eliciting any information
concerning the only person on whom suspicion has as yet alighted.
He informs me that, as I expected, and as I explained to you
yesterday"--Mr. Carruthers paused condescendingly for Clare's silent
gesture of assent--"the jury at the coroner's inquest (it closed
yesterday) have returned an open verdict--wilful murder against
some person or persons unknown; and the police have been
instructed to use all possible vigilance to bring the criminal to light."

"Have they learned anything further about the dead man?" asked
Clare, with a timid look (half of anxiety, half of avoidance) towards
the newspaper, which Mr. Carruthers had not yet opened, and which
no member of the family would have ventured to touch
unsanctioned by the previous perusal of its august head.

"About the murdered man?--no, I believe not. Mr. Dalrymple

further informs me that the fur-lined coat, and all the other less
remarkable articles of clothing found on the body, are placed in the
hands of the police, in hope of future identification. There is nothing
more to be done, then, that I can see. Can you suggest anything,
Clare!" Mr. Carruthers asked the question in a tone almost of banter,
as though there were something ridiculous in his expecting a
suggestion from such a quarter, but with very little real anxiety
nevertheless.

"I--I really do not know, uncle," returned Clare; "I cannot tell. You
are quite sure Evans told you all he knew?"

"Everything," replied Mr. Carruthers. "The clue furnished by the

coat was very slight, but it was the only one. I am convinced,
myself, that the man who wore the coat, and was last seen in
company with the murdered man, was the man who committed the
murder." Clare shivered. "But," continued Mr. Carruthers in an
argumentative tone, "the thing to establish is the identity of the man
who wore the coat with the man who bought it six weeks ago."

A bright flush rose on Clare's cheeks--a flush of surprise, of hope.

"Is there any doubt about that, uncle?" she asked. "The waiter
described the man, didn't he? Besides, no one would part with an
overcoat in six weeks."

"That is by no means certain," said Mr. Carruthers with an air of

profound wisdom. "Artists and writers, and foreigners, and generally
people of the vagabond kind, sell and barter their clothes very
frequently. The young man whom Evans describes might have been
any one, from his purposeless indistinguishable description; the
waiter's memory is clearer, as is natural, being newer."

"And what is the description he gives?" asked Clare faintly.

"You will find it in the weekly paper, my dear," returned Mr.

Carruthers, stretching his hand out towards the daily journal.
"Meantime let's see yesterday's proceedings."

Hope had arisen in Clare's heart. Might not all her fear be
unfounded, all her sufferings vain? What if the coat had not been
purchased by Paul Ward at all? She tried to remember exactly what
he had said in the few jesting words that had passed on the subject.
Had he said he had bought it at Amherst, or only that it had been