Isometries on Banach spaces function spaces 1st

Edition Richard J. Fleming pdf download

https://ebookname.com/product/isometries-on-banach-spaces-
function-spaces-1st-edition-richard-j-fleming/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Isometries on Banach spaces vector valued function

spaces and operator spaces Richard J. Fleming

https://ebookname.com/product/isometries-on-banach-spaces-vector-
valued-function-spaces-and-operator-spaces-richard-j-fleming/

Topological and Metric Spaces Banach Spaces None

https://ebookname.com/product/topological-and-metric-spaces-
banach-spaces-none/

Handbook of the Geometry of Banach Spaces None

https://ebookname.com/product/handbook-of-the-geometry-of-banach-
spaces-none/

Monastic Practices Revised Edition Charles Cummings

https://ebookname.com/product/monastic-practices-revised-edition-
charles-cummings/
Filter Theory in Electronic Engineering 1st Edition
Marya Troutman

https://ebookname.com/product/filter-theory-in-electronic-
engineering-1st-edition-marya-troutman/

Between Women Friendship Desire and Marriage in

Victorian England Sharon Marcus

https://ebookname.com/product/between-women-friendship-desire-
and-marriage-in-victorian-england-sharon-marcus/

The Party Book Jane Bull

https://ebookname.com/product/the-party-book-jane-bull/

Advances In Chromatography Volume 43 Advances in

Chromatography 1st Edition Phyllis R. Brown

https://ebookname.com/product/advances-in-chromatography-
volume-43-advances-in-chromatography-1st-edition-phyllis-r-brown/

Phase Transitions and Critical Phenomena Vol 19 1st

Edition C. Domb

https://ebookname.com/product/phase-transitions-and-critical-
phenomena-vol-19-1st-edition-c-domb/
Confronting Ecological and Economic Collapse Ecological
Integrity for Law Policy and Human Rights 1st Edition
Laura Westra

https://ebookname.com/product/confronting-ecological-and-
economic-collapse-ecological-integrity-for-law-policy-and-human-
rights-1st-edition-laura-westra/
 CHAPMAN & HALUCRC
Monographs and Surveys in
Pure and Applied Mathematics 1 29

ISOMETRIES
ON BANACH SPACES:
function spaces

O 2003 by Chapman & HallICRC

 CHAPMAN & HALLICRC
Monographs and Surveys in Pure and Applied Mathematics

Main Editors
H. Brezis, Universite' de Paris
R.G. Douglas, Texas A&M University
A. Jeffrey, University of Newcastle upon Tyne (Founding Editor)

Editorial Board
R. Aris, University of Minnesota
G.I. Barenblatt, University of California at Berkeley
H. Begehr, Freie Universitat Berlin
P. Bullen, University of British Columbia
R.J. Elliott, University of Alberta
R.P. Gilbert, University of Delaware
R. Glowinski, University of Houston
D. Jerison, Massachusetts Institute of Technology
K. Kirchgassner, Universitat Stuttgart
B. Lawson, State University of New York
B. Moodie, University of Alberta
L.E. Payne, Cornell University
D.B. Pearson, University of Hull
G.E Roach, University of Strathclyde
I. Stakgold, University of Delaware
W.A. Strauss, Brown University
J. van der Hoek, University of Adelaide

O 2003 by Chapman & HallICRC

 CHAPMAN & HALUCRC
Monographs and Surveys in
Pure and Applied Mathematics 1 29

ISOMETRIES
ON BANACH SPACES:
function spoces

RICHARD J. FLEMING
JAMES E. JAMISON

CHAPMAN & HALUCRC

A CRC Press Company
Boca Raton London NewYork Washington, D.C.

O 2003 by Chapman & HallICRC

 Library of Congress Cataloging-in-PublicationData

Fleming, Richard J.
Isometries on Banach spaces : function spaces I by Richard J. Fleming and James E. Jamison.
p. cm. - (Chapman & HallICRC monographs and surveys in pure and applied
mathematics ; 129)
Includes bibliographical references and index.
ISBN 1-58488-040-6 (alk. paper)
1. Function spaces. 2. Banach spaces. 3. Isometrics (Mathematics) I. Jamison, James E.
11. Title. 111. Series.

QA323 .F55 2002

515'.73--dc21 2002041118
CIP

This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the authors and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, microfilming, and recording, or by any information storage or
retrieval system, without prior permission in writing from the publisher.
The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC
for such copying.
Direct all inquyies to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com

O 2003 by Chapman & HallICRC

No claim to original U.S. Government works

International Standard Book Number 1-58488-040-6
Library of Congress Card Number 2002041 118
Winted in the United States of America 1 2 3 4 5 6 7 8 9 0
Printed on acid-free paper

O 2003 by Chapman & HallICRC

 Contents

Preface

Chapter 1. Beginnings
1.1. Introduction
1.2. Banach's Characterization of Isometries on C ( & )
1.3. The Mazur-Ulam Theorem
1.4. Orthogonality
1.5. The Wold Decomposition
1.6. Notes and Remarks

Chapter 2. Continuous Function Spa,ces-The Banach-Stone Theorem 25

2.1. Introduction 25
2.2. Eilenberg's Theorem 26
2.3. The Nonsurjective Case 29
2.4. A Theorem of Vesentini 39
2.5. Notes and Remarks 42

Chapter 3. The LP Spaces

3.1. Introduction
3.2. Lamperti's Results
3.3. Subspaces of LP and the Extension Theorem
3.4. Bochner Kernels
3.5. Notes and Remarks

Chapter 4. Isornetries of Spaces of Analytic Functions

4.1. Introduction
4.2. Isometries of the Hardy Spaces of the Disk
4.3. Bergman Spaces
4.4. Bloch Spaces
4.5. 9 Spaces
4.6. Notes and Remarks

Chapter 5. Rearrangement Invariant Spaces

5.1. Introduction
5.2. Lumer's Method for Orlicz Spaces
5.3. Zaidenberg's Generalization
5.4. Musielak-Orlicz Spaces

@ 2003 by Chapman & HallICRC

 5.5. Notes and Remarks
Chapter 6. Banach Algebras
6.1. Introduction
6.2. Kadison's Theorem
6.3. Subdifferentiability and Kadison's Theorem
6.4. The Nonsurjective Case of Kadison's Theorem
6.5. The Algebras and A C
6.6. Douglas Algebras
6.7. Notes and Remarks
Bibliography

O 2003 by Chapman & HallICRC

 Preface

Herman Weyl has said that in order to understand any mathematical

structure, one should investigate its group of symmetries. In the class of Ba-
nach spaces, such a goal leads naturally to a study of isometries. A principal
theme in geometry from the earliest times has been the study of transforma-
tions preserving lengths and angles. If the origins of the theory of Banach
spaces are assigned to the appearance of Banach's book in 1932, then the
study of Banach space isometries must be assigned the same starting date.
In his book, Banach included the first characterizations of all isometries on
certain classical spaces. The body of literature concerning isometries that has
grown up since that time is large, perhaps surprisingly so.
An isometry, of course, is a transformation which preserves the distance
between elements of a space. When Banach showed in 1932 that every linear
isometry on the space of continuous real valued functions on a compact metric
space must transform a continuous function x(t) into a continuous function
y(t) satisfying

where Ih(t)l - 1 and p is a homeomorphism, he was establishing a canon-

ical form characterization which fits in an astonishing number of cases. In
this volume we are interested primarily in just such explicit descriptions of
isometries. Our approach, then, will differ from that of many authors whose
interest in isometries lies in showing which spaces are isometric to each other,
or to those whose interest is in discovering properties of topological spaces on
which the functions in the Banach spaces are defined. Such interests as these
have served as excellent motivations for the types of characterizations that
interest us. There have been several excellent surveys concerning isometries,
including those by Behrends, Loomis, Jarosz, and Jarosz and Pathak, each
of which concentrates on some subset of the whole. Our intent was to pro-
vide a survey of the entire subject, and our survey article (1993) serves as an
inspiration and guide for the current work.
Our goal has been to produce a useful resource for experts in the field as
well as beginners, and also for those who simply want to acquaint themselves
with this portion of Banach space theory. We have tried to provide some
history of the subject, some of the important results, some flavor of the wide
variety of methods used in attacking the characterization problem for various
vii

O 2003 by Chapman & HallICRC

 ...
vlll PREFACE

types of spaces, and an exhaustive bibliography. We have probably underes-

timated the enormity of such a project, and perhaps what we have produced
is more of a sampler than a full-blown survey.
We have chosen to organize the material according to the different classes
of spaces under study and this is reflected in the chapter headings. The current
volume is the first of two intended volumes, and as can be seen from the table
of contents, it is primarily concerned with isometries on function spaces. The
first chapter treats some general topics such as linearity, orthogonality, and
Wold decompositions, while Chapter 6 contains material on noncommutative
C*-algebras, but the rest of the chapters treat the classical function spaces.
The second volume will include chapters with the following titles:
Chapter 7: The Banach-Stone Property
Chapter 8: Other Vector-Valued Function Spaces
Chapter 9: Orthogonal Decompositions
Chapter 10: Matrix Spaces
Chapter 11: Norm Ideals of Operators
Chapter 12: Spaces with Trivial Isometries
Chapter 13: Epilogue
In each chapter we try to include an early result of some historical impor-
tance. Other selections are made in order to expose some one of the principal
methods that have been used, or perhaps to give an account of work that
has not received much attention. The chapters, and even the sections within
the chapters, are mostly independent of each other, and the reader can begin
at any point of interest. In making our selections we have, no doubt, left
out many others just as deserving and quite possibly more important. Hope-
fully, most of these omissions in the text are mentioned in the notes and the
bibliography.
The exposition relies mostly on the original papers, and we have tried to
report faithfully on those results, with additional clarifications when possible.
Probably we have included more detail than necessary in some instances, but
we have chosen to err on that side. There are a few places where we have
given only sketchy arguments.
In each chapter we include a section on notes and remarks which give
related results and other approaches that were not included in the main text.
We hope these sections will help to soothe those who disagree sharply with
our choice of material. For the most part, all references are given in the notes
section. The exceptions are in cases where a reference is needed to justify a
statement being made in the text.
In the bibliography we give a representative selection of the works on
isometries, and a serious investigation of all such works available should prob-
ably begin with Math Sci Net using the phrases "isometries on," "isometries
in," or "isometries for." Certainly there are many papers on vector-valued
function spaces and matrix spaces which one might be expected to be men-
tioned here, but we are saving these for the second volume. That second

O 2003 by Chapman & HallICRC

 PREFACE ix

volume will include many more references which were not directly relevant in
this first one, and we also intend for Chapter 13 of Volume 2 to provide a
further guide to the literature.
We assume that our readers are familiar with the standard material in
courses in real variables, complex function theory, and functional analysis.
Terms and notation that are common in those fields we leave undefined in
the text. Page references to some of the special notation are given in the
index. Some notation, of course, serves multiple purposes which should be
understood in the context in which it appears. If some symbol or term is
encountered which is not referenced in the index, the reader should be able
to find it explained within a page or two of that location.
We have received much encouragement for this project from a number of
people over recent years, and we want to mention three people in particular
who have provided special help. Joe Diestel read portions of the work in
early stages and his kind words helped move us forward. Bill Hornor has
provided valuable advice, particularly in regard to Chapter 4 on analytic
functions. David Blecher helped immeasurably in reading much of Chapter
6, and tried to guide us in understanding the material on operator spaces and
the nonsurjective case of Kadison's theorem. However, we strongly emphasize
the fact that we alone are responsible for any existing errors.
Finally, we would both like to express our deep appreciation and love for
our wives, Diane Fleming and Jan Jamison, for their patience and devotion.

Richard J . Fleming and James E. Jamison

August 30, 2002

O 2003 by Chapman & HallICRC

 CHAPTER 1

Beginnings

1.1. Introduction
Isometries are, in the most general sense, transformations which preserve
distance between elements. Such transformations are basic in the study of
geometry which is concerned with rigid motions and properties preserved by
them. The isometries of the Euclidean plane may all be described as rotations,
translations, reflections, and glide reflections, and these transformations form
a group under the operation of composition. This group is sometimes called
the Euclidean group of the plane. Of course the Euclidean group is very large
and often certain subgroups are sought which preserve some particular subset
of the plane. If S is a subset of the Euclidean plane R 2 , the subgroup G
which consists of all isometries which map S onto itself is called the complete
symmetry group of S . A subgroup of G is called a symmetry group of S .
The symmetry group of the unit circle given by an equation

relative to a fixed coordinate system is sometimes called the orthogonal group

in the plane. It can be seen that each transformation in this group must leave
the origin fixed and is therefore a linear transformation. This is a special case
of a more general result which we will prove shortly.
The fact that transformations in the orthogonal group are linear allows
them to be represented by matrices which are of two forms:

which represents a rotation through an angle 0; and

which represents a reflection of the plane with respect to the line given by the
equation

The rotations form a subgroup of the symmetry group called the rotation
group.

O 2003 by Chapman & HallICRC

2 1. BEGINNINGS

If instead of the circle, we find the symmetry group of a regular n-gon,

we get the dihedral group D,. For example, the case n = 4 gives the sym-
metry group for the square. Clearly rotations of 7r/2 and succeeding powers
are isometries as are reflections across the four symmetry axes consisting of
lines through opposite vertices and lines passing through midpoints of oppo-
site sides. Thus D4 has eight elements whose matrix forms are particularly
nice. More generally, D, consists of 2n elements which include n rotations in
multiples of 27r/n and n reflections across symmetry axes as follows:
If n is even, there are n/2 axes passing through opposite vertices
and n/2 axes passing through midpoints of opposite sides; if n
is odd, each edge has an opposite vertex and an axis passes
through the midpoint of an edge and its opposite vertex.
We mention these well known facts simply to call attention to the basic
role of the study of isometries in geometry as well as in elementary group
theory. Before leaving this discussion of the isometries of the plane, however,
let us note that the orthogonal group (that is, the symmetry group of the
circle) is the set of all linear isometries of the real two dimensional Hilbert
space t 2 ( 2 ) . The dihedral group D4 is the group of all linear isometries of
the real two dimensional Banach space t m ( 2 ) (or for that matter, P ( 2 ) for
1 < <
p m , p # 2). By P ( n ) we mean the space of n-tuples of scalars
with the p-norm given by llxllp = (C I X ~ ~ P ) ( for <
~ ' ~ 1) p < m and Ilxllm =
sup{lxjl : j = 1 . . . n ) . The group D4 is very small in comparison with the
orthogonal group which is a common observation when comparing isometries
of a Banach space to those of a Hilbert space. In fact, a typical element of
D4 can be described by the simple formula
(1) g(t) = h(t)f ( ~ ( t ) )
where f = f (t) denotes a real function on {1,2), p is a permutation of { l , 2 ) ,
and h is a real function with Ih(t)l = 1 (unimodular). A formula something
like (1) will appear often in the sequel because descriptions of isometries are
usually given by this type of canonical form.
Herman Weyl [321, p.1441 has said that in order to fully understand a
mathematical structure one should investigate its group of symmetries. It is
not surprising, then, that mathematicians would be interested in describing
the isometries on particular Banach spaces and that this particular quest has
led to a vast literature in the subject. Nor is it surprising to learn that the
earliest results in the modern setting would be the work of Banach in his 1932
treatise on linear operators. Indeed, we open the door to our survey of the
subject by examining in detail Banach's description of the form of an isometry
on the space of continuous functions.

1.2. Banach's Characterization of Isometries on C ( Q )

Suppose that Q is a compact metric space. By C ( Q ) we mean the Banach
space of continuous real valued functions defined on Q with the supremum

O 2003 by Chapman & HallICRC

 1 . 2 . BANACH'S CHARACTERIZATION O F ISOMETRIES ON C ( Q ) 3

norm. We begin with Banach's characterization of a peak point for a function

f in relation to the existence of directional derivatives of the norm at f .
Let f
1 . 2 . 1 . LEMMA. E C(Q) a n d so E Q. I n order that

(2) If(so)l > If(s)l f o r each s E Q with s # so,

i t is necessary a n d s u f i c i e n t t h a t

lim
t+O
Ilf +tgll
t
- llfll
exists f o r each g E C(Q).
M o r e o v e r , i f f satisfies (2), w e have

lim f + t g
t+O t
- fII =g(~o)sgnf(so)

f o r each g E C(Q).
PROOF. Let us first show that ( 3 ) is necessary. If ( 2 ) holds, then 1 1 f 1 1 =
If E C(Q) and a real number t , f t g is continuous on Q
( s o )1. Now given g +
and so attains its maximum absolute value at some st E Q . Therefore,

(5) If(s0) + tg(so)l - I f ( ~ 0 ) l < Ilf + tgII - Ilfll = If(%) +tg(st)l - If(s0)l.
Also, we have
If(s0) +tg(so)l < I f ( % ) +tg(st)l
and a little manipulation yields the inequality

0 < If(s0)l - If(st)l < Itlls(s0)l + Itllg(st)l 21tlllsll.

It now follows that limt+o If (st) = 1 If 1
( s o ) and the compactness of Q allows
us to conclude that

(6) lim st = so.

t+O

Now let us first suppose that f ( s o ) < 0. By virtue of the fact that
st + so, we may choose t so small that
If (so) + tg(s0)1 - If (so) =l -f (so) - tg(s0) + f (so) = -tg(so)
and
If(%) + tg(.t)l - lf(.t)l = -f(.t) - tg(st) + f ( s t ) = -tg(st).
From these two statements and ( 5 ) we see that

-tf(so) < Ilf + t g l l - llfll < -ts(st)

for sufficiently small t which combined with ( 6 ) and the continuity of g leads
to

lim I f + tgll - f l l = -g(so),

t+O t

O 2003 by Chapman & HallICRC

4 1 . BEGINNINGS

The case where f ( s o )> 0 (which is the case considered in Banach's book)
can be treated in a similar manner to establish that

lim If +tsll - l l f l l = g ( s o ) ,
t+O t
This completes the proof of the necessity of the existence of the limit in ( 3 )
and shows that (4) must hold.
For the sufficiency let us assume that so, sl E Q with so # sl and

llfll = If(s0)l = If(s1)l > If(s)l

for all s E Q . If f ( s o ) < 0 , define g ( s ) = - d ( s , s l ) where d denotes the metric
on Q . Then

l l f + tsll - l l f l l > I f ( s 0 ) + tg(s0)l - If(s0)l = - f ( s o ) - t g ( s 0 ) + f(s0)

= t d ( s 0 ,s1)
for all sufficiently small t . We conclude that

However,

Ilf +tgll - Ilfll > I f ( s 1 ) + t d ( s 1 , s1)l - If(s1)l = 0 .

for all h , whereby we must have

lim sup I l f +tsll - llfll <0

t+O- t
The inequalities ( 7 ) and ( 8 ) show that the limit ( 3 ) cannot exist.
For the case where f ( s o )> 0 , we define f ( s ) = d ( s , s l ) and give a similar
argument.

We now state and prove the theorem of Banach for surjective isometries
on C ( Q ) spaces.

1.2.2. THEOREM.If Q and I?: are compact metric spaces then for the
spaces of real continuous functions C ( Q ) and C(I?:) to be isometrically iso-
morphic it is necessary and suficient that Q and I?: be homeomorophic. In
this case, an isometric isomorphism T from C ( Q ) onto C(I?:) must be given
by
(9) T f ( t ) = h ( t ) f( ~ ( t )f o)r t E I?:,
where p is a homeomorphism from I?: onto Q and h is a real valued unimodular
function on I?:.
PROOF.It is easy to see that if p is a homeomorphism from I?: onto Q ,
then a transformation U defined by ( 9 ) is an isometric isomorphism of C ( Q )
onto C(I?:) and thus the sufficiency of the condition is clear.

O 2003 by Chapman & HallICRC

 1 . 2 . BANACH'S CHARACTERIZATION O F ISOMETRIES ON C ( Q ) 5

For the necessity, let T be a linear isometry from C ( Q ) onto C(I?:), sup-
pose so € Q and let f € C ( Q ) be such that If ( s o )1 > If (s) 1 for all s E Q. By
Lemma 1.2.1,

lim f + rgll - f " = g(so)sgnf(so)

r+O 7'

must exist for every g E C ( Q ) . Since T is an isometry,

llTf +rTsll - IlTfll

(10) g(s0)sgnf (so) = lim
r+O 7' r+O 7'

and we apply the lemma again to conclude that there is some t o E I?: such
that

ITf(t0)l > ITf(t)l f o r all t E I?: with t # so

(It is important to note here that we can apply the lemma since Tg runs
through all of C(I?:).) Furthermore, we may conclude from (10) and (4) that

g(so)sgnf (so) = Tg(to)sgnTf (to)

If we let h(to) = sgnf(so)Tf(to), then Ih(to)l = 1 and we get

Let us define $ from Q to I?: by $(so) = t o Now $ is injective, for if

$ ( s l ) = $ ( s 2 ) , then by (11) we have Ig(sl)1 = Ig(s2) 1 for all g E C ( Q ) and so
s l = s2. To see that $ is surjective, let t o E I?: and define q on I?: by
1
q(t) =
1 + d(t,to)
where d denotes the metric on I?:. I f f = T P l q , then by (11)

for each s E Q. Since 11 f l l = llqll = 1, there exists so E Q such that I f ( s o ) l = 1.

1
Therefore, l+d($(s),to) = 1 which implies that t o = $ ( s o ) .
Finally, suppose that {s,) is a sequence in Q converging to so, g E C(I?:)
and T f = g. Since Ig($(sn) 1 = If ( s n )1 for each n and f is continuous, we
have If (sn) 1 + If ( S O )1 so that Ig($(sn)) 1 + I ~ ( $ ( s o ) ) 1 for every g E C(I?:).
By choosing g defined by g(t) = d(t, $ ( s o ) ) , we obtain

from which we conclude that $ ( s n ) + $(so). This shows that $ is continuous

and since Q is compact and I?: is Hausdorff, $ must in fact be a homeomor-
phism. Hence Q and I?: are homeomorphic and if we let p = $ - I , we get from
(11) the characterization of U given by (9).

O 2003 by Chapman & HallICRC

6 1. BEGINNINGS

1.2.3. COROLLARY. (Banach) If T is a surjective linear isometry on

C ( Q ) where Q is compact and metric, then

where Ih(t)l = 1 and p is a homeomorphism of Q onto itself.

We note that the statement in Corollary 1.2.3 gives the characterization

of all surjective linear isometries on the given Banach space and it is this type
of result that is the primary subject matter of this book. It is also important
to see that Corollary 1.2.3 describes the symmetry group of the unit ball of
the given space C ( Q ), in the language given in the introduction, and that the
elements of the symmetry group have the canonical form as given by (1). Of
course, the space P ( 2 ) discussed there is simply C ( Q ) where Q is the set
{1,2) given the discrete topology.
Banach's characterization then is truly the first attack on the general
problem of identifying the symmetry group of the unit ball of a given Banach
space; that is, characterizing the surjective, linear isometries on the space.
The proof of Banach's theorem uses in an essential way the assumptions
that U is both linear and surjective. The surjective hypothesis will be invoked
often (but not always) in remaining chapters, and the type of isometries of
interest to us are the linear ones, so that we could have included the words
"linear isometries" in the title. That the restriction to linear isometries is not
really a serious defect is the subject of the next section.

1.3. The Mazur-Ulam Theorem

In our discussion of the isometries of the plane we noted that the orthogo-
nal group consisted of linear transformations. Any linear transformation must
fix the origin and any isometry U on a normed space X can be "shifted" so
that it fixes the origin; just consider V = U - U ( 0 ) . It is a remarkable and
useful fact, then, that any surjective isometry on a real normed linear space
which fixes the origin must be linear.
To begin the discussion of the proof of this fact, let us introduce some
notation and state a couple of lemmas. If x , y are elements of a n.1.s. X ,
let H1 = H l ( x , y) denote the set of elements u E X such that IIx - ull =
lly - u I I = 211x
1
- yll. For n = 2 , 3 , . . . , let H, be the set of u E H n P 1 so that

<
I I u - vII $S(H,-~) for v E HnP1. Here, S(HnP1)denotes the diameter of
H n P 1 which is, of course, the supremum of the distances between pairs of its
elements. Clearly, S(H,) < &IIx - yII for each n . Hence, the intersection
of the H, is either empty or consists of exactly one element which is called
the (metric) center of the pair x , y.

1.3.1. LEMMA.(Mazur-Ulam) If x , y are elements of a n.1.s. X , then

$(x + y) is the center of the pair x , y.

O 2003 by Chapman & HallICRC

 1.3. T H E MAZUR-ULAM THEOREM 7

PROOF. For each u E X let 6 = x y + - u. If u E H l ( x , y), then

116 - xi1 = lly - u I I and 116 - yII = IIx - -11 SO

because u E H I . Assume t h a t 6 E H,-1 whenever u E H,-1 and let u E H,.

If v E HnP1,we have

since 6 E HnP1and u E H,. Therefore, 6 E H, as well and the upshot is

t h a t , by induction, for each positive integer n , 6 E H, whenever u E H,.
+
Next we show by induction t h a t z = $ ( x y) E H, for each n . First
we see t h a t z E H1 since llz - xll = $ 1 1 ~ - yll. Assume t h a t z E HnP1and
u E HnP1. Thus 6 E HnP1by what we proved earlier and
2llz - u I I = IIx + y - 21.1 = 116 - u I I < S(HnP1).
Hence, llz - ull <
$S(H,-~) and we have z E H,. T h e conclusion is t h a t
z E n;"H, and so is the center of x , y.

If T is an isometry from a n.1.s. X into a n.1.s. Y , then T obviously maps

Hl ( x , y) into H1( T x , T y ) , since

and similarly for IITu - Tyll. This requires no linearity nor surjectivity for T .
1.3.2. LEMMA.If T is a surjective isometry from X onto Y , then T maps
the center of any pair x , y in X to the center of the pair T x , T y in Y .
PROOF. Since T is a surjective isometry, it is clear t h a t ~ ( H l ( xy)) , =
H 1 ( T x , Ty) and S ( H l ( x , y)) = S(H1(Tx,T y ) ) . If we assume this holds for
H n P 1 ( T x ,T y ) , then for u E H,(x, y) and w E H , - l ( T x , T y ) , we have w = T v
for some v E H,- 1( x , y) while
1 1
T U- T v = Ilu - ull < - 6 ( H n - l ( ~ ,Y ) ) ~ ~ ( H ~ - I ( T x , T Y ) ) .
2
Therefore, T u E H, ( T x , T y ) . Similarly, if w E H, ( T x , T y ) , then w = T u
for some u E H, ( x , y). Thus by induction, T(H, ( x , y)) = H, ( T x , Ty) for
+
every ~ o s i t i v einteger n . By Lemma 1.3.1, $ ( x y) E H, ( x , y) for every n
+
and it follows from the statement above t h a t T ( $ ( x y)) E H,(Tx, Ty) for
every n . Hence T ( $ ( x + y)) is the unique element of n r = l H n ( T x , Ty) which
is $ (TX + Ty) by Lemma 1.3.1 again.
1.3.3. LEMMA.If T is an isometry from a n.1.s. X onto a n.1.s. Y (real
or complex), then
+
(i) T ( x y) = T x + T y - T(O),
(ii) T ( s x ) = s T x + (1 - s)T(O)for all real numbers s.

O 2003 by Chapman & HallICRC

 1. BEGINNINGS

PROOF.If x E X , then from Lemma 1.3.2, we have

so that

(la) T(2x) = 2T(x) - T(0).

+ +
Now applying 1.3.2 again, T ( x y) = T($(2x 2y)) = $ ( ~ ( 2 x ) T(2y)) and +
a double application of (12) yields (i) above.
Using (i) it is easy to show by induction that (ii) holds for all positive
integers. Since

we conclude that (ii) holds for s = -1 and therefore for any integer n . Upon
applying (ii) to T ( x ) = T ( n z ) we get that (ii) holds for all rationals and the
extension to real follows from the continuity of T .

1.3.4. THEOREM.If T is an isometry from a n.1.s. X onto a n.1.s. Y

such that llTxll = 1 for every x E X with llxll = 1, then T(0) = 0.
PROOF.Suppose z E X and T z = 0. Then
llTOll = llTz - TO11 = llz - 011 = 11z11.
Suppose z # 0. By hypothesis and Lemma 1.3.3((ii)),

from which it must be concluded that llzll = 2. However, we must also have

which is absurd. Therefore z = 0, and the theorem is proved.

1.3.5. THEOREM.(Mazur-Ulam) If T is an isometry from a n.1.s. X

onto a n.1.s. Y, and if T(0) = 0, then T is real linear.
PROOF.The proof is immediate from Lemma 1.3.3.

The assumption of surjectivity in the above theorem is necessary as can

be seen from the following easy example.
1.3.6. EXAMPLE.(Figiel) Let X = R with the absolute value norm and
Y = t m ( 2 , R ) . Then U(a) = ( a , sina) defines a nonlinear isometry from X
into Y with U(0) = (0,O).
In order to remove the surjectivity assumption, we must put a condition
on the range space. Recall that a normed linear space is strictly convex if
11x + yll = IIxII + llyll implies that x , y are linearly dependent, and in fact,
>
x = t y for s o m e t 0.

O 2003 by Chapman & HallICRC

 1.3. T H E MAZUR-ULAM THEOREM 9

1.3.7. LEMMA.(Baker) If Y is a normed linear space which is strictly

convex and x , y E Y , then H l ( x , y) is a singleton (and therefore consists of
f +
(x Y).
PROOF.We have already seen that f (x + y) E H l (x, y). Suppose u, v E
H l ( x , y). Then

Similarly, 11 y - (112) (u + v) 11 5 f llx - yll. If either inequality above is strict,

then
llx - YII 5 llx - ( 1 / 2 ) ( u + u)ll + IIY - (112)(u+ u)ll < llx - YII.
From this we can see that the inequalities must be equalities and

By the strict convexity, (x - u) = t ( x - v) for some t > 0 and since IIx - ull =
11x - vll, we have t = 1. It follows that u = v.

1.3.8. THEOREM.(Baker) If T is an isometry from a real normed linear

space X into a strictly convex, real normed linear space Y such that T(0) = 0,
then T is linear.
+
PROOF.For x , y E X , we have seen that (112) (x Y) E H i ( x , Y) and
+
T((1/2) (x y)) E H1 ( T x , Ty) . However, f (TX +
T Y ) E H I ( T x , T Y ) which is
a singleton by Lemma 1.3.7. Hence,
1
+
T((1/2)(x Y)) = -(Tx
2
TY) +
and the proof is completed as in the proof of the Mazur-Ulam Theorem.
It can be seen that if Y is not strictly convex, then there is a normed
linear space X and a nonlinear isometry T from X into Y such that T(0) = 0.
In fact, we can take X = R and if u and v are linearly independent, norm-one
+ +
vectors in Y with Ilu vll = llull IluII, define T by

The extension of these results to get complex linearity is not possible

since the isometry Uz = 2 is not complex linear on C. This simple operator is
conjugate linear but the equally simple operator T(z1, z2) = (zl , z)
is neither
linear nor conjugate linear on C2.

O 2003 by Chapman & HallICRC

10 1. BEGINNINGS

1.4. Orthogonality
Isometries on Hilbert spaces preserve the inner product and they preserve
orthogonality. Although these notions are not so natural in the Banach space
setting, they can be defined and we want to see how they relate to isometries.
1.4.1. DEFINITION.(Lumer) A semi-inner product (s.2.p.) on a complex
vector space X is a complex valued form [., .] on X x X to C which satisfies
(i) [ x , x ] > 0 if x # 0,
(ii) [ a x + p y , z I = a [ x , z ] + P [ y , z I f o r a , p E C, x , Y ,E~X ,
<
(iii) ~ [ X , Y I ~ [~x , x ] [ Y , Yfor
] all X , Y E X .
It can be shown that llxll = [x, xI1l2 defines a norm on X (with respect
to which the s.i.p. is said to be compatible) and conversely, if 11 . 11 is a norm
on X there is a s.i.p. on X which is compatible with the norm. This follows
from the Hahn-Banach Theorem which guarantees the existence of duality
maps x + p, which satisfy llp,ll = IIxII and p x ( x ) = 11x112,where p, E X * .
The functional p, is called a support functional. Such a duality map gives a
s.i.p. by means of the formula
[x, YI = P?,(x).
Of course, such maps are not unique and so there may be many semi-inner
products compatible with a given norm. It is always possible to choose a
semi-inner product which satisfies
[ s ,XYl = X[x, Yl .
1.4.2. DEFINITION.(James) In a normed linear space ( X , 11 . II), an ele-
ment x is said to be orthogonal to y (written x I y) if
llxll < IIx + Xyll for all scalars A.
1.4.3. PROPOSITION. (Giles) If [., .] is a semi-inner product compatible
with the norm of X , then [y, x] = 0 implies that x I y.
PROOF.Let X be a given scalar. Then
1 1 ~ 1 =1 ~ [x, XI = [x + XY, XI < 11x + X Y l l l l ~ l l
so that llxll < + Xyll.
1111:

The converse of the proposition is not true. For example, if x = ( 1 , l )

and y = ( 0 , l ) in t m ( 2 ) , then x is orthogonal to y but [y, x] = 112 where the
semi-inner product is defined by

assuming x = ( x l , x2) and y = (yl, y2).

We should note that in general if x I y, there is a s.i.p. compatible with
the norm such that [y, x] = 0, but this choice of s.i.p. depends on the vectors

O 2003 by Chapman & HallICRC

 1.4. ORTHOGONALITY 11

x , y. Thus we get a converse of sorts, and, in fact, the real thing when X is
smooth.

1.4.4. PROPOSITION. If x , y are elements of a n.1.s. X and x I y, then

there is a s.i.p. [., .] compatible with the norm such that [y, x] = 0. If X is
smooth, then x I y if and only if [y, x] = 0 where [., .] is the unique s.i.p.
compatible with the norm.

PROOF.Suppose x I y. Then there is a linear functional f defined on

the span of x and y such that f (x) = 1 1 ~ 1 and
1 ~ f (Y) = 0. For any scalars a ,P,

<
Hence l l f l l IIxII and it follows that l l f l l = IIxII. By the Hahn-Banach
Theorem there exists x* E X * with IIx*II = l l f l l = 11x11, x*(y) = 0 and
x*(x) = 11~11~. Let p be a duality map such that p ( x ) = x*. Then the s.i.p.
determined by p satisfies the conclusion of the proposition.
If X is smooth, then there is only one duality map from X to X * , hence
only one s.i.p. compatible with the norm and it has the desired property.

1.4.5. THEOREM.(Koehler and Rosenthal) Let X be a normed linear

space and let U be a linear operator mapping X into itself. Then U is an
isometry if and only if there is a semi-inner product [., .] such that [Ux, Uy] =
[x, YI.
PROOF.The sufficiency of the condition [Ux, Uy] = [x, y] for U to be an
isometry is trivial upon taking x = y.
On the other hand, if U is an isometry, let [., .] be any semi-inner product
and note that for any x , y E X ,

so that {[Unx,Uny]) is a bounded sequence of scalars. Let F be a linear

functional of norm 1 on trnsuch that (i) F ( ( 1 , 1 , .. .)) = 1 and (ii) F ( ( x n ) ) =
F ( ( X , + ~(i.e.,
) ) F is a Banach limit whose existence is given by Banach [21,
p.341). Then define

The properties of F guarantee that [[x,y]] defines a semi-inner compatible

with the norm and [[Ux,Uy]] = [[x,y]] follows by property (ii) of F .

In a Hilbert space it is easy to see that a linear isometry must preserve

orthogonality because it must preserve the inner product. Conversely, a non-
zero linear operator T on a Hilbert space which preserves orthogonality must
be a positive multiple of an isometry. To see that, suppose first that x is a
nonzero element of a Hilbert space X with inner product (., .) and let y be
any nonzero element for which Ty # 0 and such that (x, y) # 0. Let a # 0

O 2003 by Chapman & HallICRC

12 1. BEGINNINGS

be such that (x + a y , y) = 0. If T x = 0, and since T preserves orthogonality

we get that

from which we conclude a = 0 and (x, y) = 0 contrary to assumption. Hence

T x # 0 whenever x # 0. Now suppose x , y are any two nonzero elements
+
of X which are not orthogonal. Again choose a so that (x a y , y) = 0 and
therefore (Tx + a T y , Ty) = 0 by the given property of T . The conclusion is
that

Thus

and by interchanging the roles of x and y in (14) we get

or
( T x , T x ) - (TY,T Y )
(15)
(x, x) (Y, Y)'

If (x, y) = 0, choose z # 0 which is orthogonal to neither and apply (14)

to the pairs x , z and z, y to conclude that (15) holds for all pairs x , y which
are not zero. From this, it follows that if we fix y and let

1 .
then llTxll = rIIxII for a11 x E X so that U = -vT is the desired isometry
This argument makes heavy use of the properties of inner products and
does not carry over to the Banach space case. Yet there is a germ of an
idea here which Koldobsky has exploited to obtain the result, at least for real
Banach spaces.
1.4.6. THEOREM.(Koldobsky) Let X be a real Banach space and T a
linear operator on X to itself which preserves orthogonality. Then T = r U
where r E R and U is an isometry.
To explain the proof of this theorem we must first introduce some notation
and observe some facts. If x and y are given elements of a normed linear space
X , then continuity of the norm guarantees that there is at least one a such
that x + a y is orthogonal to y since I Ix + ryl 1 must attain its minimum value for
some scalar a . However, such a is not necessarily unique and the convexity of
the function a + 11x + ayll (and the continuity) implies that the set A(x, y)
of all a such that (x + a y ) Iy is a closed, bounded interval [a, b], in the case
where X is a real Banach space.

O 2003 by Chapman & HallICRC

 1.4. ORTHOGONALITY 13

It is useful to note here that if T is a linear operator which preserves

orthogonality, and x # 0, then T x # 0. For if T x = 0, one can find y with
Ty # 0 and x not orthogonal to y. There exists a # 0 such that y is orthogonal
+
to x a y and so Ty must be orthogonal to itself which is not possible.
For a nonzero x in X , let
D ( x ) = {x* E X * : 11x*11 = 1, x*(x) = 11x11}.
Then for every y E X ,

and the corresponding limit as a + 0- is given by

in f {x*(y) : x* E D(x)}
+
Hence, the function p defined on R by p ( a ) = 11x ayll is differentiable at a
+
if and only if x; (y) = x;(y) for every x;, x; E D ( x a y ) . Finally, we observe
here that if x , y are linearly independent, the function p is convex on R and
so differentiable almost everywhere with respect to Lebesgue measure. Let
D, (x, y) denote the set of all a at which p is differentiable; then R\D,(x, y)
is of measure zero.
It is convenient to state and prove some lemmas which are due to Koldob-
sky.
1.4.7. LEMMA. If a E D,(x, y) and P, y E R , then
(i) + +
IIx ayll = 11x ayll for all a E A(x, Y) = [a, b ] .
(ii) +
The number x*(px yy) does not depend on x* E D ( x a y ) . +
(iii) +
x + a y I px + yy if and only if x*(px yy) = 0 for every x* E
+
D(x ay).
(iv) Either x + a y I y or there exist a unique number f ( a ) E R such that
x + a y I x - f(a)y.
PROOF. +
(i) As we observed above, if x a y I y, then IIx ayll is the +
+
minimum value of I Ix r y 11 for r E R , hence is the same for all such a .
(ii) We have already seen that x*(y) is the same for all x* E S ( x a y ) . +
Since
x*(x) = x*(x + ay) - a x * ( y ) = 11x + ayll - ~x*(Y),
we see that this number is also independent of x* E S ( x + a y ) and therefore
so is the linear combination px yy. +
+
(iii) If x + a y I px yy, then by (16),

sup x* (px + yy) = lirn,+,+

(llx + i*Y + r(Px + 7Y)ll) > 0
x'€S(x+ory) r
and similarly.
inf
x*€S(x+ory)
x*(px + yy) < 0.

O 2003 by Chapman & HallICRC

14 1. BEGINNINGS

It follows from ((ii)) above t h a t

SUP X* + yy) = inf x* (Px + TY) = 0.
(px
1f x * ( p+~yY) = 0 for every x* E S(X+ a y ) , then
+ a y + r ( P x + yy)) =
X*(X + ~ Y =I 1 1 +~ ~ Y I I
X*(X

for every r and since IIx*II = 1, we have x a y I ,!?x yy. + +

+
(iv) Let x* E S ( x a y ) . If x*(y) = 0, then x a y I y by (3) above. +
Otherwise, x + a y I x - /3y if and only if x'(x - Py) = 0 so we may choose
P = f (a) = x* ( x ) l x *(Y).
As previously observed, the set A ( x , Y) of all a such t h a t x +ay I y is a
closed interval which we denote by [a, b] in the lemma below.
1.4.8. LEMMA.
(i) F o r every a > b, IIx + ayll = I I x + byllexp (JbOl(t+ f ( t ) ) - l d t ) .
(ii) F o r every a < a ,

PROOF. (i) If a E D ( x , y) with a > b and X* E S(X a ~ )we , have +

x'(x) = f ( a ) x ' (y) and x*(y) = $(a) where p(a) = 11x ayll. Then +
x*(x) = x*(x + ay) - a x * ( y ) = 11x + ayll - acp'(a)
which leads t o

Now the set R \ D ( x , y) has Lebesgue measure zero, and we can write

T h e function a
6"x
+
6 dt = "(t + f(t))-ldt for every a

In p(a) satisfies a Lipschitz condition and is absolutely

> b.

continuous. Hence

from which we obtain

x + a y l l = x + b y ~ ~ e x P L "+( t f ( t ) ) - l d t .
T h e proof of (ii) is similar.
We are now ready t o give the proof of Theorem 1.4.6.
PROOF. We assume t h a t T is a nonzero operator on X which preserves
orthogonality. Let x be such t h a t T x # 0 and suppose y is any element of X
so t h a t x and y are linearly independent. Since T preserves orthogonality, it
is clear t h a t A ( x , y) is contained in A ( T x , T y ) . Suppose a E A ( T x , T y ) . If
a E A ( x , Y ) , then by Lemma 1.4.7(iv), there exists f (a) such t h a t x a y I +

O 2003 by Chapman & HallICRC

 1.5. T H E WOLD DECOMPOSITION 15

+ +
x - f ( a ) y and so T x a T y IT x - f ( a ) T y . If x* E S ( T x a T y ) we have by
Lemma 1.4.7(iii) that x*(Ty) = 0 and x* (Tx - f ( a ) T y ) = 0 so that we must
conclude that

This is not ~ o s s i b l esince x + a y is not orthogonal to y and therefore T ( x + a y )

cannot be zero. The contradiction requires us to conclude that A(x, y) =
A(Tx, Ty) so that the numbers a , b and the function f ( a ) are the same for
both pairs (x, y) and ( T x , Ty).
As we know from Lemma 1.4.7 (i), there are constants cl, cz such that
+
11x + ayll = cl and IITx aTyll = cz for a E [a,b]. It now follows from
+
Lemma 1.4.8 that IITx + aTyll = 211x ayll for every a E R. For a = 0
we get llTxll = CZIIxII;
C1
for a > 0 we have 1
19+ Tyll = 211; + yll, and as
a + oo we get IlTyll = 2 ~ Hence,
. = for every x , y # 0 and the
conclusion of the theorem follows.

1.5. The Wold Decomposition

A linear isometry on a Hilbert space which is surjective is called a unitary
operator and this term is sometimes used to refer to surjective isometries on
general Banach spaces. It is a well known fact that every isometry on a
Hilbert space can be written as a direct sum of a unitary operator and copies
of the unilateral shift. A proof of this result makes use of the fact that if V
is an isometry on H , then M = n V n H reduces V and the restriction VIM is
unitary. If L is the orthogonal complement of V ( H ) , then V restricted to the
closed linear span N of the subspaces Vn(L) is a direct sum of copies of the
unilateral shift and H = M e N. The multiplicity of the shift is the dimension
of L.
This decomposition, sometimes referred to as the Wold Decomposition,
lays bare the structure of all isometries on Hilbert spaces. Of course, the
unitary operators are numerous and completely determined by pairings of
orthonormal bases, while the structure of unilateral shift operators is also well
understood. In the Banach space setting, isometries which are not surjective
often present difficulties, and if the goal is to find a crisp description such as
the canonical form, the search is often restricted to the surjective case.
In some instances, a type of Wold Decomposition can be obtained and
we want to say a little about that here. The given development is the work
of Campbell, Faulkner, and Sine.
1.5.1. DEFINITION. Let V be an injective linear map on a Banach space
X . Then V is said to be a unilateral shift provided there is a subspace L of X
for which X = $r
Vn (L). That is, X is the direct sum of the spaces Vn (L).
The subspace L is referred to as a wandering subspace and the dimension of
L is called the multiplicity of V.

O 2003 by Chapman & HallICRC

16 1 . BEGINNINGS

1.5.2. DEFINITION. A n isometry V of a Banach space X is said to be

a Wold isometry provided X = M , $ N , where M , = n ; " V n ( X ) and
N , = C r $ V n ( L ) . Here, the space L is a complement for the range of
V , and N , is written as a Schauder decomposition. (This means that each
element x has a unique representation as x = C r x, where x, E V n( L ) . )
Note that V restricted to M , is a surjective isometry while V restricted to
N , is a shift.

Suppose V is an isometry on X and V ( X ) is complemented in X . Let

L denote a complement of V ( X ) so that X = V ( X )$ L , and suppose PI is
a bounded projection from X onto V ( X ) . Let P2 = V P I V - ' P I and if P,
has been defined, let P,+l = VP,V-'P,. The inductively defined sequence
{P,) is an abelian family of projections with Pk projecting X onto V ~ X )
and annihilating L $ V ( L )$ V 2 ( L )$ . . . $ v"'(L). This sequence will be
referred to as a sequence of projections associated with V .

1.5.3. THEOREM.(Campbell, Faulkner, and Sine) Let V be an isometry

of a reflexive Banach space. If the range of V is complemented and V has
a uniformly bounded associated sequence of projections, then V is a Wold
isometry.

PROOF.Let X denote a reflexive Banach space and V and isometry on

X as described in the hypotheses. Let L and {P,) be as described above.
Then we are assuming that there is some real number b > 0 so that lipn 1 1 < b
for every n . Let M , = n;" V n( X ) and N , = $r
V n( L ) where L denotes the
complement of V ( X ) and the null space of P I .
+
If x E X , then x = Pnx ( I - P n ) x for each n and {P,x), { ( I - P,)x)
form bounded sequences. By the Eberlein-Smulian Theorem [89, p.4301, there
exist xo, yo which are weak limits of a subsequence of {P,x) and { ( I - P,)x)
respectively so that x = xo + yo. For a given n , P k ( x ) E V n ( X ) for all k
sufficiently large and V n( X ) is weakly closed so that xo E V n( X ) . Therefore
xo E M , and since ( I - P n ) x E N , for each n , yo E N,.
Next we see that if x E M , n N,, then we can choose a sequence I n k )
of integers with the property that x,, E L + + +
V ( L ) . . . V n k P (1L ) and
<
x,, + x . It follows that P n k ( x- x,,) = x and IIP,(x - xnk)ll bllxnk - xII.
Hence x = 0 and we conclude that X = M , $ N,. The final piece is to see
that N , can be written as a Schauder decomposition C r $ V n ( L ) .
If y E N , , then y can be approximated by elements from L+ V ( L )+. . .+
V n P 1 ( L and
) P, annihilates such elements. From this we see that P,(y) +
0 and since ( I - P,)(y) = c : - ~ ( P ~ (-~Pk+1(y))
) we may conclude that
y = C y ( P k ( y )- Pk+l(y)) where Pk(y) - Pk+l(y) E V ~ Lfor ) each k . This
expansion is unique because if z E v ~ ( L ) ~ with v ~ i(<Lj ,)then 0 = P j ( z ) =
-

An isometry whose range is 1-complemented is said to be orthocomple-

mented. The language reflects the fact that there is a projection P from X

O 2003 by Chapman & HallICRC

Exploring the Variety of Random
Documents with Different Content
Gallanol.
Gelsemium: in malarial and sthenic fevers, especially in pneumonia
and pleurisy.
Guaiacol: topically.
Hot Affusions: for headache sometimes better than cold.
Hydrastis: inferior to quinine in intermittent fever.
Ice: to suck; bag to forehead.
Lactophenin.
Lemon Juice: an agreeable refrigerant drink.
Menthol.
Mercury: small doses at the commencement of typhoid or scarlet
fever.
Musk: a stimulant in collapse; along with opium in an acute specific
fever.
Neurodin.
Opium: in typhoid delirium; with tartar emetic if furious; at the crisis
aids action of alcohol.
Phenacetin.
Phenocoll Hydrochlorate.
Phosphate of Calcium: in hectic.
Potassium Bitartrate.
Potassium Citrate.
Potassium Nitrate.
Potassium Tartrate.
Potassium and Sodium Tartrate.
Quinine: in malarial, typhoid, and septic fevers; the most generally
applicable antipyretic.
Resorcin: antipyretic and antiseptic.
Rhus Toxicodendron: in rheumatic fever, and scarlet fever with
typhoid symptoms.
Salicin: in rheumatic fevers, or in hyperpyrexia.
Salicylate of Sodium: in rheumatic fevers, or in hyperpyrexia.
Salol.
Sodium Benzoate: in infectious and eruptive fevers; antiseptic and
antipyretic.
Strychnine: subcutaneously for muscular paralysis as a sequela.
Sulphate of Magnesium: as a depletive and purgative.
Tartar Emetic: in small doses, with opium, if delirium is not greater
than wakefulness; if greater, in full doses, with small doses of
opium; diaphoretic; in ague aids quinine, also in acute.
Thermodin.
Triphenin.
Turpentine Oil: stimulant in typhoid, puerperal, and yellow, and to
stop hemorrhage in typhoid.
Valerian.
Veratrum Viride: in delirium ferox.
Warm Sponging: in the simple fevers of children.

Fibroids.—See Tumors.

Fissures.—See also, Rhagades.

Bismuth Subnitrate.
Collodion.
Creolin.
Ichthyol.
Iodoformogen.
Iodole.
Papain.
Pyoktanin.
Traumaticin.

Fissured Nipples.—See also, Rhagades.

Bismuth Oleate.
Cacao Butter.
Ichthyol.
Sozoiodole-Potassium.
Traumaticin.

Fistula.
Bismuth Oxyiodide.
Capsicum: as weak infusion locally.
Chlorine Water.
Creolin.
Diaphtherin.
Ichthyol.
Potassa.
Sanguinaria: as injection.

Flatulence.—See also, Colic, Dyspepsia.

Abstention from sugar, starchy food, tea.
Acid, Carbolic: if without acidity, etc.
Acid, Sulphurous: if due to fermentation.
Alkalies: before meals.
Ammonia: in alkaline mixture a palliative.
Asafetida: in children; simple hysterical or hypochondriacal.
Belladonna: if due to paresis of intestinal walls.
Benzo-napthol.
Bismuth: with charcoal, in flatulent dyspepsia.
Calcium Saccharate.
Calumba: with aromatics.
Camphor: in hysterical flatulence, especially at climacteric.
Carbolated Camphor.
Carlsbad Waters: if due to hepatic derangement.
Carminatives.
Charcoal.
Chloroform: pure, in drop doses in gastric flatulence.
Creosote.
Essential Oils.
Ether: in nervousness and hypochondriasis.
Eucalyptol: at climacteric, if associated with heat flushings, etc.
Galvanism.
Hot Water: between meals.
Ichthalbin.
Ipecacuanha: in constipation, oppression at epigastrium, and in
pregnancy.
Manganese Dioxide.
Mercury: when liver is sluggish.
Muscarine: in intestinal paresis.
Nux Vomica: in constipation, pain at top of head.
Oleoresin Capsicum.
Pepper.
Physostigma: in women at change of life.
Picrotoxin.
Podophyllin with Euonymin, Leptandra, Chirata and Creosote.
Potassium Permanganate: in fat people.
Rue: most efficient.
Sodium Sulphocarbolate.
Strontium Bromide.
Sulphocarbolates: when no acidity, and simple spasms.
Terebene.
Turpentine Oil: few drops internally, or as enema in fevers,
peritonitis, etc.
Valerian.
Xanthoxylum.

Fluor Albus.—See Leucorrhea.

Flushing and Heat.—See also, Climacteric Disorders.

Eucalyptol: at climacteric.
Iron: most useful.
Nitrite of Amyl: if associated with menstrual irregularity
(accompanying symptoms, cold in the extremities, giddiness,
 fluttering of the heart); inhalation, or internally in one-third of a
drop doses; effects sometimes disagreeable.
Nux Vomica: with tinct. opii in the hysteria of middle-aged women.
Ovaraden or Ovariin: at menopause.
Potassium Bromide.
Valerian.
Zinc Valerianate: at climacteric.

Fractures and Dislocations.—See also, Wounds.

Acid, Carbolic.
Arnica: internally and locally.
Calcium Glycerinophosphate: internally, to hasten union.
Chloroform.
Iodine: antiseptic dressing.
Iodoformogen.
Iodole.
Opium.
Phosphate of Calcium: internally; quickens union.
Sozoiodole-Sodium.

Freckles.
Acid, Boric.
Acid, Lactic.
Alkaline Lotions.
Benzoin.
Borax.
Copper Oleate.
Iodine.
Lime-Water.
Mercuric Chloride: locally, with glycerin, alcohol, and rose water.
Three-fourths of grn. to the oz.
Olive Oil.
Potassium Carbonate.
Resorcin.

Frost-Bite.—See also, Chilblains.

Acid, Carbolic.
Acid, Tannic.
Adeps Lanæ.
Aluminium Acetotartrate.
Camphor Cream.
Creosote.
Ichthyol.
Sozoiodole-Potassium.
Sozoiodole-Zinc.
Styrax.

Furunculus.—See Boils.

Gall Stones.—See Calculi, Biliary.

Gangrene.—See also, Wounds, Gangrenous.

Acid, Carbolic: locally in strong solution to act as caustic; as a
dressing to promote healthy action.
Acid, Chromic: local escharotic.
Acid, Citric.
Acid, Nitric: next to bromine the most useful escharotic.
Acid, Pyroligneous.
Acid, Salicylic: locally.
Ammonium Chloride.
Balsam of Peru.
Bromal.
Bromine: escharotic in hospital gangrene.
Charcoal: as poultice.
Chlorine Water: to destroy fetor.
Cinchona.
Creosote.
Eucalyptol: along with camphor in gangrene of lungs, to prevent
spread and lessen the fetor.
Lime Juice and Chlorine Water: in hospital gangrene.
Myrtol: to destroy fetor and promote healthy action.
Oakum: dressing.
Opium.
Oxygen: as a bath.
Potassa: as caustic.
Potassium Chlorate.
Potassium Permanganate.
Quinine.
Resorcin: antiseptic, antipyretic.
Sanguinaria.
Sodium Sulphate.
Tannoform.
Terebene.
Turpentine Oil: internally, and by inhalation.
Zinc Chloride.

Gastralgia.—See also, Acidity, Dyspepsia, Gastrodynia, Neuralgia.

Acetanilid.
Acid, Carbolic.
Acid, Hydrocyanic: if purely nervous.
Acid, Salicylic: used in paroxysmal form; like quinine.
Acupuncture: sometimes gives great relief.
Alkalies.
Alum: if pyrosis.
Arsenic: in small doses.
Arsenic with Iron.
Atropine: in gastric ulcer.
Belladonna.
Bismuth: in irritable gastralgia.
Bismuth and Pepsin.
Bismuth Subnitrate.
Bromides.
Cannabis Indica.
Cerium Oxalate.
Charcoal: in neuralgia.
Chloral Hydrate: to relieve pain.
Chloroform: two or three drops on sugar.
Codeine.
Cod-Liver Oil.
Counter-irritation and a vigorous revulsive, especially useful in
hysteria.
Creosote.
Diet and Hygiene.
Emesis and Purgation: when due to indigestible food.
Enemata.
Ergot.
Ether: a few drops.
Ferropyrine.
Galvanism: of pneumogastric and sympathetic.
Hot Applications.
Magnesium Oxide.
Manganese Dioxide.
Massage.
Menthol.
Methylene.
Milk Diet.
Morphine: subcutaneously, in epigastrium, very useful; or with
bismuth and milk before each meal.
Nitroglycerin: quickly eases.
Nux Vomica: to remove morbid condition on which it depends.
Opium.
Pancreatin.
Papain.
Pepsin.
Potassium Cyanide.
Potassium Nitrite.
Pulsatilla.
Quinine: if periodic in character.
Resorcin.
Silver Chloride.
Silver Iodide.
Silver Nitrate: nervine tonic.
Silver Oxide.
Sodium Salicylate.
Strontium Bromide.
Strychnine.
Suppository of Gluten, Glycerin and Soap: to overcome constipation.
Triphenin.
Valerian.
Zinc Oxide.

Gastric Dilatation.
Acid, Carbolic.
Bismuth Salicylate.
Bismuth Subnitrate.
Calcium Lactophosphate.
Charcoal.
Cod-Liver Oil or Gaduol, if due to rachitis.
Diet.
Enemas Nutrient.
Faridization of Gastric walls.
Gentian and Columba.
Ichthyol.
Iron Iodide.
Lavage.
Naphtol.
Nux Vomica.
Physostigma.
Sodium Phosphate.
Strontium Bromide.
Strychnine.

Gastric Pain.—See Gastralgia.

Gastric Ulcer.—See also, Hematemesis.

Acid, Carbolic.
Acid, Gallic.
Arsenic: in chronic ulcer it eases pain and vomiting, and improves
the appetite.
Atropine: arrests pain and vomiting.
Bismuth Oxyiodide.
Bismuth Subgallate.
Bismuth Subnitrate: in very large doses.
Cannabis Indica.
Carlsbad Salts: before meals.
Castor Oil.
Charcoal: in chronic ulcer to allay pain.
Chloroform.
Cocaine.
Codeine.
Cold Compresses.
Counter-irritation.
Creosote.
Diet and Hygiene.
Hydrogen Peroxide.
Ice-bag: to epigastrium.
Iron.
Lead Acetate: to check hematemesis.
Lime Water with Milk: and diet.
Magnesium Sulphate.
Massage and Electricity.
Mercuric Chloride: small dose before meals.
Mercury Iodide, Red.
Mercury Oxide, Red.
Methylene Blue.
Milk.
Monsel's Solution.
Morphine: like atropine.
Nutritive Enemata.
Opium.
Pepsin.
Peptonized Milk.
Potassium Iodide: with bicarbonate, to lessen flatulent dyspepsia.
Potassium Sulphite.
Resorcin.
Silver Nitrate: to relieve pain and vomiting.
Silver Oxide.
Sodium Phosphate.
Sodium Tellurate.
Spice Plaster.
Stimulants: guardedly.
Tannin.
Turpentine Oil: frequently repeated, to check hemorrhage.
Zinc Carbonate.
Zinc Oxide.
Zinc Sulphocarbolate.

Gastritis.
Acid, Hydrocyanic: to allay pain.
Acid, Tannic.
Alum: when vomiting of glairy mucus.
Ammonium Chloride: in gastric catarrh.
Arsenic: in drunkards.
Atropine: in chronic cases.
Bismuth: in catarrh.
Caffeine: especially when associated with migraine.
Calumba.
Cinchona.
Eucalyptus: in chronic catarrh.
Hydrastis.
Ice: to suck; and to epigastrium.
Ipecacuanha: in catarrh.
Lead Acetate: along with opium.
Nutrient Enemata.
Nux Vomica.
Opium.
Silver Nitrate: in chronic gastritis.
Silver Oxide.
Veratrum Viride should never be used.

Gastritis, Acute.
Belladonna.
Bismuth Subnitrate.
Calomel.
Demulcents.
Mercury.
Morphine.
Oils.
Opium.
Sodium Paracresotate.
Warm Water, internally, or Stomach Pump: to unload stomach at
onset.

Gastritis, Chronic.—See also, Dyspepsia, Gastralgia.

Alkalies.
Bismuth Salicylate.
Bismuth Subnitrate.
Bismuth and Ammonium Citrate.
Caffeine.
Calcium Salicylate.
Cinchona.
Ichthalbin: internally, as regulator and tonic.
Mercury.
Morphine.
Orexine Tannate.
Papain.
Pepsin.
Podophyllum.
Pulsatilla.
Resorcin.
Silver Nitrate: by irrigation.
Sodium Paracresotate.
Strontium Bromide.
Thymol.
Zinc Oxide.
Zinc Sulphate.

Gastrodynia.—See Gastralgia.

Gastrorrhea.—See Pyrosis.

Gingivitis.
Alum.
Aseptol.
Borax.
Myrrh.
Potassium Chlorate.
Pyoktanin.
Sozoiodole-Potassium.

Glanders and Farcy.

Acid, Carbolic.
Ammonium Carbonate.
Arsenic.
Creosote.
Escharotics.
Iodine.
Iron.
Potassium Bichromate.
Potassium Iodide.
Quinine.
Strychnine.
Sulphur Iodide.
Sulphites.

Glandular Enlargement.—See also, Bubo, Wen, Goiter, Tabes

Mesenterica, Parotitis, Tonsillitis, etc.
Acid, Carbolic: injections of a two per cent. solution.
Ammoniacum Plaster: as counter-irritant on scrofulous glands.
Ammonium Chloride.
Antimony Sulphide.
Arsenic.
Barium Chloride.
Belladonna.
Blisters: to scrofulous glands.
Cadmium Chloride.
Calcium Chloride: in enlarged and breaking-down scrofulous glands.
Calcium Sulphide: for glands behind jaw with deep-seated
suppuration.
Cod-Liver Oil.
Conium: in chronic enlargements.
Creosote.
Gaduol.
Gold Chloride: in scrofula.
Guaiacum.
Hydrastis.
Ichthalbin: internally.
Ichthyol: topically.
Iodides.
Iodine: internally; and painted around, not over the gland.
Iodoform: as a dressing to breaking-down glands.
Iodoformogen: equable and persistent in action on open glands.
Iodole: internally.
Lead Iodide: ointment.
Mercury: internally; locally the oleate of mercury and morphine.
Pilocarpine: in acute affections of parotid and submaxillary.
Potassium Iodide: ointment over enlarged thyroid and chronically
inflamed glands.
Sozoiodole-Mercury.
Sulphides.
Thiosinamine.
Valerian.

Glaucoma.
Atropine has caused this disease.
Duboisine like atropine.
Eserine: lowers intraocular tension.
Iridectomy: the only cure.
Quinine.

Glottis, Spasm of.—See Laryngismus Stridulus.

Gleet.—See also, Gonorrhea.

Acid, Tannic.
Acid, Trichloracetic.
Airol.
Aloes.
Argentamine.
Argonin.
Aristol.
Betol.
Bismuth Oxyiodide or Subnitrate: suspended in glycerin or mucilage.
Blisters: to perineum useful in obstinate gleet.
Cantharides: minim doses of tincture frequently repeated.
Copaiba: internally, and locally smeared on a bougie and introduced;
best used in chronic form.
Copper Sulphate: as injection.
Creosote.
Eucalyptol: in very chronic gleet.
Gallobromol.
Hydrastine Hydrochlorate.
Iodoform.
Iodoformogen.
Iodole.
Iron: either perchloride or sulphate as injection, along with opium.
Juniper Oil: like copaiba.
Kino.
Lead Acetate: injection is sometimes used.
Lime Water.
Mercury: half a grn. of bichloride in six ounces of water.
Naphtol.
Oil Juniper.
Oil Turpentine.
Peru, Balsam of.
Piper Methysticum.
Potassium Permanganate.
Protargol.
Salol.
Sandalwood Oil: useful both locally and generally.
Silver Citrate.
Sozoiodole-Sodium.
Tannin, Glycerite of: as injection.
Terebene.
Thalline Sulphate.
Tolu, Balsam of.
Turpentine Oil: in a condition of relaxation.
Uva Ursi.
Zinc Acetate.
Zinc Sulphate: as injection.

Glossitis.
Alum.
Bismuth: locally.
Electrolysis: in simple hypertrophy, and cystic.
Iron.
Leeches.
Purgatives.
Quinine.

Glottis, Œdema of.—See also, Croup, Laryngitis.

Acid, Tannic.
Alum.
Ammonium Carbonate: as emetic.
Conium.
Emetics.
Ethyl Iodide.
Inhalations.
Scarification.
Tracheotomy.
Glycosuria.—See Diabetes.

Goiter.—See also, Exophthalmos.

Ammonium Chloride.
Ammonium Fluoride.
Cadmium Oleate.
Electricity.
Ferric Chloride.
Iodides.
Iodine: internally, and locally as ointment or tincture, and as
injection.
Iodoform.
Iodoformogen.
Iodothyrine.
Mercuric Biniodide: as ointment, to be used in front of hot fire, or in
hot sun.
Potassium Bromide.
Potassium Iodide.
Strophanthus.
Strychnine.
Thyraden.

Gonorrhea.—See also, Chordee, Gleet, Orchitis; Rheumatism;

Gonorrheal; Urethritis, Urethral Stricture, Vaginitis.
Acid, Benzoic: internally.
Acid, Boric.
Acid, Camphoric.
Acid, Chromic.
Acid, Cubebic.
Acid, Gallic.
Acid, Tannic.
Acid, Trichloracetic.
Aconite: in acute stage.
Airol.
Alcohol not to be touched.
Alkalines: salts, or waters, as citrates or bicarbonates, to make urine
alkaline.
Alum: as an injection.
Aluminium Tannate.
Alumnol.
Antimony: if acute stage is severe.
Aristol.
Argentamine.
Argonin.
Belladonna.
Bismuth Oxyiodide.
Bismuth Subgallate.
Bismuth Subnitrate.
Buchu: more useful after acute stage.
Cadmium Sulphate: astringent injection.
Cannabis Indica: to relieve pain and lessen discharge.
Cantharides: in small doses where there is pain along urethra and
constant desire to micturate. The tincture in minim doses three
times daily in chordee.
Chloral Hydrate.
Cinnamon Oil.
Cocaine: injection to relieve the pain.
Colchicum: in acute stage.
Collinsonia.
Copaiba: after acute stage.
Copper Acetate.
Copper Sulphate.
Creolin.
Cubebs: either alone or mixed with copaiba.
Diet and Hygiene.
Ergotin.
Erigeron, Oil of.
Eucalyptus, Oil of.
Europhen.
Ferropyrine.
Formaldehyde.
Gallobromol.
Gelsemium.
Glycerite of Tannin: injection in later stage.
Hamamelis.
Hot Sitz-bath.
Hydrastine Hydrochlorate.
Hydrastis: an injection.
Hydrogen Peroxide.
Ichthyol.
Iodole.
Iron: astringent injection in later stage.
Kaolin.
Kava Kava.
Largin: very effective.
Lead Acetate.
Lead Nitrate.
Lead Subacetate, solution of.
Lead Water and Laudanum.
Mercury Benzoate.
Mercury Bichloride: weak solution, locally.
Mercury Salicylate.
Methylene Blue.
Methyl Salicylate.
Naphtol.
Opium.
Potassium Citrate.
Potassium Permanganate.
Protargol.
Pulsatilla.
Pyoktanin.
Pyridine.
Quinine: stimulant in later stage.
Quinoline Tartrate.
Resorcin.
Salol.
Sandalwood Oil: internally and locally.
Silver Nitrate: as injection, said to cut short at commencement.
Silver Oxide.
Sodium Bicarbonate.
Sodium Dithio-Salicylate.
Sodium Salicylate.
Sozoiodole-Sodium.
Sozoiodole-Zinc.
Terpin Hydrate.
Thalline Sulphate.
Turpentine Oil.
Urinating: with penis in hot water, to relieve ardor urinæ.
Veratrum Viride: in early stage of acute fever.
Warm Baths: lasting 1/2 to 2 hours, in early stage.
Zinc Permanganate.
Zinc salts in general.

Gout.—See also, Arthritis, Lithemia.

Acid, Arsenous.
Acid, Carbonic.
Acid, Di-iodo-salicylate.
Acid, Salicylic.
Aconite.
Alkalies.
Alkaline Mineral Waters.
Alkaline Poultice.
Ammonia Water.
Ammonium Benzoate.
Ammonium Phosphate.
Ammonium Tartrate.
Antipyrine.
Argentic Nitrate.
Arnica.
Arsenic.
Asaprol.
Asparagin.
Belladonna.
Blisters.
Calcium Sulphate.
Chicory.
Chloral Hydrate.
Chloroform.
Cod-Liver Oil.
Colchicine.
Colchicum.
Cold Water.
Collodion.
Colocynth with Hyoscyamus: to unload bowels.
Diet.
Diuretics and Alkaline drinks.
Ether: hypodermically.
Formin.
Fraxinus.
Gaduol.
Gentian.
Glycerinophosphates.
Guaco.
Horse Chestnut Oil.
Hydrogen Sulphide.
Ichthalbin: internally, as resolvent and alterative.
Ichthyol: topically.
Iodide of Potassium.
Iodine.
Iodoform.
Iron Iodide.
Levico Water.
Lithium Salts.
Lycetol.
Lysidine.
Magnesia.
Manganese.
Morphine.
Oil of Peppermint.
Piperazine.
Piper Methysticum.
Potassæ Liquor.
Potassium Acetate.
Potassium Bromide.
Potassium Permanganate.
Potassium Silicate.
Prunus Virginiana.
Quinine.
Rubefacients.
Salicylates: large doses.
Saliformin.
Sodium Arsenate.
Sodium Benzoate.
Sodium Bicarbonate.
Sodium Carbonate.
Sodium Chloride.
Sodium Salicylate.
Stimulants.
Strawberries.
Strontium Bromide.
Strontium Lactate.
Strontium Salicylate.
Strychnine.
Sulphides: in chronic cases.
Sulphur.
Sulphur Baths.
Sulphurated Potassa.
Tetraethyl-ammonium Hydroxide.
Trimethylamine.
Turkish Baths.
Veratrine: as ointment.
Vichy Water.
Water: distilled.

Granulations, Exuberant.
Acid, Chromic.
Alum, Dried.
Cadmium Oleate.
Copper Sulphate.
Potassium Chlorate.
Silver Nitrate.
Zinc Chloride.

Griping.—See Colic.

Growths, Morbid.—See Tumors.

Gums, Diseases of.—See also, Mouth, Sore; Scurvy, Teeth.

Acid, Carbolic.
Acid, Salicylic.
Alum.
Areca.
Catechu: as a mouth wash.
Cocaine: locally.
Ferric Chloride.
Ferropyrine.
Formaldehyde.
Hamamelis.
Iodine Tincture: locally.
Krameria.
Myrrh.
Pomegranate Bark.
Potassium Chlorate.
Potassium Iodide.
Salol.
Tannin.

Hay Fever.—See also, Asthma, Catarrh, Conjunctivitis, Influenza.

Acid, Boric.
Acid, Carbolic.
Acid, Salicylic.
Acid, Sulphurous.
Aconite.
Ammonia.
Argentic Nitrate.
Arsenic: as cigarette.
Atropine.
Brandy Vapor.
Bromine.
Camphor.
Cantharides: tincture.
Chlorate of Potassium.
Cocaine.
Coffee, strong.
Formaldehyde.
Grindelia.
Hamamelis.
Ichthyol: as spray.
Iodides.
Ipecacuanha.
Lobelia.
Morphine.
Muscarine.
Menthol.
Opium.
Pilocarpine.
Potassium Chlorate.
Potassium Iodide: internally and locally.
Quinine: locally as injection or douche.
Resorcin.
Sozoiodole salts.
Stearates.
Strychnine.
Terpin Hydrate.
Tobacco.
Turkish Baths.
Veratrum Viride.

Headache.—See also, Hemicrania.

Acetanilid.
Acid, Acetic.
Acid, Hydrobromic.
Acid, Nitrohydrochloric: for pain just above eyeballs without
constipation, also for pain at back of neck.
Acid, Phosphoric.
Acid, Salicylic.
Aconite: when circulation excited.
Actæa Racemosa.
Aloin.
Ammonia: aromatic spirits, 1/2 to 2 drams.
Ammonium Carbonate.
Ammonium Chloride: 10 to 15 grn. doses in hemicrania.
Ammonium Valerianate.
Antacids.
Antipyrine.
Arsenic: in brow ague.
Atropine: locally to eye in migraine.
Belladonna: frequently given in frontal headache, especially at
menstrual period, or if from fatigue.
Berberine.
Bismuth Valerianate.
Bleeding.
Bromides: in large doses.
Bryonia: in bilious headache.
Butyl-chloral Hydrate.
Caffeine, with antipyrine or sodium bromide.
Cajeput Oil: locally.
Camphor: internally, and saturated solution externally.
Camphor with acetanilid or antipyrine, in nervous headache.
Cannabis Indica: in neuralgic headache.
Capsicum: plaster to nape of neck.
Carbon Disulphide.
Carbon Tetrachloride.
Chamomile.
Chloralamide.
Chloroform, Spirit of: in nervous headache.
Cimicifuga: in nervous and rheumatic headache, especially at
menstrual period.
Coffee and Morphine.
Colchicum.
Cold Affusion.
Croton Oil.
Cup, to nape of neck, in congestion.
Digitalin: (German) 1/16 grn. twice a day for congestive hemicrania.
Electricity.
Ergot.
Ergotin.
Ethylene Bromide.
Erythrol Tetranitrate.
Ether Spray: locally, for frontal headache after illness or fatigue.
Eucalyptol.
Ferropyrine.
Friedrichshall Water.
Galvanism.
Gelsemium.
Guarana.
Heat: as hot water-bag or poultice to nape of neck.
Hot Sponging.
Hot Water.
Hydrastis: in congestive headache with constipation.
Hyoscyamus.
Ice-bag: applied to head, or leeches back of ears, in severe cases.
Ichthalbin: to improve digestion and nutrition.
Ignatia: in hysterical headache.
Iodide of Potassium: in rheumatic headache with tenderness of
scalp.
Iris: in supra-orbital headache with nausea.
Kola.
Lithium Bromide.
Magnesium Carbonate.
Magnesium Citrate.
Magnesium Oxide.
Magnesium Sulphate: for frontal headache with constipation.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

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

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com