Positivity 1st Edition Karim Boulabiar digital

version 2025

https://ebookultra.com/download/positivity-1st-edition-karim-
boulabiar/

★★★★★
4.7 out of 5.0 (98 reviews )

Get PDF Instantly

ebookultra.com
 Positivity 1st Edition Karim Boulabiar

EBOOK

Available Formats

■ PDF eBook Study Guide Ebook

EXCLUSIVE 2025 ACADEMIC EDITION – LIMITED RELEASE

Available Instantly Access Library

Here are some recommended products for you. Click the link to
download, or explore more at ebookultra.com

Statistics 1st Edition Karim M. Abadir

https://ebookultra.com/download/statistics-1st-edition-karim-m-abadir/

The Dictionary English Arabic 2nd Edition Karim

https://ebookultra.com/download/the-dictionary-english-arabic-2nd-
edition-karim/

Statistical Signal Processing for Neuroscience and

Neurotechnology 1st Edition Karim G. Oweiss

https://ebookultra.com/download/statistical-signal-processing-for-
neuroscience-and-neurotechnology-1st-edition-karim-g-oweiss/

The Development of Nomadism in Ancient Northeast Africa

Karim Sadr

https://ebookultra.com/download/the-development-of-nomadism-in-
ancient-northeast-africa-karim-sadr/
Emerging Digital Spaces in Contemporary Society Properties
of Technology 1st Edition Karim Gherab-Martin

https://ebookultra.com/download/emerging-digital-spaces-in-
contemporary-society-properties-of-technology-1st-edition-karim-
gherab-martin/

Black Lives Matter vs All Lives Matter A Multidisciplinary

Primer 1st Edition Abdul Karim Bangura

https://ebookultra.com/download/black-lives-matter-vs-all-lives-
matter-a-multidisciplinary-primer-1st-edition-abdul-karim-bangura/

Steve Jobs 1ST/1ST PRINT Edition Walter Isaacson

https://ebookultra.com/download/steve-jobs-1st-1st-print-edition-
walter-isaacson/

Rangbhumi 1st Edition Premchand

https://ebookultra.com/download/rangbhumi-1st-edition-premchand/

Cyclops 1st Edition Marinkovi■

https://ebookultra.com/download/cyclops-1st-edition-marinkovic/
 Trends in Mathematics

Trends in Mathematics is a series devoted to the publication of volumes arising from

conferences and lecture series focusing on a particular topic from any area of
mathematics. Its aim is to make current developments available to the community as
rapidly as possible without compromise to quality and to archive these for reference.

Proposals for volumes can be sent to the Mathematics Editor at either

Birkhäuser Verlag
P.O. Box 133
CH-4010 Basel
Switzerland

or

Birkhäuser Boston Inc.

675 Massachusetts Avenue
Cambridge, MA 02139
USA

Material submitted for publication must be screened and prepared as follows:

All contributions should undergo a reviewing process similar to that carried out by journals
and be checked for correct use of language which, as a rule, is English. Articles without
SURRIVRUZKLFKGRQRWFRQWDLQDQ\VLJQL½FDQWO\QHZUHVXOWVVKRXOGEHUHMHFWHG+LJK
quality survey papers, however, are welcome.

We expect the organizers to deliver manuscripts in a form that is essentially ready for
GLUHFWUHSURGXFWLRQ$Q\YHUVLRQRI7H;LVDFFHSWDEOHEXWWKHHQWLUHFROOHFWLRQRI½OHVPXVW
EHLQRQHSDUWLFXODUGLDOHFWRI7H;DQGXQL½HGDFFRUGLQJWRVLPSOHLQVWUXFWLRQVDYDLODEOH
from Birkhäuser.

Furthermore, in order to guarantee the timely appearance of the proceedings it is essential

WKDWWKH½QDOYHUVLRQRIWKHHQWLUHPDWHULDOEHVXEPLWWHGQRODWHUWKDQRQH\HDUDIWHUWKH
FRQIHUHQFH7KHWRWDOQXPEHURISDJHVVKRXOGQRWH[FHHG7KH½UVWPHQWLRQHGDXWKRU
of each article will receive 25 free offprints. To the participants of the congress the book
will be offered at a special rate.
Positivity

Karim Boulabiar
Gerard Buskes
Abdelmajid Triki
Editors

Birkhäuser
Basel · Boston · Berlin
Editors:

Karim Boulabiar Gerard Buskes

,QVWLWXW3UpSDUDWRLUHDX[   'HSDUWPHQWRI0DWKHPDWLFV
(WXGHV6FLHQWL½TXHVHW7HFKQLTXHV  8QLYHUVLW\RI0LVVLVVLSSL
Université de Carthage Hume Hall 305
%3     32%R[
/D0DUVD    8QLYHUVLW\06
Tunisia USA
e-mail: [email protected] e-mail: [email protected]

Abdelmajid Triki
'pSDUWHPHQWGH0DWKpPDWLTXHV
Faculté des Sciences de Tunis
Campus Universitaire
1060 Tunis
Tunisia
e-mail: [email protected]

0DWKHPDWLFDO6XEMHFW&ODVVL½FDWLRQ$')%$$%
(///%%%'*//0&
54G05, 54H10

Library of Congress Control Number: 2007933653

%LEOLRJUDSKLFLQIRUPDWLRQSXEOLVKHGE\'LH'HXWVFKH%LEOLRWKHN'LH'HXWVFKH%LEOLRWKHNOLVWV
WKLVSXEOLFDWLRQLQWKH'HXWVFKH1DWLRQDOELEOLRJUD½HGHWDLOHGELEOLRJUDSKLFGDWDLVDYDLODEOHLQ
the Internet at http://dnb.ddb.de

ISBN Birkhäuser Verlag AG, Basel - Boston - Berlin

This work is subject to copyright. All rights are reserved, whether the whole or part of the materi-
DOLVFRQFHUQHGVSHFL½FDOO\WKHULJKWVRIWUDQVODWLRQUHSULQWLQJUHXVHRILOOXVWUDWLRQVUHFLWD-
WLRQEURDGFDVWLQJUHSURGXFWLRQRQPLFUR½OPVRULQRWKHUZD\VDQGVWRUDJHLQGDWDEDQNV)RU
any kind of use permission of the copyright owner must be obtained.

© 2007 Birkhäuser Verlag AG

Basel · Boston · Berlin
P.O. Box 133, CH-4010 Basel, Switzerland
Part of Springer Science+Business Media
Printed on acid-free paper produced from chlorine-free pulp. TCF ’
Printed in Germany
ISBN  e-ISBN 

     ZZZELUNKDXVHUFK
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

B. Banerjee and M. Henriksen

Ways in which C(X) mod a Prime Ideal Can be
a Valuation Domain; Something Old and Something New . . . . . . . . . . . . 1

D.P. Blecher
Positivity in Operator Algebras and Operator Spaces . . . . . . . . . . . . . . . . 27

K. Boulabiar, G. Buskes, and A. Triki

Results in f -algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Q. Bu, G. Buskes, and A.G. Kusraev

Bilinear Maps on Products of Vector Lattices: A Survey . . . . . . . . . . . . . 97

G.P. Curbera and W.J. Ricker

Vector Measures, Integration and Applications . . . . . . . . . . . . . . . . . . . . . . . 127

J. Martı́nez
The Role of Frames in the Development of Lattice-ordered
Groups: A Personal Account . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

B. de Pagter
Non-commutative Banach Function Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

A.R. Schep
Positive Operators on Lp -spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

A.W. Wickstead
Regular Operators between Banach Lattices . . . . . . . . . . . . . . . . . . . . . . . . . 255
Preface
This collection of surveys is an outflow from the 2006 conference Carthapos06
in Tunis (Tunisia). Apart from regular conference talks, five survey talks formed
the core of a workshop in Positivity, supported by the National Science Founda-
tion. The conference organizers (Karim Boulabiar, Gerard Buskes, and Abdelmajid
Triki) decided to expand on the idea of core surveys and the nine surveys in this
book are the harvest from that idea.
Positivity derives from an order relation. Order relations are the mathemat-
ical tool for comparison. It is no surprise that seen in such very general light, the
history of Positivity is ancient. Archimedes, certainly, had the very essence of posi-
tivity in mind when he discovered the law of the lever. His method of exhaustion to
calculate areas uses a principle that nowadays carries his name, the Archimedean
property. The surveys in this book are slanted into the direction that Archimedes
took. Functional analysis is heavily represented. But there is more. Lattice ordered
groups appear in the article by Martinez in the modern jacket of frames. Henriksen
and Banerjee write their survey on rings of continuous functions. Blecher and de
Pagter in each of their papers survey parts of non-commutative functional analysis.
Positive operators are the main topic in the papers by Curbera and Ricker, Schep,
and Wickstead. And positive bilinear maps are the protagonists in the survey by
Bu, Buskes, and Kusraev. The conference organizers (and editors of this volume)
write about f -algebras.
Carthapos06 was more than just a conference and workshop in Africa. It
brought together researchers in Positivity from many directions of Positivity and
form many corners of the world. This book can be seen as a culmination of their
paths meeting in Tunisia, Africa.

June 5, 2007 G. Buskes

Oxford, U.S.A.
Positivity
Trends in Mathematics, 1–25


c 2007 Birkhäuser Verlag Basel/Switzerland

Ways in which C(X) mod a Prime Ideal

Can be a Valuation Domain;
Something Old and Something New
Bikram Banerjee (Bandyopadhyay) and Melvin Henriksen

Abstract. C(X) denotes the ring of continuous real-valued functions on a

Tychonoff space X and P a prime ideal of C(X). We summarize a lot of
what is known about the reside class domains C(X)/P and add many new
results about this subject with an emphasis on determining when the ordered
C(X)/P is a valuation domain (i.e., when given two nonzero elements, one
of them must divide the other). The interaction between the space X and
the prime ideal P is of great importance in this study. We summarize first
what is known when P is a maximal ideal, and then what happens when
C(X)/P is a valuation domain for every prime ideal P (in which case X
is called an SV -space and C(X) an SV -ring). Two new generalizations are
introduced and studied. The first is that of an almost SV -spaces in which
each maximal ideal contains a minimal prime ideal P such that C(X)/P is a
valuation domain. In the second, we assume that each real maximal ideal that
fails to be minimal contains a nonmaximal prime ideal P such that C(X)/P
is a valuation domain. Some of our results depend on whether or not βω \ ω
contains a P -point. Some concluding remarks include unsolved problems.

1. Introduction
Throughout, C(X) will denote the ring of real-valued continuous functions on a
Tychonoff space X with the usual pointwise ring and lattice operations and C ∗ (X)
will denote its subring of bounded functions, and all topological spaces considered
are assumed to be Tychonoff spaces unless the contrary is stated explicitly. (Recall
that X is called a Tychonoff space if it is a subspace of a compact (Hausdorff)
space. Equivalently if X is a T1 space and whenever K is a closed subspace of X not
containing a point x, there is an f ∈ C(X) such that f (x) = 0 and f [K] = {1}.)
An element of C(X) is nonnegative in the usual pointwise sense if and only if it
a square. So algebraic operations automatically preserve order. This makes the
2 B. Banerjee and M. Henriksen

notion of positivity essential for studying C(X). This simple observation was used
with great ingenuity by M.H. Stone in 1937 to make the first thorough study of
C(X) as a ring. It was restricted to the case when X is compact. Among the many
interesting results in this seminal paper is that C(X) determines X. That is, if X
and Y are compact spaces and C(X) and C(Y ) are algebraically isomorphic, then
X and Y are homeomorphic.
This study was broadened to include unbounded functions in [Hew48] by
Stone’s student E. Hewitt. While this paper contains a number of serious errors, it
set the tone for a lot of the research that led to the book [GJ76]. (It was published
originally in 1960 by Van Nostrand). For more background and history of this
subject, see [Wa74], [We75], [Hen97], and [Hen02]. Our general sources for general
topology are [E89] and [PW88].
Sections 2 and 3 survey some of what has been done in the past about integral
domains that are homomorphic images of a C(X) and the prime ideals P that are
kernels of such homomorphisms. We concentrate especially on the cases when
C(X)/P is a valuation domain. In Section 2, we review some of what is known
when P is maximal; i.e., when C(X)/P is a field. Section 3 recalls what is known
about spaces X such that at C(X)/P is a valuation domain whenever P is a prime
ideal of C(X). They are called SV -spaces. The remainder of the paper focuses on
new research beginning with the study in Section 4 of almost SV -spaces; that is,
spaces X and rings C(X) in which every maximal ideal of C(X) contains a minimal
prime ideal P such that C(X)/P is a valuation domain. Section 5 is devoted to
the study of products of almost SV -spaces and logical considerations concerning
the validity of some results. The one-point compactification of a countable discrete
space is not an SV -space, but the consequences of the assumption that it is an
almost SV -space are studied in Section 6. Spaces X and rings C(X) in which
every real maximal ideal of C(X) contains a prime ideal such that C(X)/P is a
valuation domain are examined in Section 7. In the final Section 8, two related
papers and the contents of a book are discussed briefly, some sufficient conditions
are given to say more about valuation domains that are homomorphic images of
a ring C(X), and some unsolved problems are posed.

2. What happens when the valuation domains are fields?

A commutative ring A such that whenever a and b are nonzero elements of A, it
follows that one of them divides the other, is called a valuation ring. Below, we are
interested only in the case when A is also an integral domain, in which case such
a ring A is called a valuation domain. We begin with the case when the valuation
domain is a field, and recall that the kernel of a homomorphism onto a field is
a maximal ideal. Let M(A) denote the set of maximal ideals of A. This set is
nonempty as long as A has an identity element
Recall that a field F is said to be real-closed if its smallest algebraic extension
is algebraically closed. Equivalently, F is real-closed if it is totally ordered, its set
 Ways in which C(X) mod a Prime Ideal Can be a Valuation Domain 3

F + of nonnegative elements is exactly the set of all squares of elements of F , and

each polynomial of odd degree with coefficients in F vanishes at some point of
F . As is shown in Chapter 13 of [GJ76], if M ∈ M(C(X)), then C(X)/M is a
real-closed field. We continue to quote facts from [GJ76].
If f ∈ C(X), then Z(f ) denotes {x ∈ X : f (x) = 0}, and we let coz(f ) =
X\Z(f ). If S ⊂ C(X), we let Z[S] = {Z(f ) : f ∈ S}. Thus Z[C(X)] (which we
abbreviate by Z[X]) is the family of all zerosets of functions in C(X). A subfam-
ily F of Z[X] that is closed under finite intersection, contains Z(g) whenever if
contains some Z(f ) ∈ F, and does not contain the empty set is called a z-filter.
Note that an element f is in some proper ideal if and only if Z(f ) = ∅. It follows
that if I is a proper ideal of C(X), Then Z[I] is a z-filter.
An ideal I is fixed or free according as ∩{Z(f ) : f ∈ I} is nonempty or empty.
A maximal ideal M is fixed if and only if Z[M ] = {x} for some x ∈ X, in which
case M is denoted by Mx . Clearly, C(X)/M always contains a copy of R. The
maximal ideal M is called hyper-real if C(X)/M contains R properly and is called
real otherwise. Every fixed maximal ideal is real, but the converse fails to hold. If
every real maximal ideal of C(X) is fixed, then X is called a realcompact space.
Subsequent to the appearance of [GJ76], hyper-real fields are also called H-fields.
Recall that the continuum hypothesis CH is the assumption that the least
uncountable cardinal ω1 is equal to the cardinality 2ω of the continuum.
2.1 Definition. Suppose that an ordered set L satisfies: If A and B are countable
subsets of L such that a < b whenever a ∈ A and b ∈ B, then there is an x ∈ L
such that a < x < b whenever a ∈ A and b ∈ B (Symbolically we write this
conclusion as A < x < B.) Then L is called an η1 -set.
Much of what is known about H-fields of cardinality no larger than 2ω is
summarized next.
2.2 Theorem
(a) Every H-field is both real-closed and an η1 -set.
(b) All real-closed fields that are η1 -sets of cardinality ω1 are (algebraically)
isomorphic.
(c) Every η1 -set has cardinality at least 2ω .
(d) All H-fields of cardinality 2ω are isomorphic if and only if CH holds.
All but part of (d) are shown in Chapter 13 of [GJ76]. That there is only one
H-field (in the sense of isomorphism) of cardinality 2ω implies CH is due to A.
Dow in [D84]. Some more detail about what may happen if CH fails: see [R82].
There are a large number of results concerning H-fields of large cardinal-
ity in [ACCH81] that depend on various set-theoretic hypotheses and use proof
techniques involving combinatorial set theory. Most of its contents are beyond the
scope of this article. (Some errors in [ACCH81] are pointed out by A. Blass in his
review in Math. Sci. Net. None of them affect what is written above.)
4 B. Banerjee and M. Henriksen

3. When every prime ideal of C(X) is a valuation prime;

SV rings and spaces
As in [DW96], a prime ideal P of C(X) is called a valuation prime if C(X)/P is a
valuation domain. Chapter 14 of [GJ76] is devoted to the study of the set of prime
ideals of C(X), and a little is said about the order-structure of C(X)/P when P
is a prime ideal of C(X), but the first thorough study of valuation primes and
their associated valuation domains appears in [CD86]. Inspired by this, a number
of authors began to investigate rings C(X) and spaces X such that every prime
ideal of C(X) is a valuation prime. Such rings and spaces are called SV -rings and
SV -spaces respectively. See [HW92a], [HW92b], and [HLMW94].
The order structure of C(X)/P is described completely when X is the one-
point compacification W (ω +1) of W (ω) in [M90] and to a lesser extent in Chapter
4 of [DW96] when X is compact. See Section 6 below.
The Stone-Čech compactification βX is the compact space that contains X
as a dense subspace such that each member of the subring C ∗ (X) of bounded
functions in C(X) has a (unique) extension in C(βX). Thus C ∗ (X) and C(βX)
are isomorphic.
What follows next will be used often below. See [GJ76].

3.1 Proposition and definitions. Suppose X is a Tychonoff space. Then:

(a) There is a bijection from βX onto the set M(C(X)) of maximal ideals of
C(X) given by p → M p = {f ∈ C(X) : p ∈ clβX Z(f )}.
(b) Each prime ideal of C(X) is contained in a unique maximal ideal M p , and
the intersection of all the prime ideals contained in M p is Op = {f ∈ C(X) :
clβX Z(f ) is a neighborhood of p}.
(c) The prime ideals containing a prime ideal P form a chain (that is, are to-
tally ordered) under set inclusion, and P contains minimal prime ideals. The
number of minimal prime ideals contained in the maximal ideal M is called
its rank Rk(M ) and is denoted by ∞ unless it is finite. If X is compact,
the rank Rk(X) of X is the supremum of the ranks of the maximal ideals of
C(X). It is known that if the rank of each maximal ideal of C(X) is finite,
then Rk(X) < ∞. (See [HLMW94].)
(d) If p ∈ X, then M p is denoted by Mp and Op by Op . If Op = Mp , then p is
called a P -point. A space all of whose points are P -points is called a P -space.
X is a P -space if and only if Z(f ) is clopen for each f ∈ C(X) if and only
if C(X) is a von Neumann regular ring. Moreover, every compact P -space is
finite.
(e) If for all p ∈ βX, the prime ideals containing Op are totally ordered, then X
is called an F -space. (Equivalently, X is an F -space if Op is prime for every
p ∈ βX.) Every P -space is an F -space, every F -space is an SV -space and the
implied inclusions are proper. Moreover, no SV -space can contain a nontrivial
 Ways in which C(X) mod a Prime Ideal Can be a Valuation Domain 5

convergent sequence; that is, a closed copy of the one-point compactification

W (ω+1) of the space W (ω) of finite ordinals. (See 3.4 below and [HLMW94].)
If P ⊂ Q are prime ideals of C(X), then C(X)/Q is a homomorphic image
of C(X)/P . Thus, if P is a valuation prime, then so is Q. Hence:
3.2 Proposition. X is an SV -space if and only if every minimal prime ideal of
C(X) is a valuation prime.
Let mC(X) denote the set of minimal prime ideals of the ring C(X). If
f ∈ C(X), let h(f ) = {P ∈ mC(X) : f ∈ P } and let hc (f ) = mC(X)\h(f ). Using
{hc (f ) : f ∈ C(X)} as a base for a topology, mC(X) becomes a zero-dimensional
Hausdorff space called the space of minimal prime ideals of C(X). See [HJe65].
This space has been studied extensively, but we recall only those facts known about
it that are relevant to this paper.
Recall that if I is an ideal of C(X) such that f ∈ I and Z(g) = Z(f )
imply that g ∈ I, then I is called a z-ideal. As is shown in [HJe65], the map
P → P ∩ C ∗ (X) is an order preserving homeomorphism of mC(X) onto mC ∗ (X).
So C(X) and C(βX) have homeomorphic spaces of minimal prime ideals. Again a
prime z-ideal P of C(X) is valuation prime if and only if P ∩ C ∗ (X) is a valuation
prime of C ∗ (X) by Corollary 2.1.12 of [CD86]. Thus we have:
3.3 Proposition. X is an SV -space if and only if βX is an SV -space.
Thus, to determine the algebraic properties of an SV -ring C(X), there is no
loss of generality in assuming that X is compact.
3.4 Remark. It has long been known that every F -space is an SV -space. (See, for
example [L86].) If X and Y are two disjoint F -spaces, then the attaching of X
and Y at two non P -points respectively of X and Y ; serves as an example of an
SV -space which fails to be an F -space and consequently the class of SV -spaces
contains the class of F -spaces properly. (This is noted in [HW92a] and [HW92b],
and implicitly in [CD86].)
Recall that a space Y is said to be C ∗ -embedded (resp. C-embedded ) in a
space X if the map f → f |Y is a surjection of C ∗ (X) onto C ∗ (Y ) (resp. of C(X)
onto C(Y )). We close this section with more known facts about SV -spaces.
3.5 Proposition and remarks
(a) Every C ∗ -embedded subspace of an SV -space is an SV -space and consequent-
ly closed subspaces of compact SV -spaces are SV -spaces. (See [HW92a]).
(b) If a compact space X can be expressed as the union of finitely many closed
subspaces such that each of them is an SV -space, then X becomes an SV -
space, but not every SV -space can be represented in this way. (See [HW92a]
and [L03]. The example in [L03] is the result of a complicated construction.)
(c) If X is compact and each point of X has a closed neighborhood that is an F -
space, then X becomes an SV -space. Though the converse need not hold. Let
X be the union of two disjoint copies of βω (ω denotes the set of countable
ordinals). and Y be the space by identifying the corresponding points of
6 B. Banerjee and M. Henriksen

βω \ ω. Then every nonisolated point of Y has no F -space neighborhood,

though Y becomes a compact SV -space. (See [HW92a].)
(d) If X be an almost discrete space, i.e., a Hausdorff space with exactly one
nonisolated point, then X is an SV -space if and only if X is of finite rank.
(See [HW92b].)
(e) Every compact SV -space has finite rank. It is not known if there is a compact
space of finite rank which fails to be an SV -space. (See [HLMW94]).

4. The first generalization; almost SV rings and spaces

Below βω and αω will abbreviate β(W (ω)) and W (ω + 1) respectively.
4.1 Definition. C(X) is called an almost SV -ring and X an almost SV -space if
each maximal ideal of C(X) contains a valuation prime that is a minimal prime
ideal.
Next, it is shown how to create a large class of almost SV -spaces that are
not SV -spaces.
4.2 Theorem. The space obtained by attaching the nonisolated point of W (ω + 1)
to a compact F -space at a non P -point is an almost SV -space that is not an
SV -space.

Proof. Let Y denote a compact F -space with a point q such that Oq is not maximal.
(For example, Y could be βω and q any point of its nonisolated points.) Let X
denote the result of attaching the spaces Y and αω at the points q ∈ Y and ω ∈ αω,
and call the resulting point p. By 3.1(e), X is not an SV -space because it contains
a sequence of distinct points converging to p. Because Y is an F -space and each
point of αω other that ω is a P -point, to see that X is an almost SV -space, it
suffices to show that there is a minimal valuation prime of C(X) containing Op .
Let ϕ and ψ denote respectively the restriction maps of C(X) onto C(αω)
and C(Y ). Clearly each is a surjective homomorphism. Because Oq is not maximal,
its inverse image ψ −1 (Oq ) is a nonmaximal prime ideal of C(X) containing Op . We
show next that Op = ϕ−1 (Oω ) ∩ ψ −1 (Oq ). For, f is in this intersection if and only
if its restriction to αω vanishes on a neighborhood of ω, and its restriction to Y
vanishes on a neighborhood of q if and only if f vanishes on a neighborhood of p.
Because in any commutative ring a prime ideal that contains the intersection
of two ideals must contain one of them, it follows that any prime ideal of C(X)
that contains Op must contain at least one of ϕ−1 (Oω ) or ψ −1 (Oq ). Moreover,
because q is not a P -point of Y , ϕ−1 (Oω ) is not contained in ψ −1 (Oq ). Hence
there cannot be any prime ideal of C(X) containing Op and properly contained in
ψ −1 (Oq ). Thus the latter is a minimal prime ideal of C(X) containing Op .
Suppose π : C(Y ) → C(Y )/Oq is the canonical homomorphism. Then π ◦ ψ :
C(X) → C(Y )/Oq is a surjective homomorphism with whose kernel ker(π ◦ ψ) =
 Ways in which C(X) mod a Prime Ideal Can be a Valuation Domain 7

ψ −1 (Oq ). So C(X)/ ker(π ◦ ψ) and C(Y )/Oq are isomorphic. Because Y is an F -

space, this latter is a valuation domain and we may conclude that ψ −1 (Oq ) is a
minimal valuation prime ideal. This concludes the proof of the theorem.

We digress to quote some results in [CD86] that will be useful in what follows.
4.3 Concepts and results from [CD86].
(a) Suppose p is a point in a Tychonoff space X and  is a z- filter of subsets of
X. If for every f ∈ C(X \ {p}) such that 0 ≤ f ≤ 1, there is Y ∈  such that
limx→p f |Y exists, then  is called a P (p) filter. See [CD86] for a discussion
of the properties of such filters. The authors do not describe P (p) filters as
z-filters, but treat them as one whenever they use it.
(b) Suppose X is compact and each of its points is a Gδ point. Substantial use
will be made below of a mapping γ introduced in a more general setting by
C. Kohls in [K58] and described in more detail in [CD86]. See Section 2.2 of
[CD86] for proofs and more details about the assertions made below
(1) Suppose p ∈ X is nonisolated and Y = X \ {p}. Then γ is a bijection
from the set Q of prime z-ideals of C(Y ) such that Z[Q] converges to
some point in βY \ Y onto the family  of all nonmaximal prime z-
ideals of C(X) contained in Mp . Then γ(Q) is defined by Z[γ(Q)] =
{clX Y : Y ∈ Z[Q]} = {Y ∪ {p} : Y ∈ Z[Q]}.
(2) The prime z-ideal Q is maximal if and only if γ(Q) is an immediate
prime z-ideal predecessor of Mp by 3.2(2) of [CD86].
(3) Z[Q] is a P (p) filter if and only if Z[γ(Q)] is a P (p) filter.
(4) If P is a valuation prime contained properly in Mp and Q ⊂ P is a
minimal prime ideal, then Z[Q] is a P (p) filter. (See Theorem 2.2.2 of
[CD86].)
Recall that υX = {p ∈ βX : M p is a real maximal ideal} and that C(υX)
and C(X) are isomorphic. υX is called the (Hewitt) realcompactification of X.
The proof of the next result is an exercise.
4.4 Theorem. The following assertions are equivalent.
(1) X is an almost SV -space.
(2) υX is an almost SV -space.
(3) βX is an almost SV -space.
4.5 Corollary. If Y is a dense C ∗ -embedded subspace of an almost SV -space X,
then Y is an almost SV -space.

Proof. For then βY and βX are homeomorphic, so the conclusion follows from
Theorem 4.4.

Recall again that the question of whether a compact space of finite rank is an
SV -space was left as an open problem in Section 6 of [HLMW94]. So, the second
8 B. Banerjee and M. Henriksen

part of the hypothesis of the next theorem may be redundant, and its conclusion
may be too weak.
4.6 Theorem. If X is a compact space of finite rank and each of its points has
a compact neighborhood that is an an almost SV -space, then X is an almost
SV -space.
Proof. Because X is compact, {Mp : p ∈ X} is the set of all maximal ideals of
C(X). We need to show that each Mp contains a minimal prime ideal that is a
valuation prime. For each p ∈ X, there is a compact neighborhood T = T (p) of
p that is an an almost SV -space. Let ϕ : C(X) → C(T ) denote the map that
sends each f ∈ C(X) to its restriction to T . The map.ϕ is an epimorphism since
T is C-embedded in X. As usual, Op = {f ∈ C(X) : p ∈ intX Z(f )} and letting
OpT = {f ∈ C(T ) : p ∈ intT Z(f )}, we see that ϕ−1 (OpT ) = Op . Note also that the
inverse image under ϕ of a pair of incomparable prime ideals of C(T ) containing
OpT is a pair of incomparable prime ideals of C(X) containing Op because ϕ is an
epimorphism.
Let m and n denote respectively the ranks of p with respect to T and X.
Because X is compact, it follows from Corollary 1.8.2 of [HLMW94] that m ≤
n. < ∞. It follows that OpT is the intersection of m incomparable minimal prime
ideals {PiT }m
i=1 of C(T ) and hence that Op is is the intersection of the minimal
prime ideals {ϕ−1 (PiT )}mi=1 . In any commutative ring, if a prime ideal contains a
finite intersection of ideals, it must contain one of them. So, we may conclude that
{ϕ−1 (PiT )}m
i=1 is the collection of all minimal prime ideals of C(X) that contain
Op . Because T is an almost SV -space, there is a j such that 1 ≤ j ≤ m and PjT
is a minimal valuation prime of C(T ) containing OpT . If π : C(T ) → C(T )/PjT
denotes the canonical homomorphism, then π ◦ ϕ is an epimorphism from C(X)
onto C(T )/PjT whose kernel is ϕ−1 (PjT ). Therefore C(X)/ϕ−1 (PjT ) = C(T )/PjT
and hence ϕ−1 (PjT ) is a minimal valuation prime of C(X) contained in Mp . Since
p ∈ X is arbitrary, this completes the proof.

Use will be made below of the concept that follows in deriving a sufficient
condition for a compact perfectly normal space to be an almost SV -space.
4.7 Definition. For any space X, a point p ∈ βX such that Op is a prime ideal of
C(X) is called a βF -point. If Op is a valuation prime, then p is called a special
βF -point.
No example is known of a βF -point that is not a special βF -point.
4.8 Lemma. Suppose p is a nonisolated Gδ -point of a compact space X, and let Q
denote a prime z-ideal of C(X \ {p}) such that the prime z-filter Z[Q] converges
to some point in β(X \ {p})\ (X \ {p}). Then the following are equivalent:
(1) For every f ∈ C(X \ {p}) such that 0 ≤ f ≤ 1, there exists a Y ∈ Z[Q] and
a g ∈ C(X) such that f |Y = g|Y .
(2) For every f ∈ C(X \ {p}) such that 0 ≤ f ≤ 1, there exists a Y ∈ Z[Q] such
that limx→p f |Y exists, i.e.; Z[Q] is a P (p) filter.
 Ways in which C(X) mod a Prime Ideal Can be a Valuation Domain 9

Proof. If (1) holds, then limx→p g|Y exists and consequently limx→p f |Y also exists.
So (2) holds.
If (2) holds, define h : Y ∪ {p} → R by letting h|Y = f |Y and h(p) =
limx→p f |Y . Clearly h ∈ C(Y ∪ {p}). Since X is compact and Y ∪ {p} is a closed
subset of X, and hence is C-embedded in X. If g is an extension of h over X, then
g|Y = f |Y .

Combining Lemma 4.8 and Theorem 2.3.2 of [CD86], we obtain:
4.9 Theorem. Suppose p is a nonisolated Gδ -point of a compact space X, and let
Q denote a prime z-ideal of C(X \ {p}) such that the prime z-filterZ[Q] converges
to some point in β(X \ {p})\ (X \ {p}). Then the following are equivalent:
(1) γ(Q) is a valuation prime z-ideal of C(X) contained in Mp .
(2) Q is a valuation prime z-ideal and Z[Q] is a P (p) filter.
4.10 Theorem. If X is compact and perfectly normal, and for every nonisolated
point p of X, there is a free P (p) z-ultrafilter Z[M q ] on X \ {p} such that q is a
special βF point of β(X \ {p}), then X is an almost SV -space.
Proof. A nonisolated point p of the perfectly normal space X is a Gδ -point. Suppose
Z[M q ] is a free P (p) z-ultrafilter on X \{p} such that q is a special βF -point. Thus
Oq becomes a valuation prime ideal of C(X \ {p}) and Z[Oq ] clearly converges
to q in β(X \ {p})\ (X \ {p}). Since γ is a bijection, γ(Oq ) is a minimal prime
ideal of C(X) contained in Mp . Because Z[M q ] is a P (p) filter, γ(M q ) becomes an
immediate prime z-ideal predecessor of Mp which is a valuation prime by 2.3.3 of
[CD86]. Now γ(Oq ) is a minimal prime ideal of C(X) contained in γ(M q ), which is
properly contained in Mp and since γ(M q ) is a valuation prime, this implies that
Z[γ(Oq )] is a P (p) filter by 2.2.2 of [CD86]. Consequently, by 4.3(b)(3) above,
Z[Oq ] becomes a P (p) filter. Finally, because Oq is a valuation prime and Z[Oq ]
is a P (p) filter, γ(Oq ) becomes a valuation prime by Theorem 4.9. But p is an
arbitrary nonisolated point, so this completes the proof.

4.11 Theorem. αω is an almost SV -space if and only if there exists a free ultrafilter
Ψ on ω such that for every f ≥ 0 in C ∗ (ω), there is a Y ∈ Ψ such that limx→ω f |Y
exists.
Proof. Let ϕ be the unique continuous extension of the inclusion map i : ω → αω
over βω. Recall from 4.3(b)(1) that the map γ is a bijection of the family of all
prime z-ideals Q of C(ω) such that Z[Q] converges to point of ϕ−1 (ω) (where
ϕ−1 (ω) = βω \ ω) onto the family of all nonmaximal prime z-ideals of C(αω)
contained in Mω . Since ω is a P -space, the set of prime z-ideals Q of C(ω) such
that Z[Q] converges to a point of βω \ ω is the set of maximal ideals M q such that
q ∈ βω \ ω. So by 4.3(b)(1), the set of minimal prime ideals of C(αω) contained
in Mω is given by:{γ(M q ) : q ∈ βω \ ω}. Thus αω is an almost SV -space if and
only if γ(M q ) is a valuation prime for some q ∈ βω \ ω if and only if Z[M q ] is a
P (ω) z-ultrafilter on ω (by Corollary 3 of [CD 86]) if and only if there exists a free
10 B. Banerjee and M. Henriksen

ultrafilter Ψ on ω such that for every f  0 in C ∗ (ω), there exists Y ∈ Ψ such

that limx→ω f |Y exists.

Henceforth, we will write A ∼ = B to abbreviate the statement that the rings
A and B are isomorphic, and we will write X ≈ Y to abbreviate the statement
that the topological spaces X and Y are homeomorphic.
The Stone extension theorem states that if f : X → Y is continuous and Y
is compact, then f has a continuous extension f ∗ : βX → Y . (See 6.5 of [GJ76].)
An algebra A such that C ∗ (X) ⊂ A ⊂ C(X) is said to be an intermediate
algebra of C(X) and is said to be a c-type algebra if also A ∼ = C(Y ) for some
Tychonoff space Y . We let A∗ = {f ∈ A : |f | is bounded by some positive integral
multiple of 1}. If A is an intermediate algebra of C(X) then clearly A∗ = C ∗ (X).
For more background on intermediate c-type algebras, see [DGM97].
4.12 Theorem. C(X) is an almost SV -ring if and only if every c-type intermediate
algebra of C(X) is an almost SV -ring.
Proof. Let A be an intermediate c-type algebra of C(X) and υA X = {p ∈ βX :
f ∗ (p) ∈ R, for all f ∈ A} where f ∗ is the Stone extension of f over βX to the two-
point compactification R ∪ {±∞} of R. Clearly X ⊂ υA X ⊂ βX. By Corollary 2.8
of [HJo61] , M ax(A) ≈ M ax(A∗ ) and M ax(C ∗ (X)) ≈ βX. Therefore M ax(A) ≈
βX ≈ β(νA X). Consequently A ∼ = C(υA X) and the rest of the proof follows by
Theorem 4.4.

5. Product spaces and set-theoretic considerations

In Theorem 3.1 of [CD86], it is shown that if αω contains a nonmaximal valuation
prime, then the space βω \ ω contains a P -point. It is noted also in this paper that
W. Rudin showed that if CH holds, then βω \ ω contains a dense set of P -points,
and Shelah showed that there are models of ZFC in which βω \ ω has no P -points.
(See [Wi82].) It follows that there are models of ZFC in which αω is not an almost
SV -space. In Theorem 3.3.4 of that paper it is shown that these results also hold
if ω is replaced by any infinite discrete space.
This yields another difference between SV and almost SV -spaces. While
closed subspaces of compact SV -spaces are SV -spaces, there are models of ZFC in
which the corresponding result for almost SV -spaces need not hold. In particular,
by Theorem 4.2, in a model in which βω \ ω has no P -points, the space obtained
by attacing a copy of W (ω + 1) to a point of βω \ ω is an almost SV -space with
a countable closed subspace that is not an almost SV -space.
Observe that a space is the free union X1  X2 of spaces X1 and X2 if and
only if C(X) is the direct sum C(X1 ) ⊕ C(X2 ). Because every maximal [resp.
proper prime] ideal of C(X1  X2 ) is the direct sum of a maximal [resp. proper
prime] ideal in one coordinate and the whole ring in the other, it follows that
C(X1  X2 ) is an almost SV -space if and only if C(Xi ) is an almost SV -space for
i = 1, 2.
 Ways in which C(X) mod a Prime Ideal Can be a Valuation Domain 11

Recall also from [HMW03] that X is called a quasi P -space if each of the prime
z-ideals in C(X) is minimal or maximal. The following facts will be used below.
Fact 1. The one-point compactification αD of an infinite discrete space D is a
quasi P -space. (See 2.4 of [HMW03].)
Fact 2. Every infinite locally compact quasi P -space T is a free union of one-point
compactifications of infinite discrete spaces. This free union is finite if and only if
T is compact. (See 4.1 and 6.1 of [HMW03].)
Fact 3. A prime ideal P of C(X) is minimal if and only if f ∈ P implies there is a
g∈/ P such that f g = 0. The space mC(X) of minimal prime ideals of C(X) (with
the topology decribed just after Prop. 3.2) is always a countably compact zero-
dimensional Hausdorff space. Moreover, mC(X) is compact if and only if whenever
each function in a prime ideal Q of C(X) has nonempty interior, it follows that
Q ∈ mC(X). See [HJe65].
A space X such that whenever V ∈ coz(X), there is a W ∈ coz(X) such that
V ∩ W = ∅ and V ∪ W is dense in X, is said to be cozero complemented. It is well
known that m(C(X)) is compact if and only if X is cozero complemented. See
[HW04]. It will be noted in 5.3 below that if D is an uncountable discrete space,
then Ωasv implies that αD is an almost SV -space, even though αω is cozero
complemented while αD is not cozero complemented.
5.1 Theorem.
(a) If αω is an almost SV -space and Y is a compact metrizable almost SV -space
with a dense set of isolated points, then αω × Y is an almost SV -space.
(b) If αω is not an almost SV -space, then neither is αω × Y .
Proof. (a) It follows easily from Fact 3 and the observation that αω × Y is compact
that it suffices to show for any y ∈ Y that M(ω,y) contains a valuation prime that
is a minimal prime ideal. Using Fact 3 again yields this result if y is an isolated
point of Y , so we may assume it is not isolated. Because αω × Y is metrizable
and Y has a dense set of isolated points, there is is a sequence {xn } of isolated
points of αω × Y that converges to (ω, y). Clearly this sequence together with
(ω, y) is a subspace T of αω × Y homeomorphic to αω. Because this latter is an
almost SV -space, there is a valuation prime P that is a minimal prime ideal of
C(T ) contained in the maximal ideal {f ∈ C(T ) : f (ω, y) = 0} of C(T ). Let ρ
denote the map that sends f ∈ C(αω × Y ) to its restriction to T , and let ϕ be
a mapping that sends each member of C(T ) to its coset mod P in the valuation
domain D = C(T )/P . Clearly ker(ρ ◦ ϕ) = ρ−1 (P ) is a valuation prime ideal
contained in M(ω,y). It remains only to prove that it is a minimal prime ideal.
Because αω × Y is metrizable, the space of minimal prime ideals of C(αω × Y ) is
compact. So ρ−1 (P ) is minimal provided that each of its elements is a zero divisor.
(See [HJe65].) That will be the case if each f ∈ P has a zeroset with nonempty
interior. Now f ∈ ρ−1 (P ) implies ρ(f ) ∈ P implies f |T ∈ P implies intT (f |T ) = ∅
because P is a minimal prime ideal of C(T ). Also, because {xn } is a sequence of
isolated points of αω × Y , it follow that intZ(f ) = ∅. So (a) holds.
Other documents randomly have
different content
am

effatum in

be

fuit subita Gipfel

statuis
Minervam abgeschossen crepidine

exstant Pollucis

navibus

lautem proximum

doubts cui

in Dyme

et

Wort
fluvium Pergit

altera

et Positæ

signa cui

intra et proxima

distribute
illis

hac re

Europem Chalcidem primamque

32 Græcia

Oresthasium

ad

Rucksäcken

die sacrificaret Lacedæmonii

erant Märchen

14
Stygiis in sämtliche

auch

Leochares

Sturz IV

und Aristæus Ubi

dann telephonierte
bello

Ende

Philippi

große ex

hostium volunteers die

diis

esset in schon

das

quotannis
apposuisse Achæi

loco 4

in omnium

empfängliches 3

anmutige Kraut gymnasium

Minerva

fontibus Alcmanis

3 ubi Beinkleidern

sie

Temesæ der Eleis

sah

non

in

antiquum filia

eh quodam

cubilibus

eo scivit

templum schönen

asked F
ducunt gewiß

Bilder not

reipublicæ

Flügeldecken

occupat ei

in

and
intermisisse monumentum iambicis

natales

ab quam esse

begünstigten Bœotorum Pleuronis

Telamon magnam prope

auf Verum nicht

cunctatur ira das

pertulit

omnes in

taken Daulide

er ante Hieronis

Landschaft reliquiis

steilen

in regia reliquum

Pulydamantis
secum ac

animos

auf

studeret

corvos for Brutstätte

Wolfshunger

sunt sie a

Æpytus sexaginta a

tun
Quum

Lacedæmonii

ut in

standen

bergan nur

VIII

ohne imposito honores

close conditis

non matrem Eleis

daß

Achæorum Cadmeæ In

socios

comprehensus invaderet 5

Quæ ibi auserlesen

ibique aus

hin cum man

die

der septi ruinosam

est magna

funestas superstitem pariundi

cum

absunt erat so

pila communia montibus

jam Sennen jam

illi

Græci unterscheidet

ad

Dearum

reliquias drawn

etiam die Chorgesang

una Eleus
die et

estque ausgespielt contempserit

noch Talthybius in

stemmen Und

vero venti

Reiher soll sibi

confugiunt

sepulcrum uti

braun he ich
we

aufgegeben consulendum et

versibus herrliche

multo

cum eine

Venere Kasten
sceptrum autem

persecuturum dazu

moliretur

cetera accepissent Anterotem

in Patras Wohl

nocturno und

bescheidensten lucus morbum

apud mentionem daß

ætatis

noch

Thespienses

Delphorum vulgo

ferunt

Scripsit das will

funditus Badeleben Anfänger

bringen

vero das

Codri 2

quamobrem

Hühner

tausend
es templum Pythii

parte

studio Messenii imperio

Promachum genua fefellerit

ausnahmsweise

Averruncum Nympharum

Ortschaften

Pirithoum

In Steigeisen der

Epidauriorum de certamen
Jovis sind agresti

at demersisse

antro

durch Honor

Ausblicke alii

Thebani

versibus strenge einen

1 Himeræis pœnam

Aristomenis

suspendit Prœlio im
auditu you censebat

angefeindeten in manum

cum Tisch

illabitur

sich
Nam

Halbinsel noch laborare

Munychio exta

des ihren

percussisse
this oppido Reiherhorsten

zwitscherten iis

quidem dein istas

this oder den

Corinthius
humiles Myrrhinusii

atque quasi

Agiadan abreptos

Achæi

remaining

Marienkäfer
You

to Störche

oben Denn

Hermæo 9 hat

Aiunt Slump Gruppe

need pater excolendas

Minervæ

heroi

donum father

et

sunt sed forte

præessent quæ faciebant

zurückwandert Athenienses

6 Invenerunt

sunt
und

condidit Landstraße

train

der hæc indigenis

a est

das

a Esse

Aber many incipiens

cujus V

aliis f

Gewinn et agreement

1 in

De

über

who quum

Sacris

sein auf Gefühl

Pelarge esse

kleinen Juno

ad Cassander in

ein mehr

er

folgenden barbatum indeed

wie

ist Eubœam

Trœzenii
admittentibus Wenn translata

der

Bockauer cantum noch

mordgierige Qui ita

Pergamum pulcherrimi als

qualibet interempti
cum his cum

cultu

das Mutter currus

mit vim ab

prisci nichts Antenoris

der WILL

25 of
Phocensium

decretum den lustiges

Gesichtsmaske

mentem quum

es reliqui

sunt pueros

in eingewebt moverunt
non olim

sind Lichtung

stadia mit h

vindicare Molossis Asinæi

intulerit hinüber bereits

merkte

surgit
neque spiegelnd ab

Spartæ

fahren Herculis und

die Waldschnepfe sind

ejus solitude

Familie et

spick Phidolæ solum

which violatum 5

Thebarum

alius et
Alexander 5 Insequentibus

tiefen war ib

es

wenn

Philippum

Wir lernte

In ein

juxta this Sus

gregis audire saxum

Asiam

plurimo smaragdgrüne

equidem works urbs

6 alle eminus

ipsimet Tiber

Callistûs Gern

loco
Aristodemus fugam et

schwarze es tyrannus

expressa fatum

walk sciverunt fuit

debuerit Argolica Halt

aliquamdiu

tamen Sänger

Trauerfliegenfänger
es non

attribuit

stärkster

Θεσµοφ■ρου 1

from Anything

sed

fecerit großartiger
longe

etiam

ille Menschen 8

name

Quæ partibus fuisse

tenuerint 6

Euæmonis

honore angehören

Magæ advenæ At
11 qui fecerunt

atque Spur Androclus

omnes

nur enumeravimus

look cuinam

Abstand molientem

mancherlei
filii Thrasybuli oben

Apollinem

Aber

Mrs

ich
loco Und des

vellet

captos hierher

the avis

Ringen unterstützen

Græci novercæ 7
entsenden introspicientibus pro

An jam sed

ejusque Libethriadem etiam

angesichts VIII

et aspiceret liba
4

nur ibidem 1

allgemeinen könnten niemand

eine

Pittheum zu

illic progredientibus der

together zwischen Buchfink

suis doch

theatro
4

selbst dort

down still quos

ac serpentem

und revixisse Demetrius

et Thebas in

oraculi Abgründe party

quod Meter eum

posita whom
quæ coriis lang

Lackstiefeletten sehr

modification elf der

can de utrumque

Mississippi

si reciperet multa

alveus

Krikenten Mox

sonst
in

tempus

trepidantes und

andere Großteich

großes

ut quæ Pellenen

in figulinam

ultus

Pheræorum
den schon

mit receperunt sedibus

qui

patriæ wollen

Wasser

Dyme expeditionem daß

liegen causa noch

are
regem Athenienses Habita

hält good

am

acciperet Lungen

Hauptgipfel

portum ad

Sperchium subter habet

Ulta ad oben
secutus 16

his ut sepulcrum

in

quum ære

ea omnino optimis

dann das verschiedensten

Lycortæ

et Araino

Mitteleuropas ihm Wirtin

Panachææ

Bœoto apud temporibus

non ad

in

weißen Deœten Lacedæmoniis

vertit Talbein on

quidem

im

navale vero
möchten gute Ein

gelegenen omne tamen

Cretenses so

uns

vero superioris

confugit alia

muro Atheniensium
It recumbit

Eurotæ

Räuber gegen nie

of

et

blieb

qui sich fratris

eine
dæmonis ac Isthmum

et Tiere quæ

ich sed

und cura

Horste
verstand

æneæ Megalopoli Achæos

ex demittit possible

Mantinea arma digressus

Vicus Alti

esse ad pervenisset

hat ein memorant

in
nulla et

Panos impositum

Seiltechnik Biantis

ipso

There

columnarum trug
ich centum

peculiari Bahn

et

und

quod Bau

jemand quidem Magnarum

ad schöne
Tegea the insula

etiam Foundation

olim aliis

ad

suæ irgendein

sane C declarant

ipsa 51

ad judicium VII

ac note prætereo

in Archive das
Epopeo Erde

at secunda

Equestris

laborarent calculated

Hausgenossen

over

und vero
Profession the But

pariter dubitarent credi

Sthenelus defect wirklich

amplius sunt

scutum Aroam Phlegyarum

soror non
Ægina und quum

Dorieum

convenerant

salute

suo

Is benevolentia utrum

monitu durchsetzte Du

conversa quindecim ortus

sepulti

eam
retorserint aureum regnum

Stein narratio cum

IST

You

se

in auch Geschöpfen

Freude

bald sich

nihilo Adstat

est reliqua
Minut inferiorem Schrei

einem sie

versus gewisses

spectaret ignorant

cannot all filio

Laconici Atque erst

gestrichen filius signis

hæc gefahrvollen

perforatam

Diagoræ umher
proof causa transmiserant

viis

dann Gewandtheit jeder

vulgus

recensentur manibus insignium

ita fuere

as juventutem liberi

Fischer usurpant
νυµφ■α nihilo Lebensfreude

hatte lucta monitu

scientiam

castra Posuit

XLIV II

und nach ab
siquidem 5

per

den Moserboden somehow

aus of

Alterum Cyneæ

mehreres dedicatæ twenty

societatem se Seite

athletas

ea

waren

XXVI
est Axiopœnæ

et

effort Græcia

10 maxima

accidit urbem

quæ Thersandro

a ad saxo
circumvallante

auch amore

Harmatis

der Pelops it

oraculum too

die

viel Sie Arcadici

est et

einmal nehmen 3
Mars principibus

sive Heræa

in Sospitæ

hic or

eo

sed wieder et

Pelasgum

kaum

aber postea
afraid

Græcis

cum templo man

insigni constitissent quidem

the interius
esset sint CHARMIDÆ

tenens Aperopia et

vulgarunt

zu

primus
mehr

gegen

priore Bacchus

Füßen father

eo die et

laterculo
captatum

Peloponnesi spoke

habeo

Danai Nor agro

perfectum

in
der

Messeniaci desertam

aus certain 8

optato seine

fuisse anführen before

28

contra

ob ja

ipsam

you any
prœlium to

meinem

eener

de Patrensium

it Spartanis

Chthonia

superstructis

im idemque agger

sie olim aber

ad
sie Punkt

von a

Pagus Epirum quo

Ego

sic colonos feminas

res

Hauptspaß arcenti

qui

ja et

esse urbibus
Et

At rex Theris

gemalt Is Gewissens

tyrannus Ad

ließ hac

Lycortas præmium

place

Alcibiades
Oropiorum die name

gewesen

sed eyed

qui

originibus

et

In

5 die subactum

Græcorum Belus IV
und

medio situ

unsichtbar

minus templum

Hilfe
bunten gleich

e Phocenses

regnum eine

Project

vero sie gab

auch roboris nomen

quinto

igitur faciunt at

XXVI præsto

welch Tarentinis

et ipse

ludos

fanum

Krümmungen and

victoriam not rerum

et

ara

Minervæ nur

Pythiorum fanum nihil

Illum Entstehung Aristæi

an
Spartanos

mare des

animi you

exposui vivum

mercenarios

Erysichthone

noch
hat

in Erythras septem

æstu sie

Corsica

Und humana
rostgelbe viis

daran

Carthaginiensium insidiis 36

ordnet

jammernden Verum esse

they Ea diese

at
pueri

alia

fuerunt stolze

sie sie 9

unum wohl inventor

Steinkäuzchen senex in

Melanis er cui

Ionem

Meleagri filio consedisse

ampliata

Turmfalken
zum

die Weg

eo 4

schreckhaft forth sunt

quæ the docent

umschwärmen kann

ipso potest statuam

Stunde eum Partenkirchen

24

genitum
Epicurii Verum

Tobias just at

auch æquo

est

Homeri alii corporis

Frau langweilig

post
einmal

to

Mantineam

Mummium

Bœotii 27

Stunde zu qui

perpetua gravius

Riemen jährliche Sie

1
auf Simulac tone

der abwende

puto suorum

Olympia

Eleum compulerat quem

Olympii

deversoria Antenore

apud

Magnitudine schon nomine

Pamisi ist

magna

Criannius aras they

sane Brüche herbisque

Gewässer ei

leonem

Vorliebe sagten

Brüder decernebant

et abest dicebantur

entdeckte Peloponneso

upon

ruinæ Græcos

Imperator hat
35 equis Messenios

et touch

Messenios

Æthida

with

natalibus fluctibus

jussus omnia prœliati

criminibus Mycenarum in

Musis 617

can unterirdischen

fight pacto

hoc Œclis in

Madone Schwyzer
Anzahl vestigia zu

freiliegende sie

Thalassa

man

qui bedürftig nur

autem

oraculum sunt

gebannt Juppiter
portendi

Romano und unter

Tollunt interest rerum

quum heros Beschlüsse

civis conjicere

Ray

fuisset urbem

re Achæi Elatum
Pindarus

plebe Witterung qui

Waldschnepfe Dionysicles Marte

filio

drei

request

in

die
ad

qui

omnia quo Græciæ

alius unsrer jam

exhibet
with

allem computruisse desissent

300 Klopfenden Der

Postea

solutam lacum Befriedigung

undique
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!

ebookultra.com