PDF Linear Systems Signal Processing and Hypercomplex Analysis Chapman University November 2017 Daniel Alpay Ebook Full Chapter

Linear Systems Signal Processing and
Hypercomplex Analysis Chapman

University November 2017 Daniel Alpay
Visit to download the full and correct content document:
https://textbookfull.com/product/linear-systems-signal-processing-and-hypercomplex-
analysis-chapman-university-november-2017-daniel-alpay/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Communications, Signal Processing, and Systems:

Proceedings of the 12th International Conference on
Communications, Signal Processing, and Systems Vol 1
Wei Wang
https://textbookfull.com/product/communications-signal-
processing-and-systems-proceedings-of-the-12th-international-
conference-on-communications-signal-processing-and-systems-
vol-1-wei-wang/
Quantum-Mechanical Signal Processing and Spectral

Analysis First Edition Belki■
https://textbookfull.com/product/quantum-mechanical-signal-
processing-and-spectral-analysis-first-edition-belkic/
Digital Signal Processing Fundamentals and Applications

3rd Edition Lizhe Tan Ph.D. Electrical Engineering
University Of New Mexico
https://textbookfull.com/product/digital-signal-processing-
fundamentals-and-applications-3rd-edition-lizhe-tan-ph-d-
electrical-engineering-university-of-new-mexico/
Cognitive Systems and Signal Processing 4th

International Conference ICCSIP 2018 Beijing China
November 29 December 1 2018 Revised Selected Papers
Part I Fuchun Sun
https://textbookfull.com/product/cognitive-systems-and-signal-
processing-4th-international-conference-iccsip-2018-beijing-
china-november-29-december-1-2018-revised-selected-papers-part-i-
Cognitive Systems and Signal Processing 4th
International Conference ICCSIP 2018 Beijing China
November 29 December 1 2018 Revised Selected Papers
Part II Fuchun Sun
https://textbookfull.com/product/cognitive-systems-and-signal-
processing-4th-international-conference-iccsip-2018-beijing-
china-november-29-december-1-2018-revised-selected-papers-part-
ii-fuchun-sun/
Signal and Image Processing Techniques for the

Development of Intelligent Healthcare Systems E. Priya
https://textbookfull.com/product/signal-and-image-processing-
techniques-for-the-development-of-intelligent-healthcare-systems-
e-priya/
Signal processing for neuroscientists Drongelen
https://textbookfull.com/product/signal-processing-for-
neuroscientists-drongelen/
Digital Image Watermarking: Theoretical and

Computational Advances (Intelligent Signal Processing
and Data Analysis) 1st Edition Borra
https://textbookfull.com/product/digital-image-watermarking-
theoretical-and-computational-advances-intelligent-signal-
processing-and-data-analysis-1st-edition-borra/
Fundamentals of Signal Enhancement and Array Signal

Processing 1st Edition Jacob Benesty
https://textbookfull.com/product/fundamentals-of-signal-
enhancement-and-array-signal-processing-1st-edition-jacob-
benesty/
Operator Theory
Advances and Applications
275
Daniel Alpay
Mihaela B. Vajiac
Editors
Linear Systems,
Signal Processing
and Hypercomplex
Analysis
Chapman University, November 2017
Operator Theory: Advances and Applications
Volume 275
Founded in 1979 by Israel Gohberg
Series Editors:
Joseph A. Ball (Blacksburg, VA, USA)
Albrecht Böttcher (Chemnitz, Germany)
Harry Dym (Rehovot, Israel)
Heinz Langer (Wien, Austria)
Christiane Tretter (Bern, Switzerland)
Associate Editors: Honorary and Advisory Editorial Board:

Vadim Adamyan (Odessa, Ukraine) Lewis A. Coburn (Buffalo, NY, USA)
Wolfgang Arendt (Ulm, Germany) Ciprian Foias (College Station, TX, USA)
B. Malcolm Brown (Cardiff, UK) J.William Helton (San Diego, CA, USA)
Raul Curto (Iowa, IA, USA) Marinus A. Kaashoek (Amsterdam, NL)
Kenneth R. Davidson (Waterloo, ON, Canada) Thomas Kailath (Stanford, CA, USA)
Fritz Gesztesy (Waco, TX, USA) Peter Lancaster (Calgary, Canada)
Pavel Kurasov (Stockholm, Sweden) Peter D. Lax (New York, NY, USA)
Vern Paulsen (Houston, TX, USA) Bernd Silbermann (Chemnitz, Germany)
Mihai Putinar (Santa Barbara, CA, USA) Harold Widom (Santa Cruz, CA, USA)
Ilya Spitkovsky (Abu Dhabi, UAE)
Subseries
Linear Operators and Linear Systems
Subseries editors:
Daniel Alpay (Orange, CA, USA)
Birgit Jacob (Wuppertal, Germany)
André C.M. Ran (Amsterdam, The Netherlands)
Subseries
Advances in Partial Differential Equations
Subseries editors:
Bert-Wolfgang Schulze (Potsdam, Germany)
Michael Demuth (Clausthal, Germany)
Jerome A. Goldstein (Memphis, TN, USA)
Nobuyuki Tose (Yokohama, Japan)
Ingo Witt (Göttingen, Germany)
More information about this series at http://www.springer.com/series/4850

Daniel Alpay • Mihaela B. Vajiac
Editors
Linear Systems,
Signal Processing
and Hypercomplex
Analysis
Chapman University, November 2017
Editors
Daniel Alpay Mihaela B. Vajiac
Schmid College of Science Schmid College of Science
and Technology and Technology
Chapman University Chapman University
Orange, CA, USA Orange, CA, USA
ISSN 0255-0156 ISSN 2296-4878 (electronic)

Operator Theory: Advances and Applications
ISBN 978-3-030-18483-4 ISBN 978-3-030-18484-1 (eBook)
https://doi.org/10.1007/978-3-030-18484-1
Mathematics Subject Classification (2010): 46E22, 35Q, 30G35, 46L54
© Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions
that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
This book is published under the imprint Birkhäuser, www.birkhauser-science.com by the registered company
Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
Editorial Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
J.A. Ball and V. Bolotnikov

Multiplicative Stieltjes Functions and Associated Pairs
of Reproducing Kernel Hilbert Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
J. Behrndt and P. Schlosser

Quasi Boundary Triples, Self-adjoint Extensions and
Robin Laplacians on the Half-space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
S. Bezuglyi and P.E.T. Jorgensen

Graph Laplace and Markov Operators on a Measure Space . . . . . . . . . 67
M. Bożejko
Conditionally Free Probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
P. Cerejeiras, U. Kähler, A. Legatiuk and D. Legatiuk

Boundary Values of Discrete Monogenic Functions over
Bounded Domains in R3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
I. Cho and P.E.T. Jorgensen

Semicircular Elements Induced by Projections on Separable
Hilbert Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
B. Fritzsche, B. Kirstein and C. Mädler

On a Backward Shifting Problem for [α, ∞)-non-negative
Definite Sequences of Complex q × q Matrices . . . . . . . . . . . . . . . . . . . . . . 211
A. Melnikov and R. Shusterman

Evolution of Nodes and their Application to Completely
Integrable PDEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
vi Contents
A. Sebbar and O. Wone

Frobenius Determinants and Bessel Functions . . . . . . . . . . . . . . . . . . . . . . 251
A. Yger
Algebraic Residue Calculus Beyond the Complex Setting . . . . . . . . . . . 289
Operator Theory:
Advances and Applications, Vol. 275, vii–viii

c Springer Nature Switzerland AG 2019
Editorial Introduction
Daniel Alpay and Mihaela Vajiac
Men only build to nothingness

Vain dreams in noble guise:
When waves to silent tomb quiesce
New waves again will rise.
Mihai Eminescu, Luceafarul
(The Legend of the Evening star),
translation by Adrian G. Sahlean.1
The present volume contains ten refereed papers written by experts in their fields
on the occasion of the conference Mathematics, Signal Processing and Linear Sys-
tems: New Problems and Directions, held at Chapman University on November
14–19, 2017. The topics of the conference were at the intersection between mathe-
matical analysis, signal processing and applications to electrical engineering prob-
lems. The research presented belongs both at the forefront and at the crossroad of
several branches of mathematics in particular complex analysis, functional analysis
and stochastic processes and electrical engineering.
This conference was part of a series of yearly mathematics conferences and
workshops held at Chapman University since 2010. The topics of this conference
were carefully chosen to further mathematical research to include applications to
engineering, due to the forthcoming opening of Chapman University’s new School
of Engineering scheduled for the Fall of 2020.
The papers can be divided in the following overlapping categories: two papers
in Schur analysis, three papers on differential operators and inverse scattering, two
papers in hypercomplex analysis and three papers in non-commutative analysis
and free probability. More precisely, we have the following classification:
Schur analysis and function theory: In the paper Multiplicative Stieltjes functions
and associated pairs of reproducing kernel Hilbert spaces, by Joseph A. Ball and
Vladimir Bolotnikov use state space methods to study pairs of reproducing
kernel Hilbert spaces associated to the (new and defined in the paper) family
1 (1993) Prospero Press, 35 Commonwealth Part, Newton, MA 02159, USA
viii D. Alpay and M. Vajiac
of multiplicative Stieltjes functions, and related interpolation problems. In On a

backward shifting problem for [α, ∞)-non-negative definite sequences of complex
q × q matrices, the authors Bernd Fritzsche, Bernd Kirstein and Conrad
Mädler combine algebraic and analytic methods to develop a Schur analysis to
matricial moment problems.
Differential operators, inverse problems and related topics: In Quasi boundary
triples, self-adjoint extensions, and Robin Laplacians on the half-space, Jussi
Behrndt and Peter Schlosser continue the study of quasi boundary triples
(introduced by Jussi Bernd and Matthias Langer in earlier papers) and ap-
ply it to the study of extensions of elliptic operators. Applications to Laplacians
with singular Robin boundary conditions are given. In Evolution of nodes and
its application to completely integrable PDEs, Andrey Melnikov and Roman
Shusterman use the notion of operator node to generalize the inverse scatter-
ing method to a wide class of non-linear equations. In Frobenius determinants
and Bessel functions, by Ahmed Sebbar and Oumar Wone, the authors study
the geometry and partial differential equations arising from the consideration of
Frobenius determinants. This leads us to address some aspects of twistor theory
as well as some extensions of Bessel functions.
Hypercomplex analysis: In Boundary values of discrete monogenic functions
over bounded domains in R3 , the authors Paula Cerejeiras, Uwe Kähler,
Anastasiia Legatiuk and Dmitrii Legatiuk develop interesting and new re-
sults in the theory of discrete monogenic functions over bounded domains. Alain
Yger, in Algebraic residue calculus beyond the complex setting presents an alge-
braic multivariate calculus in the setting of the bicomplex numbers.
Non commutative analysis: In Graph Laplace and Markov operators on a measure
space, Sergey Bezuglyi and Palle E.T. Jorgensen develop the continuous
analogue of the spectral theory of the Laplacian on infinite networks. In Con-
ditionally free probability, Marek Bozejko gives an overview of conditionally
free probability and of the non-commutative von Neumann inequality. The main
result is that the free product of a family of quantum channels is again a quan-
tum channel. In Semicircular elements induced by projections on separable Hilbert
spaces, Ilwoo Cho and Palle E.T. Jorgensen construct semicircular elements
from mutually orthogonal projections on the separable Hilbert space 2 (N0 ). They
act on semicircular elements. The corresponding free-probabilistic information is
considered.
Daniel Alpay and Mihaela Vajiac

Department of Mathematics
Chapman University
One University Drive
Orange, California 92866, USA
e-mail: alpay@chapman.edu
mbvajiac@chapman.edu
Operator Theory:
Advances and Applications, Vol. 275, 1–47

c Springer Nature Switzerland AG 2019
Multiplicative Stieltjes Functions

and Associated Pairs of
Reproducing Kernel Hilbert Spaces
Joseph A. Ball and Vladimir Bolotnikov
Abstract. Pairs of reproducing kernel Hilbert spaces associated with mul-

tiplicative Stieltjes functions are characterized in terms of shift invariance
and structural identities. We also obtain explicit state-space formulas for the
structural identities and reproducing kernels by representing the reproduc-
ing kernel Hilbert spaces as ranges of suitable frequency-domain observability
operators.
Mathematics Subject Classification (2010). 46E22; 46E40, 47A56, 47A48.

Keywords. Reproducing kernel Hilbert space, multiplicative Stieltjes function,
backward-shift invariance, structural identity, Potapov–Ginzburg transform.
1. Introduction
Let G be a Hilbert space and let L(G) stand for the set of all bounded linear
operators on G. A Hilbert space H of G-valued functions which are defined on a
domain Ω ∈ C is said to be a reproducing kernel Hilbert space if there exists a
L(G)-valued function K(z, ω) such that for every point ω ∈ Ω and every vector
c ∈ G, the function Kω c := K(· , ω)c belongs to H and has the reproducing-kernel
property for the space H:
f, Kω cH = f (ω), cG for all f ∈ H. (1.1)
The function K(z, ω) turns to be a positive kernel on Ω in the sense that

n
cj∗ K(zj , z )c ≥ 0
j,=1
for every choice of an integer n, of vectors c1 , . . . , cn ∈ G and of points z1 , . . . , zn ∈

Ω. The function K(z, ω) is furthermore uniquely determined from the reproducing
2 J.A. Ball and V. Bolotnikov
kernel property (1.1) (as is easily verified), and we then say that K is the repro-
ducing kernel of H and write H = HK . A result of Aronszajn [8] states that, for
every positive kernel K on Ω, there is a unique reproducing kernel Hilbert space
HK with K as its reproducing kernel. Moreover, the set H0 consisting of func-
tions of the form K(· , ωj )cj , where {cj } and {ωj } are finite sequences in G and
Ω, respectively, is a dense linear manifold in HK . Let us note also the following
characterization of reproducing kernel Hilbert spaces.
Theorem 1.1 (See, e.g., [8]). A given Hilbert space H consisting of G-valued func-
tions on a domain Ω has the form H = HK for some positive kernel K : Ω ×
Ω → L(G) if and only if the operator Eω : H → G induced by point-evaluation
Eω : f → f (ω) is continuous for each ω ∈ Ω.
An up-to-date account of all this information together with more recent ap-
plications (e.g., support vector machines and machine learning) can be found in
the book of Paulsen–Raghupathi [38].
1.1. Reproducing kernel Hilbert spaces with additional structure:

the multiplicative Pick class
In this paper we shall be interested in how additional properties of the positive
kernel K translate to additional structural properties of the reproducing kernel
Hilbert space HK . A specific form for the positive kernel K of interest for us can
be explained as follows.
Given a Hilbert space G, we define the unitary selfadjoint operator

0 iIG
J= ∈ L(G ⊕ G). (1.2)
−iIG 0
To distinguish the summands in the direct sum G = G ⊕ G, we identify the first

summand with the subspace G = {[ x0 ] , x ∈ G} of G and represent G as
G = G ⊕ JG. (1.3)
We choose and fix a non-empty open subset Ω ⊂ C which is symmetric about the

real axis R and consider a Hilbert space H whose elements are G-valued functions
meromorphic in Ω. Any reference to the value of a meromorphic function at α ∈ Ω
assumes that the function is analytic at α.
Definition 1.2. We say that H is a space H(Θ) if it admits a reproducing kernel

KΘ of the form
J − Θ(z)JΘ(ω)∗
KΘ (z, ω) := (1.4)
i(ω − z)
for some function Θ meromorphic on Ω, subject to
Θ(z)JΘ(z)∗ = Θ(z)∗ JΘ(z) = J for all z ∈ Ω, (1.5)
Stieltjes Functions 3
i.e., if H = HKΘ =: H(Θ) where KΘ is as in (1.4)–(1.5). On occasion it will be

useful to write KΘ (z, ω) in the equivalent aggregate form

J 0 I
I Θ(z)
0 −J Θ(ω)∗
KΘ (z, ω) =
i(ω − z)

I
I Θ(z) JMP
Θ(ω)∗ J 0
= , where JMP = . (1.6)
i(ω − z) 0 −J
We let MP(G, Ω) denote the set of all functions Θ meromorphic on Ω sat-
isfying (1.5) and such that the kernel (1.4) (or equivalently (1.6)) is positive on
Ω. Note that the positivity of the kernel KΘ implies in particular that Θ(z) is
J-contractive for points z of Ω in the upper half-plane. In the case when Ω = C,
condition (1.5) implies that Θ is J-unitary on the real axis and is therefore J-inner
in the upper half-plane.
Hilbert spaces of the type H(Θ) arise in the theory of non-selfadjoint trans-
formations (see [16], [17], [18]) as well as in the reproducing kernel approach to
classical interpolation problems of Nevanlinna–Pick type (see [21, 22] and refer-
ences therein). The study of J-contractive functions and in particular their mul-
tiplicative structure goes back to the influential work of Potapov [39]; for the
comprehensive survey of the subject, we refer to the monograph [9].
The next theorem provides a characterization of the spaces H(Θ). In its
formulation and in what follows, we will denote by Rα the backward-shift operator
centered at the point α defined by
F (z) − F (α)
Rα : F (z) → .
z−α

Theorem 1.3. A reproducing kernel Hilbert space H of G-valued functions which
are meromorphic in Ω is a space H(Θ) if and only if
1. H is Rα -invariant for each α ∈ Ω: Rα H ⊂ H.
2. The identity
Rα F, G − F, Rβ G + (β − α) Rα F, Rβ G = iG(β)∗ JF (α) (1.7)
holds for all functions F, G in H and for all α, β ∈ Ω.
If this is the case, Θ is defined uniquely up to a right J-unitary constant factor.
This theorem was established in [17], [18] under the additional condition
that the boundary of Ω contains at least one real point or the point at infinity and
functions from H are analytic at this point. In [41] this additional hypothesis was
shown to be superfluous. The same result was obtained in the paper of the first
author [10] for the case where the kernel denominator ρ(z, ω) = i(ω − z) in (1.4)
is taken instead to be ρ(z, ω) = 1 − zω with the structural identity (1.7) suitably
modified to
F, G + αRα F, G + β̄F, Rβ G − (1 − αβ̄)Rα F, Rβ G = G(β)∗ JF (α). (1.8)
The later paper of Alpay–Dym [5] worked out the result for a general setting
containing that of Theorem 1.3 and of [10] as special cases. Observe that the
left-hand side expressions in (1.7) and (1.8) can be written respectively as
Rα F, (I + βRβ )G − (I + αRα )F, Rβ G ,
(I + αRα )F, (I + βRβ )G − Rα F, Rβ G ,
disclosing the hidden role played by the operator
zF (z) − αF (α)
I + αRα : F (z) → .
z−α
As we shall see below, in the Stieltjes-class context Rα and I + αRα appear in a
combined form displaying their intertwined roles as equally important.
If Ω contains a real point μ and all functions from H(Θ) are analytic at μ,
then Θ(μ) is J-unitary (by (1.5)); since Θ is defined only up to a constant J-
unitary factor, we can normalize Θ so that Θ(μ) = IG. Letting ω = μ in (1.4)
gives
Θ(z) = IG + i(z − μ)KΘ (z, μ)J. (1.9)
To express Θ in terms of KΘ in case Ω ∩ R = ∅ requires some extra efforts, with
the resulting formulas being rather implicit (see [41]). In case μ is real, we can use
the formula (1.9) for Θ to get fairly explicit state-space realization formulas for
Θ once we have state-space realizations formulas of this type for the reproducing
kernel KΘ . This last goal can be accomplished via realizing the space H(Θ) as the
range space of some sort of observability operator, as we now explain. We shall be
particularly interested in observability operators of the form
OΠ,A,μ : x → Π(I − (z − μ)A)−1 x,
A ∈ L(X ), Π ∈ L(X , G).
The pair (Π, A) is called observable if OΠ,A,μ is injective, i.e., x = 0 whenever
OΠ,A,μ x ≡ 0. In the case when OΠ,A,μ is bounded as an operator from X to H,
we may form the gramian operator
∗
GΠ,A,μ = (OΠ,A,μ ) OΠ,A,μ ∈ L(X ).
Suppose that H is a Hilbert space of analytic functions which has a realization of
the form H = Ran OΠ,A,μ for some observable pair (Π, A) and bounded operator
OΠ,A,μ : X → H, i.e.,
H = {OΠ,A,μ x : x ∈ X } with OΠ,A,μ x2H = GΠ,A,μ x, xX . (1.10)
Then, as a consequence of the Open Mapping Theorem (see [40, Theorem III.11]),
it follows that the inverse mapping (OΠ,A,μ )−1 : H → X is also bounded, and
hence the inverse gramian operator (GΠ,A,μ )−1 is a bounded operator on X as
well. Furthermore, H is Rα -invariant for α in a neighborhood of μ, as follows from
the easily verified identity
Rα OΠ,A,μ x = OΠ,A,μ (I − (α − μ)A)−1 Ax. (1.11)
In addition, H is a reproducing kernel Hilbert space with reproducing kernel

−1
KH (z, ω) = OΠ,A,μ (z)GΠ,A,μ OΠ,A,μ (ω)∗
−1
= Π(I − (z − μ)A)−1 GΠ,A,μ (I − (ω − μ)A∗ )−1 Π∗ (1.12)
(see [8]).
Conversely, any Rα -invariant reproducing kernel Hilbert space H can be re-
alized as in (1.10) for some injective and bounded observability operator. Indeed,

given such a Hilbert space of G-valued analytic functions H, choose a point μ ∈ Ω
where each function in H is analytic. Locally in a neighborhood of μ, functions
f ∈ H have a power series representation
∞
f (z) = fn (z − μ)n .
n=0
Since H is Rμ-invariant, the function
∞
f (z) − f (μ)
Rμ : f (z) → = fn+1 (z − μ)n
z−μ n=0
is also in H. Since H is a reproducing kernel Hilbert space, the map Eμ : f → f (μ)

Then, for z in a neighborhood
of evaluation at μ is a bounded operator from H to G.
of μ, the action of the observability operator OEμ ,Rμ ,μ is given by
−1
OEμ ,Rμ ,μ f = Eμ I − (z − μ)Rμ f
∞
∞

= Eμ (Rμ ) (z − μ) f =
n n
fn (z − μ)n = f (z).
n=0 n=0
By analytic continuation we have the relation (OEμ ,Rμ ,μ f )(z) = f (z) continuing
to hold for z in all of Ω, i.e., the observability operator OEμ ,Rμ ,μ : H → H amounts
to the identity operator IH ; in particular, the pair (Eμ , Rμ ) is observable and the
gramian GEμ ,Rμ ,μ = IX is bounded and boundedly invertible.
The following theorem can be viewed as a state-space reformulation of The-
orem 1.3.

Theorem 1.4. Let H be a reproducing kernel Hilbert space of G-valued functions
which are meromorphic on a connected open set Ω ⊂ C and analytic at μ ∈ Ω.
Then:
1. H is Rμ -invariant if and only if it admits an observable realization (1.10)
One such choice of X , A, Π and GΠ,A,μ is
with A ∈ L(X ) and Π ∈ L(X , G).
X = H, A = Rμ , Π = Eμ : f → f (μ), GEμ ,Rμ ,μ = IH .
2. H is an H(Θ) space if and only if it admits an observable realization as in
part (1) and in addition, the observability gramian G = GΠ,A,μ satisfies the
Stein–Lyapunov identity
GA − A∗ G + i(μ − μ)A∗ GA = iΠ∗ JΠ. (1.13)
3. If Ω ∩ R = ∅ and ν ∈ Ω ∩ R, then Θ can be taken in the form

Θ(z) = I + i(z − ν)Π(I − (z − μ)A)−1 G −1 (I − (ν − μ)A∗ )−1 Π∗ J.
Proof. Statement (1) follows from the discussion preceding the theorem. Statement
(3) follows by combining formulas (1.9) (with μ = ν ∈ R) and (1.12) (with ω = ν).
It remains to verify statement (2); this in turn follows by showing the equivalence
of the structural identity (1.7) with the Stein–Lyapunov equation (1.13) for the
case where H is presented as the range of an observability operator OΠ,A,μ .
Let us therefore suppose that H = Ran OΠ,A,μ . It is convenient to write out
the observability operator as OΠ,A,μ : x → ΠΓ(z)x where we have set
Γ(z) := (I − (z − μ)A)−1 .
Then two generic elements F and G in H are of the form
F (z) = ΠΓ(z)x, G(z) = ΠΓ(z)y, where x, y ∈ X . (1.14)
Then the right-hand side of (1.7) can be written as
iG(β)∗ JF (α) = iJΠΓ(α)x, ΠΓ(β)yG = iΠ∗ JΠΓ(α)x, Γ(β)yG. (1.15)
On the other hand, by (1.11) and (1.14), we have
Rα F = Rα OΠ,A,μ x = OΠ,A,μ AΓ(α)x, Rβ G = Rβ OΠ,A,μ y = OΠ,A,μ AΓ(β)y,
which allows us to compute the left-hand side of (1.7) as follows:
Rα F, GH − F, Rβ GH + (β − α)Rα F, Rβ GH (1.16)
= OΠ,A,μ AΓ(α)x, OΠ,A,μ yH − OΠ,A,μ x, OΠ,A,μ AΓ(β)yH
+ (β − α)OΠ,A,μ AΓ(α)x, OΠ,A,μ AΓ(β)yH
∗−1
= (Γ(β) GΠ,A,μ A − A∗ GΠ,A,μ Γ(α)−1 + (β − α)A∗ GΠ,A,μ A)Γ(α)x, Γ(β)y X
.
Due to the arbitrariness of the vectors x and y, comparison of the results of (1.15)
and (1.16) shows us that the structural identity (1.7) holding for all F, G ∈ H is
equivalent to the operator identity
Γ(β)∗−1 GΠ,A,μ A − A∗ GΠ,A,μ Γ(α)−1 + (β − α)A∗ GΠ,A,μ A = iΠ∗ JΠ. (1.17)
Recalling now that
Γ(β)∗−1 = I − (β − μ)A∗ , Γ(α)−1 = I − (α − μ)A,
we see that the left-hand side in (1.17) is given by
(I − (β − μ)A∗ )GΠ,A,μ A − A∗ GΠ,A,μ (I − (α − μ)A) + (β − α)A∗ GΠ,A,μ A
= GΠ,A,μ A − A∗ GΠ,A,μ + (−β + μ + α − μ + β − α)A∗ GΠ,A,μ A
= GΠ,A,μ A − A∗ GΠ,A,μ + (μ − μ)A∗ GΠ,A,μ A.
With this simplification of the left-hand side of (1.17), we see that equality in
(1.17) collapses exactly to the Stein–Lyapunov identity (1.13).
Remark 1.5. Let us note that the strategy of the proof of part (2) in Theorem
1.4 foreshadows what will be done in the more complicated setting of Stieltjes
functions discussed below (see specifically Theorem 4.1).
1.2. The Pick class and connections with the multiplicative Pick class
through the Potapov–Ginzburg transform
Let us recall the Pick class P(G) (in the literature also known as Nevanlinna–
Herglotz class and sometimes also simply as R-class) consisting of L(G)-valued
functions holomorphic on the upper half-plane C+ with values there having positive
semidefinite imaginary part, i.e., the functions S : C+ → L(G) such that the kernel
S(z) − S(ω)∗
KS (z, ω) = (1.18)
z−ω
is positive on C+ . In fact, if the kernel (1.18) is positive on a domain Ω ⊂ C+ ,
it can be (uniquely) extended as a positive kernel to all of C+ due to the Pick
interpolation theorem. It is convenient (and is consistent with Nevanlinna–Herglotz
integral formula) furthermore to extend Pick functions to the lower half-plane by
reflection: define S(z) = S(z)∗ for z ∈ C− .
Let us note that the kernel KS can be rewritten in a more aggregate form as

0 iIG I
I S(z)
−iIG 0 S(ω)∗
KS (z, ω) =
i(ω − z)

I
I S(z) JP
S(ω)∗ 0 iIG
= , where JP = . (1.19)
i(ω − z) −iIG 0
In case we replace G with G = G ⊕ JG, comparison of (1.19) with (1.6) suggests

the close connection between the multiplicative Pick class MP(G) and the Pick
class over G, i.e., P(G);
the kernel KΘ built from Θ appearing in (1.6) has exactly
the same form as the kernel KS built from S appearing in (1.19), but with the
aggregate signature matrix JMP for the class MP(G) replaced by the aggregate
In fact there is a simple linear-fractional
signature matrix JP for the class P(G).
transformation TPG (called the Potapov–Ginzburg transformation (see [27]) which
maps P(G) bijectively to MP(G) and which can be derived as follows.
⊕ G),
A B ] be any operator on L(G
Let [ C written in block 2 × 2 form with
D

A, B, C, D ∈ L(G), which is (JP , JMP )-unitary, i.e., [ C
A B ] should satisfy the iden-
D
tities
A B]J
A∗ C ∗ A∗ C ∗
MP B ∗ D∗ = JP , JP [ C
[C A B]=J
D B ∗ D∗ D MP . (1.20)
One can check that one possible choice of A, B, C, D is

I 0 0 0 0 IG 0 0
A= G , B= , C= , D= . (1.21)
0 0 IG 0 0 0 0 −IG

For S a meromorphic L(G)-valued function on Ω, we say that S is in the domain
D(TPG ) of the Potapov–Ginzburg transform TPG if (A+S(z)C)−1 is a well-defined

meromorphic L(G)-valued function on Ω and then we define TPG [S] by
TPG [S](z) = (A + S(z)C)−1 (B + S(z)D). (1.22)
R
Dually we define the (right) Potapov–Ginzburg transform TPG by
R
TPG [S](z) = (iJC ∗ S(z) + iJA∗ )(iJD∗ S(z) + iJB ∗ )−1

with domain consisting of all meromorphic L(G)-valued functions on Ω such that
∗ ∗ −1
(iJD S(z) + iJB ) is also a well-defined meromorphic function on Ω. Note that
−1 ∗
A B J 0 A C∗ 0 iIG −iJC ∗ iJA∗
= = .
C D 0 −J B ∗ D∗ −iIG 0 iJD∗ −iJB ∗
The calculation

−S(z)
0 = I S(z)
I

A B −iJC ∗ −iJA∗ −S(z)
= I S(z)
C D −iJD∗ −iJB ∗ I

A B −I 0 iJC ∗ iJA∗ S(z)
= I S(z) · (−1)
C D 0 I iJD∗ iJB ∗ I

−TPG
R
(S)
= (A + S(z)C) I TPG [S](z) (−1) · (iJD∗ S(z) + iJB ∗ )−1
I
enables us to conclude that
R
TPG [S] = TPG [S] if S ∈ D(TPG ) ∩ D(TPGR
).
A B
iJC ∗
iJA∗

In case A, B, C, D are as in (1.21), then the coefficient matrix C D
= iJD∗ iJB ∗
R
for TPG works out to be

A B −A B
=
C D C −D
and hence in this case we have
TPG [S](z) = (A + S(z)C)−1 (B + S(z)D)
= (−AS(z) + B)(CS(z) − D)−1 = TPG
R
[S](z).
The following is the key property of the Potapov–Ginzburg transform TPG .

Proposition 1.6. (See [27].) Given S ∈ D(TPG ), then S is in the Pick class P(G)
if and only if Θ := TPG [S] is in the multiplicative Pick class MP(G).
With assumptions as in the
Proof. Suppose first that S is in the Pick class P(G).
statement of the theorem, we use the aggregate formula (1.6) for the MP-class
kernel KΘ with Θ = TPG [S] to compute

I
i(ω − z) · KΘ (z, ω) = I TPG [S](z) JMP
TPG [S](z)
∗
A B A C∗ I
= (A + S(z)C)−1 I S(z) JMP (A∗ + C ∗ S(ω)∗ )−1
C D B ∗ D∗ S(ω)∗

I
= (A + S(z)C)−1 I S(z) JP (A∗ + C ∗ S(ω)∗ )−1 (by (1.20))
S(ω)∗
= (A + S(z)C)−1 (i(ω − z)) · KS (z, ω)(A∗ + C ∗ S(ω)∗ )−1 (by (1.19)). (1.23)
Upon dividing both sides of the final equality by i(ω − z), we conclude that KS
implies that KΘ is a
being a positive kernel (i.e., S being in the Pick class P(G))
positive kernel (i.e., that Θ is in the multiplicative Pick class MP(G)).
Conversely, suppose that Θ = TPG [S] is in the multiplicative Pick class
MP(G). From the identity (1.23) we read off that
KS (w, ω) = (A + S(z)C)KΘ (z, ω)(A∗ + C ∗ S(ω)∗ )
where A + S(z)C is a well-defined invertible meromorphic function on Ω since we
are assuming that S ∈ D(TPG ). Thus KΘ being a positive kernel (i.e., Θ ∈ MP(G))

implies that KS is a positive kernel (i.e., that S ∈ P(G)).
Remark 1.7. Let us note that this derivation of the Potapov–Ginzburg transform
is very much in the spirit of the Grassmannian point of view toward the study of
linear-fractional transformations presented in Chapter 3 of [28].
1.3. The Stieltjes and multiplicative Stieltjes classes

An important subclass of the Pick class is the Stieltjes class denoted here by S(G),

consisting of functions S in the Pick class P(G) with analytic continuation across
the negative half-axis R− and taking positive semidefinite values on R− :
S(z) − S(z)∗
0 (z ∈ R), S(x) 0 (x < 0). (1.24)
z−z
Stieltjes functions made their first explicit appearance in [44] as continued fractions
of certain type and as Cauchy transforms of positive measures on R+ = [0, ∞).
Being special instances of absolutely monotone functions, operator monotone func-
tions and Pick functions, they have been extensively studied in various contexts
[12, 29, 30, 34, 33, 37, 43, 45]. Such functions have the alternative characterization
as being those functions S ∈ P(G) such that the function z → zS(z) is also in
(see [33] for the scalar case – the operator-valued case is similar). This leads
P(G)
to the kernel characterization of the Stieltjes class: an L(G)-valued function S is
in S(G) if and only if both kernels
S(z) − S(ω)∗ zS(z) − ωS(ω)∗
K(z, ω) = and K(z, ω) = (1.25)
z−ω z−ω
are positive on the upper half-plane.
To introduce the multiplicative version of the Stieltjes class, denoted as

MS(G), we use the Potapov–Ginzburg transform (1.22) to compute the expression
similar to that in (1.23) but with the different signature operator

0 IG
J= .
IG 0
It turns out that the operator [ CA B ] on G ⊕ G specified by (1.23) is not only
J 0 0 iIG D 0 I
J 0
0 −J , −iIG 0 -unitary (see (1.20)), but also 0 −J
, I , 0G -unitary,
G
i.e.,

A B J 0 A∗ C ∗ 0 IG
= ,
C D 0 −J B ∗ D ∗ IG 0
∗
A C∗ 0 IG A B J 0
= .
B ∗ D∗ IG 0 C D 0 −J
Therefore, if Θ = TPG [S], we have

∗
J 0 I
J − Θ(z)JΘ(ω) = I TPG [S](z)
0 −J TPG [S](ω)∗

A B J 0 A∗ C ∗ I
= (A + S(z)C) I S(z) (A∗ + C ∗ S(ω)∗ )
C D 0 −J B ∗ D∗ S(ω)∗

0 IG I
= (A + S(z)C) I S(z) (A∗ + C ∗ S(ω)∗ ). (1.26)
IG 0 S(ω)∗
and then evaluating this last expression at
Assuming that S belongs to S(G)
z = ω = x < 0 shows that
−1 −1
J − Θ(x)JΘ(x)∗ = 2 S(x)C + A S(x) C ∗ S(x)∗ + A∗ 0,
i.e., that the function Θ = TPG [S] is J-contractive on R− . We define the multi-
plicative Stieltjes class MS(G) to be the set of all functions Θ ∈ MS(G) that in
addition are J-contractive on R− (see [27]).
Remark 1.8. Besides characterizations (1.24) and (1.25), Stieltjes-class functions
are also characterized by their Herglotz integral representations (see [33])

dμ(t) dμ(t)
S(z) = γ + , where γ 0, < ∞. (1.27)
2
R+ t − z R+ 1 + t
From this representation it is clear that not only S(x) 0 for x < 0, but also

t − z
S(z) = γ + dμ(t) 0 whenever z < 0.
R+ |t − z|2
Combining this fact with (1.26) we conclude that the functions from the class
MS(G) are J-contractive on the left half-plane. It is tempting to conjecture that
the kernels
S(z) + S(ω)∗ J − Θ(z)JΘ(ω)∗
and
z+ω z+ω
are positive on the left half-plane for S ∈ S(G) and Θ ∈ SM(G), i.e., that the
matrices
N N
S(zi ) + S(zj )∗ J − Θ(zi )JΘ(zj )∗
,
zi + zj i,j=1 zi + zj
i,j=1
are positive semidefinite for all z1 , . . . , zN in the left half-plane (not just their
diagonal entries positive semidefinite). However this turns out not to be the case
in general.
we apply
To explore the alternative characterization (1.25) of the class S(G),
the Potapov–Ginzburg transform (1.22) to the function zS(z). Let P (z) be the

L(G)-valued function defined by

zIG 0
P (z) = (1.28)
0 IG
with respect to the decomposition (1.3). It is readily seen from (1.21) that
z −1 A = AP (z)−1 , C = CP (z)−1 , z −1 B = BP (z)−1 , D = DP (z)−1 .
Then
TPG [zS](z) = (A + zS(z)C)−1 (B + zS(z)D)
= (z −1 A + S(z)C)−1 (z −1 B + S(z)D)
= (AP (z)−1 + S(z)CP (z)−1 )−1 (BP (z)−1 + S(z)DP (z)−1 )
= P (z)(A + S(z)C)−1 (B + S(z)D)P (z)−1
= P (z)TPG [S](z)P (z)−1 .
We have already seen that Θ = TPG [S] is in the class MP(G) whenever S ∈ P(G).

If S belongs to the Stieltjes class S(G), then the preceding calculation shows that
the function
ΘP (z) := P (z)Θ(z)P (z)−1 (1.29)
also belongs to MP(G), that is, the kernel
J − ΘP (z)JΘP (ω)∗
KΘP (z, ω) =
i(ω − z)
J − P (z)Θ(z)P (z)−1 JP (ω)−1∗ Θ(ω)∗ P (ω)∗
= (1.30)
i(ω − z)
is positive on C+ . Observe that due to the obvious equalities
P (z)∗ JP (z) = P (z)JP (z)∗ = zJ,
it follows that the transformation Θ → P ΘP −1 preserves property (1.5), i.e., the

function (1.29) satisfies
ΘP (z)JΘP (z)∗ = ΘP (z)∗ JΘP (z) = J (z ∈ Ω). (1.31)
This leads to the alternative characterization of the class MS(G) as the set of
functions Θ that along with the associated function ΘP are multiplicative Pick
functions. We shall use this latter characterization as our formal definition of the
multiplicative Stieltjes class, a subclass of the multiplicative Pick class, as follows.
Definition 1.9. We let MS(G) denote the class of all functions in Θ ∈ MP(G)
such that the associated function ΘP defined in (1.29) also belongs to MP(G).
Equivalently, the class MS(G) consists of all functions Θ on C+ such that the two
kernels KΘ (1.4) and KΘP (1.30) are positive kernels on C+ .
1.4. Connections with interpolation theory
The importance of multiplicative Pick functions for interpolation theory arises
from the fact that the linear fractional map based on a function Θ ∈ MP(G)
maps the class P(G) into itself. Choosing Θ with a suitable pole/zero structure
then implies that the linear-fractional map based on Θ gives rise to a parametriza-
tion (with free parameter from the Pick class P(G)) of the solution set of a given
interpolation problem in the class P(G); we refer to [11, 42] for specific examples.
It turns out the multiplicative Stieltjes class MS(G, C) has similar applications
in interpolation theory for the additive Stieltjes class S(G) as the linear fractional
map based on a function Θ ∈ MS(G) not only maps the class P(G) into itself,
but also the class S(G) into itself. In the context of the Nevanlinna–Pick interpo-
lation problem, multiplicative Stieltjes functions appeared explicitly in the series
of papers [23, 25, 26]; see also [2, 13, 14, 15, 24, 25, 26] for other examples and
far-reaching generalizations. From the integral representation (1.27) for the Stielt-
jes class, we see that the Stieltjes moment problem going back to the nineteenth
century [44] can be seen as a boundary version of a Stieltjes interpolation problem.
The Stieltjes class also arises in the recent work of Agler–Tully-Doyle–Young [1]
on characterizing boundary directional derivatives of Schur-class functions on the
bidisk.
1.5. The focus here
However our focus here is not on interpolation aspects but rather on the intrinsic
structure of the associated reproducing kernel Hilbert spaces. The main objective
of the present paper is to find Stieltjes-class counterparts of Theorems 1.3 and 1.4.
Specifically, in Section 3 we shall consider the following:

Problem 1.10. Given two reproducing kernel Hilbert spaces H and H of G-valued
functions meromorphic in Ω, find necessary and sufficient conditions for the exis-
tence of a function Θ ∈ MS(G, Ω) such that H = H(Θ) and H = H(P ΘP −1 ). In
case H and H are presented as ranges of observability operators
H = Ran OΠ,A,μ and H = Ran OΠ, ,
A,μ
find necessary and sufficient conditions directly in terms of the operators Π, A, Π, A

for it to happen that H = H(Θ) and H = H(ΘP ) for some Θ.
Solutions to these problems are presented in Theorem 3.1 (the Stieltjes ana-
logue of Theorem 1.3) and Theorem 4.1 (the Stieltjes analogue of Theorem 1.4).
Finally we note that the reproducing kernel space H(Θ) determines the func-
tion Θ ∈ MP(G, Ω) only up to a unitary constant right factor Υ. While ΘΥ is
in the Pick class MP(G, Ω) whenever Θ ∈ MP(G, Ω) for any constant J-unitary
operator Υ, the corresponding property for the multiplicative Stieltjes class fails
in general. Thus it is a subtle but nontrivial point to show that, if Θ is such
that H = H(Θ) and H = H(ΘP ), then there is a choice of constant J-unitary
operators Υ and Υ so that (Θ · Υ)P = ΘP · Υ, in which case we then have
Θ := Θ · Υ ∈ MS(G, Ω) as well as H = H(Θ ) and H = H((Θ )P ). This is-
sue is addressed in Section 4.2 below.
The paper is organized as follows. Section 2 presents some material on the
simultaneous J-unitary equivalence of a pair of Kreı̆n-space operators as well as
some identities involving the operators Rα and R0α I+αR 0
α
needed in the proof
of the characterization of a pair of reproducing kernel Hilbert spaces of the form
H(Θ) and H(ΘP ). Section 3 gives an intrinsic structural characterization of pairs
of reproducing kernel Hilbert spaces of the form (H(Θ), H(ΘP )) in intrinsic geo-
metric, structural form, while in Section 4, these results are reformulated in explicit
state-space coordinates.
2. Preliminaries
In this section we present some auxiliary results. Recall that an operator M is
called J-unitary if M JM ∗ = J.
be such that
Lemma 2.1. Let operators A, B ∈ L(G)
AJA∗ = A∗ JA = BJB ∗ = B ∗ JB = 0, (2.1)
∗ ∗ ∗ ∗
B JA + A JB = AJB + BJA = J. (2.2)
Then there exist J-unitary operators M and N in L(G ⊕ G) such that

I 0 0 0
A=M G N and B = M N. (2.3)
0 0 0 IG
Proof. It follows from (2.1), (2.2) that the operator A + B is J-unitary:

(A + B)J(A + B)∗ = (A + B)∗ J(A + B) = J.
We conclude that both A + B and (A + B)∗ are surjective, and hence A + B is
invertible. We may then rewrite this last identity as
(A + B)∗−1 J(A + B)∗−1 = J = (A + B)∗−1 J(A + B)−1 .
Therefore, the operators

A = A(A + B)−1 and B = B(A + B)−1 (2.4)
enjoy all the properties (2.1) of A and B. In addition,
A + B = IG,
and we conclude from (2.1) and (2.2) that
AJ A∗ = A∗ J A = 0 and AJ + J A∗ = A∗ J + J A = J. (2.5)
Next, since A + B is J-unitary,
A = A(A + B)−1 = AJ(A∗ + B ∗ )J = AJB ∗ J
and, by (2.1), we have
A2 = AJB ∗ JAJB ∗ J = (J − BJA∗ )JAJB ∗ J = AJB ∗ J = A.
Therefore, A is a projection and
M = Ran A∗ = {x ∈ G : A∗ x = x}
Let us show next that
is a (closed) subspace of G.
Ker A = JM. (2.6)
∗
Indeed, if x ∈ M, then A x = x and by the second relation in (2.5),
AJx = (J − J A∗ )x = Jx − Jx = 0.
Thus, Jx ∈ Ker A. Let conversely, AJy = 0. Then (again by the second relation
in (2.5))
A∗ Jy = (J − J A)y = Jy
and thus, Jy ∈ M, which proves (2.6). Now it follows that
G = Ran A∗ ⊕ Ker A = M ⊕ JM = G ⊕ JG. (2.7)
Let U be a unitary transformation which maps G onto M. As every vector v ∈ G
can be represented as v = x + Jy for some x, y ∈ G, the transformation
U(x + Jy) = U x + JU y (x, y ∈ G)
defines a linear bounded operator which maps G onto itself. Moreover,
UG = M and UJ G = JM. (2.8)
Since U is unitary,
JUv, UvG = JU x + U y, U x + JU yG
= JU x, JU yG + U y, U xM
= U x, U yM + U y, U xM = x, yG + y, xG .
On the other hand,

Jv, vG = Jx + y, x + JyG
= Jx, JyG + y, xG = x, yG + y, xG .

Thus, for every vector v ∈ G,
JUv, UvG = Jv, vG
and therefore, U is J-unitary. In view of (2.6) and the second equality in (2.7),
the operator AU admits the representation

A1 0
AU = (2.9)
A2 0
Clearly,
with respect to decomposition (1.3) of G.
A1 = PG AU|G and A2 = PJG AU|G ,
where PG and PJG are the orthogonal projections onto G and JG, respectively. By
the first equality in (2.5), U∗ A∗ J AU = 0, which being combined with (1.2) and
(2.9) gives
∗ ∗
A1 A∗2 0 iIG A1 0 i(A1 A2 − A∗2 A1 ) 0
0= = ,
0 0 −iIG 0 A2 0 0 0
and we conclude that
A1∗ A2 = A∗2 A1 . (2.10)
∗
It follows from the second relation in (2.5) that A = IG − J A J and therefore,
that
AU = U − J A∗ JU. (2.11)
Let x be a vector in G. Then by (2.8), Ux ∈ M, whereas J A∗ JUx ∈ JM and
thus, by (2.7),
Ux, J A∗ JUx = 0.

G
Therefore, on account of (2.11), (2.8) and since U is unitary,
AUxG2 = Ux − J A∗ JUxG2
= UxG2 + J A∗ JUx2G ≥ Ux2G = U xM
2
= x2G .
Therefore, the operator
Δ := PG U∗ A∗ AU|G (2.12)
is strictly positive on G and has the operator norm at least one. Therefore Δ−1 is
a bounded operator on G. Comparing (2.12) and (2.9) we get
Δ = A1∗ A1 + A∗2 A2 . (2.13)
We claim that the operator

Δ−1 A∗1 Δ−1 A∗2
V= (2.14)
−A2∗ A1∗
is J-unitary. Indeed, by making use of (2.10) and the relation (2.13), we can check
that V is J-coisometric:
−1 ∗
∗ Δ 0 A1 A2∗ 0 I A1 −A2 Δ−1 0
VJV = i
0 I −A2∗ A∗1 −I 0 A2 A1 0 I
−1 ∗ ∗ ∗ ∗
−1
Δ 0 A1 A2 − A2 A1 A1 A1 + A2 A2 Δ 0
=i
0 I −A∗2 A2 − A∗1 A1 −A2∗ A1 + A1∗ A2 0 I
−1 −1
Δ 0 0 Δ Δ 0
=i
0 I −Δ 0 0 I

0 I
=i = J.
−I 0
To show that V is J-unitary given that V is J-coisometric, it suffices to show that
Ker V = {0}. Suppose therefore that V [ xy ] = [ 00 ]. Thus we have
∗
A1 A2∗ x 0
= . (2.15)
−A∗2 A∗1 y 0

Multiplying (2.15) on the left by A02 A01 then gives
A2 A∗1 x + A2 A2∗ y = 0,
−A1 A∗2 x + A1 A∗1 y = 0.
Adding these equations while recalling (2.10) then gives
Δy = (A2 A2∗ + A1 A1∗ )y = 0.
As Δ is invertible,
we conclude that y = 0. Similarly, multiplying (2.15) on the
left by A01 A02 leads to
A1 A1∗ x + A1 A2∗ y = 0,
−A2 A∗2 + A2 A1∗ y = 0.
Subtracting the second equation from the first then leads to
Δx = (A1 A1∗ + A2 A2∗ )x = 0
which then forces x = 0 as well. We conclude that V is J-unitary as claimed.
We note next that, on account of (2.9) and (2.14),

I 0
VAU = G . (2.16)
0 0
Let
α β
VBU =: B = . (2.17)
γ δ
Since the operators U, V and A + B are J-unitary, it follows from (2.1) that

∗ IG 0 ∗ IG 0
BJ B = 0 and J B + BJ = J. (2.18)
0 0 0 0
into the
By substituting the block decompositions (1.2) and (2.17) of J and B
second relation in (2.18), we see that
β = β∗, δ = IG .
This combined with the same substitution plugged into the first relation in (2.18)
then leads us to
γ = γ ∗ , α = βγ.
Thus,
= βγ β , γ = γ ∗ , β = β ∗
B (2.19)
γ IG

and conversely, any B = α β of the form (2.19) satisfies the relations (2.18).
γ δ
Since the operators β and γ are selfadjoint, the operators

I β I 0
C= G and D = G
0 IG γ IG
are J-unitary. Moreover,

IG 0 I 0 =C 0 0
=C G D and B D. (2.20)
0 0 0 0 0 IG
If we set
M = V−1 C and N = DU−1 (A + B),
then M as well as N is the product of J-unitary operators and hence are themselves
also J-unitary. By plugging in the definitions of M and N and then making use
of (2.20), (2.16), (2.17) and (2.4), we calculate

I 0 I 0
M G N = V−1 C G DU−1 (A + B)
0 0 0 0

I 0
= V−1 G U−1 (A + B) = A(A + B) = A
0 0
and
0 0 −1 0 0
M N =V C DU−1 (A + B)
0 IG 0 IG
−1 (A + B) = B(A + B) = B,
= V−1 BU
which completes the proof.
Remark 2.2. The J-unitary operators M and N in (2.3) are defined uniquely up
to, respectively, a right and left J-unitary factors of the block diagonal form. More
precisely, if
IG 0 0 0
A = M1 N and B = M1 N1 (2.21)
0 0 1 0 IG
is an another representation of A and B with J-unitary operators M1 and N1 ,
then
M1 = M T and N1 = T −1 N,
is of the form
where T ∈ L(G)

τ 0
T = , τ ∈ L(G). (2.22)
0 (τ −1 )∗
Proof. Comparing representations (2.3) and (2.21) and setting
T = M −1 M1 and T1 = N N1−1 , (2.23)
one can easily get

IG 0 I 0 0 0 0 0
T = G T and T = T . (2.24)
0 0 0 0 1 0 IG 0 IG 1
Summing up the two last equalities
we conclude
that T = T1 and that T commutes
therefore, with projections I0G 00 and 00 I0G . Therefore, T is of the block diagonal
form
τ 0
T = . (2.25)
0 τ1
Since M and M1 are J-unitary, T is also J-unitary and the blockwise comparison
in the equality

τ 0 0 iIG τ ∗ 0 0 iIG
=
0 τ1 −iIG 0 0 τ1∗ −iIG 0
leads to τ1 = (τ −1 )∗ . The latter equality together with (2.23) and (2.25) implies
the desired conclusion.
Corollary 2.3. Let
Φ(z) = Az + B (2.26)

be a linear L(G)-valued function such that
Φ(z)JΦ(z)∗ = Φ(z)∗ JΦ(z) = zJ. (2.27)
such that
Then there exist J-unitary operators M and N in L(G)
Φ(z) = M P (z)N, (2.28)
where P (z) is given by (1.28).
Proof. Upon substituting (2.26) into (2.27) we conclude that the operators A and
B satisfy relations (2.1) and (2.2). By Lemma 2.1, A and B admit representa-
tions (2.3) for some choice of J-unitary operators M and N . Substituting these
representations into (2.26), we get (2.28).
Lemma 2.4. Let Ω be a non-empty open set which is symmetric with respect to R,

let Θ be a L(G)-valued function which is meromorphic on Ω, let KΘ and KΘP be
the kernels defined in (1.4) and (1.30). Then
(ω − α)P (z)KΘ (z, ω) − (z − α)KΘP (z, ω)JP (ω)J
= −iP (α)J + iΘP (z)P (α)JΘ(ω)∗ , (2.29)

I + αRα 0 KΘP (· , ω)JP (ω)J − KΘP (· , α)JΘP (α)P (α)JΘ(ω)∗
KΘ (· , ω) = ,
0 Rα ω−α
(2.30)
∗
Rα 0 KΘ (· , ω)P (ω) − KΘ (· , α)JΘ(α)JP (α)ΘP (ω)
K (· , ω) = ,
0 I + αRα ΘP ω−α
(2.31)
KΘ (· , ω) − KΘ (· , α)JΘ(α)JΘ(ω)∗
Rα KΘ (· , ω) = , (2.32)
ω−α
KΘP (· , ω) − KΘP (· , α)JΘP (α)JΘP (ω)∗
Rα KΘP (· , ω) = (2.33)
ω−α
for any choice of points ω and α in Ω.
Proof. We tackle one formula at a time:
Proof of (2.29). From the definition of P (z) (1.28) it is easily verified that
(ω − α)P (z) − (z − α)P (ω) = (ω − z)P (α). (2.34)
The definition of ΘP (z) (1.28) together with associativity gives
P (z)Θ(z) = ΘP (z)P (z). (2.35)
A straightforward computation gives the identity
JP (ω)JP (ω) = ωIG
and hence
JP (ω)JP (ω) · Θ(ω) = Θ(ω) · JP (ω)JP (ω).
Taking adjoints gives us
Θ(ω)∗ P (ω)JP (ω)J = P (ω)JP (ω)JΘ(ω)∗
which we prefer to write as
ΘP (ω)∗ JP (ω)J = JP (ω)JΘ(ω)∗ . (2.36)
Therefore, making use of (2.34), (2.35), (2.36) gives us
(ω − α)P (z)Θ(z)JΘ(ω)∗ − (z − α)ΘP (z)JΘP (ω)∗ JP (ω)J
= (ω − α)ΘP (z)P (z)JΘ(ω)∗ − (z − α)ΘP (z)P (ω)JΘ(ω)∗
= ΘP (z) {(ω − α)P (z) − (z − α)P (ω)} JΘ(ω)∗
= (ω − z)ΘP (z)P (α)JΘ(ω)∗ .
Multiply the equality (2.34) by J on the right and then subtract from this the last
equality above; the result is
(ω − α)P (z)(J − Θ(z)JΘ(ω)∗ ) − (z − α)(J − ΘP (z)JΘP (ω)∗ )JP (ω)J
= (ω − z)(P (α)J − ΘP (z)P (α)JΘ(ω)∗ ).
Dividing this last equality by i(ω − z) and recalling the definitions (1.4) and (1.30)
of KΘ and KΘP then leads us finally to the identity (2.29).
Proof of (2.30). From (2.29) we see that

P (z)KΘ (z, ω) =
1
(z − α)KΘP (z, ω)JP (ω)J − iP (α)J + iΘP (z)P (α)JΘ(ω)∗ (2.37)
ω−α
which when evaluated at z = α gives
1
P (α)KΘ (α, ω) = {−iP (α)J + iΘP (α)P (α)JΘ(ω)∗ } . (2.38)
ω−α
Note that in general
f (z) − f (α) zf (z) − αf (α)
((I + αRα )f )(z) = f (z) + α = = (Rα g)(z)
z−α z−α
where g(z) = zf (z). More generally, for P (z) as in (1.28) we arrive at the general
identity
I + αRα 0
F = Rα (P · F ). (2.39)
0 Rα
Applying the identity (2.39) to the case where F = KΘ (·, ω) leaves us with

I + αRα 0 P (z)KΘ (z, ω) − P (α)KΘ (α, ω)
KΘ (· , ω) = .
0 Rα z−α
Plugging (2.37) and (2.38) into this last identity gives us
Θ (z)−Θ (α)
KΘP (z, ω)JP (ω)J + P i(α−z)P P (α)JΘ(ω)∗
I + αRα 0
KΘ (· , ω) = .
0 Rα ω−α
To prove (2.30), it remains only to note that
ΘP (z) − ΘP (α) Θ (z)JΘP (α)∗ − J
= P JΘP (α) = −KΘP (z, α)JΘP (α). (2.40)
i(α − z) i(α − z)
Proof of (2.31). We start with the equivalent version of (2.29) obtained by taking
adjoints followed by replacing α by α and switching z and ω:
(z − α)KΘ (z, ω)P (ω) − (ω − α)JP (z)JKΘP (z, ω) = iJP (α) − iΘ(z)JP (α)ΘP (ω)∗ .
Therefore,
1
JP (z)JKΘP (z, ω) = {(z − α)KΘ (z, ω)P (ω) − iJP (α)
ω−α
+ iΘ(z)JP (α)ΘP (ω)∗ } , (2.41)
and in particular,
1
JP (α)JKΘP (α, ω) = {−iJP (α) + iΘ(α)JP (α)ΘP (ω)∗ } . (2.42)
ω−α
Note that the identity
I 0
JP (z)J = G (2.43)
0 zIG
Another random document with
no related content on Scribd:
CHAPTER XXVIII
BETTY HAS HER OWN WAY
Don Black had not himself built the cabin where he lived. While he
was still a boy jingling his first spurs, two young Englishmen had
hewn its logs out of the untouched forest on the western slope of
Pegleg Pass. They were remittance men, exiled from their country
for the peace of mind of their families. In the casual fashion of their
class they had drifted to the Rockies to hunt for big game and, less
industriously, for elusive fortune. Long since they had returned to the
estates which Britishers of this type seem always to be inheriting
from convenient relatives.
By the simple process of moving in, Black had become owner of the
cabin. He hung his pinched-in cowboy hat on a peg in the wall and
thereby took possession. His title was perfectly good in the eyes of
the range riders who dropped in occasionally and made themselves
at home. Whether Don was or was not on the place, they were
welcome to what they found. The only obligation on them was to cut
a fresh supply of firewood in place of that they used.
One room was enough for Black’s needs. The other served as a
place in which to store old saddles, mountain-lion pelts, worn-out
boots, blankets, unused furniture, and a hundred odds and ends.
With the help of the owner, Lon Forbes set to work housecleaning.
Useless litter went flying out of doors. A vigorous broom in the hands
of Lon raised clouds of dust. In the fireplace old papers and boxes
blazed cheerfully. A Navajo rug, resurrected from the bottom of a
hingeless trunk, covered the floor in front of a walnut bed imported
by one of the Englishmen from Denver.
It took hours to make the transformation, but the foreman was quite
pleased with himself when he ushered Betty into the bedroom he
had prepared for her.
She clapped her hands softly. “My, Lon! What a fine wife some
Suffragette’s lost in you. Maybe it isn’t too late yet. You can keep
house while she—”
“Help! Help!” expostulated Forbes.
“Oh, if you’ve got your eye on one of these little flapper girls, of
course, there’s no use my saying a word,” she teased. “I know how
stubborn you are when you get ‘sot.’”
She was in a mood of happy reaction from the fears that had
oppressed her all day. Dr. Rayburn had told her—with some
reservations, to be sure—that, barring unexpected complications,
Hollister ought to get well. It would take time and nursing and good
food, but all of these the patient would get.
“You’re right I’ve got my eye on one of them li’l’ flapper girls—this
very minute,” he rapped back promptly. “An’ she’s a sure-enough
warnin’ to a fellow to play his hand out alone unless he wants to be
bossed somethin’ scandalous.”
“It would do you good to be bossed,” she told him, eyes dancing.
“The refining influence of a young woman—say about forty-five or
maybe fifty—”
“You’re pickin’ her for me, are you?” he snorted.
“She’ll do the picking when the time comes. I suppose you’ll have to
give up smoking—and you’ll have to shave every day—and probably
be a deacon in the church at Wild Horse—”
“Yes, I will not. All I got to do is look at Clint an’ see how a half-grown
kid has got a check rein on him. That scares me a plenty.” He shook
his head in mock despair, but his eyes gave him away. “Gallivantin’
into the hills, through ’steen million tons of snow, to nurse a
scalawag who—”
“He’s no scalawag, Lon Forbes.”
“Like to know why he ain’t. Nothin’ but a hobo when you first met up
with him.”
“Now, Lon, you know very well you told me you thought he was a
man from the ground up. Those were the very words you used.”
“Well, a hobo may be a man,” he defended. “Anyways, that don’t
mean you’d ought to bust up yore happy home to hike over the hills
for him.”
“Justin’s been talking to you,” she charged.
“Maybeso, an’ maybe not. That ain’t the point. While Clint’s away, it’s
up to me to run the Diamond Bar K.”
“With Justin’s help,” she cut in.
Betty thought, though she did not express it in words, that Lon would
have his hands full if he intended to take charge of her activities as a
part of the ranch. She knew that this would never have occurred to
him as included in his duties if it had not been suggested by Merrick.
“I’m not askin’ any one’s help. I reckon I’m as grown-up as I’ll ever
be. Anyways, Clint put me in charge, figurin’ I was man-size an’
competent. Question is, Would yore father want you up here?”
Betty decided to carry the war indignantly into the territory of the
enemy. “Of course, he would. After knowing Dad all these years I
should think you’d be ashamed to doubt him. Dad pays his debts.
He’s a good friend. This boy—this young fellow Hollister—tried to do
us a good turn after we had behaved pretty bad to him. You know
Dad has been looking for a chance to help him. Well, it’s come. What
are we going to do about it? Go through—or quit on the job?”
“Go through. I ain’t proposin’ anything else. But you don’t have to
stay here. I can look after him, an’ Merrick’ll see you home.”
“What do you know about nursing?” she scoffed. “Or cooking? You
know what the doctor said. He’s got to have nice things to eat after
he gets a little better. And good nursing. Dr. Rayburn told you—I
heard him say it—that he was glad I’d come because Mr. Hollister
needs a woman’s nursing.”
Lon scratched his head to help him think. It was sometimes a
laborious process. He knew cattle and crops, but chaperoning a
young woman was untried territory.
“Times has changed, Betty,” he explained. “You kinda growed up
helter-skelter an’ run wild. But you’re a young lady now, an’ you can’t
be too careful. You gotta think about what folks’ll say.”
“Fiddlesticks! What’ll they say? What can they say if you stay up
here with me? It’ll be only a day or two till Dad gets home. It’s just
that you’ve been getting notions from Justin. He’s a city man and
doesn’t know our ways. But you’ve always lived here, Lon. I’m
surprised at you.”
“O’ course there ain’t any real harm in yore stayin,” he conceded
hesitantly. “I’ll be here to look after you an’ see Prowers don’t trouble
you. An’ it won’t be long.”
“I’m staying because I really can help, Lon. Justin thinks it’s only
foolishness, but you know it isn’t. In Denver, where he lives, there
are plenty of trained nurses, but it’s different here. If Bridget could
get in, I wouldn’t say a word about staying. But she can’t. If I went
away and left this poor boy, you’d never respect me again.”
“I would, too. But there. You’re gonna stay. I see that.”
“Yes, I am.” She caught the lapels of the big foreman’s coat and
coaxed him with the smile that always had proved effective with him.
“And you know I’m right. Don’t you, Lon?”
“Nothin’ of the kind,” he blustered. “An’ you needn’t try to come it
over me. I know you too blamed well, miss. You’re bound an’
determined to have yore own way—always were since you were a
li’l’ trick knee-high to a duck. Trouble is, you’ve been spoiled.”
“Yes,” she admitted, “and you did it.”
“No such a thing. I always did tell Clint he’d find out some day what’d
come of lettin’ you boss the whole works.” To save his face he
finished with a peremptory order. “Don’s fixin’ up some supper. Soon
as you’ve had yours, why, you’ll go right straight to bed. Doc an’ me
are aimin’ to look after Hollister to-night.”
“Yes, Lon,” Betty replied meekly. She had got what she wanted, and
she was willing to propitiate him by a demure obedience calculated
to remove the sting of her victory.
Don opened the door and announced that supper was ready.
Betty saw Merrick’s eye flash a question at Forbes as they came into
the larger room. She went directly to him. Betty was a woman;
therefore complex. But she usually expressed herself simply.
“It’s all settled, Justin. Lon is going to stay with me.”
He made no answer in words, but his salient jaw set grimly. Like
many masterful men, he did not relish defeat.
They drank coffee from tin cups and ate bacon, tomatoes, and beans
served in tin plates. Don’s biscuits were appetizing, and four or five
pans of them disappeared before his guests were fed.
Betty lived up to the promise she had made Lon. She whispered with
Dr. Rayburn for a minute, then said “Good-night” to the company
generally, and vanished into her bedroom.
The day had been a full one. To come in over the snow had taxed
the strength of her muscles. She was tired, and, as she sat before
the glowing coals taking the pins out of her hair, she yawned
luxuriously.
Just now her mind was on Merrick. The vague disappointment in
their relationship had crystallized to-day into definite dissatisfaction.
To use one of her father’s expressions, Justin and she had not come
out of the same pasture. They thought in different languages.
That he had not sympathized with the urge in her to spend herself in
service for the wounded man was important beyond the immediate
question. And, even if he did not agree with her, he should have
understood her obligation to do as she thought best. It involved their
whole future. The trouble was that he did not recognize her right to
follow the guidance of her own judgment. She must defer to him,
must accept his decision as final.
Betty knew she could not do that. In essence she was a twentieth-
century woman.
CHAPTER XXIX
A CHILD OF IMPULSE
Betty went to sleep critical of Justin. She woke, in the dawn of a

new day streaming through the window, to censure of her own
conduct. Willful though the girl was, she had a capacity for
generosity that saved her from selfishness.
It was just as Lon said, her thoughts ran. She had to boss everybody
and everything, always had to have her own way without regard to
others. No wonder Justin did not like it. If she had tried hard enough,
she could have made him see that this adventure was a duty laid on
her, one she could not escape and retain her self-respect. Instead,
she had managed so badly that she had thrown him quite out of
sympathy with her point of view.
A child of impulse, she decided swiftly as she dressed to have a little
talk with him and say she was sorry. With this resolve came peace.
Everything would be all right now between them.
Hollister smiled when she came to his bedside and asked him how
he was. His face reassured her. It was very pale, but it held the look
of one who means to get well. Dr. Rayburn backed its promise.
“He’s doing fine. Fever gone down a lot. Nursing’s the thing now,
Miss Betty. You can do more for him than I can.”
“Are you going back to town to-day?” she asked.
“Got to. No two ways about that. Be back day after to-morrow
probably. Keep giving him the tablets. Every two hours. And a
teaspoon of the liquid three times a day.”
They had drawn away from the bedside and by mutual consent
passed out of the door into the sunshine. The crisp morning air was
delightful. A million glints of light sparkled from the snow.
“He’s really better, isn’t he?” she asked eagerly, and her voice
throbbed with young life.
“Better, yes. But—sometimes a man seems definitely to be on the
mend and then he relapses without any apparent cause. It’s too
soon to say he’s getting better. All I can say is that, if no unfavorable
complications set in, he ought to improve.”
“Ought you to leave him?”
He threw up his hands in an energy of exasperation. “If you had half
as much to do as I have, young lady—”
“I know, but if he’s really still in danger—”
“Danger!” fumed the doctor. “Do you think Mrs. Pillsbury can wait for
him to get out of danger?”
“I didn’t know—”
“Babies are born when they’re born,” he sputtered. “I’ve got to leave
for town right after breakfast.”
Justin came round the corner of the house. Betty almost ran to give
him her hand. Her eyes were shining wells of friendliness.
“I want to see you after breakfast,” she whispered.
He nodded, non-committally.
Black called from inside, “Yore coffee’s b’ilin’, folks.”
He gave them flapjacks and syrup.
“I love flapjacks,” Betty told him.
Their host said nothing, but he was pleased.
Lon came in late and drew up a chair beside Betty. “How’s
everything this glad mo’ning?” he asked.
“Fine as the wheat.” She added as an aside, “And the bossy little
flapper isn’t half so bossy as she sometimes lets on.”
His grin met her smile. They understood each other very well and
were still friends. Betty pushed into the back of her mind a fugitive
wish that Justin could know and appreciate her as well as good old
Lon did.
After breakfast Betty and Merrick took a short walk.
“Scrumptious day,” she commented. Then, as though it were a
continuation of the same thought: “I’m sorry, Justin.”
“You mean—?”
“I’m kinda horrid sometimes. I flare out and say ‘I will’ or ‘I won’t’ like
a spoiled kid. That’s no way to do.” She smiled at him, a little
whimsically, a little apologetically. “It keeps me busy eating humble
pie.”
He accepted her apology graciously. “Shall we forget it, Bess? It’s a
new day. We’ll turn a page of the ledger and begin again.”
Rather timidly, she went on: “I had to come. It’s not that. But if I
hadn’t been so tempery, I could have made you understand.”
He stiffened at once. “I think I understood—perfectly.”
“No, Justin. That’s just it. You didn’t, or you wouldn’t have stood in
my way. You’re fair-minded, and when you see I was doing what I
had to do—what it was my duty to do—”
“I can’t agree with you about that, Betty. I’m older than you are. I
think I know more of the world. It’s not your duty—the duty of any
unmarried girl for that matter, unless she is a trained professional
nurse—to put herself in the position you have.”
In spite of her good resolutions Betty began to feel her temper slip.
“What position have I put myself in?” she asked quietly.
“I’m an old-fashioned man,” he answered. “I believe that a young
woman must be so circumspect that nobody can find any ground to
talk about her.”
“A girl isn’t a china doll. She can’t be put away in moth balls, Justin.
Every girl is talked about some time or other by somebody if she’s
alive. It’s of no importance what gossips say.”
“It’s of the greatest importance that a girl give no chance for idle
gossip about her,” he demurred.
Betty’s irritation expressed itself in the voice, a trifle sharp. “How do
you think I can run the Quarter Circle D E without being criticized?
I’m there with the men hours and days at a time, and no other
woman on the place except old Mandy, who is deaf as a post and
can’t see six feet from her nose. If any evil-minded person wants to
talk—why, I’ll just have to let him talk.”
“On the contrary, I think you ought to have a foreman run the place
for you except for some general supervision. It’s not a girl’s
business.”
“Isn’t it? You never told me so before.”
“You never asked me.”
“For that matter, I’m not asking you now.” Her manner was
dangerously quiet. It suggested banked fires of anger. “But just the
same I’m glad to have your opinion.”
“I’m glad to give it. I’ve wanted to tell you what I think about it.
Understand me. I admire your energy, your enthusiasm, your
efficiency. I believe you are running the Quarter Circle D E better
than a good many men could do it. That’s not the question. Aren’t
you losing something you can’t afford to do without? I can’t go into
this in detail. Cattle-raising—ranching—breeding Herefords—it’s a
splendid occupation for a man. But there’s a side of it that’s—well, I’d
rather you’d turn it all over to Forbes.”
“What do you want me to do—stay at home and knit?”
“You know what I want as soon as the Sweetwater project is
finished.”
Betty side-stepped the proposed excursion into sentiment. She was
a downright young woman and wanted to know exactly where she
stood.
“I didn’t know you felt that way about the ranch, Justin. I thought you
shared my view, that I was doing something worth while when I
raised hundreds of cattle every year to help feed the world. If I had
known you thought I was degrading myself—” She stopped, a
tremolo of anger in her throat.
“I didn’t say that, Betty.”
“It’s what you meant.”
“No. No, it isn’t. I meant only that—well, there’s something very very
precious that some girls have—that you have, Betty—something
that’s like the bloom of a peach. If you lose it—well, it’s gone, that’s
all.”
“And if I do anything that’s worth while—if I pay my way in the world
by giving value received—the peach bloom is rubbed off, isn’t it?”
she retorted scornfully.
“Aren’t there different ways of giving service? We are in danger of
forgetting the home, which is the normal place for a young girl.”
“Is it? Thought you came from a city where thousands of girls go
down to offices and stores every morning to earn a living.”
They stood on a small hilltop and looked over a world blanketed in
white which flashed back countless gleams of light to the
heliographing sun, a world so virgin clean, so still and empty of life,
that it carried Betty back to the birthday of the race. She was,
miraculously, at the beginning of things again.
“You’re not in a city fortunately,” he answered. “There’s no economic
pressure on you to fight sordidly for a living.”
Her eyes sparkled. “You’re not consistent. When the city ways don’t
suit you, I’m to live like people in the country, but when you don’t
approve of ranch ways, then I’m to be like girls in Denver. I’m not to
go into business, but I’m not to be neighborly as my mother was.”
“You’re distorting what I said, Betty.”
“Am I? Didn’t you say I wasn’t to help take care of a sick man
because it wasn’t proper?”
“I said you were acting rather absurdly about this man Hollister,” he
replied tartly. “There’s no call to turn the world upside down because
he’s wounded. You want a sense of proportion.”
“I think that’s what you need, Justin,” she answered, a flush of anger
burning her cheeks. “You’ve been horrid about it from the start
without any reason.”
She moved down the hill toward the cabin. Merrick walked beside
her. His eyes were hard and his lips set close.
For the first time it dawned upon Betty that he was jealous of her
interest in another man. He was possessive, wanted to absorb all
her thoughts, intended to be the center of every activity she had.
This did not please her. It alarmed the individual in her. Marriage, as
she had dreamed it, was wonderful because it enhanced life. It was
the union of two souls, releasing all the better forces of their natures.
Through it would come freedom and not bondage. The joys of the
senses would be shared and transmuted to spiritual power. They
ought not to put chains on a man or a woman that would narrow the
horizon.
An illusion had been shattered. Justin was not the man with whom
she could walk hand in hand. She sighed, and drew the gauntlet
from her left hand.
Merrick looked at the ring she had dropped into his hand, then
straight at her with rigid gaze.
“Are you in love with this fellow Hollister? Is that what it means?” he
asked harshly.
The color in her cheeks deepened. “That’s—hateful of you, Justin,”
she said, her voice ragged with feeling.
“I’ve seen it for some time. You’re infatuated with him.”
She lifted her chin and looked at him with eyes that blazed anger.
“Now I know I’ve done right in giving you back your ring. I’m not
going to—to quarrel with you because you insult me. It’s finished.
That’s enough.”
A sob rose to her throat and choked her. She hurried on to escape
him, the trail a blurred mist through her tears.
CHAPTER XXX
FATHOMS DEEP
The days followed each other, clear, sparkling, crisp, with mornings
in which Betty’s lungs drew in a winey exhilaration of living, with
evenings which shut the cabin on the slope of Pegleg Pass from a
remote world of men and women engaged in a thousand activities.
Betty had time to think during the long winter nights after she had
retired to her room. Some of her thoughts hurt. She was shocked at
the termination of her engagement, at the manner of it. That was not
the way it should have been done at all. She and Justin should have
recognized frankly that their views of life could not be made to
harmonize. They should have parted with esteem and friendship.
Instead of which there had been a scene of which she was
ashamed.
Her cheeks burned when she recalled his crass charge that she was
infatuated with Hollister. Why hadn’t he been able to understand that
she had signed a pact of friendship with the ex-service man? If he
had done that, if he had been wise and generous and sympathetic
instead of harsh and grudging, he would (so Betty persuaded
herself) have won her heart completely. He had been given a great
chance, and he had not been worthy of it.
Merrick had humiliated her, shattered for the time at least the gallant
young egoism which made her the mistress of her world.
Her father came up as soon as he returned from Denver. She talked
over with him the break with her fiancé. Clint supported her, with
reservations that did not reach the surface.
“Never did like it,” he said bluntly, referring to her engagement.
“Merrick’s a good man in his way, but not the one for you. I been
figuring you’d see it. I’m glad this came up. His ideas about marriage
are crusted. He’d put a wife in a cage and treat her well. That
wouldn’t suit you, Bess. You’ve got to have room to try your wings.”
She clung to him, crying a little. “You don’t blame me, then, Dad?”
“Not a bit. You did right. If Merrick had been the proper man for you,
he’d have understood you well enough to know you had to come
here. Maybe it wasn’t wise to come. Maybe it was impulsive. I reckon
most folks would agree with him about that. But he’d have known his
Betty. He’d ’a’ helped you, even though it was foolish. You wouldn’t
be happy with any man who couldn’t let you fly the coop once in a
while.”
“Was it foolish to come, Dad?” she asked.
He stroked her dark hair gently. “It’s the foolishness we all love in
you, honey—that way you have of giving till it hurts.”
Betty had inherited her impulsiveness from him. He, too, could be
generous without counting the cost. He rejoiced in the eagerness
with which her spirit went out to offer the gift of herself. But he had to
be both father and mother. Generosity might easily carry her too far.
“I do such crazy things,” she murmured. “And I never know they’re
silly till afterward.”
“This wasn’t silly,” he reassured her. “I’d have figured out some other
way if I’d been home. But I wasn’t. Rayburn says your cooking an’
your nursing have helped young Hollister a lot. I’m glad you came,
now it’s over with. I reckon you’ve paid my debt in full, Bettykins.”
“He’s absurdly grateful,” she said. “I haven’t done much for him.
You’d think I’d saved his life.”
“Soon now we’ll be able to get him back to the ranch and Bridget can
take care of him. Ruth’s wearyin’ for you. I’ll be more satisfied when
we’re there. I’ve got old Jake Prowers on my mind some. Never can
tell what he’ll be up to.”
Hollister was grateful to Betty, whether absurdly so or not is a matter
of definitions. His big eyes followed her about the room as she
cooked custards with the eggs and milk brought from the Howard
place just below. “Sweet Marie” did not entrance him when Black
tunelessly sang it, but the snatches of song she hummed at her work
filled the room with melody for him. She read “David Copperfield”
aloud after he began to mend, and his gaze rested on her with the
mute admiration sick men are likely to give charming nurses
overflowing with good-will and vitality. Her laugh lifted like a lark’s
song. Even her smile had the radiant quality of one who is hearing
good news.
He noticed that she was no longer wearing Merrick’s ring, and his
thoughts dwelt on it a good deal. Was the engagement broken? He
could not see that she was unhappy. Her presence filled the place
with sunshine. It was a joy to lie there and know that she was near,
even when she was in another room and he could not see her. There
was something permeating about Betty Reed. She lit up men’s souls
as an arc-light does a dark street.
He hoped that she and Merrick had come to the parting of the ways.
The engineer was the last man in the world to make her eager spirit
happy. His strength never spent itself in rebellion. He followed
convention and would look for the acceptance of it in her. But Betty
was cast in another mould. What was important to him did not touch
her at all, or, if it did, seemed a worthless sham. She laughed at
social usage when it became mere formalism. No doubt she would
be a disturbing wife. Life with her would be exciting. That was not
what Justin Merrick wanted.
The right man for her would be one who both loved and understood.
He must be big enough to let her enthusiasms sweep over their lives
and must even give them moral support while they lasted. Also, he
must be a clean and stalwart outdoor man, not one who had been
salvaged from the yellow swamp waters of vice. This last Hollister
kept before him as a fundamental necessity. He laid hold of it to
stamp down the passionate insurgent longings that filled him.
It was an obligation on him. He must not abuse her kindness by
forgetting that he had been an outcast, had himself shut a door upon
any future that included the fine purity of her youth. An effect of her
simplicity was that he stood in constant danger of not remembering
this. There was nothing of the Lady Bountiful about Betty. Her star-
clear eyes, the song and sunlight of her being, offered friendship and
camaraderie with no assumption of superior virtue. She saw no
barrier between them. They came together on an equal footing as
comrades. The girl’s unconscious generosity enhanced her charm
and made the struggle in his heart more difficult.
Those days while he lay there and gathered strength were red-letter
ones in his life. Given the conditions, it was inevitable that he should
come to care for the gracious spirit dwelling in a form that expressed
so lovingly the mystery of maiden dreams. In every fiber of him he
cherished her loveliness and pulsed to the enticement of her.
She gave the dull cabin atmosphere. A light burned inside her that
was warm and bright and colorful. Black looked on her as he might a
creature from another world. This slip of a girl had brought
something new into the range rider’s life, something fine and spiritual
which evoked response from his long-dormant soul. He had till now
missed the joy of being teased by a girl as innocent and as vivid as
she.
Hollister was won the easier because her tenderness was for him.
Black must hunt ptarmigan for broth, Clint Reed go foraging for milk
and eggs. They submitted cheerfully to be bullied in the interest of
the patient. His needs ruled the household, since he was an invalid.
Betty pampered and petted and poked fun at him, all with a zeal that
captivated his imagination.
In the evenings they talked, three of them in a semi-circle before the
blazing logs, the fourth sitting up in the bed propped by pillows. The
talk ranged far, from cattle to Château-Thierry. It brought to the sick
man a new sense of the values of life. These people lived far from
the swift currents of urban rush and haste, but he found in them
something the world has lost, the serenity and poise that come from
the former standards of judgment. The feverish glitter of post-war
excitement, its unrest and dissatisfaction, had left them untouched.
Betty and her father were somehow anchored to realities. They did
not crave wealth. They had within themselves sources of
entertainment. The simple things of life gave them pleasure. He
realized that there must be millions of such people in the country,
and that through them it would eventually be saved from the effects
of its restlessness.
CHAPTER XXXI
BETTY MAKES A DISCOVERY
“To-morrow,” Betty said, and did a little skip-step across the floor to
put away the frying-pan she had been washing.
“You have dancing feet,” Hollister told her.
“When I have a dancing heart.”
To the man sitting before the fire she bloomed in that dark cabin like
a poppy in the desert. She was a hundred miracles each hour to him.
He saw the exquisite mystery of her personality express itself in all
she was and did—in the faint crimson just now streaming through
her cheeks beneath the warm and tawny skin, in the charmingly shy
gesture with which she had accepted his compliment, in the low,
vibrant voice that played so wonderfully on his heartstrings. Not often
is a sweet and singing soul clothed so exquisitely in a body of grace
so young and lissom and vital.
“And that’s whenever there’s an excuse for it,” he said, smiling at her.
“But why feature to-morrow? Is it your birthday?”
“We’re going home to-morrow.”
“Are you? I didn’t know.”
He fell silent, looking into the fire. It was not an unexpected
announcement. These good days could not go on forever. She had
done more for him than any other friend he had ever had. But, of
course, life made its claims on her. She had to respond to them. It
was a wholly undeserved happiness that she had stayed till he was
out of danger and on the road to health. He wanted to tell her how he
felt about it, but he would never be able to do that. Inside, he
seemed to melt to a river of tears whenever he let himself dwell on
her amazing goodness to one who had been dead when first she
gave him her little hand and now was alive again.
“Umpha, to-morrow. I’m crazy to see Ruthie, and what the boys are
doing on the ranch.”
“I’ve been an awful nuisance,” he admitted.
“Haven’t you?” The little laugh that welled out of her was sweet and
mocking. It enveloped him with her gracious and tender young
womanhood.
He liked to think that her nursing had pulled him through just as
earlier her faith had rekindled in him self-respect and courage. The
facts might not quite justify this, but he did not intend to let brutal
actualities murder a beautiful dream.
“But you can always point with pride, as the politicians say. I’m a
credit to your nursing. Off your hands in ten days.”
“You’re not off my hands yet. You’re going to the Diamond Bar K with
us.”
The blood drummed faster through his heart. He felt a stinging of the
senses. Was there no end to the goodness of this astonishing and
disturbing girl? Must she always be flinging out life lines to him?
“Good of you—awf’ly good of you.” He looked at the fire, not at her.
His voice was suspiciously low. “I might have known, knowing you.
But I can’t impose myself any longer. I’ll be all right now at the
camp.”
“Do you think Dad and I will quit in the middle of a good job? No, sir.
We’re going to get credit for finishing it. Lon’s breaking the road with
a sled to-day. You’re to go down in it. Dr. Rayburn says he hasn’t
time to go up to the camp to look after your bandages. You’ll have to
put up with us for a while.”
“I could go to the hotel at Wild Horse,” he suggested.
“You’ve never eaten a meal there. I see that. It’s impossible. No, it’s
all settled. You’re coming to the ranch.”
“Of course you know I can’t ... I can’t ... thank you.” His voice shook.
This annoyed him. He told himself savagely not to act like a baby.
“Oh, everybody works at the Diamond Bar K,” she said lightly.
“Ruthie can’t go to school through the deep snow. You claim to be a
college man. We’ll find out whether you can teach the First Reader
and two times two.”
He tried to answer in the same spirit. “If you’re going to call every
bluff I make, I’ll be more careful. Anyhow, I didn’t study the First
Reader at college. Don’t think they teach it at Massachusetts Tech.”
Later in the day he spoke to Reed on the subject. “I don’t want to
make a nuisance of myself. I can put up at the Wild Horse House till
I’m strong enough to go back to camp.”
Reed had come out of the old-time cattle days when there were
always a plate and a bed at the ranch for whoever might want them.
He still kept open house.
“Always room for one more, boy. Anyhow, if Betty’s got it fixed up
that way in her mind, you’d as well make up yours to do as she says.
It’ll be her say-so.” The owner of the Diamond Bar K grinned at him
confidentially, as one fellow victim of feminine tyranny to another.
Forbes arrived late in the afternoon and reported heavy roads. He
had brought a four-horse team, and it had been all they could do to
break through. They had dragged a sled with the body of a wagon on
the wide runners.
“Deep drifts below the rim in the cut the other side o’ Round Top. Be
all right if the wind don’t blow to-night,” he said.
The wind blew, and was still whistling when it came time to start. But
the sun was shining and the sky clear.
Betty was doubtful, on account of the patient.
“If we wrop him up good, he’ll be all right,” Lon said. The big foreman
did not want to stay in the hills until the trail he had broken was filled
up again with drifts.
“Yes,” agreed Reed.
Tug and Betty were tucked in with warm blankets. Forbes took the
reins and drove out of the draw, past the Howard place, and up the
steep hill beyond. Betty had seen to it that her patient was wrapped
to the nose in an old fur coat of her father. The whipping wind did not
distress him.
From the summit they could see the great white wastes, stretching
mile on mile. The snow was soft and heavy, and the wind had not
drifted it a great deal since Lon had driven through the previous day.
But the horses were pulling a load, and soon the sweat stood out on
their bodies.
They reached and circled Round Top, passed a treacherous dugway,
and moved into the deep drifts below the rim. Betty looked up once
and a little shudder ran down her spine. The wind up on the bluff had
a clean sweep. Over the edge yawned a great snow comb that might
at any moment loosen and come down to bury them in an immense
white mausoleum. It might, on the other hand, hang from the rim for
months.
Lon cracked his whip close to the ear of the off leader. It might
almost have been a signal. From far above came the sound of an
answering crack. Reed looked up quickly. The snow comb slid
forward, broke, and came tearing down. It gathered momentum in its
plunge, roaring down like an express train.
The cowman flung Betty into the bed of the sled and crouched over
her as a protection against the white cloud of death rushing at them.
The avalanche swept into the ravine with thunderous noise, a
hundred tons of packed snow. The bulk of its weight struck in front of
the horses, but the tail of the slide whipped a giant billow upon them
and buried team and sled.
Betty fought and scrambled her way out of the snow. From it her
father’s head was emerging a few yards away.
“Hurt?” he asked.
“No. You?”
“Jarred up. That’s all. Seen anything of Lon?”
The head and broad shoulders of the foreman pushed up. Lon shook
snow out of his hair, eyes, and ears.

PDF Linear Systems Signal Processing and Hypercomplex Analysis Chapman University November 2017 Daniel Alpay Ebook Full Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PDF Linear Systems Signal Processing and Hypercomplex Analysis Chapman University November 2017 Daniel Alpay Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

Linear Systems Signal Processing and

Hypercomplex Analysis Chapman

Communications, Signal Processing, and Systems:

Quantum-Mechanical Signal Processing and Spectral

Digital Signal Processing Fundamentals and Applications

Cognitive Systems and Signal Processing 4th

Signal and Image Processing Techniques for the

Signal processing for neuroscientists Drongelen

Digital Image Watermarking: Theoretical and

Fundamentals of Signal Enhancement and Array Signal

Founded in 1979 by Israel Gohberg

Associate Editors: Honorary and Advisory Editorial Board:

More information about this series at http://www.springer.com/series/4850

ISSN 0255-0156 ISSN 2296-4878 (electronic)

© Springer Nature Switzerland AG 2019

Editorial Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

J.A. Ball and V. Bolotnikov

J. Behrndt and P. Schlosser

S. Bezuglyi and P.E.T. Jorgensen

P. Cerejeiras, U. Kähler, A. Legatiuk and D. Legatiuk

I. Cho and P.E.T. Jorgensen

B. Fritzsche, B. Kirstein and C. Mädler

A. Melnikov and R. Shusterman

A. Sebbar and O. Wone

Men only build to nothingness

of multiplicative Stieltjes functions, and related interpolation problems. In On a

Daniel Alpay and Mihaela Vajiac

Multiplicative Stieltjes Functions

Abstract. Pairs of reproducing kernel Hilbert spaces associated with mul-

Mathematics Subject Classiﬁcation (2010). 46E22; 46E40, 47A56, 47A48.

for every choice of an integer n, of vectors c1 , . . . , cn ∈ G and of points z1 , . . . , zn ∈

1.1. Reproducing kernel Hilbert spaces with additional structure:

To distinguish the summands in the direct sum G = G ⊕ G, we identify the ﬁrst

Deﬁnition 1.2. We say that H is a space H(Θ) if it admits a reproducing kernel

i.e., if H = HKΘ =: H(Θ) where KΘ is as in (1.4)–(1.5). On occasion it will be

In addition, H is a reproducing kernel Hilbert space with reproducing kernel

is also in H. Since H is a reproducing kernel Hilbert space, the map Eμ : f → f (μ)

3. If Ω ∩ R = ∅ and ν ∈ Ω ∩ R, then Θ can be taken in the form

In case we replace G with G = G ⊕ JG, comparison of (1.19) with (1.6) suggests

kernel KΘ with Θ = TPG [S] to compute

To introduce the multiplicative version of the Stieltjes class, denoted as

it follows that the transformation Θ → P ΘP −1 preserves property (1.5), i.e., the

ﬁnd necessary and suﬃcient conditions directly in terms of the operators Π, A, Π, A

Proof. It follows from (2.1), (2.2) that the operator A + B is J-unitary:

Therefore, the operators

On the other hand,

Proof of (2.30). From (2.29) we see that

Betty went to sleep critical of Justin. She woke, in the dawn of a

You might also like

To distinguish the summands in the direct sum G = G ⊕ G, we identify the ﬁrst

is also in H. Since H is a reproducing kernel Hilbert space, the map Eμ : f → f (μ)

3. If Ω ∩ R = ∅ and ν ∈ Ω ∩ R, then Θ can be taken in the form

In case we replace G with G = G ⊕ JG, comparison of (1.19) with (1.6) suggests