Read Anytime Anywhere Easy Ebook Downloads at ebookmeta.

com

Approximation and Computation in Science and

Engineering 1st Edition Nicholas J. Daras (Editor)

https://ebookmeta.com/product/approximation-and-computation-
in-science-and-engineering-1st-edition-nicholas-j-daras-
editor/

OR CLICK HERE

DOWLOAD EBOOK

Visit and Get More Ebook Downloads Instantly at https://ebookmeta.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Exploring Mathematical Analysis, Approximation Theory, and

Optimization: 270 Years Since A.-M. Legendre’s Birth 1st
Edition Nicholas J. Daras
https://ebookmeta.com/product/exploring-mathematical-analysis-
approximation-theory-and-optimization-270-years-since-a-m-legendres-
birth-1st-edition-nicholas-j-daras/
ebookmeta.com

Computation and Approximation 1st Edition Vijay Gupta

https://ebookmeta.com/product/computation-and-approximation-1st-
edition-vijay-gupta/

ebookmeta.com

Traffic and Highway Engineering, 5th Edition Garber

Nicholas J.

https://ebookmeta.com/product/traffic-and-highway-engineering-5th-
edition-garber-nicholas-j/

ebookmeta.com

Wraith King 3 1st Edition Jack Porter

https://ebookmeta.com/product/wraith-king-3-1st-edition-jack-porter/

ebookmeta.com
The Sounds of Mandarin 1st Edition Janet Y. Chen

https://ebookmeta.com/product/the-sounds-of-mandarin-1st-edition-
janet-y-chen/

ebookmeta.com

Education Music and the Lives of Undergraduates Collegiate

A Cappella and the Pursuit of Happiness 1st Edition Roger
Mantie
https://ebookmeta.com/product/education-music-and-the-lives-of-
undergraduates-collegiate-a-cappella-and-the-pursuit-of-happiness-1st-
edition-roger-mantie/
ebookmeta.com

Managing Institutions The Survival of Minban Secondary

Schools in Mainland China 1st Edition Ying Wang

https://ebookmeta.com/product/managing-institutions-the-survival-of-
minban-secondary-schools-in-mainland-china-1st-edition-ying-wang/

ebookmeta.com

Basic Statistics in Business and Economics (ISE HED IRWIN

STATISTICS) 10th Edition Douglas A. Lind

https://ebookmeta.com/product/basic-statistics-in-business-and-
economics-ise-hed-irwin-statistics-10th-edition-douglas-a-lind/

ebookmeta.com

The Lute in the Netherlands in the Seventeenth Century

Proceedings of the International Lute Symposium Utrecht 30
August 2013 1st Edition Jan W.J. Burgers
https://ebookmeta.com/product/the-lute-in-the-netherlands-in-the-
seventeenth-century-proceedings-of-the-international-lute-symposium-
utrecht-30-august-2013-1st-edition-jan-w-j-burgers/
ebookmeta.com
Information and Communications Technology Law in Singapore
1st Edition Warren B Chik

https://ebookmeta.com/product/information-and-communications-
technology-law-in-singapore-1st-edition-warren-b-chik/

ebookmeta.com
Springer Optimization and Its Applications 180

Nicholas J. Daras
Themistocles M. Rassias Editors

Approximation
and Computation
in Science and
Engineering
Springer Optimization and Its Applications

Volume 180

Series Editors
Panos M. Pardalos , University of Florida
My T. Thai , University of Florida

Honorary Editor
Ding-Zhu Du, University of Texas at Dallas

Advisory Editors
Roman V. Belavkin, Middlesex University
John R. Birge, University of Chicago
Sergiy Butenko, Texas A&M University
Vipin Kumar, University of Minnesota
Anna Nagurney, University of Massachusetts Amherst
Jun Pei, Hefei University of Technology
Oleg Prokopyev, University of Pittsburgh
Steffen Rebennack, Karlsruhe Institute of Technology
Mauricio Resende, Amazon
Tamás Terlaky, Lehigh University
Van Vu, Yale University
Michael N. Vrahatis, University of Patras
Guoliang Xue, Arizona State University
Yinyu Ye, Stanford University
Aims and Scope
Optimization has continued to expand in all directions at an astonishing rate. New
algorithmic and theoretical techniques are continually developing and the diffusion
into other disciplines is proceeding at a rapid pace, with a spot light on machine
learning, artificial intelligence, and quantum computing. Our knowledge of all
aspects of the field has grown even more profound. At the same time, one of the
most striking trends in optimization is the constantly increasing emphasis on the
interdisciplinary nature of the field. Optimization has been a basic tool in areas
not limited to applied mathematics, engineering, medicine, economics, computer
science, operations research, and other sciences.

The series Springer Optimization and Its Applications (SOIA) aims to

publish state-of-the-art expository works (monographs, contributed volumes,
textbooks, handbooks) that focus on theory, methods, and applications of
optimization. Topics covered include, but are not limited to, nonlinear optimization,
combinatorial optimization, continuous optimization, stochastic optimization,
Bayesian optimization, optimal control, discrete optimization, multi-objective
optimization, and more. New to the series portfolio include Works at the
intersection of optimization and machine learning, artificial intelligence, and
quantum computing.

Volumes from this series are indexed by Web of Science, zbMATH, Mathematical
Reviews, and SCOPUS.

More information about this series at http://www.springer.com/series/7393

Nicholas J. Daras • Themistocles M. Rassias
Editors

Approximation and
Computation in Science
and Engineering
Editors
Nicholas J. Daras Themistocles M. Rassias
Department of Maths and Engineering Department of Mathematics, Zografou
Sciences Campus
Hellenic Military Academy National Technical University of Athens
Vari Attikis, Greece Athens, Greece

ISSN 1931-6828 ISSN 1931-6836 (electronic)

Springer Optimization and Its Applications
ISBN 978-3-030-84121-8 ISBN 978-3-030-84122-5 (eBook)
https://doi.org/10.1007/978-3-030-84122-5

Mathematics Subject Classification: 11XX, 14XX, 34XX, 37XX, 47XX, 49XX, 60XX, 65XX, 68XX,
81XX, 90XX, 91XX, 94XX, 05XX

© Springer Nature Switzerland AG 2022

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, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface

Approximation and Computation in Science and Engineering presents a wide

spectrum of research and research-survey papers in several subjects of pure and
applied mathematics with various applications. Emphasis is given to the study of
topics of approximation theory, functional equations and inequalities, fixed point
theory, numerical analysis, theory of wavelets, convex analysis, topology, opera-
tor theory, differential operators, fractional integral operators, integro-differential
equations, ternary algebras, super and hyper relators, variational analysis, discrete
mathematics, cryptography, and a variety of applications in interdisciplinary topics.
Several of these domains have a strong connection with both theories and problems
of linear and nonlinear optimization. Therefore, it is hoped that this collective
effort will be particularly useful to researchers who are focusing on applications
of theories and methods of the above-mentioned subjects for optimization. Overall,
the works published within this book will be of particular value for both theoretical
and applicable interdisciplinary research.
We would like to express our sincere thanks to the contributing authors of the
book chapters. We would also like to warmly thank the staff at Springer for their
valuable assistance throughout the preparation of this book.

Vari Attikis, Greece Nicholas J. Daras

Athens, Greece Themistocles M. Rassias

v
Contents

Rearrangements, L-Superadditivity and Jensen-Type Inequalities . . . . . . . . 1

Shoshana Abramovich
Approximate Generalized Jensen Mappings in 2-Banach Spaces . . . . . . . . . . 17
Muaadh Almahalebi, Themistocles M. Rassias, Sadeq Al-Ali,
and Mustapha E. Hryrou
The Asymptotic Expansion for a Class of Convergent Sequences
Defined by Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Dorin Andrica and Dan Ştefan Marinescu
Weak Pseudoprimality Associated with the Generalized Lucas
Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Dorin Andrica, Ovidiu Bagdasar, and Michael Th. Rassias
Finite Shift-Invariant Subspaces of Periodic Functions:
Characterization, Approximation, and Applications. . . . . . . . . . . . . . . . . . . . . . . . . 77
Nikolaos Atreas
Generalized Intensity-Dependent Multiphoton
Jaynes–Cummings Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
V. Bartzis, N. Merlemis, M. Serris, and G. Ninos
Functional Inequalities for Multi-additive-Quadratic-Cubic Mappings . . . 103
Abasalt Bodaghi and Themistocles M. Rassias
Generalizations of Truncated M-Fractional Derivative Associated
with (p, k)-Mittag-Leffler Function with Classical Properties . . . . . . . . . . . . . 127
Mehar Chand and Praveen Agarwal
On Hyers-Ulam-Rassias Stability of a Volterra-Hammerstein
Functional Integral Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Sorina Anamaria Ciplea, Nicolaie Lungu, Daniela Marian,
and Themistocles M. Rassias

vii
viii Contents

Analysis of Electroencephalography (EEG) Signals Based on the

Haar Wavelet Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Y. Contoyiannis, P. Papadopoulos, S. M. Potirakis, M. Kampitakis,
N. L. Matiadou, and E. Kosmidis
Perov-Type Contractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Marija Cvetković, Erdal Karapınar, Vladimir Rakočević,
and Seher Sultan Yeşilkaya
On a Logarithmic Equation by Primes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
S. I. Dimitrov
Hermite-Hadamard Trapezoid and Mid-Point Divergences . . . . . . . . . . . . . . . . 227
Silvestru Sever Dragomir
Hermite-Hadamard-Type Integral Inequalities for Perspective
Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Silvestru Sever Dragomir
On the Maximum Value of a Multi-variable Function . . . . . . . . . . . . . . . . . . . . . . . 273
Soheila Emamyari and Mehdi Hassani
Image Reconstruction for Positron Emission Tomography Based
on Chebyshev Polynomials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
George Fragoyiannis, Athena Papargiri, Vassilis Kalantonis,
Michael Doschoris, and Panayiotis Vafeas
Approximation by Mixed Operators of Max-Product–Choquet Type . . . . . 297
Sorin G. Gal and Ionut T. Iancu
On the Approximation of Extinction Time for the Discrete-Time
Birth–Death Circuit Chains in Random Environments . . . . . . . . . . . . . . . . . . . . . 333
Chrysoula Ganatsiou
Some Hyperstability Results in Non-Archimedean 2-Banach
Space for a σ -Jensen Functional Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Rachid EL Ghali and Samir Kabbaj
A Characterization for the Validity of the Hermite–Hadamard
Inequality on a Simplex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
Allal Guessab
On the Stability of the Triangular Equilibrium Points in the
Photogravitational R3BP with an Oblate Infinitesimal and
Triaxial Primaries for the Binary Lalande 21258 System . . . . . . . . . . . . . . . . . . . 397
Jessica Mrumun Gyegwe, Aguda Ekele Vincent, and Angela E. Perdiou
Normalized Symmetric Differential Operators in the Open Unit Disk . . . . 417
Rabha W. Ibrahim
Contents ix

New Hermite–Hadamard Inequalities Concerning Twice

Differentiable Generalized ψ-Convex Mappings via Conformable
Fractional Integrals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Artion Kashuri and Rozana Liko
Some New Fractional Inequalities Using n-Polynomials s-Type
Convexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
Artion Kashuri, Themistocles M. Rassias, and Rozana Liko
Hyperstability of Orthogonally 3-Lie Homomorphism: An
Orthogonally Fixed Point Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
Vahid Keshavarz and Sedigheh Jahedi
Some New Inequalities for Fractional Integral Operators . . . . . . . . . . . . . . . . . . 487
Jichang Kuang
New Generalized Convexity and Their Applications . . . . . . . . . . . . . . . . . . . . . . . . . 509
Jichang Kuang
Ternary Biderivations and Ternary Bihomorphisms in
C ∗ -Ternary Algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
Jung Rye Lee, Choonkil Park, and Themistocles M. Rassias
Hyers–Ulam Stability of an Additive-Quadratic Functional Equation . . . . 561
Jung Rye Lee, Choonkil Park, and Themistocles M. Rassias
Orthogonal Dirichlet Polynomials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573
Doron S. Lubinsky
Generalizations and Improvements of Approximations of Some
Analytic Functions: A Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 589
Branko Malešević, Tatjana Lutovac, and Marija Rašajski
Some Classes of Meir–Keeler Contractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609
Laura Manolescu, Paşc Găvruţa, and Farshid Khojasteh
Interpolation of the Zech’s Logarithm: Explicit Forms . . . . . . . . . . . . . . . . . . . . . 619
Gerasimos C. Meletiou and Michael Th. Rassias
Numerical Calculations on Multi-Photon Processes in Alkali
Metal Vapors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627
Nikolaos Merlemis, Andreas Lyras, Georgios Papademetriou,
Dionysios Pentaris, and Thomas Efthimiopoulos
General Preinvex Functions and Variational-Like Inequalities . . . . . . . . . . . . . 643
Muhammad Aslam Noor, Khalida Inayat Noor, Bandar Mohsen, Michael
Th. Rassias, and Andrei Raigorodskii
A Variational Formulation of Network Games with Random
Utility Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667
Mauro Passacantando and Fabio Raciti
x Contents

Fixed Point Theory in Graph Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679

A. Petruşel and G. Petruşel
Approximate Solution of Fredholm Integral and
Integro-Differential Equations with Non-Separable Kernels . . . . . . . . . . . . . . . 693
E. Providas
Ordinary, Super and Hyper Relators Can Be Used To Treat the
Various Generalized Open Sets in a Unified Way . . . . . . . . . . . . . . . . . . . . . . . . . . . . 709
Themistocles M. Rassias and Árpád Száz
Applications of Apostol-type Numbers and Polynomials:
Approach to Techniques of Computation Algorithms in
Approximation and Interpolation Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 783
Yilmaz Simsek
Spectra of Signed Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 861
Irene Triantafillou
Perturbed Geometric Contractions in Ordered Metric Spaces . . . . . . . . . . . . . 875
Mihai Turinici
On G(σ,h) -Convexity of the Functions and Applications to
Hermite-Hadamard’s Inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 927
Muhammad Uzair Awan, Muhammad Aslam Noor, Khalida Inayat Noor,
Yu-Ming Chu, and Sara Ellahi
Rearrangements, L-Superadditivity
and Jensen-Type Inequalities

Shoshana Abramovich

Abstract We deal here with the minimum and the maximum of


n
F (a2i−1 , a2i ) , (a) ∈ R2n
i=1

and of


n
F (ai , ai+1 ) , an+1 = a1 , (a) ∈ Rn
i=1

obtained by using rearrangement techniques. The results depend on the arrangement

of (a) and are used in proving Jensen-type inequalities.

1 Introduction

We deal  here with the minimum and the maximum of ni=1 F (a2i−1 , a2i ) , (a) ∈
R2n and ni=1 F (ai , ai+1 ) , where an+1 = a1 , (a) ∈ Rn when F (x, y), (x, y) ∈ R2
is L-superadditive function. These extrema are obtained by using rearrangement
techniques. The results are used in proving Jensen-type inequalities.
In [5], the authors prove the following theorem:

 A ([5, Lemma2]) Let yi ∈ R+ , i = 1, . . . , 2n. Then, for t > 0, the

Theorem
sum ni=1 ln (y2i−1 y2i + t) gets its maximum value when (y) = (y1 , . . . , y2n ) is
arranged in a decreasing order.

S. Abramovich ()
Department of Mathematics, University of Haifa, Haifa, Israel

© Springer Nature Switzerland AG 2022 1

N. J. Daras, Th. M. Rassias (eds.), Approximation and Computation in Science
and Engineering, Springer Optimization and Its Applications 180,
https://doi.org/10.1007/978-3-030-84122-5_1
2 S. Abramovich

A generalization of this result is as follows:

Theorem B ([1, Theorem 2]) Let (a) = (a1 , . . . , am ) and (y) = (y1 , . . . , y2n ) be
sets
n of non-negative
 numbers, where (y) is given except
 its arrangements. Then,
i=1 ln (y y
2i−1 2i ) m
+ a (y y
1 2i−1 2i )m−1
+ . . . + am attains its maximum when
(y) is arranged in a decreasing order.
Definition 1 ([8, Page 150]) A real-valued function F defined on R2 is said to be
L-superadditive if it satisfies the condition that

F (x + α, y) − F (x, y) is increasing in y for all x and all α > 0.

∂
Corollary 1 ([7]) Let F have second partial derivatives. Then, ∂y F (x, y) is
∂2
increasing in x, which is equivalent to ∂x∂y F (x, y) ≥ 0, if and only if F is L-
superadditive function.
For clarity, R+ and R++ are [0, ∞) and (0, ∞), respectively.
In the sequel, we show that the following arrangements of a given set of real
numbers have a role in getting our extrema.
Definition 2 For any (x) = (x1 , . . . , xn ) ∈ Rn , let

x[1] ≥ . . . ≥ x[n]

denote the terms of (x) in decreasing order, and let

   
x↓ = x[1] . . . x[n]

denote the decreasing rearrangement of (x).

Similarly, let

x(1) ≤ . . . ≤ x(n)

denote the terms of (x) in increasing order, and let

   
x↑ = x(1) . . . x(n)

denote the increasing rearrangement of (x).

Definition 3 ([4, 6]) An ordered set (x) = (x1 , . . . , xn ) of n real numbers is
arranged in symmetrical decreasing order if

x1 ≤ xn ≤ x2 ≤ . . . ≤ x[(n+2/2)] (1)
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 3

or if

xn ≤ x1 ≤ xn−1 ≤ . . . ≤ x[(n+1/2)] . (2)

Definition 4 ([4, 6]) A circular rearrangement of an ordered set (x) is a cyclic

rearrangement of (x) or a cyclic rearrangement followed by inversion. For example,
the circular rearrangements of the ordered set (1, 2, 3, 4) are the sets

(1, 2, 3, 4) , (2, 3, 4, 1) , (3, 4, 1, 2) , (4, 1, 2, 3) ,

(4, 3, 2, 1) , (1, 4, 3, 2) , (2, 1, 4, 3) , (3, 2, 1, 4) .

Definition 5 ([4, 6]) A set (x) is arranged in circular symmetrical order if one of
its circular rearrangements is symmetrically decreasing.
Definition 6 ([4]) An ordered set (x) = (x1 , . . . , xn ) of n real numbers is arranged
in an alternating order if

x1 ≤ xn−1 ≤ x3 ≤ xn−3 ≤ . . . ≤ x n+1 ≤ . . . ≤ x4 ≤ xn−2 ≤ x2 ≤ xn , (3)

2

or if

xn ≤ x2 ≤ xn−2 ≤ x4 ≤ . . . ≤ x n+1 ≤ . . . ≤ xn−3 ≤ x3 ≤ xn−1 ≤ x1 . (4)

2

Definition 7 ([4]) A set (x) is arranged in a circular alternating order if one of

its circular rearrangements is arranged in an alternating order.
       
We denote in the sequel x↑ = x(1) . . . x(n) , x↓ = x[1] . . . x[n] , (x) =
(x1 , x2 , . . . , xn ) and (x) = (x1 , x2 , . . . , xn ) to be the increasing order, the decreas-
ing order, the circular alternating order and the circular symmetrical order of a given
set (x) ∈ Rn , respectively.

2 L-Superadditivity Applications to Rearrangements

We extend Theorems A and B as follows:

Theorem 1 Let F (u, v) be a real function symmetric in u and v defined on α <
u, v < β, −∞ ≤ α < β ≤ ∞, and assume that

G (u, v, w) = F (u, w) − F (u, v) , α < u, v, w < β, (5)

is increasing in u for w ≥ v, that is, F is L-superadditive function. Then, for any

set (x) =(x1 , x2 , . . . , x2n−1 , x2n ), α < xi < β, i = 1, . . . , 2n, given except its
4 S. Abramovich

arrangement, the inequalities


n
  n
 
F x(i) , x[i] = F x(i) , x(2n−i+1) (6)
i=1 i=1


n
≤ F (x2i−1 , x2i )
i=1


n
  n
 
≤ F x(2i−1) , x(2i) = F x[2i−1] , x[2i]
i=1 i=1
   
hold when x↑ = x(1) , x(2) , . . . , x(2n−1)  , x(2n) is
 the increasing
 rearrangement
of (x) =(x1 , x2 , . . . , x2n−1 , x2n ), and x↓ = x[1] . . . x[2n] is the decreasing
rearrangement of (x).

Proof We first prove that the minimum of ni=1 F (x2i−1 , x2i ) is obtained when
x2i−1 = x(i) , i = 1, . . . , n and x2i = x(2n−i+1) , i = 1, . . . , n.
We may assume that x1 = x(1) . If x2 = x(2n) , and x(2n) appears as x2j −1 or x2j ,
we may assume that x2j = x(2n) because of the symmetry of F (u, v). We now rear-
range (x) = (x1 , x2 , . . . , x2n−1 , x2n ) and get x = x1 , x2 , . . . , x2n−1 , x2n =
 
x(1) , x(2n) , . . . , x2j −1 , x2 , x2j +1 , x2j +2 , . . . , x2n . We exchange here only the loca-
tions of x2 with x2j = x(2n) and compute the difference


n 
n
F (x2i−1 , x2i ) − F x2i−1 , x2i (7)
i=1 i=1
        
= F x(1) , x2 + F x2j −1 , x(2n) − F x(1) , x(2n) + F x2j −1 , x2
        
= F x2j −1 , x(2n) − F x2j −1 , x2 − F x(1) , x(2n) − F x(1) , x2 .

As x2j −1 ≥ x(1) and x(2n) ≥ x2 , we get by using the conditions of the theorem on
G (u, v, w) that
        
F x2j −1 , x(2n) − F x2j −1 , x2 ≥ F x(1) , x(2n) − F x(1) , x2 . (8)

From (7) and (8), we get that


n 
n
F (x2i−1 , x2i ) ≥ F x2i−1 , x2i .
i=1 i=1
 
As a result, we get that the two terms x(1) and x(2n) appear as F x(1) , x(2n) in
n n
i=1 F x2i−1 , x2i without increasing the sum i=1 F (x2i−1 , x2i ). We continue

now with the other 2n − 2 terms in x and by the same procedure bring x(2) and
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 5

x(2n−1) next to each other in F . After at most n steps, we get the left-hand side of
inequalities of (6). 
To get the maximum of ni=1 F (x2i−1 , x2i ), we use the same procedure as in
the proof of its minimum.
Without loss of generality, we may assume that the minimal term in (x) is
x1 = x(1) . If x2 = x(2) , but x2j = x(2) , we create a new rearrangement x
= x1 , x2 , . . . , x2n−1 , x2n in which x2 = x(2) = x2j , and x2j = x2 , and all other
terms are as in (x).
We compute now the difference


n 
n
F x2i−1 , x2i − F (x2i−1 , x2i ) (9)
i=1 i=1
        
= F x(1) , x(2) + F x2j −1 , x2 − F x(1) , x2 + F x2j −1 , x(2)
        
= F x2j −1 , x2 − F x2j −1 , x(2) − F x(1) , x2 − F x(1) , x(2) .

As it is given in (5) that G (u, v, w) is increasing in u when w ≥ v and as x2j −1 ≥

x(1) and x2 ≥ x(2) , we get that
        
F x2j −1 , x2 − F x2j −1 , x(2) − F x(1) , x2 − F x(1) , x(2) ≥ 0. (10)

From (9) and (10), we obtain that


n 
n
F x2i−1 , x2i − F (x2i−1 , x2i ) ≥ 0,
i=1 i=1
 
which means that the two smallest terms x(1) and x(2) appear as F x(1) , x(2) in
n n
i=1 F x2i−1 , x2i without decreasing i=1 F (x2i−1 , x2i ).
Repeating now the procedure by bringing the two minimal terms x(3) and x(4) of
the remaining 2n − 2 xi th to be in the same F . After at most n similar steps, we get
the right-hand side of inequalities of (6).
This completes the proof of the theorem.
It is easy to see that the following functions satisfy Theorem 1.
Example 1 Let F (u, v) = 1
uv+1 , then according to Theorem 1, we get by simple
computations that


n
1 
n
1  n
1
≤ ≤
x(i) x(2n+1−i) + 1 x2i−1 x2i + 1 x(2i−1) x(2i) + 1
i=1 i=1 i=1

when (x) ≥ (1).

6 S. Abramovich

Example 2 Let f : R+ → R+ be increasing and convex function. Then,

F (u, v) = f (uv) satisfies Theorem 1.
Remark 1 Theorems A and B are special cases of the right-hand side of (6). Also,
in [10], the author shows that for L-superadditive function F the right-hand side of
(6) can be derived from [7].
Lately, Theorems 2 and 3 were proved in [4]. However, it was proved earlier in
[1] that the maximum of ni=1 F (xi , xi+1 ) , xn+1 = x1 , is attained when (x) is
arranged in a circular symmetrical order (which generalizes a theorem in [6], there
on f (|x − y|) for concave decreasing f ). The proof uses the induction procedure.
It relies on the fact that if the n numbers (x1 , x2 , . . . , xn−1 , xn ) satisfy (1), then the
n − 1 numbers (x2 , . . . ,  xn−1 , xn ) satisfy (2).
The arrangement of ni=1 F (xi , xi+1 ) , xn+1 = x1 , for which its minimum
value is attained is stated in Theorem 2 and proved in [4, Theorem 1]. A brief
outline of this proof appears also in [2]. The proof uses the induction procedure.
The validity of the minimum value is assumed for the set of n − 2 numbers, and
it is shown that this implies its validity for the set of n numbers. This relies on the
fact that if the n numbers (x1 , x2 , . . . , xn−1 , xn ) satisfy (3), then the n − 2 numbers
(x2 , . . . , xn−1 ) satisfy (4).
Theorem 2 ([4, Theorem 1]) Let F (u, v) be differentiable and symmetric real
function in u and v defined on α ≤ u, v, w ≤ β, and assume that

∂F (v, u) ∂F (v, w)
≤
∂v ∂v
for u ≤ min (w, v).
Then, for any set (x) = (x1 , x2 , . . . , xn ) , α ≤ xi ≤ β, i = 1, . . . , n, given except
its arrangements, the sum


n
F (xi , xi+1 ) , xn+1 = x1 ,
i=1

is maximal if (x) is arranged in a circular symmetrical order and minimal if (x) is

arranged in the circular alternating order as defined above.
Theorem 3 ([4, Theorem 2]) Let F = F (u, v) be a real function defined on α ≤
u, v ≤ β, which is symmetric in u and v. Then, ni=1 F (xi , xi+1 ) , where xn+1 =
x1 , is maximal if (x) is arranged in a circular symmetrical order and minimal if (x)
is arranged in the circular alternating order in each of the following cases:
Case (a) F (x, y) = f (x + y), where f is convex on R+ ,
Case (b) F (x, y) = f (|x − y|) , 
wheref is concave and decreasing on R+ .
Case (c) F (x, y) = f x+y 2 + f  x−y2
 , where f is convex and differentiable
on R+ and f (0) = 0,
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 7

 x+y   x−y 2
Case (d) F (x, y) = f 2 +C× 2 , where the constant C satisfies
xm +xj
C ≤ ϕ 2 , with 0 ≤ xm ≤ xj ≤ xi , i = m, j, i = 1, . . . , n, and
f (x) = xϕ (x) , and ϕ is twice differentiable and convex function on 0 ≤ x ≤ b.
     x−y 2
Case (e) F (x, y) = f x+y 2 + ϕ x+y 2 2 , where ϕ and ϕ are twice
differentiable and convex on R+ and f (x) = xϕ (x).
 
Also, it has been proved lately in [2] that F (x, y) = xϕ yx + yϕ xy when ϕ is
 
concave on R+ and lim xϕ (x) − ϕ (x) = 0 and in particular when F (x, y) =
x→0+
x s y t + x t y s , x, y, t, s ∈ R+ satisfy Theorem 2.
Theorem  2.7]) If ϕ is a concave differentiable function on R+ and
 4 ([2, Theorem
lim xϕ (x) − ϕ (x) = 0. Then, the inequalities
x→0+

n 
    
xi+1 xi
xi ϕ + xi+1 ϕ (11)
xi xi+1
i=1
n 
    
xi+1 xi
≤ xi ϕ + xi+1 ϕ
xi xi+1
i=1
n 
    
xi+1 xi
≤ xi ϕ + xi+1 ϕ
xi xi+1
i=1

hold where xn+1 = x1 , xn+1 = x1 and xn+1 = x1 .  

If ϕ is a convex differentiable function on R+ and lim xϕ (x) − ϕ (x) = 0,
x→0+
then the reverse of inequalities of (11) holds.
Corollary 2 ([2, Theorem 2.8]) Let F (x, y) = x s y t + x t y s .
(a) If x, y, s, t ∈ R+ , then for (x) ∈ Rn+ , the inequalities


n
 t s 
xi xi+1 + xi+1
t
xis (12)
i=1


n
 s t 
≤ xi xi+1 + xi+1
t
xis
i=1


n
 t s 
≤ xi xi+1 + xi+1
t
xis ,
i=1

hold, where xn+1 = x1 , xn+1 = x1 , xn+1 = x1 , and (x) is the circular

alternating order of (x) and (x) is the circular symmetrical order of (x). In
particular, (12) holds when t + s = 1, 0 ≤ t ≤ 1.
8 S. Abramovich

(b) If s ≤ 0, t ≥ 0, the reverse of (12) holds for (x) ∈ Rn++ . In particular, the
reverse of (12) holds when t + s = 1, t ≥ 1.
Remark 2 It is easy to verify that in addition to Examples 1 and 2, the functions
F (x, y), which appear in Theorem 3 cases (a), (d) and (e), Theorem 4 and Corollary
2 F (x,y)
2, are twice differentiable L-superadditive functions satisfying ∂ ∂x∂y ≥ 0, and
therefore also Theorem 1 holds.

3 Jensen-Type Inequalities and Rearrangements

Using the results of Sect. 2, we get in this section Jensen-type inequalities. The
methods employed here can be obtained similarly to those derived in [4, Section 3]
and [2] using Theorem 3 [4, Theorem 2].
We demonstrate and prove two refinements of Jensen-type inequalities and quote
one from [4]. For more inequalities related to Theorem 2, see [4, Section 3] and [2,
Theorem 2.11].
We first refine the Jensen inequality for convex function f : I → R, I ⊂ R,
which reads
 n 
 n 
ai f (xi ) ≥ f ai xi
i=1 i=1
n
for all ai ≥ 0, xi ∈ I , i = 1, . . . , n, i=1 ai = 1. The refinement is obtained by
using Theorem 1 for the L-superadditive function F (x, y) = f (x + y) when f is
a convex function.
Theorem 5 Let f be a convex function on an interval, and let (x1 , . . . , x2n ) be a
given 2n real numbers. Then,

  
1
2n n
f (xi ) x(2i−1) + x(2i)
≥ f (13)
2n n 2
i=1 i=1
   
1 1
n n
x2i−1 + x2i x(i) + x(2n+1−i)
≥ f ≥ f
n 2 n 2
i=1 i=1
 
2n
i=1 xi
≥f .
2n

2n f (xi )  +f x
= n1 ni=1 ( (2i−1) )2 ( (2i) ) , as a result
f x
Proof From the identity i=1 2n
of Jensen inequality, we get the first inequality in (13), and because of the L-
superadditivity of F (x, y) = f (x + y), we get the second and third inequalities
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 9

of Equation (13). The last inequality follows again from the Jensen inequality for
convex functions.
We quote now some definitions and lemmas that we need for the theorems
presented below.
The following is a definition of the Jensen-type inequality for strongly convex
functions with modulus C.
Definition 8 ([9]) Let n ∈ N. The function f : I → R, I ⊂ R, is called strongly
convex
n with modulus C if for all xi ∈ I , and all ai ≥ 0, i = 1, . . . , n, such that
i=1 ai = 1, the Jensen-type inequality


n 
n
ai f (xi ) − f (x) ≥ C ai (xi − x)2 ,
i=1 i=1
n
where C ≥ 0 and x = i=1 ai xi , holds.
Definition 9 ([3]) A real-valued function ψ1 defined on an interval [a, b) with 0 ≤
a < b ≤ ∞ is called 1-quasiconvex if it can be represented as the product of a
convex function ϕ and the function p (x) = x.
Corollary 3 A 1-quasiconvex function ψ1 , as defined in Definition 9, satisfies the
inequalities


n 
n
ai ψ1 (xi ) ≥ ψ1 (x) + ϕ (x) ai (xi − x)2 (14)
i=1 i=1


n
≥ ψ1 (x) + C ai (xi − x)2 ,
i=1

 
where C ≤ min ϕ (xi ), ai ≥ 0, i = 1, . . . , n, ni=1 ai = 1 and x = ni=1 ai xi .
If, in addition, ϕ is increasing, then ψ1 is also a strongly convex function.
Theorems 6, 7 and 8 show the use of rearrangements for refinements of Jensen-
type inequality for 1-quasiconvex functions by using Corollary 1 and Theorem 1 for
the function
    2
x+y x+y x−y
F (x, y) = f +ϕ ,
2 2 2

when f (x) = xϕ (x) and where ϕ and ϕ are convex. In this case, F (x, y) is L-
superadditive and therefore satisfies Theorem 1 as well as Theorem 2.
Theorem 6 Let xi , i = 1, . . . , 2n, n ∈ N, be a sequence of real non-negative
numbers, and let ϕ and ϕ be convex on R+ and f (x) = xϕ (x), x ∈ R+ . Then,
10 S. Abramovich

2n
denoting x = 1
2n j =1 xj , it yields that

(15)

2n 
2n
1
f (xi ) − (2n − 1) f (x) − (2n − 1) ϕ (x) (xi − x)2
2n
i=1 i=1
      
1
n
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
≥ f +ϕ
n 2 2 2
i=1
      
1
n
x2i−1 + x2i x2i−1 + x2i x2i − x2i−1 2
≥ f +ϕ
n 2 2 2
i=1
      
1
n
x(i) + x(2n+1−i) x(i) + x(2n+1−i) x(2n+1−i) − x(i) 2
≥ f +ϕ .
n 2 2 2
i=1

If also

(16)
       
1
n
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
f +ϕ
n 2 2 2
i=1

1  2n
≥ f (x) + ϕ (x) (xi − x)2 ,
2n
i=1

then


2n 
2n
f (xi ) − 2nf (x) − ϕ (x) (xi − x)2 (17)
i=1 i=1
      
1
n
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
≥ f +ϕ
n 2 2 2
i=1


2n
1
−f (x) − ϕ (x) (xi − x)2 ≥ 0
2n
i=1

refines Inequality (14).

Rearrangements, L-Superadditivity and Jensen-Type Inequalities 11

Proof


2n 
2n
 
f (xi ) = f x(i) (18)
i=1 i=1

1    2n − 1 
2n 2n
= f x(i) + f (xi )
2n 2n
i=1 i=1
   
1  f x(2i−1) + f x(2i) 2n − 1 
n 2n
= + f (xi )
n 2 2n
i=1 i=1
n 
    
1 x(2i−1) + x(2i) x(2i−1) + x(2i)
≥ f +ϕ
n 2 2
i=1
  
x(2i) − x(2i−1) 2
×
2

2n − 1 
2n
+ f (xi )
2n
i=1
      
1
n
x2i−1 + x2i x2i−1 + x2i x2i − x2i−1 2
≥ f +ϕ
n 2 2 2
i=1

2n − 1 
2n
+ f (xi )
2n
i=1
n 
    
1 x(i) + x(2n+1−i) x(i) + x(2n+1−i)
≥ f +ϕ
n 2 2
i=1
  
x(2n+1−i) − x(i) 2
×
2

2n − 1 
2n
+ f (xi ) .
2n
i=1

Indeed, Inequality (18) follows from Inequality (14) for ai = 12 , i = 1, 2.

12 S. Abramovich

Using again Inequality (14), this time for ai = 1

2n , i = 1, . . . , 2n, we get that the
inequalities


2n
f (xi )
i=1
      
1
n
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
≥ f +ϕ
n 2 2 2
i=1

2n − 1 
2n
+ f (xi )
2n
i=1
      
1
n
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
≥ f +ϕ
n 2 2 2
i=1
2n 
 
1
+ (2n − 1) f (x) + ϕ (x) (xi − x) 2
2n
i=1

      
1
n
x2i−1 + x2i x2i−1 + x2i x2i − x2i−1 2
≥ f +ϕ
n 2 2 2
i=1
2n 
 
1
+ (2n − 1) f (x) + ϕ (x) (xi − x)2
2n
i=1
      
1
n
x(i) + x(2n+1−i) x(i) + x(2n+1−i) x(2n+1−i) − x(i) 2
≥ f +ϕ
n 2 2 2
i=1
2n 
 
1
+ (2n − 1) f (x) + ϕ (x) (xi − x)2
2n
i=1

hold, from which Inequality (15) follows.

The first inequality in (15) can be rewritten as Inequality (17). Therefore, when
(16) holds, the inequality (17) refines (14) for ai = 2n 1
, i = 1, . . . , 2n. This
completes the proof of the theorem.
Example 3 shows a case where (16) holds, and therefore in this case, (17) is a
refinement of (14):
     x−y 2
Example 3 F (x, y) = f x+y 2 + ϕ x+y2 2 satisfies Theorem 3 Case (e)
for f (x) = x = xϕ (x), where ϕ (x) = x . It is easy to check that for xi = i,
4 3
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 13

i = 1, 2, 3, 4,
      
1
2
x(2i−1) + x(2i) x(2i−1) + x(2i) x(2i) − x(2i−1) 2
f +ϕ
2 2 2 2
i=1

1  4
≥ f (x) + ϕ (x) (xi − x)2 ,
4
i=1

and therefore (17) is a refinement of (14) in this case.

Similarly to Theorem 6, but this time using the L-superadditivity of F (x, y) =
     x−y 2
f x+y2 + ϕ x+y 2 2 as appears in Theorem 3 case (e), we get the following
theorem:
Theorem 7 ([4, Theorem 5]) Let xi , i = 1, . . . , n, n ∈ N, be a sequence of real
 and let ϕ and ϕ , be convex on R+ where f (x) = xϕ (x),
non-negative numbers,
x ∈ R+ and x = n1 nj=1 xj . Then,


n 
n
1
f (xi ) − (n − 1) f (x) − (n − 1) ϕ (x) (xi − x)2
n
i=1 i=1
     2 
1 
n
xi + xi+1 xi + xi+1 xi − xi+1
≥ f +ϕ
n 2 2 2
i=1
     2 
1
n
xi + xi+1 xi + xi+1 xi − xi+1
≥ f +ϕ
n 2 2 2
i=1
     2 
1
n
xi + xi+1 xi + xi+1 xi − xi+1
≥ f +ϕ ,
n 2 2 2
i=1

where xn+1 = x1 , xn+1 = x1 , xn+1 = x1 .

If also
      
1
n
xi + xi+1 xi + xi+1 xi − xi+1 2
f +ϕ
n 2 2 2
i=1

1 n
≥ f (x) + ϕ (x) (xi − x)2 ,
n
i=1
14 S. Abramovich

then


n 
n
f (xi ) − nf (x) − ϕ (x) (xi − x)2
i=1 i=1
      
1
n
xi + xi+1 xi + xi+1 xi − xi+1 2
≥ f +ϕ
n 2 2 2
i=1


n
1
−f (x) − ϕ (x) (xi − x)2 ≥ 0.
n
i=1

Theorem 8 (Using [4, Theorem 2, Case (d)]) Let xi , i = 1, . . . , n, n ∈ N, be a

sequence of real non-negative numbers, and let ϕ be convex on x ≥ 0 and f (x) =
xϕ (x). Let C ≤ min ϕ (xi ), i = 1, . . . , n. Then, the 1-quasiconvex function f
(which is strongly convex when C ≥ 0) satisfies
 n 

n
j =1 xj
f (xi ) − (n − 1) f
n
i=1
 n  n  n 2
n−1 j =1 xj
 j =1 xj
− ϕ xi −
n n n
i=1
     
1
n
xi + xi+1 xi − xi+1 2
≥ f +C×
n 2 2
i=1
     
1
n
xi + xi+1 xi − xi+1 2
≥ f +C×
n 2 2
i=1
     
1
n
xi + xi+1 xi − xi+1 2
≥ f +C× ,
n 2 2
i=1

where xn+1 = x1 , xn+1 = x1 , xn+1 = x1 .

If, in addition,
     
1
n
xi + xi+1 xi − xi+1 2
f +C×
n 2 2
i=1
⎛    n 2 ⎞
n
j =1 xj n
1 j =1 xj
− ⎝f +C xi − ⎠ ≥ 0,
n n n
i=1
Rearrangements, L-Superadditivity and Jensen-Type Inequalities 15

then
 n   n 2

n
j =1 xj

n
j =1 xj
f (xi ) − nf −C× xi −
n n
i=1 i=1
     
1
n
xi + xi+1 xi − xi+1 2
≥ f +C×
n 2 2
i=1
⎛    n 2 ⎞
n
=1 xj n
1 =1 xj
− ⎝f ⎠.
j j
+C xi −
n n n
i=1

References

1. S. Abramovich, The increase of sums and products dependent on (y1 , . . . , yn ) by rearrange-

ment of this set. Israel J. Math. 5, 177–181 (1967)
2. S. Abramovich, Extended Heinz and Jensen type inequalities and Rearrangements, Operator
Theory, Advances and Applications, 282, 1–14 (2021)
3. S. Abramovich, L.-E. Persson, Fejer and Hermite-Hadamard type inequalities for N-
quasiconvex functions. Math. Notes 102(5), 599–609 (2017)
4. S. Abramovich, L-E. Persson, Rearrangements and Jensen type inequalities related to convex-
ity, superquadracity, strong convexity and 1-quasiconvexity. J. Math. Inequal 14(3), 641–659
(2020)
5. R.J. Daffin, A.C. Shaeffer, Refinement of an inequality of the brothers Markoff. Trans. Amer.
Math. Soc. 50, 517–528 (1941)
6. A.L. Lehman, Results on rearrangements. Israel J. Math. 1(1), 22–28 (1963)
7. G.G. Lorentz, An inequality for rearrangement. Amer. Math. Monthly 60, 176–179 (1953)
8. A.W. Marshal, I. Olkin, Inequalities: Theory of Majorization and Its Applications (Academic,
Cambridge, 1979)
9. C.P. Niculescu, L.E. Persson, Convex Functions and Their Applications: A Contemporary
Approach. CMS Books in Mathematics, 2nd edn. (Springer, Berlin, 2018)
10. Y. Rinott, Multivariate majorization and rearrangement inequalities with some applications to
probability and statistics. Israel J. Math. 15, 60–77 (1973)
Approximate Generalized Jensen
Mappings in 2-Banach Spaces

Muaadh Almahalebi, Themistocles M. Rassias, Sadeq Al-Ali,

and Mustapha E. Hryrou

Abstract Our aim is to investigate the generalized Hyers-Ulam-Rassias stability

for the following general Jensen functional equation:


n−1
f (x + bk y) = nf (x),
k=0

where n ∈ N2 , bk = exp( 2iπ n ) for 0 ≤ k ≤ n − 1, in 2-Banach spaces by

k

using a new version of Brzdȩk’s fixed point theorem. In addition, we prove some
hyperstability results for the considered equation and the general inhomogeneous
Jensen equation


n−1
f (x + bk y) = nf (x) + G(x, y).
k=0

1 Introduction and Preliminaries

In the middle of the 1960s, S. Gähler [10, 11] introduced the basic concepts of linear
2-normed spaces. In the following definitions and lemmas, we present some facts
concerning the linear 2-normed spaces.

M. Almahalebi () · S. Al-Ali · M. E. Hryrou

Department of Mathematics, Ibn Tofail University, Kenitra, Morocco
Th. M. Rassias
Department of Mathematics, Zografou Campus, National Technical University of Athens, Athens,
Greece

© Springer Nature Switzerland AG 2022 17

N. J. Daras, Th. M. Rassias (eds.), Approximation and Computation in Science
and Engineering, Springer Optimization and Its Applications 180,
https://doi.org/10.1007/978-3-030-84122-5_2
18 M. Almahalebi et al.

Definition 1 Let X be a real linear space with dimX > 1 and ., . : X × X −→

[0, ∞) be a function satisfying the following properties:
1. x, y = 0 if and only if x and y are linearly dependent,
2. x, y = y, x ,
3. λx, y = |λ| x, y ,
4. x, y + z ≤ x, y + x, z ,
for all x, y, z ∈ X and λ ∈ R. Then the function ., . is called a 2-norm on X, and
the pair (X, ., . ) is called a linear 2-normed space. Sometimes, the condition (4)
is called the triangle inequality.
Example 1 For x = (x1 , x2 ), y = (y1 , y2 ) ∈ X = R2 , the Euclidean 2-norm
x, y R2 is defined by

x, y R2 = |x1 y2 − x2 y1 | .

Lemma 1 Let (X, ., . ) be a 2-normed space. If x ∈ X and x, y = 0, for all

y ∈ X, then x = 0.
Definition 2 A sequence {xk } in a 2-normed space X is called a convergent
sequence if there is an x ∈ X such that

lim xk − x, y = 0,
k→∞

for all y ∈ X. If {xk } converges to x, write xk −→ x with k −→ ∞, and call x the

limit of {xk }. In this case, we also write limk→∞ xk = x.

Definition 3 A sequence {xk } in a 2-normed space X is said to be a Cauchy

sequence with respect to the 2-norm if

lim xk − xl , y = 0,
k,l→∞

for all y ∈ X. If every Cauchy sequence in X converges to some x ∈ X, then X

is said to be complete with respect to the 2-norm. Any complete 2-normed space is
said to be a 2-Banach space.
Now, we state the following results as lemma (see [15] for the details).
Lemma 2 Let X be a 2-normed space. Then,
 
1.  x, z − y, z  ≤ x − y, z for all x, y, z ∈ X,
2. if x, z = 0 for all z ∈ X, then x = 0,
3. for a convergent sequence xn in X,
 
 
lim xn , z =  lim xn , z
n−→∞ n−→∞

for all z ∈ X.
Approximate Generalized Jensen Mappings in 2-Banach Spaces 19

The stability problem of functional equations originated from a question of S.

M. Ulam [18] concerning the stability of group homomorphisms. D. H. Hyers [13]
gave a first affirmative partial answer to the question of Ulam for Banach spaces.
Hyers’ theorem was generalized by T. Aoki [2] for additive mappings and by Th. M.
Rassias [16] for linear mappings by considering an unbounded Cauchy difference.
A generalization of the Th. M. Rassias theorem was obtained by Găvruţa [12] by
replacing the unbounded Cauchy difference by a general control function in the
spirit of Th. M. Rassias’ approach.
Throughout this chapter, we will denote the set of natural numbers by N, the set
of real numbers by R with R+ := [0 , ∞), and the set of complex numbers by C.
By Nm , m ∈ N, we will denote the set of all natural numbers greater than or equal
to m.
Let R+ = [0, ∞) the set of nonnegative real numbers. We write B A to mean the
family of all functions mapping from a nonempty set A into a nonempty set B.
n
Definition 4 Let X be a nonempty set, (Y, d) be a metric space, ε ∈ RX
+ , and F1 ,
n n
F2 be operators mapping from a nonempty set D ⊂ Y into Y . We say that the
X X

operator equation

F1 ϕ(x1 , . . . , xn ) = F2 ϕ(x1 , . . . , xn ), (x1 , . . . , xn ∈ X) (1)

is ε-hyperstable provided that every ϕ0 ∈ D which satisfies

d (F1 ϕ0 (x1 , . . . , xn ), F2 ϕ0 (x1 , . . . , xn )) ≤ ε(x1 , . . . , xn ), (x1 , . . . , xn ∈ X)

fulfills Eq. (1).

J. Brzdȩk et al. [5] proved the fixed point theorem for a nonlinear operator in metric
spaces and used this result to study the Hyers-Ulam stability of some functional
equations in non-Archimedean metric spaces. In this work, they also obtained the
fixed point result in arbitrary metric spaces as follows:
Theorem 1 ([5]) Let X be a nonempty set, (Y, d) be a complete metric space, and
Λ : Y X → Y X be a non-decreasing operator satisfying the hypothesis

lim Λδn = 0
n→∞

for every sequence {δn }n∈N in Y X with

lim δn = 0.
n→∞

Suppose that T : Y X → Y X is an operator satisfying the inequality

   
d Tξ(x), Tμ(x) ≤ Λ Δ(ξ, μ) (x), ξ, μ ∈ Y X , x ∈ X, (2)
20 M. Almahalebi et al.

where Δ : Y X × Y X → RX
+ is a mapping which is defined by
 
Δ(ξ, μ)(x) := d ξ(x), μ(x) ξ, μ ∈ Y X , x ∈ X. (3)

If there exist functions ε : X → R+ and ϕ : X → Y such that

d (Tϕ)(x), ϕ(x) ≤ ε(x) (4)

and
 
ε∗ (x) := Λn ε (x) < ∞ (5)
n∈N0

for all x ∈ X, then the limit

 
lim (T n ϕ) (x) (6)
n→∞

exists for each x ∈ X. Moreover, the function ψ ∈ Y X defined by

 
ψ(x) := lim (T n ϕ) (x) (7)
n→∞

is a fixed point of T with

 
d ϕ(x), ψ(x) ≤ ε∗ (x) (8)

for all x ∈ X.
In 2013, Brzdȩk [4] gave the fixed point result by applying Theorem 1 as follows:
Theorem 2 Let X be a nonempty set, (Y, d) a complete metric space, and
f1 , . . . , fs : X → X and L1 , . . . , Ls : X → R+ be given mappings. Let Λ : RX
+ →
RX
+ be a linear operator defined by


s
Λδ(x) := Li (x)δ(fi (x)), (9)
i=1

for δ ∈ RX
+ and x ∈ X. If T : Y → Y is an operator satisfying the inequality
X X

  s
 
d Tξ(x), Tμ(x) ≤ Li (x)d ξ(fi (x)), μ(fi (x)) , ξ, μ ∈ Y X , x ∈ X,
i=1
Approximate Generalized Jensen Mappings in 2-Banach Spaces 21

and a function ε : X → R+ and a mapping ϕ : X → Y satisfy

d (Tϕ(x), ϕ(x)) ≤ ε(x), (x ∈ X),

∞

ε∗ (x) := Λk ε(x) < ∞, (x ∈ X),
k=0

then for every x ∈ X, the limit

ψ(x) := lim T n ϕ(x)

n→∞

exists, and the function ψ ∈ Y X so defined is a unique fixed point of T with

d (ϕ(x), ψ(x)) ≤ ε∗ (x), (x ∈ X).

In the following theorem, we extend the fixed point theorem (Theorem 2) in 2-

Banach spaces.
 
Theorem 3 Let X be a nonempty set, Y, ·, · be a 2-Banach space, g : X → Y
be a mapping such that the set g(X) ⊆ Y contains two linearly independent vectors,
and f1 , . . . , fr : X → X and L1 , . . . , Lr : X → R+ be given mappings. Suppose
that T : Y X → Y X and Λ : RX×X + → RX×X
+ are two operators satisfying the
conditions

   r      
Tξ(x) − Tμ(x), g(z) ≤  
Li (x)ξ fi (x) − μ fi (x) , g(z) (10)
i=1

for all ξ, μ ∈ Y X , x, z ∈ X and


r
 
Λδ(x, z) := Li (x)δ fi (x), z , δ ∈ RX×X
+ , x, z ∈ X. (11)
i=1

If there exist functions ε : X × X → R+ and ϕ : X → Y such that

 
 
Tϕ(x) − ϕ(x), g(z) ≤ ε(x, z) (12)

and
∞
  n 
ε∗ (x, z) := Λ ε (x, z) < ∞ (13)
n=0
22 M. Almahalebi et al.

for all x, z ∈ X, then the limit

 
lim (T n ϕ) (x) (14)
n→∞

exists for each x ∈ X. Moreover, the function ψ : X → Y defined by

 
ψ(x) := lim (T n ϕ) (x) (15)
n→∞

is a fixed point of T with

 
ϕ(x) − ψ(x), g(z) ≤ ε∗ (x, z) (16)

for all x, z ∈ X.
There are two other versions of Theorem 3 in 2-Banach space given in [1] and
[6]. In addition, Brzdȩk et al. [7] gave important related results in generalized metric
spaces. In this chapter, we discuss the generalized Hyers-Ulam-Rassias stability
problem for the following generalization of Jensen functional equation:


n−1
f (x + bk y) = nf (x), (17)
k=0

where n ∈ N2 and bk = exp( 2iπ n ) for 0 ≤ k ≤ n − 1, in 2-Banach spaces by using

k

Theorem 3 as a basic tool. The general solution and stability of this equation and its
generalizations were studied by numerous researchers; see, for example, [3, 8, 9, 14]
and [17].

2 Main Results

Let X be a complex normed space. We will denote by Aut (X) the family of all
automorphisms of X. Moreover, for each u ∈ XX , we write ux := u(x) for x ∈ X,
and we define u by u x := x − ux for x ∈ X.
The following theorem is the main result concerning the stability of the functional
equation (17).
3
Theorem 4 Let X be a C-normed space, Y be a 2-Banach space, ε : X \ {0} →
R+ , and

l(X) := u ∈ Aut (X) : u , (u + bk u) ∈ Aut (X),


n−1 
αu := nλ(u ) + λ(u + bk u) < 1 = ∅ (18)
k=1
Approximate Generalized Jensen Mappings in 2-Banach Spaces 23

where
 
λ(u) := inf t ∈ R+ : ε(ux, uy, z) ≤ tε(x, y, z), ∀x, y, z ∈ X \ {0}

for all u ∈ Aut (X). Assume that f : X −→ Y satisfies the inequality

 
 
n−1 
 
f (x + y) − nf (x) + f (x + bk y) , g(z) ≤ ε(x, y, z) (19)
 
k=1

for all x, y, z ∈ X \ {0} such that x + bk y = 0 for 0 < k < n − 1, where g : X → Y

is a mapping such that the set g(X) ⊆ Y contains two linearly independent vectors.
Then, for each nonempty subset U ⊂ l(X) such that

u ◦ v = v ◦ u, ∀u, v ∈ U, (20)

there exists a unique function J : X −→ Y which satisfies Eq. (17) and

f (x) − J (x) , g(z) ≤ ε̃(x, z) x, z ∈ X \ {0}, (21)

where
 ε(u x, ux, z) 
ε̃(x, z) := inf :u∈U x, z ∈ X \ {0}.
1 − αu

Proof Write X0 := X \ {0}, and let us fix u ∈ U. Replacing x with u x and y with
ux in (19), we get
 
 
n−1
  
 
f (x) − nf (u x) + f (u + bk u)x , g(z) ≤ ε(u x, ux, z) := εu (x, z)
 
k=1
(22)

0 ×X0
for all x, z ∈ X0 . We define the operators Tu : Y X0 → Y X0 and Λu : RX
+ →
X0 ×X0
R+ by


n−1
 
Tu ξ(x) := nξ(u x) − ξ (u + bk u)x , (23)
k=1

  n−1
Λu δ(x, z) := nδ u x, z + δ (u + bk u)x, z
k=1
24 M. Almahalebi et al.

0 ×X0
for all x, z ∈ X0 , ξ ∈ Y X0 and δ ∈ RX
+ . Then the inequality (22) becomes
 
 
f (x) − Tu f (x) , g(z) ≤ εu (x, z)

for all x, z ∈ X0 and u ∈ U.

The operator Λu has the form given by (11) with s = n and fn (x) = u x,
fi (x) = (u + bi u)x, Ln (x) = n, Li (x) = 1, i ∈ {1, 2, . . . , n − 1} for all x ∈ X0
and u ∈ U.
Further,

   
n−1
 
  
Tu ξ(x) − Tu μ(x) , g(z) = nξ(u x) − ξ (u + bk u)x − nμ(u x)
k=1


n−1
  

+ μ (u + bk u)x , g(z)
k=1
 
 
≤ nξ(u x) − μ(u x) , g(z)
n−1 
 
     
+ ξ (u + bk u)x − μ (u + bk u)x , g(z)
k=1

for all x, z ∈ X0 , u ∈ U, and ξ, μ ∈ Y X0 .

Note that, in view of the definition of λ(u),

ε(ux, uy, z) ≤ λ(u)ε(x, y, z), x, y, z ∈ X0 .

By mathematical induction on s ∈ N, it is easy to show that

Λsu εu (x, z) ≤ αus ε(u x, ux, z),

for all x, z ∈ X0 and all u ∈ U, where


n−1
αu = nλ(u ) + λ(u + bk u).
k=1

Hence,
∞
 ∞
 ε(u x, ux, z)
ε∗ (x, z) := Λru εu (x, z) ≤ ε(u x, ux, z) αur = < ∞ (24)
1 − αu
r=0 r=0
Approximate Generalized Jensen Mappings in 2-Banach Spaces 25

for all x, z ∈ X0 . According to Theorem 3, there exists a unique solution Ju : X →

Y of the equation


n−1
 
Ju (x) = nJu (u x) − Ju (u + bk u)x (25)
k=1

for all x ∈ X0 and all u ∈ U, which is a fixed point of Tu such that

  ε(u x, ux, z)
 
Ju (x) − f (x) , g(z) ≤ (26)
1 − αu

for all x ∈ X0 and all u ∈ U. Moreover,

Ju (x) = lim Tru f (x)

r→∞

for all x ∈ X0 and all u ∈ U.

To prove that Ju satisfies the functional equation (17) on X0 , we just prove the
following inequality:

 
n−1 
 r 
T
 u f (x + y) − nTu
r
f (x) + Tur f (x + bk y) , g(z) ≤ αur ε(x, y, z) (27)
k=1

for all r ∈ N, all u ∈ U and all x, y, z ∈ X0 such that x + bk y = 0.

Indeed, if r = 0, then (27) is simply (19). So, take r ∈ N1 , and suppose that (27)
holds for r and x, y, z ∈ X0 . Then, by using (23) and the triangle inequality, we
have

 
n−1 
 r+1 
Tu f (x + y) − nTur+1 f (x) + Tur+1 f (x + bk y) , g(z)
k=1

 
n−1

= nTu f u (x + y) + nTu f bk u(x + y) −
r r
Tur f (u + bk u)(x + y)

k=1


n−1
−n2 Tur f (u x) + n Tur f (u + bk u)x
k=1
  

n−1 
n−1 

+ nTru f u (x + bk y) − Tru f (u + bk u)(x + bk y) , g(z)

k=1 k=1
 
 
n−1 
 r 
≤ n Tu f u (x + y) − nTu f (u x) +
r
Tu f u (x + bk y) , g(z)
r
 
k=1
26 M. Almahalebi et al.

n−1 
  r
− Tu f u + bk u)(x + y) − nTur f (u + bk u)x
k=1


n−1 

+ Tur f (u + bk u)(x + bk y) , g(z)

k=1
 

n−1
≤ αur nε(u x, u y, z) + ε (u + bk u)x, (u + bk u)y, z
k=1
 

n−1
≤ αur nλ(u ) + λ(u + bk u) ε(x, y, z)
k=1

= αur+1 ε(x, y, z).

By induction, we have shown that (27) holds for all r ∈ N, all u ∈ U, and all
x, y, z ∈ X0 such that x + bk y = 0. Letting r → ∞ in (27), we get


n−1
Ju (x + bk y) = nJu (x)
k=0

for all x, y ∈ X0 such that x + bk y = 0 for 0 ≤ k ≤ n − 1. Thus, we have proved

that for every u ∈ U, there exists a function Ju : X0 → Y which is a solution of the
functional equation (17) on X0 and satisfies
  ε(u x, ux, z)
 
f (x) − Ju (x) , g(z) ≤
1 − αu

for all x, z ∈ X0 . Next, we prove that each solution J : X → Y of (17) satisfying

the inequality

f (x) − J (x) , g(z) ≤ L ε(v x, vx, z), x, z ∈ X0 (28)

with some L > 0 and v ∈ U is equal to Jw for each w ∈ U. So, fix v, w ∈ U, L > 0
and J : X → Y a solution of (17) satisfying (28). Note that, by (26) and (28), there
is L0 > 0 such that

J (x) − Jw (x) , g(z) ≤ J (x) − f (x) , g(z) + f (x) − Jw (x) , g(z)

∞
  
≤ L0 ε(v x, vx, z) + ε(w x, wx, z) · r
αw (29)
r=0

for all x, z ∈ X0 . In other side, J and Jw are solutions of (25) because they satisfy
(17).
Approximate Generalized Jensen Mappings in 2-Banach Spaces 27

We show that, for each j ∈ N,

J (x) − Jw (x) , g(z)

∞
  
≤ L0 ε(v x, vx, z) + ε(w x, wx, z) · r
αw , (x, z ∈ X0 ). (30)
r=j

The case j = 0 is exactly (29). So fix γ ∈ N0 , and assume that (30) holds for
j = γ . Then, in view of definition of λ(u),

   
n−1
 
J (x) − Jw (x) , g(z) = 
nJ (w x) − J (w + bk w)x
k=1


n−1
  

−nJw (w x) + Jw (w + bk w)x , g(z)
k=1
 
≤ nJ (w x) − Jw (w x) , g(z)

n−1
     
+ J (w + bk w)x − Jw (w + bk w)x , g(z)
k=1
∞
  
≤ n L0 ε(v w x, vw x, z) + ε(w w x, ww x, z) · r
αw
r=γ


n−1
 
+L0 ε v (w + bk w)x, v(w + bk w)x, z
k=1
∞
  
+ε w (w + bk w)x, w(w + bk w)x, z · r
αw
r=γ


n−1 ∞

 
≤ L0 ε(v x, vx, z) + ε(w x, wx, z) nλ(w ) + λ(w + bk w) · r
αw
k=1 r=γ
∞

= L0 ε(v x, vx, z) + ε(w x, wx, z) · r
αw .
r=γ +1

Hence, we have shown (30). Now, letting j → ∞ in (30), we get

J (x) = Jw (x) ∀x ∈ X0 . (31)

28 M. Almahalebi et al.

By similar method, we also prove that Ju = Jw for each u ∈ U, which yields

ε(u x, ux, z)
f (x) − Jw (x) , g(z) ≤ x, z ∈ X0 , u ∈ U.
1 − αu

This implies (21) with J := Jw and the uniqueness of J is given by (31).

In the following theorem, we prove the hyperstability of Eq. (17) in 2-Banach
spaces.
Theorem 5 Let X, Y , and ε be as in Theorem 4. Suppose that there exists a
nonempty set U ∈ l(X) such that u ◦ v = v ◦ u ∀u, v ∈ U and

infu∈U ε(u x, ux, z) = 0 ∀x, z ∈ X0 , ∀u ∈ U
(32)
supu∈U αu < 1.

Then every f : X → Y satisfying (19) is a solution of (17).

Proof Suppose that f : X → Y satisfies (19). Then, by Theorem 4, there exists a
mapping J : X → Y which satisfies (17) and

f (x) − J (x) , g(z) ≤ ε̃(x, z) ∀x, z ∈ X0 .

In view of (32), ε̃(x, z) = 0 ∀x, z ∈ X0 . This means that f (x) = J (x) ∀x ∈ X0 ,

whence


n−1
f (x + bk y) = nf (x),
k=0

for all x, y ∈ X0 such that x + bk y = 0 for 0 < k < n − 1 which implies that f
satisfies the functional equation (17) on X0 .

3 Applications

In this section, we discuss some hyperstability results for Eq. (17) and the inhomo-
geneous functional equation


n−1
f (x + bk y) = nf (x) + G(x, y). (33)
k=0

Namely, from Theorems 4 and 5, we can obtain the following corollaries as natural
results.
Approximate Generalized Jensen Mappings in 2-Banach Spaces 29

Corollary 1 Let X, Y , and ε be as in Theorem 4 and G : X2 → Y . Suppose that

 
G(x, y) , g(z) ≤ ε(x, y, z), x, y, z ∈ X0 , (34)

where g : X → Y is a mapping such that the set g(X) ⊆ Y contains two linearly
independent vectors, G(x0 , y0 ) = 0 for some x0 , y0 ∈ X0 , and there exists a
nonempty U ⊂ l(X) such that (20) and (32) hold. Then the inhomogeneous equation


n−1
f (x + bk y) = nf (x) + G(x, y) (35)
k=0

for all x, y ∈ X0 such that x + bk y = 0 for 0 < k < n − 1 has no solutions in the
class of functions f : X → Y .
Proof Suppose that f : X → Y is a solution to (35). Then
n−1 
   
 
 f (x + bk y) − nf (x) , g(z) = nf (x) + G(x, y) − nf (x) , g(z)
 
k=0
 
= G(x, y) , g(z)
≤ ε(x, y, z),

for all x, y, z ∈ X0 such that x + bk y = 0 for 0 < k < n − 1. Consequently, by

Theorem 5, f is solution of (17). Therefore,


n−1
G(x0 , y0 ) = f (x0 + bk y0 ) − nf (x0 ) = 0,
k=0

which is contradiction.
Corollary 2 Let X, Y , and ε be as in Theorem 4. Assume that G : X2 → Y and
f : X → Y satisfy the inequality
 
 
n−1 
 
f (x + y) − nf (x) + f (x + bk y) − G(x, y) , g(z) ≤ ε(x, y, z) (36)
 
k=1

for all x, y, z ∈ X \ {0} such that x + bk y = 0 for 0 < k < n − 1, where g : X → Y

is a mapping such that the set g(X) ⊆ Y contains two linearly independent vectors
and there exists a nonempty U ⊂ l(X) such that (20) and (32) hold. If the functional
equation (35) has a solution f0 : X → Y , then f is a solution to (35).
30 M. Almahalebi et al.

Proof In view of (36), we obtain that J := f − f0 satisfies (19). According to

Theorem 5, we conclude that J is a solution to Eq. (17). Therefore,


n−1 
n−1 
n−1
f (x + bk y) − nf (x) − G(x, y) = J (x + bk y) + f0 (x + bk y)
k=0 k=0 k=0
−nJ (x) − nf0 (x) − G(x, y)
= 0,

for all x, y ∈ X0 such that x + bk y = 0 for 0 < k < n − 1 which means that f is a
solution to (35).
Corollary 3 Let X and Y be a C-normed space and a 2-Banach space, respectively.
Assume that p, q ∈ R, p < 0, q < 0 and θ, r ≥ 0. If f : X → Y satisfies
n−1 
 
 
 f (x + bk y) − nf (x) , g(z) ≤ θ x p
+ y q
z r
(37)
 
k=0

for all x, y, z ∈ X0 such that x + bk y = 0, then f satisfies the functional equation

(17) on X0 .
Proof The proof follows from Theorem 5 by taking

ε(x, y, z) = θ x p
+ y q
z r, x, y, z ∈ X0 ,

with some real numbers θ, r ≥ 0, p < 0, and q < 0. For each m ∈ N, define
um : X0 → X0 by um x := −mx and um : X0 → X0 by um x := (1 + m)x. Then

ε(um x, u y, z) = ε(−mx, −y, z)

=θ −mx p
+ −y q
z r

= θ mp x p
+ q y q
z r

 
≤ mp + q θ x p + y q
z r

 
= mp + q ε(x, y, z)

for all x, y, z ∈ X0 such that x + bk y = 0 with , m ∈ N. Hence,

lim ε(um x, um y, z) ≤ lim (1 + m)p + mq ε(x, y, z)

m→∞ m→∞
=0
Approximate Generalized Jensen Mappings in 2-Banach Spaces 31

for all x, y, z ∈ X0 such that x +bk y = 0. Then (32) is valid with λ(um ) = mp +mq
for m ∈ N, and there exists n0 ∈ N such that m ≥ n0 and


n−1
   
n (1 + m)p + (1 + m)q + 1 + m + bk mp + 1 + m + bk mq < 1.
k=1

So, it easily seen that (18) is fulfilled with

 
U := um ∈ Aut X : m ∈ Nn0 .

Therefore, by Theorem 5, every f : X → Y satisfying (37) is a solution of the

functional equation (17) on X0 .
Corollary 4 Let X and Y be a C-normed space and a 2-Banach space, respectively.
Assume that p, q ∈ R, p + q < 0, and θ, r ≥ 0. If f : X → Y satisfies
n−1 
 
 
 f (x + bk y) − nf (x) , g(z) ≤ θ x p
y q
z r
 
k=0

for all x, y, z ∈ X0 such that x + bk y = 0 for 0 < k < n − 1, then f satisfies the
functional equation (17) on X0 .
Proof It is easily seen that the function ε given by

ε(x, y, z) = θ x p
y q
z r
x, y, z ∈ X\{0}

satisfies (32) and

ε(mx, y, z) = θ mx p
y q
z r

= θ mp q x p
y q
z r

= mp q ε(x, y, z)

for all x, y, z ∈ X0 such that x + bk y = 0 with , m ∈ N and m = 0.

The remainder of the proof is similar to the proof of Corollary 3.
By an analogous conclusion, the function ε given by

ε(x, y, z) = θ x p
+ y q
+ x p
y q
z r
x, y, z ∈ X0 ,
Other documents randomly have
different content
a warehouse on land to facilitate the passage of the floating
warehouses of their vessels” according to a contemporary report.

We can visualize the busy river front at this point on the east bank,
with merchants plying back and forth in skiffs, and larger batteaux
bringing loads of the high quality English goods, including cloth,
cutlery, farming utensils—and slaves. New Orleans businessmen who
dared not indulge in this illicit trade protested to the Spanish
authorities who continued to wink at the British until the American
revolution made anti-British sympathies fashionable. Then Bernardo
Galvez, the Spanish governor, on April 17, 1777, seized the English
boats and warehouses, and the illegal trade stopped. But it is safe to
assume that the commercial identity of what became Lafayette’s
riverfront persisted, perhaps even its later competition with New
Orleans having begun in that early illicit trade.

The good business prospects in Lafayette caused men to say that the
city might possibly “capture” New Orleans some day. The Texas cattle
trails ended their long treks over the prairies at Gretna. Cattle boats
brought the Stock across the river to the slaughter houses with
special landings for the pounding, snorting beasts, skittering down
the gangplanks into the pens, as if entirely cognizant that the end of
the trail had come now for certain.

If one could stand the smell, it was a curious and exciting vista
watching the steamboats unload cattle at Lafayette. Satellite
industries lined the river front, the tallow renderers, the soap-boilers,
the hide merchants, the tanners, the disposers of unused parts, and
the bone grinders, whose odoriferous products made tasty 13
vegetables and sweet sugar cane grow, paradoxical as that
seemed.

King Cotton considered the wharves at Lafayette one of his royal

ports. The puffing, graceful white swans of the Mississippi began to
nudge the rows of flatboats from the river front. In the Forties the
city council decreed exclusive facilities between First and Second
Streets for steamboats, with inclined landing areas. The city supplied
heavy, thick planks, bound on each end with iron bands and proudly
branded “L”, for the use of the steamboats calling at its wharves. By
1850 twenty piers had been constructed to accommodate the
packets.

The “breadbasket” of the upper valley dumped its cargoes at

Lafayette wharves also. The first grain elevator on the lower river was
built at the foot of Harmony Street. Flour was an important
commodity, with busy factors waiting to trade as the clumsy
broadhorns, floating on the current, edged into their moorings. So
crowded were the flatboat moorings—above Second Street—that
flatboat captains received a $10 fine for not immediately removing
the large steering oars on each side. Twenty-four hours was the time
limit for unloading.

At first, 80-foot sections of the river front had been set aside for the
flatboats, once delivered of their cargoes, to be broken up. But this
became a nuisance, and by 1845, it was forbidden along the entire
Lafayette river front.

There was always a ready market for the timbers from broken up
flatboats, or “gunwales”, as the long, heavy fore-and-aft planks were
called. Many of the early houses were built of these excellent,
weathered timbers from the virgin forests of the upper valley. Most of
the streets of Lafayette, until after the mid-nineteenth century, were
“paved” with them, as were the sidewalks, or “banquettes”. The long
boards would not disappear so quickly into the mud, as would rocks
and bricks. Numerous cottages remaining in various sections of New
Orleans near the river are built of these sturdy, enduring timbers.

Lafayette City was a complete entity in every respect, with the

exception of a bank. There was a branch of the Carrollton Bank to
serve its citizens, at the corner of Jackson and Levee, but it was not
their own. No doubt a Bank of Lafayette was high on the list of these
enterprising citizens, when annexation took place.
Row of buildings at Seventh and Laurel in 1866. Second from right is
tannery. Note bridges over deep gutters. Some are still in use.

James H. Caldwell, the theatrical impresario, entrepreneur, the man

most responsible for the development of the “American” section, the
former Faubourg Ste. Marie, left his enterprising mark on Lafayette.
In 1847, the City Council granted him the sole right of vending gas
lighting under the name of the Lafayette Gas Light Company. 14
He could lay pipes and conduits at the company’s expense in
the city streets. For the privilege, the company had to supply gas to
public lamps throughout the city as well as in public buildings at
special rates. The first home reported to have gas illumination was
that of Mr. E. S. Miles on Nayades (St. Charles) between Sixth and
Seventh.
 This unique building housed the Sixth Precinct on Rousseau near
Jackson, on site of earlier Jefferson Parish courthouse and prison.
Egyptian style building still stands, now serves as city sign shop.

Lafayette was the scene of a celebrated legal case involving large and
valuable sections of the city, second in local court annals only to the
Gaines litigation. The original name of the faubourg which later
became known as Lafayette was Faubourg Panis, after its owner, the
Widow Panis. She first had this property subdivided into lots and
streets. Her daughter Mme. Rousseau, a widow, inherited the
faubourg. In 1818 she sold the remaining property for $100,000 to
John Poultney, who died before he could pay for it. His creditors, who
had advanced him part of the money to make the purchase, paid the
notes and proceeded to sell lots in honest belief of clear title.
Poultney’s wife, on behalf of herself and her minor children, had
renounced their rights to the property. The name of the faubourg was
changed at that time to Lafayette, in honor of the French patriot who
had visited New Orleans.
Later the Poultney heirs claimed that their tender age and legal
incapacity prevented them from accepting the property at the time of
the succession. The suit was instigated in 1832 and rambled through
the courts until 1855 when the United States Supreme Court upheld
the Louisiana tribunal against the plaintiffs. Interestingly, one of the
disappointed claimants was the then Major G. T. Beauregard,
engineer in charge of construction of the New Orleans Customhouse.
Mrs. John Poultney had been Emilie Toutant-Beauregard.

The city was not without its theatrical attractions. The only actual
theater built for stage presentations was the Lafayette on Rousseau
between Philip and Soraparu Streets. It was in the center of the
block, on the lake side, directly across from Terpsichore Hall, the
favorite salle à danser. The theater opened in late December, 1848,
and according to the Statesman, “has already increased property
values near it.” It had 100 feet of depth, was 55 feet wide, 40 15
feet high, and its stage was 35 feet deep, said the newspaper
account. Sol Smith, one of the pioneer actors who penetrated the
“frontier” communities from the East, along with Noah Ludlow and
his troupe, played the Lafayette and left this comment, dated
February, 1849:

“A theater in Lafayette, a suburb of New Orleans, was opened under

the management of Mr. Oliver this season. The prevalence of the
cholera blighted any prospects there might have been of success.
This company was composed principally of new beginners and their
salaries were paid in various commodities, such as the manager
stipulated to receive of the citizens for tickets. It was a stipulation in
each article of agreement (so the manager told me) that every actor
should take a portion of his salary in coffins, should he need any!—
that is to say, if he should die during the season, he should be buried
on account; the style of coffin, number of carriages, and so forth, to
be regulated by the amount due at the time of his demise.

“I had a fellow feeling for this manager, and when he asked me to act
one night for him, assuring me that I could fill the house at double
prices, I could not refuse him, though I doubted very much whether
my acting would add anything to his receipts. Manager Oliver was
right, however, and I had the pleasure of playing the Mock Duke in
the Honey Moon to one of the most crowded audiences I have ever
acted to. Of course, under the circumstances, I would take no pay for
my night’s services, though the grateful manager offered me a clear
half of the receipts.

“The season failed totally, the manager left for parts unknown and
next season, after a vain attempt by one Hickey to resuscitate the
drama by presenting some horrible representations (or
misrepresentations rather) of Yankee character, the theater took fire
one day and was burned to the ground. Lafayette is too near New
Orleans to give an efficient support to the theater.”

The earliest homes in Lafayette naturally were built on streets closest

to the river. As early as 1842, the crusty editor of the Daily Picayune
in New Orleans was rhapsodic over the beautiful cottages in Lafayette
City with their handsome architecture and lovely gardens. The
Lafayette Spectator, by 1850, was equally enthusiastic. “The City of
Lafayette”, wrote John McMillin, “at no previous time could boast of
so many valuable buildings in progress as at present. Styles, finishes
and materials being so vastly improved.” The cottages were becoming
mansions at this point, getting away from the flatboat gunwales.

“Such is the demand for lots,” continued McMillin, “in the back part of
the city that they are selling for nearly double the price of those three
or four squares from the river. Lots on or near the railroad (St.
Charles Avenue) sell for $1,800. Those on Jersey (Annunciation),
$800 or $900. Cheapest lots are on Jersey and Laurel Streets.

“To become independent here,” he advised, “it is necessary to

purchase a few lots only, at a low rate and keep them a few years
when the fortunate owner finds himself well off in the world. We
believe for the next five years real estate will increase 20% per year.”
In 1852, Lafayette City counted a population of 12,651 with 1,539
slaves added. This was short of the anticipated total because the city
had just come through a particularly devastating yellow fever
epidemic in which some 2,000 souls had been lost. One journalist felt
that the census takers had not been thorough in their tally. 16

The city burial grounds, the Lafayette Cemetery on Washington

between Coliseum and Prytania, laid out in 1833, was hard put to
find space for the bodies. Apparently most of the deceased were part
of the huge drifting population, newly arrived immigrants, the
flatboatmen and others, for only 389 citizens of Lafayette could be
accounted for by the census taker among those buried in the
cemetery.

The prospects of the Garden District of Lafayette were also favorably

mentioned by the editor of the Spectator, a Whig newspaper
published there during the mid-century era:

“It is already the seat of fashionable residences. The property in the

rear of the district has been greatly sought by merchants and bankers
and professional men. Little or none has been held for speculation. It
will maintain its value.”

The appearance of the Greek Revival mansions now rising from the
vicinity of Nayades and in toward Magazine supported the claim that
at this time, New Orleans had more per capita wealth than New York.

It appears from the best sources that the first house of consequence
to be built in what is now the Garden District proper was that of
Thomas Toby in 1838. He came from Philadelphia, and his father’s
ships brought some of the materials for the house from his native
city. This house is still standing at Prytania and First Streets. Others
who built in the same general neighborhood in the following decade
were F. B., T. B., and Charles Conrad, P. N. Wood, Judge R. F. Ogden,
Captain Thomas Ivey, and Charles Briggs. The Fifties saw the
greatest activity of construction of the great houses of the Garden
District.
This section, peopled as it was chiefly by those who conducted their
businesses in New Orleans, but who enjoyed the shaded gardens for
their residences, rapidly developed a life apart from the teeming
waterfront of Lafayette City. Inevitably a rivalry began. The catalyst
was the constant annoyance of pounding hooves and the odors of
the slaughterhouses and tanneries. Eventually, the early residents of
the Garden District were instrumental in getting the Lafayette City
Council to pass restrictive measures which removed the cattle
landing. The important commerce of this trade moved upriver to the
neighboring town of Jefferson, which caused the fiery editor of the
Spectator, the outspoken champion of the city’s growth, to howl from
his columns that the city had lost a million and a half dollars in trade
a year by this act.

He belabored particularly the aldermen from the “rear of the city”

who “turned up the whites of their eyes and stopped their delicate
noses as they passed by with their white gloves on and exclaimed,
‘What a nuisance!’”

Tomb of members of Jefferson Fire Company No. 22 in Lafayette

Cemetery. “Ready at the First Sound” was motto. Relief shows 1832
engine.

Despite these earmarks of a brewing donnybrook, one couldn’t 17

exactly blame the owners of the fine gardens for objecting to
“the great numbers of horses and mules running at large, particularly
at night, occupying the sidewalks to the danger of the passers-by and
racing up and down streets, disturbing the rest of the families.” Not
only horses and mules, but goats, too! “If a gate is left open for a
minute, choice rose bushes suffer and the rare plants of the most
careful training are ruined. Our feed stores are compelled to keep an
extra clerk to protect the corn sacks and bales of hay from these bold
plunderers.” At least there was a law passed in 1841 which prohibited
the keeping of bears in Lafayette City.

One of Garden District’s great houses was the Stauffer mansion on

Jackson, corner of Prytania, shown in Archives drawing dated 1870.

In the early days of Lafayette and the Garden District, “the war”
referred only to the very real and fresh memories of Jackson’s battle
with the British at Chalmette hardly twenty years previously. In the
Forties, it referred to the War with Mexico, in which many Lafayette
citizens took part. Troops were encamped and trained on some of the
vacant lots. The Rev. Jerome Twichell held Presbyterian services for
them, and a government warehouse near the river on Washington
dispensed supplies to troops coming down the Mississippi for Mexican
service.

Then, in 1861, “the war” took on a present and terrifying meaning.

Although no actual fighting was reported in or around the Garden
District, the coming and going of troops, the warships passing on the
river, the shortages because of the blockade, the restrictions of the
occupation, the loss of sons, brothers, husbands and fathers, the
surrender of proud homes for quartering of Union officers—these and
other tangible evidences left deep scars.

Perhaps the Garden District’s most distinguished and colorful figure to

wear the gray uniform was Bishop Leonidas Polk, bishop of the
Episcopal Diocese of Louisiana and rector of Trinity Church. A West
Point graduate, he answered the call to the priesthood in 1831. When
war came, after repeated urgings from his former West Point
classmate, Jefferson Davis, he “buckled the sword over the gown”, as
he phrased it, and accepted a commission as major general. In June,
1864, while he was reconnoitering near Etowah, Ga., this gallant
figure was stilled by a cannon ball, leaving memories at Trinity which
persist to this day.

After the war two other prominent figures in the Confederacy were
closely associated with the Garden District. Jefferson Davis often
visited his friend Judge Charles Fenner and died in the Fenner 18
house on the corner of First and Camp Streets. General John B.
Hood, “The Gallant Hood”, had his family home on the corner of Third
and Camp.
Bishop Leonidas Polk of Trinity Church became Confederate general.

Calvary Episcopal Church’s resolute minister, the Rev. John Fulton,

was one of the three Episcopal ministers who defied General
Benjamin “Silver Spoons” Butler. In morning prayer this trio omitted
the prayer for the President of the United States and all in civil
authority. They instead invited their congregations to join in silent
prayer. This enraged Butler, and after several verbal altercations with
them, he exiled the group to a New York prison.

Butler quartered officers in several Garden District houses, including

that of General Wirt Adams on Chestnut and Josephine, now owned
by Trinity Church and called Copeland House. For his own use Butler
cast his covetous eye on the fabulous Washington Avenue “Italian
villa” built by James Robb and later owned and occupied by John
Burnside, wealthy merchant and planter. It is told that Butler and his
retinue approached the front door, to be met by Burnside. The Union
general not only was refused use of the house but was not even
admitted. And the refusal stuck. The reason: Burnside was a British
citizen. So Butler took the lovely home of Confederate (late U.S.A.)
General David E. Twiggs, on Camp near Calliope Street, which still
stands today as St. Theresa’s school.

The homes of the elite attracted their share of celebrities to the

hospitable, high-ceilinged drawing rooms and parlors, and to the
dining tables so immaculately set and served with viands to please a
nabob. Culture, travel and education were hallmarks of most of the
inhabitants of the great houses. Delightful, spirited discussions on a
wide variety of subjects kept visiting authors, poets, artists and
correspondents for the eastern magazines enthralled.

19

House erected in 1860, said to have been built by James Robb for his
daughter, at corner Washington, Camp.
 House as it appears today, with galleries, ironwork added in 1870’s.

Strangely, these distinguished writers and authors went back to their

offices in the East and proceeded to turn out bales of copy about
New Orleans but with only side references to the Garden District.
Passing mention was made of the luxury and beauty of the homes of
this area, but the French Quarter was the subject of all the sketches
and engraved illustrations. Rare is the surviving sketch, tintype or
glass plate photo of amateur or professional.
Yet the district captivated the earliest of many visitors who put their
sentiments down for posterity. The Rev. Theodore Clapp, beloved
parson of the mid-nineteenth century, wrote of his arrival in 1822,
before Lafayette was so named:

“On a beautiful morning near the close of February we were landed

at Lafayette where the boat stopped to discharge a part of her cargo,
about three miles above New Orleans. The passengers, impatient of
delay, concluded to walk to the city. Leaving the levee, we took a
circuitous route through unenclosed fields, which a few years before
had belonged to a large sugar plantation. They were adorned with a
carpet of green grass, where herds and flocks grazed in common.
Here and there we passed a farm house in the midst of gardens,
luxuriant shrubbery and orange groves.... The air was cool, inspiring
and scented with the flowers of early spring. The music of the thrush
and various other species of singing birds, saluted our ears with their
sweetest notes. All things, so far as our eyes could reach, seemed
like a paradise. These suburbs, then so radiant with rural charms, are
now the site of a large portion of the buildings belonging to New
Orleans.”

Walt Whitman, a writer for the New Orleans Crescent in 1848, 20

living on Washington Street near the river and travelling to and
from his desk in New Orleans by omnibus, must have been impressed
by the large live oak trees in Lafayette City. In a later edition of
“Leaves of Grass”, he refers to the live oak as “rude, unbending, lusty,
made me think of myself.”

Commenting on the ways one kept cool in summer, Julian Ralph, in

Dixie, or Southern Scenes and Sketches, related:

“... when I rode through the Garden District—the new part of the
town—my lady friends pointed to the galleries and said: ‘You should
see them in the summer, before the people leave or after they come
back. The entire population is out-of-doors in the air, and the galleries
are loaded with women in soft colors, mainly white. They have white
dresses by the dozen. They go about without their hats, in carriages
and in street cars, visiting up and down the streets. In-doors, one
must spend one’s whole time and energy in vibrating a fan.’”

Writing of the Garden District, Mark Twain said: “All the dwellings are
of wood ... and all have a comfortable look. Those in the wealthy
quarter are spacious; painted snowy white, usually, and generally
have wide verandas, or double-verandas, supported by ornamental
columns. These mansions stand in the center of large grounds and
rise, garlanded with roses, out of the midst of swelling masses of
shining green foliage and many-colored blossoms. No houses could
well be in better harmony with their surroundings, or more pleasing
to the eye, or more home-like and comfortable-looking....”

“The galleries are loaded with women in soft colors, mainly white.”
George Washington Cable, who is credited with introducing the
French Quarter’s charms to the world, was born on Annunciation
Square, just below Lafayette City and later grew up and spent many
years in various homes in the Garden District. Cable was
internationally celebrated in his day for his Creole stories. His house
on Eighth Street, between Chestnut and Coliseum, still standing
today, was a mecca for visiting authors. Public education had its start
in Lafayette City shortly before it was started in New Orleans.
However, Cable gives a delightful glimpse of the wild carefree
youngsters of Lafayette in the 1830’s before the free 21
educational institutions were established:
 Mark Twain and George W. Cable posed for this picture during a
lecture tour. Twain wrote of evening at Cable’s Eighth Street home.
“... The mass of educable youth—the children who played ‘oats, peas,
beans’ with French, German and Irish accents, about the countless
sidewalk doorsteps of a city of one and two-story cottages (it was
almost such); the girls who carried their little brothers and sisters on
one elbow and hip and stared in at weddings and funerals; the boys
whose kite-flying and games were full of terms and outcries in
mongrel French, and who abandoned everything at the wild clangor
of bells and ran to fires where volunteer firemen dropped the hose
and wounded and killed each other in pitched battles; the ill-kept lads
who risked their lives daily five months of the year swimming in the
yellow whirlpools of the Mississippi among the wharves and flat-
boats, who, naked and dripping, dodged the dignified police that
stalked them among the cotton bales, who robbed mocking-birds’
nests and orange and fig trees, and trapped nonpareils and cardinals,
orchard-orioles and indigo-birds in the gardens of Lafayette and the
suburban fields—these had not been reached and had not been
sought by the educator.”

Visualize Twain, Cable and Charles Dudley Warner of Harper’s

Magazine at Cable’s Eighth Street home. Add Lafcadio Hearn and Joel
Chandler Harris for very good measure. You have the principals of a
scene which actually took place, well documented by Cable’s children
who were also present as youngsters, and described delightfully by
Mark Twain in Life on the Mississippi. Briefly, Twain, Warner and
Hearn had come to join the host in welcoming the famous “Uncle
Remus”. A literary evening ensued, but to the dismay of the 22
children, not only was “Uncle Remus” white, but he didn’t talk
the dialect of which he was the undisputed master. Harris was so very
shy that Twain read the “Tar Baby” for him to assuage the feelings of
the disappointed youngsters. Then the authors read from their own
works; Cable played his guitar and sang his celebrated Creole songs.
Twain’s amusing passage describing the scene has an equally
humorous sketch showing himself reading while the others are sound
asleep.
 Lovely raised cottage on Eighth Street was Cable’s home, scene of
many literary gatherings during late 19th century in New Orleans.

Great sports figures knew the Garden District. The Southern Athletic
Club, at Washington Avenue and Prytania Street, now Behrman
Gymnasium, was a center of athletic endeavor for the elite of the
area, and its volunteer military units had headquarters there. Among
the sports luminaries who used its facilities was the great Jake Kilrain.
He trained there in 1889 for his bout with John L. Sullivan at
Richburg, Miss. In 1892 “Gentleman Jim” Corbett trained there for his
celebrated fight with Sullivan at the Olympic Club, and to the
Southern he returned triumphant for a victory celebration. The S.A.C.
had New Orleans’ first Turkish bath. In 1878, the Lawn Tennis Club
had the city’s first tennis court at Jackson Avenue and Prytania.
The most discussed showplace in an area of palatial homes was the
Renaissance-inspired house of James Robb on Washington Street,
now Avenue. His dream house deserved all the adjectives lavished
upon it. The one-story brick and plaster mansion was surrounded by
gardens rivalling those of Europe’s royal estates. He brought over a
German gardener to design and maintain them. Statuary by
European and American masters embellished the grounds.

Some contemporary observers found the severe classical exterior a

bit plain, but inside there was a lavishness of detail which made even
these carpers wax enthusiastic. The house contained frescoes by the
celebrated Dominique Canova, priceless European pictures, furniture,
rugs and objects. The most famous art work was probably Hiram
Powers’s Greek Slave, the daring marble beauty which had shocked
New York. Robb allowed it to be exhibited in several cities before
bringing it to his home. Everywhere it aroused controversy. Today it is
in Washington’s Corcoran Gallery of Art.

Robb, millionaire businessman, president of the first trunk line 23

railroad to New Orleans, the New Orleans, Jackson and Great
Northern (later part of the Illinois Central) lost his fortune and in
1860 the great house which had been dubbed “Robb’s Folly” was
acquired by another millionaire, John Burnside. Under his ownership
the beauties of the dwelling were preserved. The noted octagonal
room, decorated in the Pompeian fashion, with its arrangement of
mirrors which reflected the scene ad infinitum, continued to excite
admiration.

In 1890, Mrs. Josephine Newcomb purchased the three-acre square

and its buildings for the girls’ college which she had endowed some
years before. In converting it into a school, great care was taken to
preserve the architectural beauty. When Newcomb College moved to
its present campus in 1918, the old campus was acquired by the
Baptist Bible Institute, later the Baptist Theological Seminary. They
used the site until 1955, when they, too, moved to larger quarters.
The Baptists extended Conery Street through the square, divided the
property into lots, and sold them. Fine new homes have arisen there.

The people of Lafayette were notably deep in their religious faiths

and in love for their fellow men. This is shown by their early church
organizations, by their solicitation for the welfare of the indigent and
the orphans of the immigrants devastated by cholera and yellow
fever epidemics, and by their inauguration of public education,
lyceum programs and a library.

It is interesting to note that, with the background of Germans and

Irish, it was a Protestant church which first was erected for a
Lafayette City congregation. In a building on St. Mary Street, near
Fulton (now St. Thomas), as early as 1831 the Methodists were
meeting. Some 10 years later the same denomination built a new
church on Magazine Street out of flatboat gunwales, and this was
known for years as the “Flatboat Church”. Later it became identified
with a young pastor, Elijah Steele, who had died of yellow fever. As
Steele’s Chapel it united with the St. Mary Street Church and the
Andrew Chapel, which had been built on Dryades and Felicity in
1835, to form the Felicity Methodist Church.

Although a parish was chartered for Lafayette Roman Catholics in

1836, they had no church and no priest until 1843. That year Father
Peter Chakert, of the Redemptorist order, gathered the faithful in
Kaiser’s Hall on Chippewa and Josephine Streets for masses on
Sunday morning after the past evening’s dance had ceased. The
following year saw the start of their first church on Josephine Street,
St. Mary’s Assumption. This lovely little wooden chapel, with its bell
which was cast at Des Allemands, was later replaced by the present
structure. However, the first building is still standing, moved to St.
Joseph’s cemetery on Washington Avenue, where it serves, all white
and clean, as a mortuary chapel, 117 years old.

St. Mary’s Assumption first served all the Roman Catholics of

Lafayette with sermons alternating in German, French and English. In
1850 St. Alphonsus Church was completed across the Street, chiefly
for the Irish, and nine years later, this remarkable tri-lingual parish
opened a church for the French people on Jackson Avenue. This was
taken down in 1925, but is perpetuated in the Chapel of Our Mother
of Perpetual Help at Third and Prytania, the old Lonsdale-McStea
house.

24

Surveyor’s drawing of Livaudais Plantation, or “Faubourg”, divided

into squares for sale of lots following purchase from Mme. Livaudais.
She retained square with her house and garden, near river. Note how
this square governed size of entire row of squares, as they are today.

High-resolution Map
 A SKETCH OF THE PLAN OF THE
FAUBOURG LIVAUDAIS,
Drawn at the request of Messrs. L. PEIRCE, W.
H. CHASE, M. MORGAN and S. J. PETERS,
By B. BUISSON, Surveyor for the Parish of
Jefferson—March, 1832.
PRINTED BY BENJAMIN LEVY, CHARTRES-STREET.

The year 1840 saw the Presbyterians organized in Lafayette 26

City under the popular Rev. Jerome Twichell. Their church,
completed in 1843, on Fulton between Josephine and Adele, was also
occasionally used by the Society of Friends. Henry Clay attended
services once in the church soon after it was opened. The Prytania
Presbyterian Church, where George W. Cable worshipped and sang in
the choir, was founded in 1846.

Episcopal services began in a room on the corner of Washington and

Laurel streets in 1847. Later that year, construction started for the
Church of the Holy Trinity at the corner of Live Oak (now Constance)
and Second streets. In 1851, the Rev. Alexander F. Dobb, a dynamic
churchman, began working for the construction of a handsome new
edifice at Jackson and Coliseum Streets. Trinity Church, as its name
was shortened, was occupied in 1853. Unfortunately, Mr. Dobb and
his wife died in the tragic yellow fever epidemic of that year and
never saw the completed church.

Congregation Gates of Prayer, the Jewish synagogue, originally

worshipped in a building near the corner of Sixth and Tchoupitoulas
Streets, but in 1854 it moved to a building, still standing though no
longer used for that purpose, on Jackson Avenue.

Missions for the German Protestants were provided by the

Evangelical, Methodist, Presbyterian and Episcopal churches in
various locations.
Lafayette City is no more. Its heritage is two-fold: the sturdy Irish-
German stock of its riverfront section; and the great houses and
cultural heritage of the Garden District, its fine residential section. Of
the former, volumes could be written; of the latter, the following
pages will attempt to touch the high spots. If this small book
encourages the reader to visit the scenes described, if it provides a
setting for the better appreciation of the great houses, the many
hours of patient research and writing will be well rewarded.
Mansion on Prytania, between Philip and Jackson, typifies great days
of Garden District, was once home of authoress, historian Grace King.
Frances Jones was Miss King’s illustrator.

27
LOUISE S. McGEHEE SCHOOL
2343 Prytania Street

The Main Building

Formerly one of the most lavish private homes in the Garden District,
this mansion now serves as the main building of the Louise S.
McGehee School, for almost half a century one of the outstanding
private schools for girls in the South. Amid architectural surroundings
which bespeak a bygone age of leisure, work and study now prevail
as the students pursue their exacting college preparatory curriculum.

Designed in the splendid free Renaissance style by James Freret, the

mansion was constructed in 1872 for Bradish Johnson, a young man
of wealth and discrimination whose family fortune was based on
sugar plantations. Its erection marked the second great period of
affluence for the Garden District. According to tradition it was built at
a cost of one hundred thousand dollars and its furnishings were as
lavish as the house itself. Always beautifully maintained by the
Johnsons and the Walter Denègre family, its later owners, the
architectural features of the building have been carefully preserved
by the school corporation. Of undiminished loveliness are the fluted
Corinthian columns, lofty ceilings and elaborate moldings embellished
with classical motifs. An outstanding feature of the building is the
winding staircase which rises at the rear of the marble-floored
entrance hall. This stairway of unsurpassed beauty has been
frequently honored as a masterpiece of design and craftsmanship.

A curious fact about the building is that neither a marriage, a birth,

nor a death has ever taken place within its walls. However, since its
acquisition by McGehee school in 1929 it has been the scene of many
scholastic triumphs. The school features an honor system and 28
student self government, the first high school in the city to
establish this type of government. Nearly all of the school’s graduates
have gone to college and most of the alumnae are active in civic
affairs.
 Magnificent spiral staircase in marble-floored central hall of former
Bradish Johnson mansion has mahogany railing, stained glass
skylight. Johnson fortune was based on large sugar plantations. City
house was showplace.

A stroll around the grounds on the First Street side gives a good view
of the former servants’ wing, which extends to the rear, looking today
much as it did when the house was new. The beautiful grounds are
particularly lovely in the spring when myriads of azaleas are in bloom
as well as the large wisteria vine which drapes the arch of the front
gate. Aged and majestic are the many magnolia trees, the largest of
which some years ago was declared by E. H. Sargent, then curator of
the Arnold Arboretum, to be the most magnificent specimen of
magnolia grandiflora in the United States.