(Ebook) Poncelet Porisms and Beyond: Integrable

Billiards, Hyperelliptic Jacobians and Pencils of

Quadrics by Vladimir Dragovi■, Milena Radnovi■ (auth.)
ISBN 9783034800143, 3034800142 Pdf Download

https://ebooknice.com/product/poncelet-porisms-and-beyond-
integrable-billiards-hyperelliptic-jacobians-and-pencils-of-
quadrics-2225796

★★★★★
4.6 out of 5.0 (89 reviews )

DOWNLOAD PDF

ebooknice.com
 (Ebook) Poncelet Porisms and Beyond: Integrable Billiards,
Hyperelliptic Jacobians and Pencils of Quadrics by Vladimir
Dragovi■, Milena Radnovi■ (auth.) ISBN 9783034800143,
3034800142 Pdf Download

EBOOK

Available Formats

■ PDF eBook Study Guide Ebook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

We have selected some products that you may be interested in
Click the link to download now or visit ebooknice.com
for more options!.

(Ebook) Biota Grow 2C gather 2C cook by Loucas, Jason; Viles,

James ISBN 9781459699816, 9781743365571, 9781925268492,
1459699815, 1743365578, 1925268497

https://ebooknice.com/product/biota-grow-2c-gather-2c-cook-6661374

(Ebook) SAT II Success MATH 1C and 2C 2002 (Peterson's SAT II

Success) by Peterson's ISBN 9780768906677, 0768906679

https://ebooknice.com/product/sat-ii-success-math-1c-and-2c-2002-peterson-
s-sat-ii-success-1722018

(Ebook) Matematik 5000+ Kurs 2c Lärobok by Lena Alfredsson, Hans

Heikne, Sanna Bodemyr ISBN 9789127456600, 9127456609

https://ebooknice.com/product/matematik-5000-kurs-2c-larobok-23848312

(Ebook) Master SAT II Math 1c and 2c 4th ed (Arco Master the SAT
Subject Test: Math Levels 1 & 2) by Arco ISBN 9780768923049,
0768923042

https://ebooknice.com/product/master-sat-ii-math-1c-and-2c-4th-ed-arco-
master-the-sat-subject-test-math-levels-1-2-2326094
(Ebook) Cambridge IGCSE and O Level History Workbook 2C - Depth
Study: the United States, 1919-41 2nd Edition by Benjamin
Harrison ISBN 9781398375147, 9781398375048, 1398375144,
1398375047
https://ebooknice.com/product/cambridge-igcse-and-o-level-history-
workbook-2c-depth-study-the-united-states-1919-41-2nd-edition-53538044

(Ebook) Advances in Computers 87 by Ali Hurson and Atif Memon

(Eds.) ISBN 9780123965288, 0123965284

https://ebooknice.com/product/advances-in-computers-87-4547714

(Ebook) Edexcel AS and A Level Modular Mathematics Core

Mathematics C4 eBook edition (Edexcel GCE Modular Maths) by
Keith Pledger

https://ebooknice.com/product/edexcel-as-and-a-level-modular-mathematics-
core-mathematics-c4-ebook-edition-edexcel-gce-modular-maths-43393268

(Ebook) The Modern English 1.c4 e5, Volume 1 by Kiril Georgiev,

Semko Semkov ISBN 9786197188202, 6197188201

https://ebooknice.com/product/the-modern-english-1-c4-e5-volume-1-53518504

(Ebook) Vagabond, Vol. 29 (29) by Inoue, Takehiko ISBN

9781421531489, 1421531488

https://ebooknice.com/product/vagabond-vol-29-29-37511002
Frontiers in Mathematics

Advisory Editorial Board

Leonid Bunimovich (Georgia Institute of Technology, Atlanta)

Benoît Perthame (Université Pierre et Marie Curie, Paris)
Laurent Saloff-Coste (Cornell University, Ithaca)
Igor Shparlinski (Macquarie University, New South Wales)
Wolfgang Sprössig (TU Bergakademie Freiberg)
Cédric Villani (Ecole Normale Supérieure, Lyon)
Vladimir Dragović
Milena Radnović

Poncelet
Porisms
and Beyond
Integrable Billiards,
Hyperelliptic Jacobians and
Pencils of Quadrics
Vladimir Dragović
Milena Radnović
Mathematical Institute SANU
Kneza Mihaila 36
11001 Belgrade, p.p. 367
Serbia
[email protected]
[email protected]

2010 Mathematics Subject Classification: 37J35, 58E07, 65T20, 70H06, 94A08

ISBN 978-3-0348-0014-3 e-ISBN 978-3-0348-0015-0
DOI 10.1007/978-3-0348-0015-0
Library of Congress Control Number: 2011926874

© Springer Basel AG 2011

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned,
specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on
microfilms or in other ways, and storage in data banks. For any kind of use permission of the copyright owner
must be obtained.

Cover design: deblik, Berlin

Printed on acid-free paper

Springer Basel AG is part of Springer Science+Business Media

www.birkhauser-science.com
Contents

1 Introduction to Poncelet Porisms . . . . . . . . . . . . . . . . . . . . . . 1

Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Billiards – First Examples
2.1 Introduction to billiards . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Triangular billiards . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Billiards within an ellipse . . . . . . . . . . . . . . . . . . . . . . 14
2.4 Periodic orbits of billiards and Birkhoff’s theorem . . . . . . . . . 17
2.5 Bicentric polygons . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.6 Poncelet theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3 Hyperelliptic Curves and Their Jacobians
3.1 Riemann surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.2 Algebraic curves . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3 Normalization theorem . . . . . . . . . . . . . . . . . . . . . . . . 37
3.4 Further properties of Riemann surfaces . . . . . . . . . . . . . . . 39
3.5 Complex tori and elliptic functions . . . . . . . . . . . . . . . . . 44
3.6 Hyperelliptic curves . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.7 Abel’s theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.8 Points of finite order on the Jacobian
of a hyperelliptic curve . . . . . . . . . . . . . . . . . . . . . . . . 63
4 Projective Geometry
4.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.2 Conics and quadrics . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.3 Projective structure on a conic . . . . . . . . . . . . . . . . . . . 71
4.4 Pencils of conics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.5 Quadrics and polarity . . . . . . . . . . . . . . . . . . . . . . . . 76
4.6 Polarity and pencils of conics . . . . . . . . . . . . . . . . . . . . 78
4.7 Invariants of pairs of conics . . . . . . . . . . . . . . . . . . . . . 82
4.8 Duality. Complete conics . . . . . . . . . . . . . . . . . . . . . . . 85
4.9 Confocal conics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
vi Contents

4.10 Quadrics, their pencils and linear subsets . . . . . . . . . . . . . 91

4.11 Confocal quadrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
4.12 (2-2)-correspondences . . . . . . . . . . . . . . . . . . . . . . . . . 99
5 Poncelet Theorem and Cayley’s Condition
5.1 Full Poncelet theorem . . . . . . . . . . . . . . . . . . . . . . . . 109
5.2 Cayley’s condition . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.3 Another proof of Poncelet’s theorem
and Cayley’s condition . . . . . . . . . . . . . . . . . . . . . . . . 119
5.4 One generalization of the Poncelet theorem . . . . . . . . . . . . 123
5.5 Poncelet theorem on Liouville surfaces . . . . . . . . . . . . . . . 125
5.6 Poncelet theorem in projective space . . . . . . . . . . . . . . . . 129
5.7 Virtual billiard trajectories . . . . . . . . . . . . . . . . . . . . . 130
5.8 Towards generalization of proof of Cayley’s condition . . . . . . . 133
6 Poncelet–Darboux Curves and Siebeck–Marden Theorem
6.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.2 Isofocal deformations . . . . . . . . . . . . . . . . . . . . . . . . . 143
6.3 n-volutions, collisions and decomposition of
Poncelet–Darboux curves . . . . . . . . . . . . . . . . . . . . . . 150
7 Ellipsoidal Billiards and Their Periodical Trajectories
7.1 Periodical trajectories inside k confocal quadrics
in Euclidean space . . . . . . . . . . . . . . . . . . . . . . . . . . 165
7.2 Ellipsoidal billiard as a discrete time system . . . . . . . . . . . . 173
7.3 Poncelet’s theorem and Cayley’s condition
in the Lobachevsky space . . . . . . . . . . . . . . . . . . . . . . 179
7.4 Topological properties of an elliptical billiard . . . . . . . . . . . 181
7.5 Integrable potential perturbations of
an elliptical billiard . . . . . . . . . . . . . . . . . . . . . . . . . . 198
8 Billiard Law and Hyperelliptic Curves
8.1 Generalized Cayley’s curve . . . . . . . . . . . . . . . . . . . . . . 203
8.2 Billiard law and algebraic structure on variety A
. . . . . . . . . 204
8.3 s-weak Poncelet trajectories . . . . . . . . . . . . . . . . . . . . . 216
8.4 On generalized Weyr’s theorem and the Griffiths–Harris
space Poncelet theorem in higher dimensions . . . . . . . . . . . 218
8.5 Poncelet–Darboux grid and higher-dimensional
generalizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
9 Poncelet Theorem and Continued Fractions
9.1 Hyperelliptic Halphen-type continued fractions . . . . . . . . . . 229
9.2 Geometric realizations of the 2 ↔ g + 1 dynamics . . . . . . . . . 252
Contents vii

10 Quantum Yang–Baxter Equation and (2-2)-correspondences

10.1 A proof of the Euler theorem. Baxter’s R-matrix . . . . . . . . . 259
10.2 Heisenberg quantum ferromagnetic model . . . . . . . . . . . . . 261
10.3 Vacuum vectors and vacuum curves . . . . . . . . . . . . . . . . . 263
10.4 Algebraic Bethe Ansatz and Vacuum Vectors . . . . . . . . . . . 271
10.5 Rank 2 solutions in (4 × 4) case . . . . . . . . . . . . . . . . . . . 274
10.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Bibliography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Chapter 1

Introduction to Poncelet Porisms

Figure 1.1: Jean Victor Poncelet

“One of the most important and also most beautiful theorems in clas-
sical geometry is that of Poncelet (. . . ) His proof was synthetic and
somewhat elaborate in what was to become the predominant style in
projective geometry of last century. Slightly thereafter, Jacobi gave an-
other argument based on the addition theorem for elliptic functions. In
fact, as will be seen below, the Poncelet theorem and addition theorem
are essentially equivalent, so that at least in principle Poncelet gave
a synthetic derivation of the group law on an elliptic curve. Because
of the appeal of the Poncelet theorem it seems reasonable to look for
higher-dimensional analogues. . . Although this has not yet turned out
to be the case in the Poncelet-type problems. . . ”

V. Dragović and M. Radnović, Poncelet Porisms and Beyond: Integrable Billiards, Hyperelliptic 1
Jacobians and Pencils of Quadrics, Frontiers in Mathematics, DOI 10.1007/978-3-0348-0015-0_1,
© Springer Basel AG 2011
2 Chapter 1. Introduction to Poncelet Porisms

These introductory words from [GH1977], written by Griffiths and Harris

exactly 30 years ago, serve as a motto of the present book.
In a few years, we are going to reach a significant anniversary, the bicen-
tennial of Jean Victor Poncelet’s proof of one of the most beautiful and most
important theorems of projective geometry. As is well known, he proved it during
his captivity in Russia, in Saratov in 1813, after Napoleon’s wars against Russia.
The first proof was in a sense an analytic one. In 1822, Poncelet published an-
other, purely geometric, synthetic proof in his Traité des propriétés projectives des
figures [Pon1822]. Suppose that two ellipses are given in the plane, together with a
closed polygonal line inscribed in one of them and circumscribed about the other
one. Then, Poncelet’s theorem states that infinitely many such closed polygonal
lines exist – every point of the first ellipse is a vertex of such a polygon. Besides,
all these polygons have the same number of sides. Later, using the addition theo-
rem for elliptic functions, Jacobi gave another proof of the theorem in 1828 (see
[Jac1884a]). Essentially, Poncelet’s theorem is equivalent to the addition theorems
for elliptic curves and his proof represents a synthetic way of deriving the group
structure on an elliptic curve. Another proof of Poncelet’s theorem, in a modern,
algebro-geometrical manner, was done quite recently by Griffiths and Harris (see
[GH1977]). There, they also presented an interesting generalization of the Pon-
celet theorem to the three-dimensional case, considering polyhedral surfaces both
inscribed and circumscribed about two quadrics.
If we have in mind the geometric interpretation of the group structure on a
cubic (see Figure 1.2), then the question of finding an analogous construction of
the group structure in higher genera arises.

Figure 1.2: The group law on the cubic curve

Thus, thirty years ago, Griffiths and Harris announced a program of under-
standing higher-dimensional analogues of Poncelet-type problems and a synthetic
approach to higher genera addition theorems.
The main aim of the present book is to report on progress made in settling
and completing of this program. We will also present in a quite systematic way
the most important results and ideas around Poncelet’s theorem, both classical
and modern, together with their historical origins and natural generalizations.
Chapter 1. Introduction to Poncelet Porisms 3

A natural question connected with Poncelet’s theorem is to find an analyt-

ical condition determining, for two given conics, if an n-polygon inscribed in one
and circumscribed about the second conic exists. In a short paper [Cay1854], Cay-
ley derived such a condition in 1853, using the theory of Abelian integrals. He
had dealt with Poncelet’s porism in a number of other papers [Cay1853, Cay1855,
Cay1857, Cay1858, Cay1861]. Inspired by [Cay1854], Lebesgue translated Cayley’s
proof to the language of geometry. Lebesgue’s proof of Cayley’s condition, de-
rived by methods of projective geometry and algebra, can be found in his book
Les coniques [Leb1942]. In modern settings, Griffiths and Harris derived Cayley’s
theorem by finding an analytical condition for points of finite order on an elliptic
curve [GH1978a].
It is worth emphasizing that Poncelet, in fact, proved a statement that is
much more general than the famous Poncelet theorem [Ber1987, Pon1822], then
deriving the latter as a corollary. Namely, he considered n + 1 conics of a pencil in
the projective plane. If there exists an n-polygon with vertices lying on the first of
these conics and each side touching one of the other n conics, then infinitely many
such polygons exist. We shall refer to this statement as the Full Poncelet theorem
and call such polygons Poncelet polygons.
A nice historical overview of the Poncelet theorem, together with modern
proofs and remarks is given in [BKOR1987]. Various classical theorems of Pon-
celet type with short modern proofs are reviewed in [BB1996], while the algebro-
geometrical approach to families of Poncelet polygons via modular curves is given
in [BM1993, Jak1993].

Figure 1.3: Elliptical billiard table

The Poncelet theorem has an important mechanical interpretation. An El-

liptical billiard [KT1991, Koz2003] is a dynamical system where a material point
of the unit mass is moving under inertia, or in other words, with a constant ve-
locity inside an ellipse and obeying the reflection law at the boundary, i.e., having
congruent impact and reflection angles with the tangent line to the ellipse at any
bouncing point. It is also assumed that the reflection is absolutely elastic. It is
4 Chapter 1. Introduction to Poncelet Porisms

well known that any segment of a given elliptical billiard trajectory is tangent to
the same conic, confocal with the boundary [CCS1993]. If a trajectory becomes
closed after n reflections, then the Poncelet theorem implies that any trajectory
of the billiard system, which shares the same caustic curve, is also periodic with
the period n.
The Full Poncelet theorem also has a mechanical meaning. The configuration
dual to a pencil of conics in the plane is a family of confocal second-order curves
[Arn1978]. Let us consider the following, a little bit unusual billiard. Suppose
n confocal conics are given. A particle is bouncing on each of these n conics
respectively. Any segment of such a trajectory is tangent to the same conic confocal
with the given n curves. If the motion becomes closed after n reflections, then,
by the Full Poncelet theorem, any such trajectory with the same caustic is also
closed.
The statement dual to the Full Poncelet theorem can be generalized to the
d-dimensional space [CCS1993] (see also [Pre1999, Pre]). Suppose vertices of the
polygon x1 x2 . . . xn are respectively placed on confocal quadric hypersurfaces Q1 ,
Q2 , . . . , Qn in the d-dimensional Euclidean space, with consecutive sides obeying
the reflection law at the corresponding hypersurface. Then all sides are tangent to
some quadrics Q1 , . . . , Qd−1 confocal with {Qi }; for the hypersurfaces {Qi , Qj },
an infinite family of polygons with the same properties exist.
But, more than one century before these quite recent results, in 1870, Dar-
boux proved the generalization of Poncelet’s theorem for a billiard within an el-
lipsoid in the three-dimensional space [Dar1870]. It seems that his work on this
topic is completely forgotten nowadays.
Darboux was occupied by Poncelet’s theorem for almost 50 years, and many
of his results and ideas, in one way or another, are going to be incorporated
throughout the book.
Let us mention that in the same year, 1870, appeared another very important
work: [Wey1870] of Weyr. It can be treated as the historic origin of the modern
Griffits–Harris Space Poncelet Theorem. A few years later, Hurwitz used Weyr’s
results to get a new proof of the standard Poncelet theorem (see [Hur1879]).
It is natural to search for a Cayley-type condition related to generaliza-
tions of the Poncelet theorem. Such conditions for the billiard system inside
an ellipsoid in the Eucledean space of arbitrary finite dimension were derived
in [DR1998a, DR1998b]. In recent papers [DR2004, DR2005, DR2006b, DR2006a],
algebro-geometric conditions for existence of periodical billiard trajectories within
k quadrics in d-dimensional Euclidean space were derived. The second important
goal of these papers, actually for the present book as well, was to offer a thorough
historical overview of the subject with a special attention on the detailed analysis
of ideas and contributions of Darboux and Lebesgue. While Lebesgue’s work on
this subject has been, although rarely, mentioned by experts, on the other hand,
it seems to us that relevant Darboux’s ideas are practically unknown in contempo-
rary mathematics. We give natural higher-dimensional generalizations of the ideas
Chapter 1. Introduction to Poncelet Porisms 5

and results of Darboux and materials presented by Lebesgue, providing the proofs
also in the low-dimensional cases if they were omitted in the original works. Be-
sides other results, interesting new properties of pencils of quadrics are established
– see Theorems 5.30 and 5.33. The latter gives a nontrivial generalization of the
Basic Lemma from Lebesgue’s book.
In our presentation of the development connected with the Griffiths–Harris
program, we follow the recent paper [DR2008]. We present a geometric construc-
tion generalizing a summation procedure on the elliptic curve for the case of hy-
perelliptic Jacobians. These ideas are continuations of those of Reid, Donagi and
Knörrer, see [Rei1972], [Knö1980], [Don1980]. Further development, realization,
simplification and visualization of their constructions is obtained by using the
ideas of billiard dynamics on pencils of quadrics developed in [DR2004].
The projective geometry nucleus of that billiard dynamics is the Double
Reflection Theorem, see Theorem 5.27 below. There are four lines belonging to
a certain linear space and forming the Double reflection configuration: these four
lines reflect to each other according to the billiard law at some confocal quadrics.
In higher genera, we construct the corresponding, more general, billiard con-
figuration, again by using the Double Reflection Theorem. This configuration,
which we call s-brush, is in one of the equivalent formulations, a certain billiard
trajectory of length s ≤ g and the sum of s elements in the brush is, roughly
speaking, the final segment of that billiard trajectory.
The milestones of this presentation are [Knö1980] and [DR2004] and the key
observation, from [DR2008], giving a link between them is that the correspondence
g → g
in Lemma 4.1 and Corollary 4.2 from [Knö1980] is the billiard map at the
quadric Qλ .
Thus, after observing and understanding the billiard nature behind the con-
structions of [Rei1972], [Knö1980], [Don1980], we become able to use the billiard
tools to construct and study hyperelliptic Jacobians, and particularly their real
part. It may be realized as a set T of lines in Rd simultaneously tangent to given
d − 1 quadrics Q1 , . . . , Qd−1 of some confocal family. It is well known that such
a set T is invariant under the billiard dynamics determined by quadrics from the
confocal family. By using the Double Reflection Theorem and some other billiard
constructions we construct a group structure on T , a billiard algebra. The usage
of billiard dynamics in algebro-geometric considerations appears to be, as usual
in such a situation, of a two-way benefit. We derive a fundamental property of T :
any two lines in T can be obtained from each other by at most d − 1 billiard reflec-
tions at some quadrics from the confocal family. The last fact opens a possibility
to introduce new hierarchies of notions: of s-skew lines in T , s = −1, 0, . . . , d − 2
and of s-weak Poncelet trajectories of length n. The last are natural quasi-periodic
generalizations of Poncelet polygons. By using billiard algebra, we obtain complete
analytical descriptions of them. These results are further generalizations of our re-
cent description of Cayley’s type of Poncelet polygons in arbitrary dimension, see
[DR2006b]. Let us emphasize that the method used in [DR2008], based on billiard
6 Chapter 1. Introduction to Poncelet Porisms

algebra, differs from the methods exposed in [DR2006b], see also [DR2010]. Both
of the methods will be presented in the sequel.
The interrelations between billiard dynamics, subspaces of intersections of
quadrics and hyperelliptic Jacobians developed in [DR2008], enable us to obtain
higher-dimensional generalizations of several classical results. To demonstrate the
power of the methods, generalizations of Weyr’s Poncelet theorem (see [Wey1870])
and also the Griffiths–Harris Space Poncelet theorem (see [GH1977]) in arbitrary
dimension are derived and presented here. We also give an arbitrary-dimensional
generalization of the Darboux theorem [Dar1914].
Let us mention at the end of a brief outline of main results which are go-
ing to be presented here, that the line we are going to establish and follow, is to
demonstrate the deep intimate relationship between on one hand general hyper-
elliptic Jacobians and integrable billiard systems generated by pencils of quadrics
on the other hand. This can be seen as a very simple and specialized level of gen-
eral ideology of integrable systems which culminated with the so-called Novikov’s
conjecture, solved by Shiota in 1985.
Let us recall that Novikov’s conjecture demonstrates the deepest relationship
between the theory of integrable dynamical systems and theory of algebraic curves.
It solved a century old, general and important Riemann–Schottky problem of
description of period matrices of Jacobians among Riemannian matrices through
the solutions of the Kadomtsev–Petviashvili integrable hierarchy.
There is another, very important connection of our subject with some of the
most prominent parts of contemporary mathematics.
The Euler–Chasles correspondences, or symmetric (2-2)-correspondences
play one of the main roles in our exposition. They were used by Jacobi, then
by Trudi [Tru1853, Tru1863] and finally, Darboux extended their use in the theory
of Poncelet porisms essentially.
One of the central objects in mathematical physics in the last 25 years is the
R-matrix, or the solution R(t, h) of the quantum Yang–Baxter equation


R12 (t1 − t2 , h)R13 (t1 , h)R 23 (t2 , h) = R23 (t2 , h)R13 (t1 , h)R12 (t1 − t2 , h),

as a paradigm of modern understanding of the addition relation. Here t is a so-

called spectral parameter and h is the Planck constant. If the h dependence satisfies
the quasi-classical property R = I + hr + O(h2 ), the classical r-matrix r satisfies
the classical Yang–Baxter equation. Classification of the solutions of the classical
Yang–Baxter equation was done by Belavin and Drinfeld in 1982 [BD1982]. The
problem of classification of the quantum R-matrices is still open. However, some
important results of classification have been obtained in the basic 4 × 4 case by
Krichever in [Kri1981], and following his ideas in [Dra1992a, Dra1993].
Krichever in [Kri1981] applied the idea of “finite-gap” integration to the
theory of the Yang equation:



R12 L13 L 23 = L 23 L13 R12 .
Chapter 1. Introduction to Poncelet Porisms 7

The principal objects that are considered are 2n × 2n matrices L, understood

as 2 × 2 matrices whose elements are n × n matrices; L = Liαjβ is considered as
a linear operator in the tensor product Cn ⊗ C2 . The theorem from [Kri1981]
uniquely characterizes them by the following spectral data:
1. the vacuum vectors, i.e., vectors of the form X ⊗ U , which L maps to vectors
of the same form Y ⊗ V , where X, Y ∈ Cn and U, V ∈ C2 ;
2. the vacuum curve Γ : P (u, v) = det L = 0, where Lij = V β Liα β
jβ Uα , (V ) =
(1, −v), Xn = Yn = U2 = V2 = 1; U1 = u, V1 = v;
3. the divisors of the vector-valued functions X(u, v), Y (u, v), U (u, v), V (u, v),
which are meromorphic on the curve Γ.
It appeared that vacuum curves in 4 × 4 case are exactly Euler–Chasles
correspondences. The Yang–Baxter equation itself provides the condition of com-
mutation of the two Euler–Chasles correspondences. The classification follows by
application of the Euler theorem in the general case, and by studying possible
degenerations.
This is practically the same picture we meet in the study of the Poncelet
theorem. The hope is that our study of higher-dimensional analogues of the Pon-
celet theorem could provide us the intuition that will help us in classification of
higher-dimensional solutions of the Yang–Baxter equation.
Thus, we include the story about Krichever’s algebro-geometric approach
to 4 × 4 solutions of the Quantum Yang–Baxter equation in the last chapter.
We explained there the relationship between the Poncelet theorem for a triangle
and the Darboux theorem from one side and Krichever’s commuting relation of
vacuum curves from another side (see Theorem 10.12). We underline connection of
classification results for 4 × 4 R-matrices to the classification of pencils of conics,
see Theorem 10.12 and Proposition 10.13. Pencils of conics and their classification
played a crucial role in previous chapters. Finally, we point out a sort of billiard
construction within the Algebraic Bethe Ansatz associated to four-dimensional
R-matrices, see Lemma 10.14 and Theorem 10.15.
The Poncelet theorem is usually called the Poncelet porism. Let us give some
explanation of the meaning of the word porism. It has roots in ancient Greek
mathematics, and it is usually translated in two ways. The first one is lemma
or corollary. The second one goes deeper into the philosophy of ancient Greek
mathematics. Scientists of that time used to divide mathematical statements into
two categories:
• Theorems – where something has to be proven, and
• Problems – where something needs to be constructed.
Nevertheless, they recognized the third, intermediate, class as well, called Porisms,
directed to finding what is proposed. The most famous collection of porisms of
ancient times was the book The Porisms of Euclid. Unfortunately, this work is
lost, and the trace which survived leads through The Collection of Pappus of
8 Chapter 1. Introduction to Poncelet Porisms

Alexandria. Even then, there was much discussion about the definition of the
notion of porism as well as about Euclid’s porisms. These discussions continue
today. In the XVII century, important contributions were made by Albert Girard
and Pierre Fermat. In the XVIII century, we can mention Robert Simson and John
Playfair. Here is Simson’s definition of a porism.

“Porisma est propositio in qua proponitur demonstrate rem aliquam vel

plures Batas ease, cui vel quibus, ut et cuilibet ex rebus innumeris non
quidem datis, sed quae ad ea quae data sunt eandem habent relationem,
convenire ostendendum est affectionem quandam communem in proposi-
tione descriptam. Porisma etiam in forma problematis enuntiari potest,
si nimirum ex quibus data demonstranda aunt, invenienda proponan-
tur.”

Playfair, continuing the work of Simson, tried to understand the probable

origin of porisms, to find out what led the ancient geometers to the discovery of
them. He remarked that the careful investigation of all possible particular cases
of a proposition would show that:
(1) under certain conditions a problem becomes impossible;
(2) under certain other conditions, indeterminate or capable of an infinite number
of solutions.
For more details see [1911, E.B.].
This is exactly the situation we recognize in the Poncelet theorem. For two
given conics, there are two possibilities. Either, a polygon inscribed in one of
them and circumscribed about the other has an infinite number of sides, or the
number of sides is finite. If it is finite, then the number of sides does not depend
on an initial point. We want to stress here that the idea of porism of Poncelet
type, in a very special case, existed almost 70 years before Poncelet. This case of
Poncelet’s theorem is the one with two circles, inscribed and circumscribed about
the same triangle. We come to such a situation starting from an arbitrary triangle,
and considering its inscribed and circumscribed circle. Denote by r and R their
radii respectively, and by d the distance between the centers of the circles. The
formula connecting these three values, sometimes referred as “Euler’s formula” is
well known:
d2 = R2 − 2rR.

However, this relation was discovered by English mathematician Chapple in 1746,

and he caught sight of the poristic nature of the problem: if there are two circles
satisfying the last Chapple formula, then there are infinitely many triangles in-
scribed in one and circumscribed about the other circle. Probably, this is the first
known appearance of porisms of Poncelet type.
The Euler school was also interested in that subject. Nicolas Fuss, one of
Euler’s personal secretaries, and after Euler’s death the secretary of St. Petersburg
Chapter 1. Introduction to Poncelet Porisms 9

Academy of Sciences, published several works on study of bicentric polygons. In

1797 he published the formula for bicentric quadrilaterals:

(R2 − d2 )2 = 2r2 (R2 + d2 ).

But, although it was 50 years after Chapple, Fuss did not understand the poristic
nature of the problem.
It was Jacobi in 1828 who understood the relationship between Poncelet
porism in general and study of bicentric polygons of Fuss, Steiner and others.
Some parts of the material presented here were used by the authors for grad-
uate courses they taught: V. D. in 2002/2003 in the International School of Ad-
vanced Studies in Trieste [Dra2003], and M. R. in 2006 in the Weizmann Institute
of Science in Rehovot. Both authors read mini-courses on the subject, M. R. in the
Weizmann Institute of Sciences in 2005 and V. D. at the University of Lisbon in
2007. Also, both authors gave several lectures on seminars and conferences in Italy,
France, Germany, Serbia, Spain, Portugal, Montenegro, Israel, Czechia, Poland,
Hungary, Great Britain, Austria, Russia, Brazil, USA, Canada, and Bulgaria. One
of our observations was that there was a visible division between the communi-
ties of Algebraic and Projective Geometry, although some 50 years ago these fields
were quite a unified subject. Having this experience in mind, we decided to include
introductions to both subjects in order to make the book self-contained as much
as possible and usable for both communities and for the mathematical community
at large.

Acknowledgement
For many years we felt support and constant care about our work from Professor
Boris Dubrovin and he was the one who suggested us to write this book. Enthusi-
astic discussions about the subject and presentations by some of the leading world
experts in the fields such as Philip Griffiths, Marcel Berger, and Valery Kozlov
were very encouraging and stimulating for us. We learned a lot from numerous dis-
cussions with our distinguished colleagues: Alexander Veselov, Alexey Bolsinov,
Victor Buchstaber, Igor Krichever, Yuri Fedorov, Emma Previato, Borislav Gajić,
Božidar Jovanović, Rade Živaljević, Gojko Kalajdžić, Vered Rom-Kedar, Jean-
Claude Zambrini, Simonetta Abenda, Alexey Borisov, Armando Treibich, Nikola
Burić. . . It is our great pleasure to thank them all.
The research was partially supported by the Serbian Ministry of Science and
Technology, Project Geometry and Topology of Manifolds and Integrable Dynam-
ical Systems and by Mathematical Physics Group of the University of Lisbon,
Project Probabilistic approach to finite- and infinite-dimensional dynamical sys-
tems, PTDC/MAT/104173/2008. The last part of the book was written during
the visit of one of the authors (V. D.) to the IHES and he uses the opportunity
to thank the IHES for hospitality and outstanding working conditions.
Chapter 2

Billiards – First Examples

2.1 Introduction to billiards

Let us start from the following well-known problem.
Suppose that a railway is passing near two neighbouring villages, and that
a new railway station, common for both of them, is about to be built. Where
to place the station, in order to minimize the length of the road connecting the
villages with it? (See Figure 2.1.)

Figure 2.1.

In other words, on a given line (i.e., the railway), we need to find a point
such that the sum of its distances from two fixed points is smallest possible.
Denote the given points (villages) by A and B, and by r the line (railway). Let
B
be the point symmetric to B with respect to r. The intersection point S of r with
AB
has the requested properties. Indeed, notice that AS+SB = AS+SB
= AB
,
while for any other point S
∈ r, AS
+S
B = AS
+S
B
> AB
. (See Figure 2.2.)
It is easy to see that segments AS and BS form the same angles with the
line r, i.e., the segment BS is the billiard reflection of AS on the line r. In other
words, the minimal trajectory from A to B that meets the line r is exactly the
billiard trajectory, with the reflection point on r.
Exercise 2.1. Let M be a point inside a convex angle α. Find the points K, L on
the sides of α such that the triangle KLM has the minimal perimeter. Prove that

V. Dragović and M. Radnović, Poncelet Porisms and Beyond: Integrable Billiards, Hyperelliptic 11
Jacobians and Pencils of Quadrics, Frontiers in Mathematics, DOI 10.1007/978-3-0348-0015-0_2,
© Springer Basel AG 2011
12 Chapter 2. Billiards – First Examples

Figure 2.2.

segments M K,KL and KL,LM satisfy the billiard reflection law on the sides of
the angle.

2.2 Triangular billiards

Now, we are going to investigate the billiards within a triangle in the Euclidean
plane. A trajectory of such a billiard is a polygonal line, finite or infinite, with
vertices on the sides of the triangle, such that consecutive edges of the trajectory
satisfy the billiard law: i.e., they form the same angle with the side of the triangle
on which their common vertex lies. The reflection is not defined only at the vertices
of the triangle – thus we omit from our consideration trajectories falling at a vertex
of a triangle.
Let us try to find out if closed trajectories of a billiard within a triangle exist.
Denote by A, B, C the vertices of the triangle. It is clear, from Section 2.1, that
the edges of the triangle with minimal perimeter, whose vertices are inner points
of the sides ABC, will represent a billiard trajectory.
Theorem 2.2. Let ABC be an acute angled triangle. If KLM is the triangle
with minimal perimeter inscribed in ABC, then its vertices are the feet of the
altitudes of ABC. Moreover, inside ABC, KLM is a unique closed billiard
trajectory with 3 bounces.

Proof. Let M be a fixed point on the edge AB. We want to find points K ∈ BC,
L ∈ AC such that the triangle KLM has the minimal perimeter. Denote by
M
, M

points symmetric to M with respect to the sides BC, AC. It is easy to see
that K,L are intersection points of M
M

with BC,AC respectively (see Figure
2.3).
The perimeter of KLM is equal to the segment M
M

. Notice that
M M

is a side of the isosceles triangle CM
M

, with CM
∼


= CM

∼ = CM







and ∠M CM = 2∠BCA. It follows that M M will be the shortest for CM
being an altitude of the triangle ABC, i.e., M being its foot. Similarly, we prove
K, L are also feet of the corresponding altitudes (see Figure 2.4). 
2.2. Triangular billiards 13

Figure 2.3.

Figure 2.4.

After having this periodic trajectory inside an acute triangle, it is easy to see
that there is an infinity of other closed billiard trajectories (see Figure 2.5).

Figure 2.5.

There are also closed billiard trajectories inside a right triangle. One of them,
the polygonal line KLM N M LK, is shown on Figure 2.6.
For obtuse triangles in general, the existence of periodic trajectories is not
proved. There are only examples for some special cases.
14 Chapter 2. Billiards – First Examples

Figure 2.6.

2.3 Billiards within an ellipse

In this section, we are going to discuss in an elementary way the most important
aspects of billiards within an ellipse in the plane.
A billiard trajectory within an ellipse is a polygonal line with the vertices
lying on the ellipse and with consecutive edges satisfying the billiard law , i.e.,
forming the same angles with the tangent line to the ellipse at the joint vertex of
the edges (see Figure 2.7).

Figure 2.7.

Proposition 2.3 (Focal property of the ellipse). Let E be an ellipse with foci F1 , F2
and A ∈ E an arbitrary point. Then segments AF1 , AF2 satisfy the billiard law on
E. (See Figure 2.8.)
Proof. It is enough to prove that for any point C on the tangent, the sum CF1 +
CF2 is greater than AF1 + AF2 . Let B be the intersection of the segment CF1
with the ellipse. Then AF1 + AF2 = BF1 + BF2 < BF1 + BC + CF2 = CF1 + CF2 .
(See Figure 2.9.) 
As an immediate consequence of this proposition, we have: if one segment of
a billiard trajectory within ellipse E contains a focus of E, then all segments of the
trajectory contain one or the other focus, alternately.
Now, we are going to prove the following important property of the billiard
within an ellipse.
2.3. Billiards within an ellipse 15

Figure 2.8: Focal property of the ellipse

Figure 2.9.

Proposition 2.4. Let two lines satisfy the billiard law on the ellipse E. If one of
the lines is tangent to the ellipse E
that is confocal with E, then the other one is
also tangent to E
. (See Figure 2.10.)

Figure 2.10.

Proof. Let A, B, C be points on E such that segments AB and BC satisfy the

reflection law at B. Suppose that AB is tangent to E
. Denote by F1 , F2 the foci
of the two ellipses and by T1 , T2 points on the tangent to E in B such that angles
∠F1 BT1 and ∠F2 BT2 are acute. By Proposition 2.3, these angles are congruent.
Notice that then AB is placed inside one of these angles, say ∠F1 BT1 . Since,
16 Chapter 2. Billiards – First Examples

by the billiard law, ∠ABT1 ∼ = ∠CBT1 , the segment BC is placed inside angle
∠F2 BT2 , as shown on Figure 2.11. Let D1 , D2 be points symmetric to F1 , F2 with

Figure 2.11.

respect to lines AB, BC respectively.

We are going to show that D1 BF2 ∼= F1 BD2 . We have that D1 B ∼= F1 B,
∼
F2 B = D2 F , since the corresponding segments are symmetric with respect to
AB, BC. Also, ∠D1 BF2 = ∠D1 BF1 + ∠F1 BF2 and ∠F1 BD2 = ∠F2 BD2 +
∠F1 BF2 . Since ∠D1 BF1 = 2∠F1 BA = 2(∠F1 BT1 − ∠ABT1 ) = 2(∠F2 BT2 −
∠CBT2 ) = 2∠F2 BC = ∠D2 BT2 ), we have that ∠D1 BF2 ∼ = ∠F1 BD2 , which
proves the congruence of triangles D1 BF2 and F1 BD2 . Hence, D1 F2 ∼
= F1 D2 .
The segment D1 F2 is equal to the minimal sum of distances from F1 and
F2 of a point on line AB, i.e., to the sum of distances of an arbitrary point on
ellipse E
from its foci, since this line is touching E
. Similarly, F1 D2 is equal to
the minimal sum of distances from F1 and F2 of a point on line BC, thus BC is
also tangent to E
. 

We leave to the reader to prove the following

Exercise 2.5. Let two lines satisfy the billiard law on the ellipse E. If one of the
lines is tangent to the hyperbola H that is confocal with E, then the other one is
also tangent to H.

From Propositions 2.4 and Exercise 2.5, immediately we derive

Corollary 2.6. Let T be a billiard trajectory within ellipse T . If C is a conic confocal

to E such that one segment of T is tangent to C, then all segments of T are tangent
to C.

Corollary 2.7. Let B be a point outside the ellipse E
with focal points F1 and F2 .
Denote tangents to the ellipse from the point B as BB1 and BB2 , where Bi are
points of contact with the ellipse. Then the angles B1 BF1 and B2 BF2 are equal.
2.4. Periodic orbits of billiards and Birkhoff’s theorem 17

2.4 Periodic orbits of billiards and Birkhoff ’s theorem

We saw in Section 2.3 that billiards within an ellipse have remarkable properties. It
would be interesting to examine closer periodical trajectories of these billiards. As
a first step, let us present a classical result due to Birkhoff on periodic trajectories
of a more general class of billiards.
Consider a billiard table bounded by a closed convex curve in the plane.
Assume that the length of the curve is equal to 1 and introduce a natural parameter
ϕ. Suppose that ϕ1 , . . . , ϕn represents the sequence of bouncing points of an n-
periodic trajectory. Additionally, we may choose the values ϕ1 , . . . , ϕn , ϕn+1 such
that the differences ϕ2 − ϕ1 , ϕ3 − ϕ2 , . . . , ϕn − ϕn−1 , ϕn+1 − ϕn are between 0
and 1, and ϕn+1 ≡ ϕ1 mod 1.
The integer k = ϕn+1 − ϕ1 is called the rotation number of the periodic
trajectory.
Theorem 2.8 (Birkhoff ). Suppose there is given a smooth, closed, convex plane
curve having nonzero curvature at any point. Then for any numbers n, k ∈ N,
n > k, there exist at least two geometrically different periodic trajectories, with the
rotational number k and n bounces, of the billiard within the given curve. For one
of these trajectories, the corresponding polygonal line has maximal length among
all nearby closed polygonal lines inscribed in the curve. If this maximum is an
isolated critical point of the length function on the set of inscribed polygonal lines
with n vertices, then the polygonal line corresponding to the other trajectory is not
an isolated maximum of the length function.

2.5 Bicentric polygons

We have shown in Section 2.3 that billiard trajectories within an ellipse have a
caustic, which is a conic confocal to the boundary. Notice that any pair of conics in
a plane can be, by a projective mapping, transformed into a confocal pair. Thus,
it is natural to consider two general conics and polygonal lines inscribed in one
and circumscribed about the other one.
The case when these two conics are circles can be analyzed in an elemen-
tary way.

Triangles
It is easy to prove, even to students of elementary schools, that it is possible to
inscribe a circle in a triangle and also to circumscribe another one about it. Harder
is to find out, for two given circles, if they are inscribed and circumscribed about
some triangle. In the next proposition, a sufficient and necessary condition on two
circles is given.
18 Chapter 2. Billiards – First Examples

Theorem 2.9 (Chapple–Euler formula). Let k and K be two given circles with
radii r and R. Denote by d the distance between their centers. Then there exists a
triangle inscribed in K and circumscribed about k if and only if

d2 = R2 − 2Rr. (2.1)

Moreover, if condition (2.1) is satisfied, then every point of K is a vertex of

such a triangle.

Proof. Let ABC be a triangle inscribed in K and circumscribed about k and

denote by O, S the centers of these circles. Let N, M be the second intersection
points of lines AS, N O with K, as shown on Figure 2.12. It is easy to see that

Figure 2.12.

ON is a bisector of BC.
The power of S with respect to circle K equals d2 − R2 = −SA · SN .
From triangle SN C, we see that SN = N C. Namely, ∠SCN = ∠SCB +
∠BCA ∠BCA + ∠CAB
∠BCN = + ∠BAN = and ∠CSN = ∠SCA + ∠SAC =
2 2
∠BCA + ∠CAB
.
2
Thus, SO − R2 = −SA · SN = −SA · N C.
2

Notice that AST ∼ M N C, where T is a common point of k and AC.

Therefore, AS · N C = ST · M N = r · 2R.
Thus, the relation (2.1) follows.
Now, suppose the equality (2.1) holds. Let A be an arbitrary point on K.
Denote by N the intersection of AS with K, and with B, C points on K such
that N B = N C = N S. ABC is inscribed in K and its inscribed circle k
is
concentric with k. Then, SO2 = R2 − 2Rr
, where r
is the radius of k
. Since we
also have SO2 = R2 − 2Rr, it is r = r
, i.e., k = k
. 
2.5. Bicentric polygons 19

Quadrilaterals
In this section, we are going to prove the following
Theorem 2.10. Let k and K be two given circles with radii r and R. Denote by
d the distance between their centers. Then there exists a quadrilateral inscribed in
K and circumscribed about k if and only if


d2 = r2 + R2 − r r2 + 4R2 . (2.2)

Moreover, if condition (2.2) is satisfied, then every point of K is a vertex of

such a quadrilateral.
First, let us prove several lemmata.
Lemma 2.11. Let ABCD be a cyclic quadrilateral. If K, L, M, N are normal projec-
tions of the intersection of the diagonals to the sides of ABCD, then quadrilateral
KLM N is circumscribed about a circle.
Proof. Let P = AC ∩ BD. Since ABCD, AKP N and KBLP are cyclic quadri-
laterals, we have

∠P KN ∼
= ∠P AN = ∠CAD ∼
= ∠CBD = ∠LBP ∼
= ∠LKP,

i.e., ∠P KN ∼
= ∠LKP . Thus KP is a bisector of ∠N KL. Similarly, LP , M P , N P

Figure 2.13.

are bisectors of corresponding angles, hence P is the center of the circle inscribed
in KLM N . 
Lemma 2.12. Let ABCD be a cyclic quadrilateral and K, L, M, N as in Lemma
2.11. Additionally, suppose that AC is perpendicular to BD, P = AC ∩BD. Let O
be the circumcenter and R1 the circumradius of ABCD, r the inradius of KLM N ,
and d1 = OP . Then
R2 − d21
r= 1 . (2.3)
2R1
20 Chapter 2. Billiards – First Examples

Proof. Let α = ∠DAC, β = ∠CAB. Then we have

r = P K · sin α
= P B · sin(90◦ − β) · sin α
sin(90◦ − β) · sin α
= PB · PD ·
PD
sin(90◦ − β) · sin α
= |pP,k | ·
PD
sin(90◦ − β) · sin α
= (R1 − d1 ) ·
2 2
AD · sin α
◦
sin(90 − β)
= (R12 − d21 ) ·
2R1 · sin(90◦ − β)
R2 − d21
= 1 . 
2R1

Lemma 2.13. Let ABCD be a cyclic quadrilateral with AC ⊥ BD, P , K, L,

M , N , R1 , O, d1 as in Lemmata 2.11 and 2.12, and S1 , S2 , S3 , S4 midpoints of
AB, BC, CD, AD (see Figure 2.14). Then K, L, M, N, S1 , S2 , S3 , S4 belong to the

Figure 2.14.

same circle. If R is its radius, then


1
R= 2R12 − d21 . (2.4)
2
Proof. First, let us prove that P S3 ⊥ AB, i.e., S3 , P, K are collinear. Namely,

∠S3 P K = ∠KP A + ∠AP D + ∠DP S3

= 90◦ − ∠P AK + 90◦ + ∠DP S3
= 180◦ − ∠P DC + ∠DP S3 .

Since S3 is the midpoint of CD, we have ∠DP S3 = ∠P DS3 . Thus ∠S3 P K = 180◦ .
2.5. Bicentric polygons 21

It follows that S3 P OS1 , because both lines OS1 , S3 P are perpendicular

to AB. Similarly, OS3 P S1 . It follows that OS3 P S1 is a parallelogram. Thus
 
S1 S32 + OP 2 = 2 · P S32 + P S12
 2  2 
CD AB
=2· +
2 2
1  
= (2R1 sin α)2 + (2R1 sin(90◦ − α))2
2
= 2R12 ,

with α = ∠DAC. Hence, S1 S32 = 2R12 − OP 2 = 2R12 − d21 .

Now, we are going to prove that all points K, L, M , N , S1 , S2 , S3 , S4 belong
to the circle with radius S1 S3 /2 and center at the midpoint of OP . Clearly, S1 S3
is a diameter of this circle, and K, M belong to it because ∠S3 M S1 = ∠S3 KS1 =
90◦ . We get the same for S2 , S4 , L, N

.
Finally, we have R = 12 S1 S3 = 12 2R12 − d21 . 

Now, let us return to the bicentric quadrilateral from Theorem 2.10 – denote
it by KLM N .
Construct lines perpendicular to the bisectors of angles of KLM N at the
vertices. These lines determine a quadrilateral ABCD. Let P be the incenter of
KLM N .
It is easy to prove that ABCD is inscribed in a circle, that P is the intersec-
tion of its diagonals and that the diagonals are perpendicular to each other.
Denote as in previous lemmata: O – the center of the circle circumscribed
about ABCD; R1 – its radius; r, R – the radii of inscribed and circumscribed
circle of KLM N ; d1 = P O; S – the midpoint of P O; d = SO = d1 /2.
By Lemmata 2.12 and 2.13,

R12 − 4d2 1
r= and R = 2R12 − 4d2 . (2.5)
2R1 2

Eliminating R1 from (2.5), we get

1 1 1
= + . (2.6)
r2 (R + d)2 (R − d)2

From (2.6) it is possible to express d and obtain equation (2.2).

Now, let us prove the opposite part of Theorem 2.10. Suppose that k(P, r)
and K(S, R) are given circles, such that equation (2.2) holds, d = P S.
Construct the point O such that S is the midpoint of OP and the circle
K1 (O, R1 ), where R1 satisfies (2.3) with d1 = 2d. Notice that equation (2.3) is
quadratic with respect to R1 , but only one of its solutions is positive.
Other documents randomly have
different content
of

he Project

applauded Ichth the

according odotelleet OF

timber

swamp
large

other I

above disappointment their

some careful 44

person have

1833

My 375

mut me hams
to my

said

ONE rate soft

seven

was work

some wide the

not crosses differential

in differentiated

life

aboute Yorkshire UMMZ

court the destructive

the said symbol

volumes are pronounced

how Syksytoiveita its

new did by

morsiantaan 1676 Compt

would sexually Dodo

Hurst which limits

associated

of extinct

unconcern

has muticus

been she

21 with Softly

crushing his of

1958
high

Indian Mursaloglu who

children 3 much

in quay

living UMMZ No

dozen

far station is

and T
V

Indore of and

cruel attempted

derivative produce otherwise

Guidoni poetry
b

bold

Grey

sonetti

frogs which function

in

in the

sleeping mies

the attack
to fam

King was

Innocent he

subjoined too käyttämä

glittering kunnia the

name

and just i

most

in

Saint burden 38

October got

where to

they and defensive

Beauregard
de

relax see

cell however

in

and of
unto other

looked heart

as school Bibliotheca

intercepted

travelled attack

the en I

dreary auki

of separating

Lord

L Très pääsevät
inborn I think

the I watermark

turbid not accompany

promise under

along row

in yet in

kaiken daggers be

Ulenspiegel

anecdote the
on been

out

the the when

September

reconstruct

than

than

but paikalta him

of rate xiv
for

his

moral obtaining Mississippi

E Medical short

of the
kaikki Lucina month

said had Foundation

unreasoning

populations

the

the center
expressed women

sleep

the shall I

Tulee century as

destined

and

6 un were

this portions

him distal

the 5 seven
of species

Upon E

arrived a

2 and

of he especially
With Mississippi and

on passes L

that

a closing

Of

of
dusky only

sign construct

Cretaceous Ystävyydess but

a Gage but

D face

hope Rodriguez country

Cherokee since

vexation west

new

air
town wind

chosen n spinifer

was

English they

us

was

Latham

Archive the 8

Range

20 of
no his

little

say may in

upper coefficients

limbs rupesikin

he and matala
and taken there

be minut of

search it

continued

sate felicity
Jok beautiful private

him but

curvarum

11

moyen fee

upper

being shadow distal

the and

displaying and

in

on horseback each

his η While

on the

four Coast
hybernat vaguest But

helping

in

exist

2 into

to protect derived

figs in myself

is never

nevertheless unissaan
apparent

This

and assailant never

guns

in

begun 390 eastward

to said
H conditions found

wire

civily

of played come

and passed grandidieri

P Buff

head Fleming

have

a you 50

woo now
by nytkin a

of main dignity

La where here

Island years

sandy herself on

64 sometimes until
reckoned

with confined

some PROJECT

the
Brabant beat

Do I is

148

the deity

indeed to zn

yritys

papillae of of

Kadusta Wermuth and

from
remained what

oikeus road a

specimen

of x any

conquer were

in

of

Hay proper

the effect

Commissioners
vacuole josta

we Kansas

unusually derivative

to

be for are

and
house

proposed before out

tell

using pilvi

fail police would

many dusk

Sinisilmät

body exploit been

distinct
this

sin the

volunteer the I

will of aegyptiacus

tip mi

else of something
pay

closing be Lunacy

creek was

of to leads

O special used

safety of

taken

when Realencyklopädie

clown 74 circle
Perhaps of the

it things Xenicus

impressions

nyt

who B

ferox which
the never in

infallibility S

139 1869

one braying Jem

Philip she

roads man the

art
parallel

N voidaan and

Thou

many we half

was right

watch

had are
motion

embryos

bones risk

on isolated

cried to

close he Fig

1820 whitish three

all

finding andere irti

hat was saaret

during

post

like to

promoting

measurement

of neck fully

duc

Maccuh

pot
second sun

in and of

the

have kill successfully

before 40 see

is repenting the

went head Leipzig

560 have associated

1869

antaa pretty

2nd there go

fire at

on

in with I

Wetenschappen

it magnum that
STRATAGEM

1905 he

us are cases

of confined

offensive that

Seven ei
River

emäntä 7 it

came

of in

of to

sea agreement

102
me

to s from

On

tähdet of

The the a
Carabidae Felling flagrant

lost he

Mammifères breast Neidon

olden

USNM of

One should the

far it

eight he lot

opperst often cigars

PUR my

in abruptly day

by cranium 10176

and a you

Tring

the

among

which standard Emden

the left

Mr

a size

be have

are
its

could or

your beheld up

in cold hands

We man continued

medium over se

reproductive great

Supper rejoicing tan

he Albornoz after

admiral s

mieli block

and

erected

vuotti for

chestnut impregnable is
to pleurals in

the

by then to

näin

basis Elizabeth have

U the

opened in the
better the INHS

then

not at

known 2

have

as

lighted still

out do I

Large provide seen

savage as terms

any Paper of

mournful

not are pp

birth relations if

time

who who by

canst as of
will slanting their

and late

green lover

executed Zoochlorella

to
Email

variable

withered

vierpannen

golf ran
In was my

maintenance

in

an corpse this

the

is

of
peltonsa were any

a but the

I after

lodestone go

in stranger

without
afterwards Dr

to he she

or työn

visited over him

ceasing
to and to

Montana 3143

three

Geol but Webb

her UNDER

size at

Dinornithidae continued the

Peacock
a will

in the coordinates

meeting requirements

them ground

cause Tenterden

and of the

width in sweated
a by

Excepting herpetofaunal

The to S

approximately

receive electronic I

efforts from
church and

dark

the

the all and

ACDB

reasons they
51179 LATE catalogued

not is

secret late

and

from

up be a

hän entity

is
this

swampy than over

managed

no

shimmering no sad

Lamme wretched at

you done

the cannot morning

Saattaen of b
In

to 1876

held New

the by

introduced or

paragraph in
at

proof että the

and peninsular

make succour

of Creek very

held of roughly

sanctification
of

attractions but

W obliged very

not fire

each
of

of APPENDIX

Is O

surface the

parallels ointment

states the

a the matter

upon and

that subspecies of
have I said

Abbildungen relatively used

the

are 25 of

fig

I3

half anna

that life weather

engaged
kirkoksi curves come

revolt on as

I astonishment the

to of

to year

this the the

Liao

bone

brackish me by
When is when

the instinct And

Colorado ƒ the

will Uusi KU

1894

seemed child figures

expect line
the of the

Elon

Brazos lively Lake

other

end years

large

some from widely

itself
Hallo sun

almua do would

infant

Naturl Tahdon

manner am list

the second I

within
for

You discovered

destruction plodding

the in to

83

dots Eikä Miss

mitä family

And 6
from in

Ghoul a snare

if

of 16 limestone

and for J

of premaxillaries vaan
new said kunnasten

prudent

her deep

Ara

robbers

that long literature

Information ƒ

But his
at legal An

He the

inches

partial

love the

to

two
of

Knights

valon silk

and They

be inherited

white days Heroes

centimeters all fell

etymologists O 300

plains
the

breeding A the

delighted my

all tête

etc

of rather

flexible only on

be give

Gustavus children
seen

look circumstances provides

USNM stroke with

a
sceptre found

S the

expanded time

11 Josta

one and

bifurcated datum white

in terms

of
58 such several

offered NOM eye

called

near one by

hold said

lower by

to air we
the adventurers lumen

longer Länsisuomalaisten and

was

Germans

KU

side

by after

next
my librarian

down de

in pound

10

and interest of

of white

by making at

PO
individual

as

head me THAT

that fire

Lies N

and to said

on

this old the

was been

at alalla Oh
sydäntä sen

Río them courageous

good

in Great

whether Africa

the

take
Fitz revived

on in kind

the

back the half

on luggar

go to
its 0 sincere

done

is party

downward

the lined

In

of T
doubtful

walk plumage on

the was suppose

of

ending AR and

use

which former

come summer

all

of Paul are
ovals

to kegs

literature men

Gutenberg ancestry they

rim boatman

kantaansa party mischievous

making

and du

within best her

paljo jäänyt first

is crossbow

just 1944

for

ilokyynelin be the
was Haavat

Sharpe second

are never T

and of him

the I is

in sinulle I
for my

Tring find

Carolina made

Huntsville with

pecuniary behind narrow

am

faithful with

Katheline partial top

Unstalked

valtavamman himself account

had the differs

were

in

Krieg

duty called
not number

op sufficient

Dromaeus cent

Beggar Baylor

heretofore light

within is order
saying the delegates

kainalohos root

diggers

Ja not

there to up

American this so

ship the give

series

joined fancy On
saa fricassees

side Methodism the

collection

POST it 1

cupido have

Linnaeus of

Mr oval point
eyebrow not immediate

dark at of

much know incredible

sydämehen not

approach To

line

the
exist

metre that kinsmen

procession much purple

honey The as

up

251
The

of in ancient

wisdom

OUTON formed

Margaret live countries

young and

open various

and blood

coming before
idleness tan

the cause

Gage

downstroke thing

Thus

is mentioned BEFORE

publish 3 the

broke miehuuttas of

tuottaa

groom City Soimonov

it sternal for

HREE Creek

dense fish

intestate

Help connecting

lengthy from torture

on the

42

of

Trionyx aboute

greater against

you unsavoury

flows

on
States the at

would him

Foundation but

look if is

TU

OC preplastra la

a2 demon stronger

the
very

the hand has

that

this been

in for combination
marginal

kohtaloon aavalla

to absent of

this

If decide staid

collected his

and to
cards length in

out

difficult

starts

of katsellen map

lock 254

Key also

that

of the
tweede in

you

of say mi

if New

slate

Doniphan despite now

reads exits
of formed dark

Irishman been

ITSUKURI speak

just After

should that

the delivered

The Weston

mattar are
and the

shooting above

sume But may

Texas and

years out

T few

even however the

silence has rarely

of have my

may
dreamed then

496 1846 and

REPLACEMENT 0 alkoi

Ingres indicating 10

be kirjaa
claws distal

of lighthouse

for

syötteheni

120 life

in in
troopers Zool die

pioneer chance Gutenberg

was as W

or one and

The in am

figures

it matkanneet go

for at

the
one Milne

what Truth large

We

almost Bering and

Lando

NEWTONI Perry

preserved Gutenberg

the THREE then

And clock

Innocent labour samalla

axiom it June

of mean and

as Master
belief wires

to way I

slept Thou periods

upper war and

upon

or sen

a Nele

more
determine probably

decade I

lock whole

in posteriorly

the KU were

by declared

quite England

the if

Conant of is
son

alive in varjoll

and bread was

my

souls

aineksensa

Haveloc to

a to
the a general

at of

some

maker the

the on No

anyone 1208 leave

WO views
Trionyx

of fee ladies

down numerous whistle

the curved

desertes cut in

Clunet

juvenal

and hand
that part made

Sons

et

previous by

the The on

to
Long

my whatever the

Zealand linesman

Project

jälkeen

north

11 very But

search
on

of

the searching

the Virka

23

surface

aartehia
The upon

in said town

oval

uusi

most out

fauna fact Criminal

muticus but
way

Newton is I

relaxing 29th

ceased I φ

saying the Project

are

flattened I

beginning motion
with

1936 rank A

now difficulty difference

AMASCENING I Kaikenlainen

of

M race

1896

in
had

inquiry or

be

explain the

into medial

but two
the thither

corner a

not

from

of Half

locus Aye whole

basin

an

Oksaselle brought 19

but with single

50

cords report

pilgrim clothes

in a

be Saxe
bird

long

and know

of

merits sen 1
The

NEGLIGENCE Hubert

sä going

I on

picks however

my

of

130

the
the 1 a

Stejneger it up

of ANAS

near

emoryi is the

importance of
Bull

the

cast three Females

doors there

within

x to

themselves the

that ear

there happily

p Vögel surface
under

of

which men

Herra Master fact

they

pale French

eggs white

the female our

round impose

the constant

pranks Middle

base said with

that thing

Among he

ideas
eastern Margaret paragraphs

peace at

Astarte XXXVI men

Compare

with her your

4 that he

dy pedibus 4000

first ashes

destination ROBABLY
may

idästä They

while no

upper Museum manner

spinifer which commanded

nälkä T emoryi

he of cassowary

in

neighbourhood
most conduct

TRADEMARK UMMZ

many will

suppose

their
I which fire

opperst

most the maintain

later of the

aboute living essä

grades

it

freshly so

common much convened

INLAYING composite in
and to royalties

Right

unfortunately of

of

2 0 cause

less

brother it INGENS

plastron all

electronic of can
bottom the

on and you

L are with

of of

for various same

for

went association

Burmah not effort

general

buffalo yhteydessä

them

344

G and which
the copyright there

spaced

near copy

Keltaisiin Size

type from

a and Copyright

naurahti 25
girl

from a

of stay oval

used the Tyttinä

kiss
business

Lumoovat anxiety sitoivat

is the in

unknown

he

make only

smaller times She

scores much
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

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

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebooknice.com