Content-Type: multipart/mixed; boundary="-------------0606181305916" This is a multi-part message in MIME format. ---------------0606181305916 Content-Type: text/plain; name="06-185.comments" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="06-185.comments" 12 pages ---------------0606181305916 Content-Type: text/plain; name="06-185.keywords" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="06-185.keywords" Chaos, Ergodic Hypothesis, Poincare' Recurrence, SRB distribution, Hyperbolic Systems, Time reversal, Fluctuation theorem, Chaotic hypothesis ---------------0606181305916 Content-Type: text/plain; name="vienna06.blg" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="vienna06.blg" This is BibTeX, Version 0.99c (Web2C 7.4.5) The top-level auxiliary file: vienna06.aux The style file: apsrev.bst Database file #1: 0Bibcaos.bib Warning--I didn't find a database entry for "endnote53" Warning--I didn't find a database entry for "endnote54" Reallocated wiz_functions (elt_size=4) to 6000 items from 3000. You've used 52 entries, 3411 wiz_defined-function locations, 1468 strings with 15997 characters, and the built_in function-call counts, 28071 in all, are: = -- 1578 > -- 558 < -- 134 + -- 403 - -- 146 * -- 3700 := -- 3239 add.period$ -- 52 call.type$ -- 52 change.case$ -- 208 chr.to.int$ -- 41 cite$ -- 52 duplicate$ -- 2726 empty$ -- 2684 format.name$ -- 384 if$ -- 5316 int.to.chr$ -- 12 int.to.str$ -- 53 missing$ -- 486 newline$ -- 171 num.names$ -- 156 pop$ -- 775 preamble$ -- 1 purify$ -- 208 quote$ -- 0 skip$ -- 969 stack$ -- 0 substring$ -- 881 swap$ -- 2079 text.length$ -- 88 text.prefix$ -- 0 top$ -- 0 type$ -- 264 warning$ -- 0 while$ -- 131 width$ -- 0 write$ -- 524 (There were 2 warnings) ---------------0606181305916 Content-Type: application/x-tex; name="vienna06.bbl" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="vienna06.bbl" \begin{thebibliography}{52} \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi \expandafter\ifx\csname bibnamefont\endcsname\relax \def\bibnamefont#1{#1}\fi \expandafter\ifx\csname bibfnamefont\endcsname\relax \def\bibfnamefont#1{#1}\fi \expandafter\ifx\csname citenamefont\endcsname\relax \def\citenamefont#1{#1}\fi \expandafter\ifx\csname url\endcsname\relax \def\url#1{\texttt{#1}}\fi \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi \providecommand{\bibinfo}[2]{#2} \providecommand{\eprint}[2][]{\url{#2}} \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{a}})}]{Bo66} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{{\"U}ber die mechanische {B}edeutung des zweiten {H}aupsatzes der {W\"a}rmetheorie}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{a}}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{b}})}]{Bo871a} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{{\"U}ber das {W\"a}rmegleichgewicht zwischen mehratomigen {G}asmolek{\"u}len}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{b}}). \bibitem[{\citenamefont{Brush}(2003)}]{Br03} \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Brush}}, \emph{\bibinfo{title}{History of modern physical sciences: The kinetic theory of gases}} (\bibinfo{publisher}{Imperial College Press}, \bibinfo{address}{London}, \bibinfo{year}{2003}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{c}})}]{Bo868} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Studien {\"u}ber das Gleichgewicht der le-\\bendigen Kraft zwischen bewegten materiellen Punkten}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen-\\{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{c}}). \bibitem[{\citenamefont{Boltzmann}(2000)}]{Fl00} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Entropie und Wahrscheinlichkeit}}, vol. \bibinfo{volume}{286} of \emph{\bibinfo{series}{Ostwalds Klassiker der Exacten Wissenschaften, Ed. D. Flamm}} (\bibinfo{publisher}{Verlag Harri Deutsch, ISBN 3-8171-3286-7}, \bibinfo{address}{Frankfurt am Main}, \bibinfo{year}{2000}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{d}})}]{Bo871b} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Einige allgemeine s{\"a}tze {\"u}ber {W\"a}rme-\\gleichgewicht}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{d}}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{e}})}]{Bo877a} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Bemerkungen {\"u}ber einige Probleme der mechanischen {W}{\"a}rmetheo\-rie}}, vol.~\bibinfo{volume}{2} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{e}}). \bibitem[{\citenamefont{Helmholtz}(1895{\natexlab{a}})}]{He884a} \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Helmholtz}}, \emph{\bibinfo{title}{Prinzipien der Statistik monocyklischer Systeme}}, vol. \bibinfo{volume}{III} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen}} (\bibinfo{publisher}{Barth}, \bibinfo{address}{Leipzig}, \bibinfo{year}{1895}{\natexlab{a}}). \bibitem[{\citenamefont{Helmholtz}(1895{\natexlab{b}})}]{He884b} \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Helmholtz}}, \emph{\bibinfo{title}{Studien zur Statistik monocyklischer Systeme}}, vol. \bibinfo{volume}{III} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen}} (\bibinfo{publisher}{Barth}, \bibinfo{address}{Leipzig}, \bibinfo{year}{1895}{\natexlab{b}}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{f}})}]{Bo884} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{{\"U}ber die {E}igenshaften monozyklischer und anderer damit verwandter {S}ysteme}}, vol.~\bibinfo{volume}{3} of \emph{\bibinfo{series}{{W}issen-\\schaftliche {A}bhandlungen}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New-York}, \bibinfo{year}{1968}{\natexlab{f}}). \bibitem[{\citenamefont{Gallavotti}(2000)}]{Ga00} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \emph{\bibinfo{title}{Statistical Mechanics. A short treatise\\}} (\bibinfo{publisher}{Springer Verlag}, \bibinfo{address}{Berlin}, \bibinfo{year}{2000}). \bibitem[{\citenamefont{Gallavotti}(1995)}]{Ga95a} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{Journal of Statistical Physics} \textbf{\bibinfo{volume}{78}}, \bibinfo{pages}{1571} (\bibinfo{year}{1995}). \bibitem[{\citenamefont{Fisher}(1964)}]{Fi64} \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Fisher}}, \bibinfo{journal}{Archive for Rational Mechanics and Analysis} \textbf{\bibinfo{volume}{17}}, \bibinfo{pages}{377} (\bibinfo{year}{1964}). \bibitem[{\citenamefont{Ruelle}(1968)}]{Ru68} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{Communications in Mathematical Physics} \textbf{\bibinfo{volume}{9}}, \bibinfo{pages}{267} (\bibinfo{year}{1968}). \bibitem[{\citenamefont{Brush}(1976)}]{Br76} \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Brush}}, \emph{\bibinfo{title}{The kind of motion that we call heat, (I, II)}} (\bibinfo{publisher}{North Holland}, \bibinfo{address}{Amsterdam}, \bibinfo{year}{1976}). \bibitem[{\citenamefont{Ehrenfest and Ehrenfest}(1990)}]{EE11} \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Ehrenfest}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Ehrenfest}}, \emph{\bibinfo{title}{The conceptual foundations of the statistical approach in Mechanics}} (\bibinfo{publisher}{Dover}, \bibinfo{address}{New York}, \bibinfo{year}{1990}). \bibitem[{\citenamefont{Boltzmann}(2003)}]{Bo96} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Reply to Zermelo's Remarks on the theory of heat}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{History of modern physical sciences: The kinetic theory of gases, ed. {S. B}rush}} (\bibinfo{publisher}{Imperial College Press}, \bibinfo{address}{London}, \bibinfo{year}{2003}). \bibitem[{\citenamefont{Boltzmann}(1964)}]{Bo96a} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{{L}ectures on gas theory, English edition annotated by S. Brush}} (\bibinfo{publisher}{University of California Press, Berkeley}, \bibinfo{year}{1964}). \bibitem[{\citenamefont{Thomson}(1874)}]{Th74} \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Thomson}}, \bibinfo{journal}{Proceedings of the Royal Society of Edinburgh} \textbf{\bibinfo{volume}{8}}, \bibinfo{pages}{325} (\bibinfo{year}{1874}). \bibitem[{\citenamefont{Levesque and Verlet}(1993)}]{LV93} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Levesque}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Verlet}}, \bibinfo{journal}{Journal of Statistical Physics} \textbf{\bibinfo{volume}{72}}, \bibinfo{pages}{519} (\bibinfo{year}{1993}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{g}})}]{Bo877b} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{{\"U}ber die {B}eziehung zwischen dem zwei\-ten {H}aupt\-satze der mechanischen {W}{\"a}rmetheo\-rie und der {W}ahrscheinlichkeitsrechnung, respektive den {S}{\"a}tz\-en {\"u}ber das {W}{\"a}rme\-gleichgewicht}}, vol.~\bibinfo{volume}{2} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{g}}). \bibitem[{\citenamefont{Ruelle}(1996)}]{Ru96} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{Journal of Statistical Physics} \textbf{\bibinfo{volume}{85}}, \bibinfo{pages}{1} (\bibinfo{year}{1996}). \bibitem[{\citenamefont{Gallavotti et~al.}(2004)\citenamefont{Gallavotti, Bonetto, and Gentile}}]{GBG04} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Bonetto}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gentile}}, \emph{\bibinfo{title}{Aspects of the ergodic, qualitative and statistical theory of motion}} (\bibinfo{publisher}{Springer Verlag}, \bibinfo{address}{Berlin}, \bibinfo{year}{2004}). \bibitem[{\citenamefont{Gallavotti}(1996)}]{Ga96} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{Journal of Statistical Physics} \textbf{\bibinfo{volume}{84}}, \bibinfo{pages}{899} (\bibinfo{year}{1996}). \bibitem[{\citenamefont{Adler and Weiss}(1970)}]{AW70} \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Adler}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{B.}~\bibnamefont{Weiss}}, \emph{\bibinfo{title}{Similarity of automorphism of the tours}}, vol.~\bibinfo{volume}{98} of \emph{\bibinfo{series}{Memoirs of the American Mathematical Society}} (\bibinfo{publisher}{American Mathematical Society}, \bibinfo{year}{1970}). \bibitem[{\citenamefont{Sinai}(1968{\natexlab{a}})}]{Si68a} \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Sinai}}, \bibinfo{journal}{Functional Analysis and Applications} \textbf{\bibinfo{volume}{2}}, \bibinfo{pages}{64} (\bibinfo{year}{1968}{\natexlab{a}}). \bibitem[{\citenamefont{Sinai}(1968{\natexlab{b}})}]{Si68b} \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Sinai}}, \bibinfo{journal}{Functional analysis and Applications} \textbf{\bibinfo{volume}{2}}, \bibinfo{pages}{70} (\bibinfo{year}{1968}{\natexlab{b}}). \bibitem[{\citenamefont{Sinai}(1972)}]{Si72} \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Sinai}}, \bibinfo{journal}{Russian Mathematical Surveys} \textbf{\bibinfo{volume}{27}}, \bibinfo{pages}{21} (\bibinfo{year}{1972}). \bibitem[{\citenamefont{Bowen}(1970)}]{Bo70a} \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Bowen}}, \bibinfo{journal}{American Journal of Mathematics} \textbf{\bibinfo{volume}{92}}, \bibinfo{pages}{725} (\bibinfo{year}{1970}). \bibitem[{\citenamefont{Bowen and Ruelle}(1975)}]{BR75} \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Bowen}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{Inventiones Mathematicae} \textbf{\bibinfo{volume}{29}}, \bibinfo{pages}{181} (\bibinfo{year}{1975}). \bibitem[{\citenamefont{Ruelle}(1976)}]{Ru76} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{American Journal of Mathematics} \textbf{\bibinfo{volume}{98}}, \bibinfo{pages}{619} (\bibinfo{year}{1976}). \bibitem[{\citenamefont{Ruelle}(1980)}]{Ru80} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{Annals of the New York Academy of Sciences} \textbf{\bibinfo{volume}{357}}, \bibinfo{pages}{1} (\bibinfo{year}{1980}). \bibitem[{\citenamefont{Gallavotti and Cohen}(1995)}]{GC95} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Cohen}}, \bibinfo{journal}{Physical Review Letters} \textbf{\bibinfo{volume}{74}}, \bibinfo{pages}{2694} (\bibinfo{year}{1995}). \bibitem[{\citenamefont{Ruelle}(1973)}]{Ru73} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \emph{\bibinfo{title}{Ergodic theory}}, vol. \bibinfo{volume}{Suppl X} of \emph{\bibinfo{series}{The Boltzmann equation, ed. E.G.D Cohen, W. Thirring, Acta Physica Austriaca}} (\bibinfo{publisher}{Springer}, \bibinfo{address}{New York}, \bibinfo{year}{1973}). \bibitem[{\citenamefont{Evans et~al.}(1993)\citenamefont{Evans, Cohen, and Morriss}}]{ECM93} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Evans}}, \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Cohen}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Morriss}}, \bibinfo{journal}{Physical Review Letters} \textbf{\bibinfo{volume}{70}}, \bibinfo{pages}{2401} (\bibinfo{year}{1993}). \bibitem[{\citenamefont{Gentile}(1998)}]{Ge98} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gentile}}, \bibinfo{journal}{Forum Mathematicum} \textbf{\bibinfo{volume}{10}}, \bibinfo{pages}{89} (\bibinfo{year}{1998}). \bibitem[{\citenamefont{Gallavotti}(2004{\natexlab{a}})}]{Ga04b} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{Chaos} \textbf{\bibinfo{volume}{14}}, \bibinfo{pages}{680} (\bibinfo{year}{2004}{\natexlab{a}}). \bibitem[{\citenamefont{Gallavotti}(2001)}]{Ga01} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{Communication in Mathematical Physics} \textbf{\bibinfo{volume}{224}}, \bibinfo{pages}{107} (\bibinfo{year}{2001}). \bibitem[{\citenamefont{Goldstein and Lebowitz}(2004)}]{GL03} \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Goldstein}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Lebowitz}}, \bibinfo{journal}{Physica D} \textbf{\bibinfo{volume}{193}}, \bibinfo{pages}{53} (\bibinfo{year}{2004}). \bibitem[{\citenamefont{Evans and Morriss}(1990)}]{EM90} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Evans}} \bibnamefont{and} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Morriss}}, \emph{\bibinfo{title}{Statistical Mechanics of Non{\-}equilibrium Fluids}} (\bibinfo{publisher}{Academic Press}, \bibinfo{address}{New-York}, \bibinfo{year}{1990}). \bibitem[{\citenamefont{Gallavotti}(2006)}]{Ga06} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{Chaos} \textbf{\bibinfo{volume}{16}}, \bibinfo{pages}{023130 (+7)} (\bibinfo{year}{2006}). \bibitem[{\citenamefont{Bonetto et~al.}(2006{\natexlab{a}})\citenamefont{Bonetto, Gallavotti, Giuliani, and Zamponi}}]{BGGZ05} \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Bonetto}}, \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Giuliani}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Zamponi}}, \bibinfo{journal}{Journal of Statistical Mechanics} \textbf{\bibinfo{volume}{123}}, \bibinfo{pages}{39} (\bibinfo{year}{2006}{\natexlab{a}}). \bibitem[{\citenamefont{Bonetto et~al.}(2006{\natexlab{b}})\citenamefont{Bonetto, Gallavotti, Giuliani, and Zamponi}}]{BGGZ06} \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Bonetto}}, \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Giuliani}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{F.}~\bibnamefont{Zamponi}}, \bibinfo{journal}{Journal of Statistical Mechanics (cond-mat/0601683)} p. \bibinfo{pages}{P05009} (\bibinfo{year}{2006}{\natexlab{b}}). \bibitem[{\citenamefont{Jarzynski}(1999)}]{Ja99} \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Jarzynski}}, \bibinfo{journal}{Journal of Statistical Physics} \textbf{\bibinfo{volume}{98}}, \bibinfo{pages}{77} (\bibinfo{year}{1999}). \bibitem[{\citenamefont{J.P.Eckmann et~al.}(1999)\citenamefont{J.P.Eckmann, Pillet, and Bellet}}]{EPR99} \bibinfo{author}{\bibnamefont{J.P.Eckmann}}, \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Pillet}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{L.~R.} \bibnamefont{Bellet}}, \bibinfo{journal}{Communications in Mathematical Physics} \textbf{\bibinfo{volume}{201}}, \bibinfo{pages}{657} (\bibinfo{year}{1999}). \bibitem[{\citenamefont{Ruelle}(2006)}]{Ru06} \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Ruelle}}, \bibinfo{journal}{Communications in Mathematical Physics} \textbf{\bibinfo{volume}{??}}, \bibinfo{pages}{??} (\bibinfo{year}{2006}). \bibitem[{\citenamefont{Boltzmann}(1968{\natexlab{h}})}]{Bo72} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Weitere {S}tudien {\"u}ber das {W\"a}rmegleich-\\gewicht unter {G}asmolek{\"u}len}}, vol.~\bibinfo{volume}{1} of \emph{\bibinfo{series}{{W}is\-sen\-schaft\-li\-che {A}bhandlungen, ed. {F}. {H}asen{\"o}hrl}} (\bibinfo{publisher}{Chelsea}, \bibinfo{address}{New York}, \bibinfo{year}{1968}{\natexlab{h}}). \bibitem[{\citenamefont{Klein}(1973)}]{Kl73} \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Klein}}, \emph{\bibinfo{title}{The development of Boltzmann statistical ideas}}, vol. \bibinfo{volume}{Suppl X} of \emph{\bibinfo{series}{The Boltzmann equation, ed. E.G.D Cohen, W. Thirring, Acta Physica Austriaca}} (\bibinfo{publisher}{Springer}, \bibinfo{address}{New York}, \bibinfo{year}{1973}). \bibitem[{\citenamefont{Garrido et~al.}(2005)\citenamefont{Garrido, Goldstein, and Lebowitz}}]{GGL04} \bibinfo{author}{\bibfnamefont{P.~L.} \bibnamefont{Garrido}}, \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Goldstein}}, \bibnamefont{and} \bibinfo{author}{\bibfnamefont{J.~L.} \bibnamefont{Lebowitz}}, \bibinfo{journal}{Physical Review Letters} \textbf{\bibinfo{volume}{92}}, \bibinfo{pages}{050602 (+4)} (\bibinfo{year}{2005}). \bibitem[{\citenamefont{Gallavotti}(2004{\natexlab{b}})}]{Ga04} \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Gallavotti}}, \bibinfo{journal}{cond-mat/0402676} (\bibinfo{year}{2004}{\natexlab{b}}). \bibitem[{\citenamefont{Boltzmann}(1974)}]{Bo874} \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Boltzmann}}, \emph{\bibinfo{title}{Theoretical Physics and philosophical writings, ed. B. Mc Guinness}} (\bibinfo{publisher}{Reidel}, \bibinfo{address}{Dordrecht}, \bibinfo{year}{1974}). \bibitem[{\citenamefont{Gibbs}(1902)}]{Gi02} \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Gibbs}}, \emph{\bibinfo{title}{Elementary principles in statistical mechanics (reprint)}} (\bibinfo{publisher}{Schribner}, \bibinfo{address}{Cambridge (USA)}, \bibinfo{year}{1902}). \end{thebibliography} ---------------0606181305916 Content-Type: application/x-tex; name="vienna06.tex" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="vienna06.tex" %**start of header %\documentclass[10pt,twocolumn]{article} \documentclass[pra,twocolumn,showpacs,superscriptaddress,floatfix]{revtex4} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%% LETTERE GRECHE E LATINE IN NERETTO %%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % lettere greche e latine in neretto italico - pag.430 del manuale \font\cs=cmcsc10\font\sc=cmcsc10\font\css=cmcsc8% \font\ss=cmss10\font\sss=cmss8% \font\crs=cmbx8% \font\tenmib=cmmib10 \font\sevenmib=cmmib7 \font\fivemib=cmmib5 \font\msytw=msbm9 scaled\magstep1% \textfont5=\tenmib\scriptfont5=\sevenmib\scriptscriptfont5=\fivemib \mathchardef\Ba = "050B %alfa \mathchardef\Bb = "050C %beta \mathchardef\Bg = "050D %gamma \mathchardef\Bd = "050E %delta \mathchardef\Be = "0522 %varepsilon \mathchardef\Bee = "050F %epsilon \mathchardef\Bz = "0510 %zeta \mathchardef\Bh = "0511 %eta \mathchardef\Bthh = "0512 %teta \mathchardef\Bth = "0523 %varteta \mathchardef\Bi = "0513 %iota \mathchardef\Bk = "0514 %kappa \mathchardef\Bl = "0515 %lambda \mathchardef\Bm = "0516 %mu \mathchardef\Bn = "0517 %nu \mathchardef\Bx = "0518 %xi \mathchardef\Bom = "0530 %omi \mathchardef\Bp = "0519 %pi \mathchardef\Br = "0525 %ro \mathchardef\Bro = "051A %varrho \mathchardef\Bs = "051B %sigma \mathchardef\Bsi = "0526 %varsigma \mathchardef\Bt = "051C %tau \mathchardef\Bu = "051D %upsilon \mathchardef\Bf = "0527 %phi \mathchardef\Bff = "051E %varphi \mathchardef\Bch = "051F %chi \mathchardef\Bps = "0520 %psi \mathchardef\Bo = "0521 %omega \mathchardef\Bome = "0524 %varomega \mathchardef\BG = "0500 %Gamma \mathchardef\BD = "0501 %Delta \mathchardef\BTh = "0502 %Theta \mathchardef\BL = "0503 %Lambda \mathchardef\BX = "0504 %Xi \mathchardef\BP = "0505 %Pi \mathchardef\BS = "0506 %Sigma \mathchardef\BU = "0507 %Upsilon \mathchardef\BF = "0508 %Fi \mathchardef\BPs = "0509 %Psi \mathchardef\BO = "050A %Omega \mathchardef\BDpr = "0540 %Dpr \mathchardef\Bstl = "053F %* \def\BK{\bf K} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%% SIMBOLI VARI %%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \let\a=\alpha \let\b=\beta \let\g=\gamma \let\d=\delta \let\e=\varepsilon \let\z=\zeta \let\h=\eta \let\th=\theta \let\k=\kappa \let\l=\lambda \let\m=\mu \let\n=\nu \let\x=\xi \let\p=\pi \let\r=\rho \let\s=\sigma \let\t=\tau \let\f=\varphi \let\ph=\varphi\let\c=\chi \let\ch=\chi \let\ps=\psi \let\y=\upsilon \let\o=\omega\let\si=\varsigma \let\G=\Gamma \let\D=\Delta \let\Th=\Theta\let\L=\Lambda \let\X=\Xi \let\P=\Pi \let\Si=\Sigma \let\F=\Phi \let\Ps=\Psi \let\O=\Omega \let\Y=\Upsilon %\def\\{\hfill\break} \def\*{\vglue0.3truecm} \let\==\equiv \let\0=\noindent \def\ie{{\it i.e.\ }} \def\rhs{{\it r.h.s.}\ } \def\tende#1{\,\vtop{\ialign{##\cr\rightarrowfill\cr \noalign{\kern-1pt\nointerlineskip} \hskip3.pt${\scriptstyle #1}$\hskip3.pt\cr}}\,} \def\otto{\,{\kern-1.truept\leftarrow\kern-5.truept\to\kern-1.truept}\,} \def\media#1{{\langle#1\rangle}} \def\defi{\,{\buildrel def\over=}\,} \def\RRR{\hbox{\msytw R}} \def\EE{{\cal E}}\def\NN{{\cal N}}\def\FF{{\cal F}}\def\CC{{\cal C}} \newcommand\revtex{{R\kern-0.24mm\lower0.5mm\hbox{E}\kern-0.6mm V\kern-0.5mm% \lower0.5mm\hbox{T}\kern-0.4mm E\kern-.2mm \lower0.5mm\hbox{X}}} \def\V#1{{\bf#1}} \def\lis#1{\overline#1} \def\eg{{\it e.g.\ }} \def\etc{{\it etc.\ }} \def\ap{{\it a priori\ }} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \newdimen\xshift \newdimen\xwidth \newdimen\yshift \newdimen\ywidth \def\ins#1#2#3{\vbox to0pt{\kern-#2\hbox{\kern#1 #3}\vss}\nointerlineskip} \def\eqfig#1#2#3#4#5{ \par\xwidth=#1 \xshift=\hsize \advance\xshift by-\xwidth \divide\xshift by 2 \yshift=#2 \divide\yshift by 2% %\line {\hglue\xshift \vbox to #2{\vfil #3 \special{psfile=#4.eps} }\hfill\raise\yshift\hbox{#5}}} %%%%%%%%%%%%%%%%%%%%%%%%%%% \font\tenmib=cmmib10 \font\eightmib=cmmib8 \font\sevenmib=cmmib7\font\fivemib=cmmib5 \font\ottoit=cmti8\font\fiveit=cmti5\font\sixit=cmti6%% \font\fivei=cmmi5\font\sixi=cmmi6\font\ottoi=cmmi8 \font\ottorm=cmr8\font\fiverm=cmr5\font\sixrm=cmr6 \font\ottosy=cmsy8\font\sixsy=cmsy6\font\fivesy=cmsy5%% \font\ottobf=cmbx8\font\sixbf=cmbx6\font\fivebf=cmbx5% \font\ottott=cmtt8% \font\ottocss=cmcsc8% \font\ottosl=cmsl8% \def\ottopunti{\def\rm{\fam0\ottorm}\def\it{\fam6\ottoit}% \def\bf{\fam7\ottobf}% \textfont1=\ottoi\scriptfont1=\sixi\scriptscriptfont1=\fivei% \textfont2=\ottosy\scriptfont2=\sixsy\scriptscriptfont2=\fivesy% %\textfont3=\tenex\scriptfont3=\tenex\scriptscriptfont3=\tenex% \textfont4=\ottocss\scriptfont4=\sc\scriptscriptfont4=\sc% %\scriptfont4=\ottocss\scriptscriptfont4=\ottocss% \textfont5=\eightmib\scriptfont5=\sevenmib\scriptscriptfont5=\fivemib% \textfont6=\ottoit\scriptfont6=\sixit\scriptscriptfont6=\fiveit% \textfont7=\ottobf\scriptfont7=\sixbf\scriptscriptfont7=\fivebf% %\textfont\bffam=\eightmib\scriptfont\bffam=\sevenmib% %\scriptscriptfont\bffam=\fivemib% \setbox\strutbox=\hbox{\vrule height7pt depth2pt width0pt}% \normalbaselineskip=9pt\rm} \let\nota=\ottopunti %**end of header \begin{document} \voffset0.5truecm \title{Entropy, Nonequilibrium, Chaos and Infinitesimals} \* \author{Giovanni Gallavotti} \affiliation{Dipartimento di Fisica and INFN, Universit\`a di Roma {\em La Sapienza}, P.~A.~Moro 2, 00185, Roma, Italy} \date{\today} \begin{abstract} \it A survey of the approach to Statistical Mechanics following Boltzmann's theory of ensembles and ergodic hypothesis leading to chaoticity as a unifying principle of equilibrium and nonequilibrium Statistical Mechanics. \end{abstract} \pacs{47.52.+j, 05.45.-a, 05.70.Ln, 05.20.-y} \maketitle \section{Boltzmann and Entropy} \* Since the earliest works {\cs Boltzmann} aimed at a microscopic interpretation or ``proof'' of the newly formulated second Law of Thermodynamics and of the associated concept of {\it entropy}, \cite{Bo66}. At the time, since the works of {\cs Bernoulli, Avogadro, Herapath, Waterstone, Kr\"onig, Clausius} it was well established that there should be an identification between absolute temperature and average kinetic energy at least in gas theory. {\cs Boltzmann} starts in \cite{Bo66} by stating clearly that in general (\ie not only for gases) temperature and average kinetic energy must be identified and gives a derivation of the second law for isocore (\ie constant volume) transformations. This first derivation makes use of a periodicity assumption on the motion of (each) gas particle to obtain the existence of the time average of the kinetic energy and seems to fail if the motions are not periodic. Nevertheless {\cs Boltzmann} insisted in conceiving aperiodic motions as periodic with infinite period: see \cite{Bo66} where on p. 30 one finds that ``{\it... this explanation is nothing else but the mathematical formulation of the theorem according to which the paths that do not close themselves in any finite time can be regarded as closed in an infinite time}''. He therefore pursued the implied mechanical proof of Thermodynamics to the extreme consequences. A ``proof'' of the second law meant to look for properties of the trajectories in phase space of a mechanical equation of motion, like time averages of suitable observables, which could have the interpretation of thermodynamic quantities like pressure $p$, volume $V$, energy $U$, temperature $T$, entropy $S$ {\it and be related} by the thermodynamic relations, namely \begin{eqnarray}dS=\frac{dU\,+\,p\,dV}{T}\label{1.1}\end{eqnarray} % where $dU,dV,dS$ are the variations of $U,V,S$ when the control parameters of the system are infinitesimally changed. The extreme consequence was the {\it ergodic hypothesis} which was first mentioned by {\cs Boltzmann} around 1870, see p. 237 in \cite{Bo871a}, for the internal motion of atoms in single molecules of a gas (``{\it it is clear that the various gas molecules will go through all possible states of motion}'' which, however, could possibly be understood from the context to be different from the ergodic hypothesis, see \cite{Br03}, because the molecules undergo from time to time collisions). See also p. 96 in \cite{Bo868} and p. xxxvii in the recent collection by {\cs Flamm} \cite{Fl00}. Considering a collection of copies of the system alike to a large molecule, p. 284 in \cite{Bo871b}, the same assumption became what is often referred as the ergodicity property of the entire gas. It implied that, by considering all motions periodic, kinetic energy equipartition would follow and, better (see p. 287 in \cite{Bo871b}), even what we call now the microcanonical distribution would follow (as well as the canonical distribution). The hypothesis was taken up also by {\cs Maxwell} (1879), see p. 506 in \cite{Br03}).\footnote{[The paper \cite{Bo871b} is a key work, albeit admittedly obscure: in modern notations it considers a system of equations of motion with dimension $n$ and $0$ divergence admitting $n-k$ constants of motion, $\f_{k+1},\ldots,\f_n$, and are decribed by coordinates $s_1,\ldots,s_n$. Then the distribution proportional to $\prod_{j=k+1}^n \d(\f_j-a_j)\cdot \prod_{i=1}^n ds_i$ is {\it invariant} and can be written ${\prod_{i=1}^k ds_i}\cdot\frac1{|\det \partial(\f_{n-k+1},\ldots,\f_n)|}$ where the last denominator denotes the Jacobian determinant (``last multiplier'') of $\f_{n-k+1},\ldots,\f_k$ with respect to $s_{k+1},\ldots,s_n$ evaluated at the given values $a_{k+1},\ldots,a_n$ of the constants of motion ({\cs Boltzmann} calls this an instance of the ``last multiplier principle'' of {\cs Jacobi}). If the system has only one constant of motion, namely the energy $H=\chi+\psi$ with $\chi\,=$ potential energy and $\ps\,=$ kinetic energy, this is the microcanonical distribution, as also recognized by {\cs Gibbs} in the introduction of his book, \cite{Gi02} (where he quotes \cite{Bo871b}, but giving to it the title of its first section).]} In this way {\cs Boltzmann} was able to derive various thermodynamic consequences and a proof of \ref{1.1}, see \cite{Bo877a}, and was led to exhibiting a remarkable example of what later would be called a ``{\it thermodynamic analogy}'' (Sec.III of \cite{Bo877a}). This meant the existence of quantities associated with the phase space of a mechanical equation of motion (typically defined as time averages over the solutions of the equations of motion), which could be given thermodynamic names like equilibrium state, pressure $p$, volume $V$, energy $U$, temperature $T$, entropy $S$ {\it and be related} by the thermodynamic relations that are expected to hold between the physical quantities bearing the same name, namely \ref{1.1}. The notion of mechanical thermodynamic analogy was formulated and introduced by {\cs Helmoltz} for general systems admitting only periodic motions (called {\it monocyclic}), \cite{He884a,He884b}. The proposal provided a new perspective and generated the new guiding idea that the Thermodynamic relations would hold in {\it every mechanical system}, from the small and simple to the large and complex: in the first cases the relations would be trivial identities of no or little interest, just {\it thermodynamic analogies}, but in the large systems they would become nontrivial and interesting being relations of general validity. In other words they would be a kind of symmetry property of Hamiltonian Mechanics. The case of spatially confined systems with one degree of freedom was easy (easier than the example already given in \cite{Bo877a}): with all motions periodic, the microscopic state was indentified with the phase space point $(p,q)$ representing the full mechanical state of the system, the {\it macroscopic state} in the corresponding thermodynamic analogy was identified with the energy surface $H(p,q)=\frac1{2m} p^2+\f_V(q)=U$, where $m$ is the mass and $\f_V$ is the potential energy which confines the motion in position space, \ie in $q$, and depends on a parameter $V$. The state is completely determined by two parameters $U,V$. Average kinetic energy $T=\lim_{\t\to\infty}\frac1\t\int_0^\t K(p(t)) dt$ is identified with temperature; energy is identified with $U$: then if pressure $p$ is defined as the time average $\lim_{\t\to\infty} -\frac1\t\int_0^\t \partial_V \f_V(q(t))dt$ the quantities $T,p$ become functions $p=p(U,V)$,$T=T(U,V)$ of the parameters $U,V$ determining the state of the system and the \ref{1.1} should hold. Indeed the limits as $\t\to\infty$ exist in such a simple case, in which all motions are periodic and confined between $q_\pm=q_\pm(U,V)$ (where $U=\f_V(q_\pm)$); it is $dt\=\frac{dq}{|\dot q|}=\frac{dq}{\sqrt{2(U-\f_V(q))/m}}$ and the period of the oscillations is $\t_0=\t_0(U,V)=2\int_{q_-}^{q_+} \frac{dq}{\sqrt{2(U-\f_V(q))/m}}$, hence (p. 127 in \cite{Bo884} and Ch. I in \cite{Ga00}), \begin{eqnarray}\label{1.2} &T=\frac2{\t_0} \int_{q_-(U,V)}^{q_+(U,V)} \frac{m}{2}{ \sqrt{\frac 2m(U-\f_V(q))}}dq, \qquad \\ &p=\frac2{\t_0} \int_{q_-(U,V)}^{q_+(U,V)} \frac{\partial_V\f_V(q)}{\sqrt{\frac m2(U-\f_V(q))}}dq\nonumber \end{eqnarray} % and it is immediate to check, as in \cite{Bo877a}, that \ref{1.1} is fulfilled by setting \begin{eqnarray}\label{1.3} &S(U,V)\,= \,2\, \log \int_{H=U} pdq=\nonumber\\ &=\,2\,\log \int_{q_-(U,V)}^{q_+(U,V)} \sqrt{2m (U-\f_V(q))}\,dq\end{eqnarray} % The case of the central motion studied in \cite{Bo877a} was another instance of {\it monocyclic} systems, \ie systems with only periodic motions. Then in the fundamental paper \cite{Bo884}, following and inspired by the quoted works of {\cs Helmoltz}, {\cs Boltzmann} was able to achieve what I would call the completion of his program of deducing the second law (\ref{1.1}) from Mechanics. If \* \0(1) the {\it absolute temperature} $T$ is identified with the average kinetic energy over the periodic motion following the initial datum $(\V p,\V q)$ of a macroscopic collection of $N$ identical particles interacting with a quite {\it arbitrary} pair interaction, and \0(2) the {\it energy} $U$ is $H(\V p,\V q)$ sum of kinetic and of a potential energy, \0(3) the {\it volume} $V$ is the volume of the region where the positions $\V q$ are confined (typically by a hard wall potential), \0(4) the {\it pressure} $p$ is the average force exercized on the walls by the colliding particles, \* \0then, from the assumption that each point would evolve periodically visiting every other point on the energy surface (\ie assuming that the system could be regarded as monocyclic, see \cite{Ga00} Appendix 9.3 for details) it would follow that the quantity $p$ could be identified with the $\media{-\partial_V\f_V}$, time average of $-\partial_V \f_V$, and \ref{1.1} would follow as a {\it heat theorem}. The heat theorem would therefore be a consequence of the general properties of monocyclic systems. This led {\cs Boltzmann} to realize, in the same paper, that there were a large number of mechanical models of Thermodynamics: the macroscopic states could be identified with regions of phase spaces invariant under time evolution and their points would contribute to the average values of the quantities with thermodynamic interpretation (\ie $p,V,U,T$) with a weight (hence a probability) also invariant under time evolution. Hence imagining the weights as a density function one would see the evolution as a motion of phase space points leaving the density fixed. Such distributions on phase space were called {\it monodic} (because they keep their identity with time or, as we say, are invariant): and in \cite{Bo884} several collections of weights or {\it monodes} were introduced: today we call them collections of invariant distributions on phase space or {\it ensembles}. Among the ensembles $\EE$, \ie collections of monodes, {\cs Boltzmann} singled out the ensembles called {\it orthodes} (``behaving correctly''): they were the families of probability distributions depending on a few parameters (normally $2$ for simple one component systems) such that the corresponding averages $p,V,U,T$, defined in (1-4) above, would vary when the parameters were varied causing variations $dU,dV$ of average energy and volume in such a way that the \rhs of \ref{1.1} would be an exact differential, thereby defining the {\it entropy} $S$ as a function of state, see \cite{Ga95a,Ga00}. The ergodic hypothesis yields the ``orthodicity'' of the ensemble $\EE$ that today we call {\it microcanonic} (in \cite{Bo884} it was named {\it ergode}): but ergodicity, \ie the dynamical property that evolution would make every phase space point visit every other, was not necessary to the orthodicity proof of the ergode. In fact in \cite{Bo884} the relation \ref{1.1} is proved directly without recourse to dynamical properties (as we do today, see \cite{Fi64,Ru68,Ga00}); and in the same way the orthodicity of the {\it canonical ensemble} (called {\it holode} in \cite{Bo884}) was obtained and shown to generate a Thermodynamics which is equivalent to the one associated with the microcanonical ensemble. \footnote{[Still today a different interpretation of the word ``ensemble'' is widely used: the above is based on what {\cs Boltzmann} calls ``{\it Gattung von Monoden}'', see p.132, l. 14 of \cite{Bo884}: unfortunately he is not really consistent in the use of the name ``monode'' because, for instance in p. 134 of the same reference, he clearly calls ``monode'' a collection of invariant distributions rather than a single one; further confusion is generated by a typo on p. 132, l. 22, where the word ``ergode'' is used instead of ``holode'' while the ``ergode'' is defined only on p. 134. It seems beyond doubt that ``holode'' and ``ergode'' were intended by Boltzmann to be {\it collections} $\EE$ of invariant distributions (parameterized respectively by $U,V$ or by $(k_B T)^{-1},V$ in modern notations): Gibbs instead called ``ensemble'' each single invariant distribution, or at least that is what is often stated. It seems that the original names proposed by {\cs Boltzmann} are more appropriate, but of course we must accept calling ``microcanonical ensemble'' the ergode and ``canonical ensemble'' the holode, see \cite{Ga00}.]} In the end in \cite{Bo871b} and, in final form, in \cite{Bo884} the theory of ensembles and of their equivalence was born without need of the ergodic property: the still important role of the ergodic hypothesis was to guarantee that the quantities $p,V,U,T,S$ defined by orthodic averages with respect to invariant distributions on phase space had the physical meaning implied by their names (this was true for the microcanonical ensemble by the ergodic hypothesis, and for the other ensembles by the equivalence). At the same time entropy had received a full microscopic interpretation consistent with, but independent of, the one arising from the {\it Boltzmann's equation} in the rarefied gases case, which can be seen as a quite independent development of {\cs Boltzmann}'s work. Furthermore it became clear that the entropy could be identified, up to a universal proportionality constant $k_B$, with the volume of phase space enclosed by the energy surface. Unfortunately the paper \cite{Bo884} has been overlooked until quite recently by many, actually by most, physicists possibly because it starts, in discussing the thermodynamic analogy, by giving the Saturn rings as an ``example'': a brilliant one, certainly but perhaps discouraging for the suspicious readers of this deep and original paper on Thermodynamics. See p.242 and p. 368 in \cite{Br76} for an exception, possibly the first. \section{Boltzmann's discrete vision of the ergodic problem} The ergodic hypothesis could not possibly say that every point of the energy surface in phase space visits in due time (the {\it recurrence time}) every other, see also p. 505 and following in \cite{Br03}. But this statement was attributed to {\cs Boltzmann} and criticized over and over again (even by Physicists, including in the influential book, \cite{EE11}, although enlightened mathematicians could see better, see p.385 in \cite{Br76}): however for {\cs Boltzmann} phase space was discrete and points in phase space were {\it cells} $\D$ with finite size, that I will call $h$. And time evolution was a permutation of the cells: ergodicity meant therefore that the permutation was a {\it one cycle permutation}. This conception, perfectly meaningful mathematically, was apparently completely misunderstood by his critics: yet it was clearly stated in one of the replies to {\cs Zermelo}, \cite{Bo96}, and in the book on gases, \cite{Bo96a}, see also \cite{Ga95a} and the {\cs de Courtenay}'s communication in this Symposium. In order to explain how a reversible dynamics could be compatible with the irreversibility of macroscopic phenomena he had, in fact, to estimate the recurrence time. This was done by multiplying the typical time over which a microscopic event (\ie a collision) generates a variation of the coordinates of an order of magnitude appreciable on microscopic scales (\ie a time interval of $\sim10^{-12}$s and a coordinate variation of the order of $1^o$A) times the number of cells into which phase space was imagined to be subdivided. The latter number was obtained by dividing the phase space around the energy surface into equal boxes of a size $h$ equal to the $3N$-th power of $\r^{-\frac13}$ times $\sqrt{m k_BT}$ with $\r$ the numerical density and $k_B$ Boltzmann's constant and $T$ temperature. With the data for $H_2$ at normal conditions in $1{\rm cm}^3$ an ealier estimate of Thomson, \cite{Th74}, was rederived (and a recurrence time scale so large that it would be immaterial to measure it in seconds or in ages of the Universe). Of course conceiving phase space as discrete is essential to formulate the ergodicity property in an acceptable way: it does not, however, make it easier to prove it even in the discrete sense just mentioned (nor in the sense acquired later when it was formulated mathematically for systems with continuous phase space). It is in fact very difficult to be \ap\ sure that the dynamics is an evolution which has only one cycle. Actually this is very doubtful: as one realizes if one attempts a numerical simulation of an equation of motion which generates motions which are ergodic in the mathematical sense. And the difficulty is already manifest in the simpler problem of simulating differential equations in a way which rigorously respects the uniqueness theorem. In computers the microscopic states are rigorously realized as cells (because points are described by integers, so that the cells sizes are limited by the precision of hardware and software) and phase space is finite. By construction simulation programs map a cell into another: but it is extremely difficult, and possible only in very special cases (among which the only nontrivial that I know is \cite{LV93}) without dedicating an inordinate computing time to insure a $1-1$ correspondence between the cells. Nevertheless the idea that phase space is discrete and motion is a permutation of its points is very appealing because it gives a privileged role to the {\it uniform distribution} on the phase space region in which the motion develops (\ie the energy surface, if the ergodic hypothesis holds). However it is necessary, for consistency, that the phase space cells volume does not change with time, see Ch. 1 in \cite{Ga00}: this is a property that holds for Hamiltonian evolutions and therefore allows us to imagine the ergodic hypothesis as consistent with the predictions of Statistical Mechanics. \* \section{Boltzmann's heritage} \* The success of the ergodic hypothesis has several aspects. One that will not be considered further is that it is not necessary: this is quite clear as in the end we want to find the relations between a very limited number of observables and we do not need for that an assumption which tells us the values of all possible averages, most of which concern ``wild'' observables (like the position of a tagged particle). The consequence is that the ergodic hypothesis is intended in the sense that confined Hamiltonian systems ``can be regarded as ergodic for the purpose of studying their equilibrium properties''. What is, perhaps, the most interesting aspect of the hypothesis is that it can hold for systems of any size and lead to relations which are essentially size independent as well as model independent and which become interesting properties when considered for macroscopic systems. {\it Is it possible to follow the same path in studying nonequilibrium phenomena?} The simplest such phenomena arise in stationary states of systems subject to the action of nonconservative forces and of suitable heat removing forces (whose function is to forbid indefinite build up of energy in the system). Such states are realized in Physics with great accuracy for very long times, in most cases longer than the available observation times. For instance it is possible to keep a current circulating in a wire subject to an electromotive force for a very long time, provided a suitable cooling device is attached to the wire. As in equilibrium, the stationary states of a system will be described by a collection of probability distributions on phase space $\EE$, invariant with respect to the dynamics, which I call {\it ensemble}: the distributions $\m$ in $\EE$ will be parameterized by a few parameters $U,V,E_1,E_2,\ldots$ which have a physical interpretation of (say) average energy, volume, intensity of the nonconservative forces acting on the system (that will be called ``external parameters''). Each distribution $\m$ will describe a macroscopic state in which the averages of the observables will be their integrals with respect to $\m$. The equations of motion will be symbolically written as \begin{eqnarray}\dot{\V x}=\V f(\V x)\label{3.1}\end{eqnarray} % and we shall assume that $\V f$ is smooth, that it depends on the external parameters and that the phase space visited by trajectories is bounded (at fixed external parameters and initial data). Since we imagine that the system is subject to nonconservative forces the phase space volume (or any measure with density with respect to the volume) will not be preserved by the evolution and the divergence \begin{eqnarray}\s(\V x)=-\sum_{i}\partial_{x_i} f_i(\V x)\label{3.2} \end{eqnarray} % will be {\it different} from $0$. We expect that, in interesting cases, the time average $\s_+$ of $\s$ will be positive: \begin{eqnarray}\s_+\defi \lim_{\t\to\infty}\frac1\t \int_0^\t \s(S_tx)\,dt\,>\,0\,.\label{3.3}\end{eqnarray} % and, with few exceptions, $x$--independent. This means that there cannnot be invariant distributions with density with respect to the volume. And the problem to find even a single invariant distribution is notrivial, except possibly for the ones concentrated on periodic orbits. The problem can be attacked, possibly, by following {\cs Boltzmann}'s view of dynamics as discrete, (``{\it die Zahl der lebendigen Kr\"aft ist eine diskrete}'', see p. 167 in \cite{Bo877b}). \* \section{Extending Boltzmann's ergodic hypothesis.} \* Consider a generic ``chaotic'' system described by equations like \ref{3.1} which generate motions confined in phase space. Under very general conditions it follows that $\s_+\ge0$, \cite{Ru96}, and we concentrate on the case $\s_+>0$. The suggestion that phase space should be regarded as discrete, and motion should simply be a one-cycle permutation of the ``cells'' $\D$ representing the phase space points is still very appealing as it would lead to the unambiguous determination of the invariant distribution $\m$ describing the statistical properties of the stationary states. In fact this is an assumption implicit in any claim of physical relevance of a simulation: as already mentioned above, a computer program defines a map on small cells in phase space. Already in the case of Hamiltonian systems (\ie in equilibrium theory) a simulation will not respect the uniqueness of solutions of the equation of motion because the map between the cells will not be invertible: it is extremely hard to write a program which avoids that two distinct cells are mapped into the same cell (see above). When $\s_+>0$ so that, in the average, phase space volume contracts the uniqueness problem becomes essentially unsurmountable (and not only in simulations); and there will be very many cells that eventually evolve into the same cell: thus the evolution will not be a permutation of the cells. It will, however, become {\it eventually} a permutation of a {\it subset} of the initial set of cells. This reflects the fact that the orbits of the solutions of the differential equation \ref{3.1} will ``cluster'' on an {\it attractor} which is a set of $0$ volume. The conclusion is that the statistics of the motions will still be a well defined probability distribution on phase space {\it provided} the ergodic hypothesis is extended to mean that the permutation of the cells on the attractor is a one-cycle permutation: it will be, in this case, still the uniform distribution concentrated on the cells lying on the attractor. This viewpoint unifies the conception of the statistics of equilibrium and of stationary nonequilibrium: the {\it statistics} $\m$ of the motions, \ie the probability distribution $\m$ such that, in the continuous version of the models, \begin{eqnarray}\lim_{\t\to\infty}\frac1\t\int_0^\t F(S_tx)\,dt\,= \,\int F(y) \,\m(dy)\,,\label{4.1}\end{eqnarray} % for all smooth observables $F$ and for all but a set of zero volume of points $x$ on phase space, can be considered, {\it in equilibrium as well as in stationary non equilibrium} states, as a probability distribution which is uniform on the attractor. The key obstacle to the above conception of Statistical Mechanics for stationary states is that phase space cells cannot be supposed to evolve, under the evolution assigned by \ref{3.1} when $\s_+>0$, keeping a constant volume. Therefore regarding evolution as a map between cells of a discretized version of phase space contains new sources of possible errors. Besides the error that is present in equilibrium theory due to the cells deformations which leads to violations of the uniqueness, \cite{Ga00}, there is an error due to their contraction $\s_+>0$. In equilibrium the first error can be reduced by reducing the cells size and the time intervals at which the observations (to be interpolated into the estimate of the integral in \ref{4.1}) are taken. This is a nontrivial source of errors that can be estimated to be physically acceptable, at least for the evaluation of the averages of the few observables relevant for Thermodynamics, only in certain regions of the phase diagrams, see Ch. I in \cite{Ga00}. But at least in such regions the discrete interpretation of the ergodic hypothesis leads us to a consistent representation of the evolution as a permutation between discrete elements of a partition of phase space into small cells. Out of equilibrium the further source of discretization error due to the actual reduction of phase space volume implies that it is not consistent to view the motion as a permutation of cells of a discretization of phase space into small equal volume elements. A possible way out is to restrict attention to systems that show strongly chaotic behavior. For instance systems which are transitive (\ie admit a dense orbit) and hyperbolic, see \cite{GBG04} for a formal definition, are typically chaotic systems which are also quite well understood. To enter into some detail it is convenient to look at the time evolution by drawing a few surfaces $\Si_1,\Si_2,\ldots,\Si_s$ transversal to the phase space trajectories, and such that the trajectories cross some of the surfaces over and over again (\ie each trajectory crosses the surfaces infinitely many times both in the future and in the past). Let $\Si=\cup_j \Si_j$ (usually called a ``Poincar\'e's section'') and let $S$ be the map which transforms a point $\x\in \Si$ (\ie on one of the surface elements $\Si_1,\Si_2,\ldots,\Si_s$) into the point $S\x$ where the orbit of $\x$ meets again for the first time $\Si$ (\ie it is again on one of the surface elements defining $\Si$). The points in phase space can therefore be described by pairs $x=(\x,\th)$ if $\x$ is the point in $\Si$ last visited by the trajectory starting at $x$ and $\th$ is the time elapsed since that moment. It is possible to partition $\Si$ into regions $P_1,P_2,\ldots,P_n$ with the property that the symbolic dynamics histories $\Bs=\{\s_i\}_{i=-\infty}^\infty$ on the sets $P_\s$, $\s=1,\ldots,n$, has a {\it Markov property}, in the sense that (1) there is a suitable matrix $M_{\s,\s'}$ with entries $0$ or $1$, such that if $M_{\s_i,\s_{i+1}}\=1$ for all $i$ then there is a unique point $x$ such that $S^ix\in P_{\s_i}$: the point $x$ is said to be ``coded'' by the sequence $\Bs$. And (2) calling {\it compatible} a sequence $\Bs$ with $M_{\s_i,\s_{i+1}}\=1$ then for all points $x$ there is at least one compatible sequence $\Bs$ which codes $x$ and for all but a set of zero volume relative to $\Si$ the sequence $\Bs$ is unique (\ie much as it is the case in the binary representation of real numbers). The partition $P_1,P_2,\ldots,P_n$ is then called a {\it Markov partition}: since the set of exceptions in the correspondence $x\otto\Bs$ has zero volume, the volume distribution can be represented as a probability distribution $\m_0$ over the space of compatible sequences. And the statistics of the evolution of data $\x$ chosen at random with respect to the distribution $\m_0$, which is the main object of interest, will therefore be represented also by a $S$--invariant probability distribution on the space $\O$ of the compatible sequences $\Bs$, \cite{GBG04}. The sets $P_1,P_2,\ldots,P_n$ can be used to represent conveniently the microscopic states of the system: given a precison $h>0$ it is possible to find $N_h$ such that the sets \begin{eqnarray}P_{\s_{-N_h},\ldots,\s_{N_h}}=\bigcap_{j=-N_h}^{N_h} S^{-j} P_{\s_j}\label{4.2}\end{eqnarray} % have a diameter $0$, and \\(b) {\it time reversal symmetry} in the sense that there is a smooth isometry $I$ such that $IS=S^{-1}I$. \\Define \begin{eqnarray}p=\frac1\t\sum_{j=0}^{\t-1} \frac{\s(S^jx)}{\s_+}\label{6.2} \end{eqnarray} % then the following theorem holds, \cite{GC95}, \* {\bf Fluctuation theorem:} {\it With respect to the SRB distribution the observable $p$ satisfies a large deviation property ({\rm see below}) with a rate function $\z(p)$ which is analytic and convex in an interval $(-p^*,p^*)$, for a suitable $p^*\ge 1$, where it exhibits the symmetry property \begin{eqnarray}\z(-p)=\z(p)-p\s_+\label{6.3}\end{eqnarray} % } \* This means that the probability, with respect to the SRB distribution $\m$ of $(\FF,S)$, that $p$ is inside an interval $[a,b]\subset (-p^*,p^*)$ is $P_{a,b}$ with $\lim_{\t\to\infty} \frac1\t\log P_{a,b}=\max_{p\in[a,b]} \z(p)$. Existence and analyticity of $\z(p)$ is part of the quoted general results of {\cs Sinai}, while the symmetry \ref{6.3} was pointed out in \cite{GC95} in an attempt to explain the numerical results of an earlier computer experiment \cite{ECM93}. The interest of the theorem lies in the fact that it is a symmetry property: hence it holds without any free parameter. The theorem can be extended to mixing Anosov flows, \cite{Ge98}, and therefore, via the chaotic hypothesis and if $\s(x)$ is the phase space contraction rate defined in \ref{3.2}, it becomes a property of essentially any system which is {\it chaotic, dissipative and reversible}. \* \section{Entropy?} \* Interest in the properties of the observable $\s(x)$, \ref{3.2} for flows and \ref{6.2} for maps, arose in several molecular dynamics simulations in which it was naturally related to the {\it entropy creation rate}. A natural question is whether a definition of entropy can be extended to nonequilibrium stationary states in analogy with the corresponding definition for equilibrium states (which are a very special case of stationary states). The identification between the SRB distributions and distributions giving equal probability to the microcells in the attractor shows that it should be possible, at least, to define a function which is a Lyapunov function for the approach to stationarity: this would be an extension of the $H$-theorem of Boltzmann. However equality between the $H$ function evaluated in equilibrium states and thermodynamic entropy might be a coincidence, important but not extendible to non equilibrium (hence not necessary). Arguments in this direction can be found in the literature, \cite{Ga04b,Ga01}, and here the controversial aspects of this matter will not be touched, \cite{GL03}. It will be worth however to enter into more details about why $\s(x)$ has been called entropy creation rate. This is simply because in several experiments it had such an interpretation, being the ratio between a quantity that could be identified with the work per unit time done by the noncoservative forces stirring the system divided by a quantity identified with temperature of the thermostat providing the forces that extract the energy input from the stirring forces. The experiments were simulations and from many sides critiques were expressed because the interpretation seemed closely tied to the explicit form of the thermostats models, often considered ``unphysical''. Furthermore the explicit dependence on the equations of motion makes the identification of $\s(x)$ with the entropy creation rate quite useless if the aim is to compare the theory with experiments different from simulations because in real experiments (\ie on experiments on matter distinct from impressive arrays of transistors) there usually is no explicit model of thermostat force and it is difficult to evaluate $\s(x)$. And it might turn out that the identification of $\s(x)$ with entropy creation rate is closely related to the special models considered. A simple, but quite general, model of thermostatted system may be useful to show that, while we should expect that there is a relation between entropy creation rate and phase space contraction, still the two notions are quite different. The system consists in $N\equiv N_0$ particles in a container $\CC_0$ and of $ N_a$ particles in $n$ containers $\CC_a$ which play the role of {\it thermostats}: their positions will be denoted $\V X_a,\,a=0,1,\ldots,n$, and $\V X\defi(\V X_0,\V X_1,\ldots,\V X_n)$. Interactions will be described by a potential energy \begin{eqnarray} W(\V X)=\sum_{a=0}^{n} U_a(\V X_a) +\sum_{a=1}^n W_a(\V X_0,\V X_a) \label{7.1}\end{eqnarray} % {\it i.e.} thermostats particles only interact indirectly, via the system. All masses will be $m=1$, for simplicity. The particles in $\CC_0$ will also be subject to external, possibly nonconservative, forces $\V F(\V X_0,\BF)$ depending on a few strength parameters $\BF=(E_1,E_2,\ldots)$. It is convenient to imagine that the force due to the confining potential determining the region $\CC_0$ is included in $\V F$, so that one of the parameters is the volume $V=|\CC_0|$. See Fig.2 below. %\input fig2 \eqfig{110pt}{90pt}{}{fig2}{} \0{\nota Fig.2 The reservoirs occupy finite regions outside $C_0$, \eg sectors $C_a\subset R^3$, $a=1,2\ldots$. Their particles are constrained to have a {\it total} kinetic energy $K_a$ constant, by suitable forces $\Bth_a$, so that the reservoirs ``temperatures'' $T_a$, see \ref{7.3}, are well defined.\vfil} \* \kern-3mm The equations of motion will be, assuming the mass $m=1$, \begin{eqnarray}\label{7.2} &\ddot{\V X}_{0i}=-\partial_i U_0(\V X_0)-\sum_{a} \partial_i U_a(\V X_0,\V X_i)+\V F_i\\ &\ddot{\V X}_{ai}=-\partial_i U_a(\V X_a)- \partial_i U_a(\V X_0,\V X_i)-\a_a \dot{\V X}_a\nonumber\end{eqnarray} % where the last force term $-\a_a \dot{\V X}_a$ is a phenomenological force that implies that the thermostats particles keep constant kinetic energies: \begin{eqnarray}K_a=\sum_{j=1}^{N_a} \frac12\, (\dot{\V X}^a_j)^2\defi \frac32 N_a k_B T_a\defi \frac32 N_a\b_a^{-1}\label{7.3}\end{eqnarray} % where the parameters $T_a$ should define the thermostats {\it temperatures} and $\a_a$ can, for instance, be defined by \begin{eqnarray}-\a_a \,\defi\,\frac{L_a-\dot U_a} {3N_a k_B T_a}\label{7.4} \end{eqnarray} % where $L_a=-\partial_{\V X_a} W_a(\V X_0,\V X_a)\cdot \dot{\V X}_a$ is the work done per unit time by the forces that the particles in $\CC_0$ exert on the particles in $\CC_a$. The exact form of the forces that have to be added in order to insure the kinetic energies constancy should not really matter, within wide limits. But this is a property that is not obvious and which is much debated. The above thermostatting forces choice is dictated by Gauss' {\it least effort} principle for the constraints $K_a=const$, see appendix 9.4 in \cite{Ga00}: this is a criterion that has been adopted in several simulations, \cite{EM90}. Independently of Gauss' principle it is immediate to check that if $\a_a$ is defined by \ref{7.4} then the kinetic energies $K_a$ are strictly constants of motion. The work $L_a$ in \ref{7.4} will be interpreted as {\it heat} $\dot Q_a$ ceded, per unit time, by the particles in $\CC_0$ to the $a$-th thermostat (because the ``temperature'' of $\CC_a$ remains constant, hence the thermostats can be regarded in thermal equilibrium). The {\it entropy creation rate} due to heat exchanges between the system and the thermostats can, therefore, be naturally defined as \begin{eqnarray}\s^0(\dot{\V X},\V X)\defi\sum_{a=1}^{N_a} \frac{\dot Q_a}{k_B T_a}\label{7.5}\end{eqnarray} % It should be stressed that here {\it no entropy notion} is introduced for the stationary state: only variation of the thermostats entropy is considered and it should not be regarded as a new quantity because the thermostats should be considered in equilibrium at a fixed temperature. The question is whether there is any relation between $\s_0$ and the phase space contraction $\s$ of \ref{3.2}. The latter can be immediately computed and is (neglecting $O(\min_{a>0} N_a^{-1})$) \begin{eqnarray}\s^\G(\dot{\V X},\V X)={\mathop\sum\limits_{a>0}} {\frac{3N_a-1}{3 N_a}} \frac{\dot Q_a-\dot U_a}{k_B T_a} = {\mathop\sum\limits_{a>0}} \frac{\dot Q_a}{k_B T_a}-\dot U\label{7.6} \end{eqnarray} % where $U=\sum_{a>0} \frac{3N_a-1}{3 N_a} \frac{U_a}{k_B T_a}$. Hence in this example in which the thermostats are ``external'' to the system volume (unlike to what happens in the common examples in which they act inside the volume of the system), the phase space contraction is not the entropy creation rate, \cite{Ga06}. {\it However it differs from the entropy creation rate by a total derivative}. The latter remark implies that if the chaotic hypothesis is accepted for the system in Fig.2 then, assuming $U_a$ bounded (for simplicity, see \cite{BGGZ05,Ga06} for more general cases) it is $\s_+=\media{\s_0}$ because the derivative $\dot U$ contributes $\frac1\t(U(\t)-U(0))\tende{\t\to\infty}0$ and also the observable $p$, in the continuous time extension of \ref{6.2}, \cite{Ge98}, has the same rate function as the observable $p=\frac1\t\int_0^\t \s_0(S_tx)\,dt\=\frac1\t\int_0^\t \s(S_tx)\,dt+O(\t^{-1})$. Since the equations of motion \ref{7.2} are time reversible (a rather general property of Gaussian constraints, with $I$ being here simply velocity reversal) it follows that {\it the ``physical entropy creation'' \ref{7.5} has a fluctuations rate $\z(p)$ satisfying the fluctuation relation \ref{6.3}.} This is relevant because the definition \ref{7.5} has meaning independently of the equations of motions and can therefore be suitable for experimental tests. \cite{BGGZ06,Ga06}. The above is just a model of thermostats: other interesting models have been proposed based on purely Hamiltonian interactions at the price of relying on thermostats of infinite size, see \cite{Ja99,EPR99,Ru06}. \* \section{Extensions of Boltzmann's $H$-theorem} \* The above analysis {\it does not require a notion of entropy} to be defined for stationary states. There is, however, another key contribution of Boltzmann to Statistical Mechanics, briefly mentioned above. This is the Boltzmann's equation and the relative {\it H-theorem}, \cite{Bo72}. The theorem has attracted deep interest because of its philosophical implications. For our purposes it is important because it provides a theory of approach to equilibrium and therefore it is one of the first results on nonequilibrium. It is useful to stress that the definition of $H$ is given in the context of the approach to equilibrium and {\cs Boltzmann} never applied it (nor, perhaps, meant to apply it) to the approach to other stationary states and to their theory. The equality of the value of $H$ with the thermodynamic entropy when evaluated on the equilibrium state raised the hope that it could be possible to define entropy for systems out of equilibrium and even if not in stationary state. The idea emerged clearly already from the foundational papers on the Boltzmann equation (``{\it $\log P$ was well defined whether or not the system is in equilibrium, so that it could serve as a suitable generalization of entropy}'', p. 82 in \cite{Kl73} and p. 218 in \cite{Bo877b}) and many attempts can be found in the literature to define entropy for systems out of equilibrium in stationary states or even in macroscopically evolving states. Strictly speaking the implication that can be drawn from the works on the Boltzmann's equation is that a rarefied gas started in a given configuration evolves in time so that the average values of the observables, at least of the few of interest, acquire an asymptotic value which is the same as the one that can be computed from a probability distribution maximizing a function $H$. The acquisition of an asymptotic value by the averages of the observables is a property expected to hold also when the asymptotic state is a nonequilibrium stationary state. And it is natural to think that also in such cases there will be a function that approaches monotonically an asymptotic value signaling that the few observables of interest approach their asynptotic average. As remarked above the SRB distribution is a uniform distribution over the attractor: therefore it verifies a variational property and this can be used to define a Lyapunov function that evolves towards a maximum, \cite{Ga04b}. Let $\Bh=(\s_{-N_h},\ldots,\s_{N_h})$ and $H\defi\frac1\t\sum_{\Bh} -p_{\Bh}\big(\log p_{\Bh}\ +\log \L_\t(\Bh^{-1})\big)$ where $p_{\Bh}$ denotes the fraction of microcells that can be found in the cell $\D=P_{\Bh}=\cap_{k=-N_h}^{N_h}S^{-k} P_{\h_k}$ after a time of $\t$ units has elapsed starting from an initial distribution $p^0_\Bh$ (typically a uniform distribution over the microcells in a single cell $\D^0$). This is a quantity that tends to a maximum as time evolves (reaching it when the $ p_\Bh$ have the value of the SRB distribution and the maximum equals, therefore, the logarithm of the number of microcells on the attractor). Therefore the quantity $H$ tends in the average to a maximum and it can be regarded as an instance of an $H$--function. However the maximum depends on the precision $h$ of the coarse graining defined by the partition of phase space by the cells $\D$. Changing the precison several changes occur which have to be examined if a meaning other than that of a Lyapunov function has to be given to $H$. The analysis in \cite{Ga04b} points out that $H$ changes with the precision $h$ in a trivial way (\ie by an additive constant, independent of the control parameters of the system and depending only on the precision $h$) if the SRB state on which it is evaluated is an equilibrium state. In the latter case it is proportional to the logarithm of the phase space volume that can be visited. In the nonequilibrium cases however $H$ changes when the precision $h$ changes by additive quantities that {\it are not just functions of $h$} but depend on thermodynamic quantities, (like average energy, temperatures, {\etc.}), \cite{Ga04b}. This indicates that while not excluding the possibility of existence of Lyapunov functions, see \cite{GGL04}, indicating the approach to equilibrium (within a given precision $h$) the identity of the $H$ function with entropy, \ie its identity with a function of the state parameters of the system, is possible only when the state is in an equilibrium state. My interpretation of this analysis, based once more on a discrete point of view on the problem, is that one should not insist in looking for a notion of entropy in systems out of equilibrium, \cite{Ga04b}. If so once again {\cs Boltzmann}'s attitude to consider phase space as discrete and in general to deny reality to the continua might have led to insights into difficult questions. \* \section{Conclusion} \* {\cs Boltzmann}'s contribution to the theory of ensembles and to the mechanical interpretation of heat and Thermodynamics was based on a discrete conception of the continuum: his staunch coherence on this view has been an essential aspect of the originality of his thought. It is in fact a method of investigation which is still very fruitful and used in various forms when ``cut--offs'' or ``regularizations'' are employed in the most diverse fields. In my view it has been and still is important in the recent developments in the theory of nonequilibrium stationary states. The Fluctuation theorem and its various interpretations, extensions and applications (to Onsager reciprocity at non zero forcing, to Green-Kubo formulae, to fluid Mechanics, Turbulence and Intermittency, see \cite{Ga00,Ga04,GBG04}) is, hopefully, only an example. It is interesting in this context recall a few quotes from {\cs Boltzmann} \* ``{\it Through the symbols manipulations of integral calculus, which have become common practice, one can temporarily forget the need to start from a finite number of elements, that is at the basis of the creation of the concept, but one cannot avoid it}''; \0see p. 227 in \cite{Bo874}, or in the same page: ``{\it Differential equations require, just as atomism does, an initial idea of a large finite number of numerical values and points ...... Only afterwards it is maintained that the picture never represents phenomena exactly but merely approximates them more and more the greater the number of these points and the smaller the distance between them. Yet here again it seems to me that so far we cannot exclude the possibility that for a certain very large number of points the picture will best represent phenomena and that for greater numbers it will become again less accurate, so that atoms do exist in large but finite number.}'' \0and, see p. 55 in \cite{Bo874}: ``{\it This naturally does not exclude that, after we got used once and for all to the abstraction of the volume elements and of the other symbols {\rm[of Calculus]} and once one has studied the way to operate with them, it could look handy and luring, in deriving certain formulae that Volkmann calls formulae for the coarse phenomena, to forget completely the atomistic significance of such abstractions. They provide a general model for all cases in which one can think to deal with $10^{10}$ or $10^{10^{10}}$ elements in a cubic millimeter or even with billions of times more; hence they are particularly invaluable in the frame of Geometry, which must equally well adapt to deal with the most diverse physical cases in which the number of the elements can be widely different. Often in the use of all such models, created in this way, it is necessary to put aside the basic concept, from which they have overgrown, and perhaps to forget it entirely, at least temporarily. But I think that it would be a mistake to think that one could become free of it entirely.}'' \* And the principle was really applied not only in the conception of the ergodic hypothesis, \cite{Bo871a,Bo871b}, but also in the deduction of the Boltzmann's equation which {\cs Boltzmann} felt would be clarified by following discretization methods (in energy) inspired by those employed in the ``{\it elegant solution of the problem of string-vibrations}'' of {\cs Lagrange}, or in {\cs Stefan}'s study of diffusion or in {\cs Riemann}'s theory of mean curvature, \cite{Bo72} and in various discussions of the heat theorem, \cite{Bo877b}. The above conception of the infinitesimal quantities, rooted in the early days of Calculus when ``$dx$'' was regarded as {\it infinitely small and yet still of finite size} (in apparent, familiar, logical contradiction), is an important legacy that should not be forgotten in spite of the social pressure that induces all of us to identify clarity of physical understanding with continuous models of reality. \* \0{Appendix: \it Temperature and kinetic energy, \cite{Br03,Br76}} \* The first attempts at a kinetic explanation of the properties of gases came following the experiments by {\cs Boyle}, (1660), on the gas compression laws. The laws established that ``air'' had elastic properties and that there was inverse proportionality between pressure and volume: a theory that was considered also by {\cs Newton}. It was {\cs D. Bernoulli}, (1720), who abandoned the view, espoused by {\cs Newton}, that the atoms were arranged on a kind of lattice repelling each other (with a force inversely proportional to their distances to agree with Boyle's law, but extending only to the nearest neighbors). {\cs Bernoulli} imagined the atoms to be free and that pressure was due to the collisions with the walls and proportional to the square of the average speed proposing that a correct definition of temperature should be based on this property. In 1816 {\cs Avogadro} established that, for rarefied gases, the ratio $pV/T$ is proportional to the number of atoms or molecules via a universal constant. This was a striking result, explaining the anomalies in the earlier theory of {\cs Dalton} and allowing, besides the definition of the {\it Avogadro's number}, the correct determination of the relative molecular and atomic weights. It openend the way to the definition of absolute temperature, independently of the special gas-thermometer employed, and to the principle of energy equipartition and to the later works of {\cs Waterston, Clausius, Boltzmann}, among others. The attempt of {\cs Laplace}, (1821), proposed an elaborate scheme in which the atoms, still essentially fixed in space at average distance $r$ would contain a quantity $c$ of {\it caloric} and would interact with a short range force proportional to the product of their quantity of caloric and depending on the distance. Identifying the caloric $c$ with the a fixed amount contained in each atom would have led to a gas law with $p$ proportional to the square of the density $\r$, \ie to $\r^2 c^2$; but this was avoided by supposing that the amount of caloric $c$ in each molecule was determined by an equilibrium between the amount of caloric emitted by a molecule and the caloric received by it (emitted from the other molecules) which was supposed to depend only on the temperature, see \cite{Br03}. %The caloric emitted was supposed to be proportional to %both the amount $c$ of caloric in the molecule and the amount of %caloric in the surroundings $\r c$, \ie proportional to $\r c^2$: so %$\r c^2$ had to depend only on the temperature and the equation of %state $p=const \r^2 c^2$ became $p=const \r f(T)$ with $f(T)$ function %of the temperature alone (hence it could be supposed to be propotional %to $T$ defining conveniently $T$). The theory of {\cs Laplace} did not sound convincing and the work of {\cs Bernoulli} went unnoticed; the same was the fate of the work of {\cs Herapath}, (1820), who again proposed, without knowing {\cs Bernoulli}'s theory, that the atoms were free and pressure was due to collisions with the walls; however he assumed that pressure was proportional to the average momentum rather than kinetic energy obtaining an incorrect definition of absolute temperature. In any event his work was rejected by the {\it Philosophical transactions of the Royal Society of London} and published on the {\it Annals of Philosophy} falling into oblivion for a while. In 1845 {\cs Waterstone}, unaware of both {\cs Bernoulli} and {\cs Herapath} but (likely) familiar with {\cs Avogadro}'s work, proposed the theory of gases with the correct identification of pressure as proportional to the average kinetic energy and the density, introducing also a rather detailed conception of te interatomic forces taking up ideas inspired by {\cs Mossotti} (who probably had also made {\cs Avogadro} and Italian science better known in England during his political exile). Unfortunately he submitted it to the {\it Philosophical transactions of the Royal Society} which readily rejected it and remained unpublished, until it was rediscovered much later (1892, by {\cs Raileigh}). In the 1840's, through the work of {\cs Meyer, Joule, Helmoltz} and others the energy conservation principle was established with the consequent identification of heat as a form of energy convertible into mechanical work forcing (reasonable) physicists to abandon the hypothesis of the existence of caloric as a conserved entity. The theory of gases begun to be really accepted with the work of {\cs Kr\"onig}, (1856), who clearly proposed identifying temperature with average kinetic energy of molecules. His work became well known as it appeared to have prompted the publication of {\cs Clausius}'s paper of (1857), who had independently reached the same conclusions and gone much further. Not only {\cs Clausius} went quite far in establishing energy equipartition (completed by {\cs Maxwell} in 1860) but he introduced a basic concept of kinetic theory: the mean free path. Thus making clear the role of collisions in gas theory: they lead to prediction and to a first understanding of the phenomenon of diffusion, explaining the apparent paradoxes linked to the earlier assumptions that in rarefied gases collisions could be simply neglected, and also initiate the theory of the transport coefficients. The latter papers, one century after the too far in advance (over his time) work of {\cs Bernoulli}, gave origin to kinetic theory in the sense we intend it still now, and stimulated also the related investigations of {\cs Maxwell}. Therefore {\cs Maxwell} (1859) and a little later {\cs Boltzmann} (1866) could start their work taking for granted the well established identity between temperature and average kinetic energy for gases extending it to hold in all systems in equilibrium (rarefied or not). This key view was not destined to have a long life: the advent of Quantum Mechanics would prove that proportionality between average kinetic energy and temperature could only be approximate and to hold if quantum corrections to Atomic Mechanics were negligible, see \cite{Ga00}. Nevertheless the identification of temperature and kinetic energy plaid (and still plays, whenever quantum effects are negligible) an essential role not only in classical Statistical Mechanics but also in the discovery of Quantum Mechanics, which was heralded by the failure of the related equipartition of energy. \* \0{\it Source of the talk at the {\cs Boltzmann's Legacy} international symposium at ESI, Vienna, 7-9 June, 2006} \nota %\bibliography{0Bibcaos} \bibliographystyle{apsrev} %\baselineskip=0.2mm\parskip=0mm\bibliographystyle{unsrt} \let\em=\nota \input vienna06.bbl %\end{thebibliography} \revtex \end{document} ---------------0606181305916 Content-Type: application/postscript; name="fig1.eps" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="fig1.eps" %!PS-Adobe-2.0 EPSF-1.2 %%Title: %%Creator: Ghostscript ps2epsi from %%CreationDate: 2006-05-26 00:20 fig1.ps %%For:giovanni giovanni %%Pages: 1 %%DocumentFonts: Times-Roman %%BoundingBox: 0 0 151 31 %%BeginPreview: 151 31 8 124 % 00000000008888888888888888888888888888888888888888888888888888888888880000000000 % 00000000000000000000000000000000000000008888888888888888888888888888888888888888 % 88888888888888888888888888888888888888888888888888888888888888888888888888888888 % 88888888888888888888888888888888888888888888888888888888888800 % 00000000aac888888888888888888888888888888888888888888888888888888888d88800000000 % 00000000000000000000000000000000000000aac888888888888888888888888888888888888888 % 888888888888888888f2e688888888888888888888888888888888888888888888888888888888f2 % e688888888888888888888888888888888888888888888888888888888d888 % 00000000cc5500000000000000000000000000000000000000000000000000000000cc5500000000 % 00000000000000000000000000000000000000cc5500000000000000000000000000000000000000 % 000000000000000000f58e00000000000000000000000000000000000000000000000000000000f5 % 8e00000000000000000000000000000000000000000000000000000000cc55 % 00000000ff3300000000000000000000000000000000000000000000000000000000ff3300000000 % 00000000000000000000000000000000000000ff3300000000000000000000000000000000000000 % 000000000000000000ff5c00000000000000000000000000000000000000000000000000000000ff % 5c00000000000000000000000000000000000000000000000000000000ff33 % 00000033ff4444444444444444444444444444444444444444444444444444444469ff0000000000 % 00000000220000000000000000000000000033ff0000000000000000000000000000000000000000 % 00000000000000005cff000000000000000000000000000000000000000000000000000000005cff % 0000000000000000000000000000000000000000000000000000000033ff00 % 00000055e8bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbd2d70000000000 % 000099cccc7700000000000000000000000055cc0000000000000000000000000000000000000000 % 00000000000000008ef5000000000000000000000000000000000000000000000000000000008ef5 % 0000000000000000000000000000000000000000000000000000000055cc00 % 00000088aa0000000000000000000000000000000000000000000000000000000088aa0000000000 % 0000ccaa114400000000000000000000000088aa0000000000000000000000000000000000000000 % 0000000000000000c8e300000000000000000000000000000000000000000000000000000000c8e3 % 0000000000000000000000000000000000000000000000000000000088aa00 % 000000aa8800000000000000000000000000000000000000000000000000000000aa880000000000 % 00000066cc66000000000000000000000000aa880000000000000000000000000000000000000000 % 0000000000000000e3c800000000000000000000000000000000000000000000000000000000e3c8 % 00000000000000000000000000000000000000000000000000000000aa8800 % 000000cc5500000000000000000000000000000000000000000000000000000000cc550000000000 % 0000441111ff000000000000000000000000cc550000000000000000000000000000000000000000 % 0000000000000000f58e00000000000000000000000000000000000000000000000000000000f58e % 00000000000000000000000000000000000000000000000000000000cc5500 % 000000ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff330000000000 % 000077eeee88000000000000000000000000ff330000000000000000000000000000000000000000 % 0000000000000000ff5c00000000000000000000000000000000000000000000000000000000ff5c % 00000000000000000000000000000000000000000000000000000000ff3300 % 000033ff0000000000000000000000000000000000000000000000000000000033ff000000000000 % 000000002200000000000000000000000033ff000000000000000000000000000000000000000000 % 000000000000005cff000000000000000000000000000000000000000000000000000000005cff00 % 00000000000000000000000000000000000000000000000000000033ff0000 % 000055cc0000000000000000000000000000000000000000000000000000000055cc000000000000 % 00000000ff99220000000000000000000055cc000000000000000000000000000000000000000000 % 000000000000008ef5000000000000000000000000000000000000000000000000000000008ef500 % 00000000000000000000000000000000000000000000000000000055cc0000 % 000088aa0000000000000000000000000000000000000000000000000000000088aa000000000000 % 00000000ffffff9922000000000000000088aa000000000000000000000000000000000000000000 % 00000000000000c8e300000000000000000000000000000000000000000000000000000000c8e300 % 00000000000000000000000000000000000000000000000000000088aa0000 % 0000bbe0bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbe988000000002244 % 44444444ffffffffff9922000000000000bbe0bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb % bbbbbbbbbbbbbbfaf1bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbfaf1bb % bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbe9880000 % 0000d08244444444444444444444444444444444444444444444444444444444da550000000066bb % bbbbbbbbffffffffffffbb110000000000d082444444444444444444444444444444444444444444 % 44444444444444f9ac44444444444444444444444444444444444444444444444444444444f9ac44 % 444444444444444444444444444444444444444444444444444444da550000 % 0000ff3300000000000000000000000000000000000000000000000000000000ff33000000000000 % 00000000ffffffffbb4400000000000000ff33000000000000000000000000000000000000000000 % 00000000000000ff5c00000000000000000000000000000000000000000000000000000000ff5c00 % 000000000000000000000000000000000000000000000000000000ff330000 % 0033ff0000000000000000000000000000000000000000000000000000000033ff00000000000000 % 00000000ffffbb44000000000000000033ff00000000000000000000000000000000000000000000 % 0000000000005cff000000000000000000000000000000000000000000000000000000005cff0000 % 000000000000000000000000000000000000000000000000000033ff000000 % 0055da4444444444444444444444444444444444444444444444444444444482d000000000000000 % 00000000bb440000000000000000000055da44444444444444444444444444444444444444444444 % 444444444444acf944444444444444444444444444444444444444444444444444444444acf94444 % 444444444444444444444444444444444444444444444444444482d0000000 % 0088e9bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbe0bb00000000000000 % 0000000000000000000000000000000088e9bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb % bbbbbbbbbbbbf1fabbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbf1fabbbb % bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbe0bb000000 % 00aa8800000000000000000000000000000000000000000000000000000000aa8800000000000000 % 00000000000000000000000000000000aa8800000000000000000000000000000000000000000000 % 000000000000e3c800000000000000000000000000000000000000000000000000000000e3c80000 % 0000000000000000000000000000000000000000000000000000aa88000000 % 00cc5500000000000000000000000000000000000000000000000000000000cc5500000000000000 % 00000000000000000000000000000000cc5500000000000000000000000000000000000000000000 % 000000000000f58e00000000000000000000000000000000000000000000000000000000f58e0000 % 0000000000000000000000000000000000000000000000000000cc55000000 % 00ff3300000000000000000000000000000000000000000000000000000000ff3300000000000000 % 00000000000000000000000000000000ff3300000000000000000000000000000000000000000000 % 000000000000ff5c00000000000000000000000000000000000000000000000000000000ff5c0000 % 0000000000000000000000000000000000000000000000000000ff33000000 % 33ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff0000000000000000 % 00000000000000000000000000000033ffffffffffffffffffffffffffffffffffffffffffffffff % ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff % ffffffffffffffffffffffffffffffffffffffffffffffffffffff00000000 % 55cc0000000000000000000000000000000000000000000000000000000055cc0000000000000000 % 00000000000000000000000000000055cc0000000000000000000000000000000000000000000000 % 00000000008ef5000000000000000000000000000000000000000000000000000000008ef5000000 % 0000000000000000000000000000000000000000000000000055cc00000000 % 88aa0000000000000000000000000000000000000000000000000000000088aa0000000000000000 % 00000000000000000000000000000088aa0000000000000000000000000000000000000000000000 % 0000000000c8e300000000000000000000000000000000000000000000000000000000c8e3000000 % 0000000000000000000000000000000000000000000000000088aa00000000 % aa8800000000000000000000000000000000000000000000000000000000aa880000000000000000 % 000000000000000000000000000000aa880000000000000000000000000000000000000000000000 % 0000000000e3c800000000000000000000000000000000000000000000000000000000e3c8000000 % 00000000000000000000000000000000000000000000000000aa8800000000 % dff3ededededededededededededededededededededededededededededf5550000000000000000 % 000000000000000000000000000000cc550000000000000000000000000000000000000000000000 % 0000000000f58e00000000000000000000000000000000000000000000000000000000f58e000000 % 00000000000000000000000000000000000000000000000000cc5500000000 % ff9176767676767676767676767676767676767676767676767676767676ff330000000000000000 % 000000000000000000000000000000ff330000000000000000000000000000000000000000000000 % 0000000000ff5c00000000000000000000000000000000000000000000000000000000ff5c000000 % 00000000000000000000000000000000000000000000000000ff3300000000 % ff0000000000000000000000000000000000000000000000000000000033ff000000000000000000 % 000000000000000000000000000033ff000000000000000000000000000000000000000000000000 % 000000005cff000000000000000000000000000000000000000000000000000000005cff00000000 % 00000000000000000000000000000000000000000000000033ff0000000000 % cc0000000000000000000000000000000000000000000000000000000055cc000000000000000000 % 000000000000000000000000000055cc000000000000000000000000000000000000000000000000 % 000000008ef5000000000000000000000000000000000000000000000000000000008ef500000000 % 00000000000000000000000000000000000000000000000055cc0000000000 % d888888888888888888888888888888888888888888888888888888888c8aa000000000000000000 % 000000000000000000000000000088d8888888888888888888888888888888888888888888888888 % 88888888e6f288888888888888888888888888888888888888888888888888888888e6f288888888 % 888888888888888888888888888888888888888888888888c8aa0000000000 %%EndImage %%EndPreview save countdictstack mark newpath /showpage {} def /setpagedevice {pop} def %%EndProlog %%Page 1 1 %PS!Giovanni /punto { % x y punto gsave 3 0 360 newpath arc fill stroke grestore} def /puntox {% x P1 P2 puntox : punto su segmento P1 P2 a frazione x /y2 exch def /x2 exch def /y1 exch def /x1 exch def /x exch def x1 x2 x1 sub x mul add y1 y2 y1 sub x mul add} def /linea { %x1 y1 x2 y2 gsave 4 2 roll moveto lineto stroke grestore} def /origine1assexper2 { 4 2 roll 2 copy translate exch 4 1 roll sub 3 1 roll exch sub 2 copy atan rotate 2 copy exch 4 1 roll mul 3 1 roll mul add sqrt } def /punta0{0 0 moveto dup dup 0 exch 2 div lineto 0 lineto 0 exch 2 div neg lineto 0 0 lineto fill stroke } def /dirpunta{ gsave origine1assexper2 0 translate 7 punta0 grestore} def /d {5} def /H {30} def /dd {H 7 div} def /U {/P0 {0 0} def /PH {d H} def /PD {d H add H} def /Ph {H 0} def /Q1 {1 7 div P0 PH puntox} def /Q2 {2 7 div P0 PH puntox} def /Q3 {3 7 div P0 PH puntox} def /Q4 {4 7 div P0 PH puntox} def /Q5 {5 7 div P0 PH puntox} def /Q6 {6 7 div P0 PH puntox} def /QQ1 {1 7 div Ph PD puntox} def /QQ2 {2 7 div Ph PD puntox} def /QQ3 {3 7 div Ph PD puntox} def /QQ4 {4 7 div Ph PD puntox} def /QQ5 {5 7 div Ph PD puntox} def /QQ6 {6 7 div Ph PD puntox} def P0 PH linea PH PD linea PD Ph linea Ph P0 linea Q1 QQ1 linea Q2 QQ2 linea Q3 QQ3 linea Q4 QQ4 linea Q5 QQ5 linea Q6 QQ6 linea Q1 QQ1 linea} def U H dd 2 mul add H 1.8 div H dd 3.3 mul add H 1.8 div linea H dd 2 mul add H 1.8 div H dd 3.3 mul add H 1.8 div dirpunta /TR {/FO exch def /Times-Roman findfont FO scalefont setfont} def 8 TR H dd 2.7 mul add H 1.4 div moveto (S) show /UU {/P0 {0 0} def /PH {d H} def /PD {d H add H} def /Ph {H 0} def /Q1 {1 7 div P0 PH puntox} def /Q2 {2 7 div P0 PH puntox} def /Q3 {3 7 div P0 PH puntox} def /Q4 {4 7 div P0 PH puntox} def /Q5 {5 7 div P0 PH puntox} def /Q6 {6 7 div P0 PH puntox} def /QQ1 {1 7 div Ph PD puntox} def /QQ2 {2 7 div Ph PD puntox} def /QQ3 {3 7 div Ph PD puntox} def /QQ4 {4 7 div Ph PD puntox} def /QQ5 {5 7 div Ph PD puntox} def /QQ6 {6 7 div Ph PD puntox} def P0 PH linea PH PD linea PD Ph linea Ph P0 linea %Q1 QQ1 linea Q2 QQ2 linea Q3 QQ3 linea Q4 QQ4 linea %Q5 QQ5 linea %Q6 QQ6 linea } def H H 1.2 div add 0 translate UU H 0 translate UU H 0 translate UU %%Trailer cleartomark countdictstack exch sub { end } repeat restore %%EOF ---------------0606181305916 Content-Type: application/postscript; name="fig2.eps" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="fig2.eps" %!PS-Giovanni /linea { %x1 y1 x2 y2 gsave 4 2 roll moveto lineto stroke grestore} def /punto { % x y punto gsave 3 0 360 newpath arc fill stroke grestore} def /R0 {15} def /R1 {50} def /C {R1 R1} def /A1 {-10} def /A2 {60} def /B1 {90} def /B2 {130} def /C1 {160} def /C2 {190} def /D1 {220} def /D2 {255} def /X1 {A1 cos R0 mul R1 add A1 sin R0 mul R1 add} def /X2 {A1 cos R1 mul R1 add A1 sin R1 mul R1 add} def /XX1 {A2 cos R0 mul R1 add A2 sin R0 mul R1 add} def /XX2 {A2 cos R1 mul R1 add A2 sin R1 mul R1 add} def %X1 X2 punto punto /Y1 {B1 cos R0 mul R1 add B1 sin R0 mul R1 add} def /Y2 {B1 cos R1 mul R1 add B1 sin R1 mul R1 add} def /YY1 {B2 cos R0 mul R1 add B2 sin R0 mul R1 add} def /YY2 {B2 cos R1 mul R1 add B2 sin R1 mul R1 add} def /Z1 {C1 cos R0 mul R1 add C1 sin R0 mul R1 add} def /Z2 {C1 cos R1 mul R1 add C1 sin R1 mul R1 add} def /ZZ1 {C2 cos R0 mul R1 add C2 sin R0 mul R1 add} def /ZZ2 {C2 cos R1 mul R1 add C2 sin R1 mul R1 add} def /W1 {D1 cos R0 mul R1 add D1 sin R0 mul R1 add} def /W2 {D1 cos R1 mul R1 add D1 sin R1 mul R1 add} def /WW1 {D2 cos R0 mul R1 add D2 sin R0 mul R1 add} def /WW2 {D2 cos R1 mul R1 add D2 sin R1 mul R1 add} def /g {.7} def .8 .8 scale g setgray newpath C R0 A1 A2 arc XX2 lineto X2 lineto closepath fill stroke newpath XX2 moveto X2 lineto C R1 A1 A2 arc closepath fill stroke g setgray newpath C R0 B1 B2 arc YY2 lineto Y2 lineto closepath fill stroke newpath YY2 moveto Y2 lineto C R1 B1 B2 arc closepath fill stroke g setgray newpath C R0 C1 C2 arc ZZ2 lineto Z2 lineto closepath fill stroke newpath ZZ2 moveto Z2 lineto C R1 C1 C2 arc closepath fill stroke g setgray newpath C R0 D1 D2 arc WW2 lineto W2 lineto closepath fill stroke newpath WW2 moveto W2 lineto C R1 D1 D2 arc closepath fill stroke 0 setgray C R0 0 360 arc stroke /TR {/FO exch def /Times-Roman findfont FO scalefont setfont} def /SI {/FO exch def /Symbols findfont FO scalefont setfont} def /TRB {/FO exch def /Times-Roman-Bold findfont FO scalefont setfont} def /SIB {/FO exch def /Symbols-Bold findfont FO scalefont setfont} def /espressioner{% ####1_####2 P /y exch def /x exch def 13 TRB x y moveto exch show x 7 add y 3 sub moveto 10 TRB show} def (T) (1) 36 84 espressioner (T) (2) 78 67 espressioner (T) (3) 29 21 espressioner /espressioner{% ####1_####2 P /y exch def /x exch def 13 TRB x y moveto exch show x 10 add y 3 sub moveto 10 TRB show} def (C) (0) 43 48 espressioner ---------------0606181305916 Content-Type: application/octet-stream; name="vienna06.aux" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="vienna06.aux" XHJlbGF4IApcY2l0YXRpb257Qm82Nn0KXGNpdGF0aW9ue0JvNjZ9ClxjaXRhdGlvbntCbzY2 fQpcY2l0YXRpb257Qm84NzFhfQpcY2l0YXRpb257QnIwM30KXGNpdGF0aW9ue0JvODY4fQpc Y2l0YXRpb257RmwwMH0KXGNpdGF0aW9ue0JvODcxYn0KXGNpdGF0aW9ue0JvODcxYn0KXGNp dGF0aW9ue0JyMDN9ClxjaXRhdGlvbntlbmRub3RlNTN9ClxjaXRhdGlvbntCbzg3N2F9Clxj aXRhdGlvbntCbzg3N2F9ClxjaXRhdGlvbntIZTg4NGF9ClxjaXRhdGlvbntIZTg4NGJ9Clxj aXRhdGlvbntCbzg3N2F9ClxuZXdsYWJlbHtGaXJzdFBhZ2V9e3t9ezF9e317fXt9fQpcQHdy aXRlZmlsZXt0b2N9e1xjb250ZW50c2xpbmUge3NlY3Rpb259e1xudW1iZXJsaW5lIHtJfUJv bHR6bWFubiBhbmQgRW50cm9weX17MX17fX0KXG5ld2xhYmVsezEuMX17ezF9ezF9e317fXt9 fQpcY2l0YXRpb257Qm84ODR9ClxjaXRhdGlvbntHYTAwfQpcY2l0YXRpb257Qm84NzdhfQpc Y2l0YXRpb257Qm84NzdhfQpcY2l0YXRpb257Qm84ODR9ClxjaXRhdGlvbntHYTAwfQpcY2l0 YXRpb257Qm84ODR9ClxjaXRhdGlvbntHYTAwfQpcY2l0YXRpb257R2E5NWF9ClxjaXRhdGlv bntCbzg4NH0KXGNpdGF0aW9ue0JvODg0fQpcY2l0YXRpb257R2EwMH0KXGNpdGF0aW9ue0Zp NjR9ClxjaXRhdGlvbntSdTY4fQpcY2l0YXRpb257Qm84ODR9ClxjaXRhdGlvbntlbmRub3Rl NTR9ClxjaXRhdGlvbntCbzg3MWJ9ClxjaXRhdGlvbntCbzg4NH0KXG5ld2xhYmVsezEuMn17 ezJ9ezJ9e317fXt9fQpcbmV3bGFiZWx7MS4zfXt7M317Mn17fXt9e319ClxjaXRhdGlvbntC bzg4NH0KXGNpdGF0aW9ue0JyNzZ9ClxjaXRhdGlvbntCcjAzfQpcY2l0YXRpb257RUUxMX0K XGNpdGF0aW9ue0JyNzZ9ClxjaXRhdGlvbntCbzk2fQpcY2l0YXRpb257Qm85NmF9ClxjaXRh dGlvbntHYTk1YX0KXGNpdGF0aW9ue1RoNzR9ClxjaXRhdGlvbntMVjkzfQpcY2l0YXRpb257 R2EwMH0KXEB3cml0ZWZpbGV7dG9jfXtcY29udGVudHNsaW5lIHtzZWN0aW9ufXtcbnVtYmVy bGluZSB7SUl9Qm9sdHptYW5uJ3MgZGlzY3JldGUgdmlzaW9uIG9mIHRoZSBlcmdvZGljIHBy b2JsZW19ezN9e319ClxAd3JpdGVmaWxle3RvY317XGNvbnRlbnRzbGluZSB7c2VjdGlvbn17 XG51bWJlcmxpbmUge0lJSX1Cb2x0em1hbm4ncyBoZXJpdGFnZX17M317fX0KXGNpdGF0aW9u e0JvODc3Yn0KXGNpdGF0aW9ue1J1OTZ9ClxjaXRhdGlvbntHYTAwfQpcY2l0YXRpb257R2Ew MH0KXG5ld2xhYmVsezMuMX17ezR9ezR9e317fXt9fQpcbmV3bGFiZWx7My4yfXt7NX17NH17 fXt9e319ClxuZXdsYWJlbHszLjN9e3s2fXs0fXt9e317fX0KXEB3cml0ZWZpbGV7dG9jfXtc Y29udGVudHNsaW5lIHtzZWN0aW9ufXtcbnVtYmVybGluZSB7SVZ9RXh0ZW5kaW5nIEJvbHR6 bWFubidzIGVyZ29kaWMgaHlwb3RoZXNpcy59ezR9e319ClxuZXdsYWJlbHs0LjF9e3s3fXs0 fXt9e317fX0KXGNpdGF0aW9ue0dCRzA0fQpcY2l0YXRpb257R0JHMDR9ClxuZXdsYWJlbHs0 LjJ9e3s4fXs1fXt9e317fX0KXGNpdGF0aW9ue0dhMDB9ClxjaXRhdGlvbntHYTk2fQpcY2l0 YXRpb257QVc3MH0KXGNpdGF0aW9ue0dCRzA0fQpcY2l0YXRpb257U2k2OGF9ClxjaXRhdGlv bntTaTY4Yn0KXGNpdGF0aW9ue0dCRzA0fQpcY2l0YXRpb257U2k3Mn0KXGNpdGF0aW9ue0Jv NzBhfQpcY2l0YXRpb257QlI3NX0KXGNpdGF0aW9ue1J1NzZ9ClxjaXRhdGlvbntSdTc2fQpc Y2l0YXRpb257UnU4MH0KXGNpdGF0aW9ue0dDOTV9ClxjaXRhdGlvbntHYTAwfQpcY2l0YXRp b257UnU3M30KXGNpdGF0aW9ue1J1ODB9ClxjaXRhdGlvbntSdTczfQpcQHdyaXRlZmlsZXt0 b2N9e1xjb250ZW50c2xpbmUge3NlY3Rpb259e1xudW1iZXJsaW5lIHtWfUEgYml0IG9mIGhp c3RvcnkufXs2fXt9fQpcQHdyaXRlZmlsZXt0b2N9e1xjb250ZW50c2xpbmUge3NlY3Rpb259 e1xudW1iZXJsaW5lIHtWSX1EZXZlbG9wbWVudHMuIH17Nn17fX0KXGNpdGF0aW9ue0dDOTV9 ClxjaXRhdGlvbntHQzk1fQpcY2l0YXRpb257RUNNOTN9ClxjaXRhdGlvbntHZTk4fQpcY2l0 YXRpb257R2EwNGJ9ClxjaXRhdGlvbntHYTAxfQpcY2l0YXRpb257R0wwM30KXG5ld2xhYmVs ezYuMX17ezl9ezd9e317fXt9fQpcbmV3bGFiZWx7Ni4yfXt7MTB9ezd9e317fXt9fQpcbmV3 bGFiZWx7Ni4zfXt7MTF9ezd9e317fXt9fQpcQHdyaXRlZmlsZXt0b2N9e1xjb250ZW50c2xp bmUge3NlY3Rpb259e1xudW1iZXJsaW5lIHtWSUl9RW50cm9weT99ezd9e319ClxuZXdsYWJl bHs3LjF9e3sxMn17N317fXt9e319ClxjaXRhdGlvbntHYTAwfQpcY2l0YXRpb257RU05MH0K XGNpdGF0aW9ue0dhMDZ9ClxjaXRhdGlvbntHYTA2fQpcY2l0YXRpb257QkdHWjA1fQpcY2l0 YXRpb257R2U5OH0KXGNpdGF0aW9ue0dhMDZ9ClxjaXRhdGlvbntCR0daMDZ9ClxjaXRhdGlv bntKYTk5fQpcY2l0YXRpb257RVBSOTl9ClxjaXRhdGlvbntSdTA2fQpcY2l0YXRpb257Qm83 Mn0KXG5ld2xhYmVsezcuMn17ezEzfXs4fXt9e317fX0KXG5ld2xhYmVsezcuM317ezE0fXs4 fXt9e317fX0KXG5ld2xhYmVsezcuNH17ezE1fXs4fXt9e317fX0KXG5ld2xhYmVsezcuNX17 ezE2fXs4fXt9e317fX0KXG5ld2xhYmVsezcuNn17ezE3fXs4fXt9e317fX0KXEB3cml0ZWZp bGV7dG9jfXtcY29udGVudHNsaW5lIHtzZWN0aW9ufXtcbnVtYmVybGluZSB7VklJSX1FeHRl bnNpb25zIG9mIEJvbHR6bWFubidzICRIJC10aGVvcmVtfXs4fXt9fQpcY2l0YXRpb257S2w3 M30KXGNpdGF0aW9ue0JvODc3Yn0KXGNpdGF0aW9ue0dhMDRifQpcY2l0YXRpb257R2EwNGJ9 ClxjaXRhdGlvbntHYTA0Yn0KXGNpdGF0aW9ue0dHTDA0fQpcY2l0YXRpb257R2EwNGJ9Clxj aXRhdGlvbntHYTAwfQpcY2l0YXRpb257R0JHMDR9ClxjaXRhdGlvbntHYTA0fQpcY2l0YXRp b257Qm84NzR9ClxAd3JpdGVmaWxle3RvY317XGNvbnRlbnRzbGluZSB7c2VjdGlvbn17XG51 bWJlcmxpbmUge0lYfUNvbmNsdXNpb259ezl9e319ClxjaXRhdGlvbntCbzg3NH0KXGNpdGF0 aW9ue0JvODcxYX0KXGNpdGF0aW9ue0JvODcxYn0KXGNpdGF0aW9ue0JvNzJ9ClxjaXRhdGlv bntCbzg3N2J9ClxjaXRhdGlvbntCcjAzfQpcY2l0YXRpb257QnI3Nn0KXGNpdGF0aW9ue0Jy MDN9ClxjaXRhdGlvbntHYTAwfQpcYmliY2l0ZXtCbzY2fXt7MX17MTk2OHt9fXt7Qm9sdHpt YW5ufX17e319fQpcYmliY2l0ZXtCbzg3MWF9e3syfXsxOTY4e319e3tCb2x0em1hbm59fXt7 fX19ClxiaWJjaXRle0JyMDN9e3szfXsyMDAzfXt7QnJ1c2h9fXt7fX19ClxiaWJjaXRle0Jv ODY4fXt7NH17MTk2OHt9fXt7Qm9sdHptYW5ufX17e319fQpcYmliY2l0ZXtGbDAwfXt7NX17 MjAwMH17e0JvbHR6bWFubn19e3t9fX0KXGJpYmNpdGV7Qm84NzFifXt7Nn17MTk2OHt9fXt7 Qm9sdHptYW5ufX17e319fQpcYmliY2l0ZXtCbzg3N2F9e3s3fXsxOTY4e319e3tCb2x0em1h bm59fXt7fX19ClxiaWJjaXRle0hlODg0YX17ezh9ezE4OTV7fX17e0hlbG1ob2x0en19e3t9 fX0KXGJpYmNpdGV7SGU4ODRifXt7OX17MTg5NXt9fXt7SGVsbWhvbHR6fX17e319fQpcYmli Y2l0ZXtCbzg4NH17ezEwfXsxOTY4e319e3tCb2x0em1hbm59fXt7fX19ClxiaWJjaXRle0dh MDB9e3sxMX17MjAwMH17e0dhbGxhdm90dGl9fXt7fX19ClxiaWJjaXRle0dhOTVhfXt7MTJ9 ezE5OTV9e3tHYWxsYXZvdHRpfX17e319fQpcYmliY2l0ZXtGaTY0fXt7MTN9ezE5NjR9e3tG aXNoZXJ9fXt7fX19ClxiaWJjaXRle1J1Njh9e3sxNH17MTk2OH17e1J1ZWxsZX19e3t9fX0K XGJpYmNpdGV7QnI3Nn17ezE1fXsxOTc2fXt7QnJ1c2h9fXt7fX19ClxiaWJjaXRle0VFMTF9 e3sxNn17MTk5MH17e0VocmVuZmVzdCBhbmQgRWhyZW5mZXN0fX17e319fQpcYmliY2l0ZXtC bzk2fXt7MTd9ezIwMDN9e3tCb2x0em1hbm59fXt7fX19ClxiaWJjaXRle0JvOTZhfXt7MTh9 ezE5NjR9e3tCb2x0em1hbm59fXt7fX19ClxiaWJjaXRle1RoNzR9e3sxOX17MTg3NH17e1Ro b21zb259fXt7fX19ClxiaWJjaXRle0xWOTN9e3syMH17MTk5M317e0xldmVzcXVlIGFuZCBW ZXJsZXR9fXt7fX19ClxiaWJjaXRle0JvODc3Yn17ezIxfXsxOTY4e319e3tCb2x0em1hbm59 fXt7fX19ClxiaWJjaXRle1J1OTZ9e3syMn17MTk5Nn17e1J1ZWxsZX19e3t9fX0KXGJpYmNp dGV7R0JHMDR9e3syM317MjAwNH17e0dhbGxhdm90dGkgZXR+YWwufX17e0dhbGxhdm90dGks IEJvbmV0dG8sIGFuZCBHZW50aWxlfX19ClxiaWJjaXRle0dhOTZ9e3syNH17MTk5Nn17e0dh bGxhdm90dGl9fXt7fX19ClxiaWJjaXRle0FXNzB9e3syNX17MTk3MH17e0FkbGVyIGFuZCBX ZWlzc319e3t9fX0KXGJpYmNpdGV7U2k2OGF9e3syNn17MTk2OHt9fXt7U2luYWl9fXt7fX19 ClxAd3JpdGVmaWxle3RvY317XGNvbnRlbnRzbGluZSB7c2VjdGlvbn17XG51bWJlcmxpbmUg e31SZWZlcmVuY2VzfXsxMX17fX0KXGJpYmNpdGV7U2k2OGJ9e3syN317MTk2OHt9fXt7U2lu YWl9fXt7fX19ClxiaWJjaXRle1NpNzJ9e3syOH17MTk3Mn17e1NpbmFpfX17e319fQpcYmli Y2l0ZXtCbzcwYX17ezI5fXsxOTcwfXt7Qm93ZW59fXt7fX19ClxiaWJjaXRle0JSNzV9e3sz MH17MTk3NX17e0Jvd2VuIGFuZCBSdWVsbGV9fXt7fX19ClxiaWJjaXRle1J1NzZ9e3szMX17 MTk3Nn17e1J1ZWxsZX19e3t9fX0KXGJpYmNpdGV7UnU4MH17ezMyfXsxOTgwfXt7UnVlbGxl fX17e319fQpcYmliY2l0ZXtHQzk1fXt7MzN9ezE5OTV9e3tHYWxsYXZvdHRpIGFuZCBDb2hl bn19e3t9fX0KXGJpYmNpdGV7UnU3M317ezM0fXsxOTczfXt7UnVlbGxlfX17e319fQpcYmli Y2l0ZXtFQ005M317ezM1fXsxOTkzfXt7RXZhbnMgZXR+YWwufX17e0V2YW5zLCBDb2hlbiwg YW5kIE1vcnJpc3N9fX0KXGJpYmNpdGV7R2U5OH17ezM2fXsxOTk4fXt7R2VudGlsZX19e3t9 fX0KXGJpYmNpdGV7R2EwNGJ9e3szN317MjAwNHt9fXt7R2FsbGF2b3R0aX19e3t9fX0KXGJp YmNpdGV7R2EwMX17ezM4fXsyMDAxfXt7R2FsbGF2b3R0aX19e3t9fX0KXGJpYmNpdGV7R0ww M317ezM5fXsyMDA0fXt7R29sZHN0ZWluIGFuZCBMZWJvd2l0en19e3t9fX0KXGJpYmNpdGV7 RU05MH17ezQwfXsxOTkwfXt7RXZhbnMgYW5kIE1vcnJpc3N9fXt7fX19ClxiaWJjaXRle0dh MDZ9e3s0MX17MjAwNn17e0dhbGxhdm90dGl9fXt7fX19ClxiaWJjaXRle0JHR1owNX17ezQy fXsyMDA2e319e3tCb25ldHRvIGV0fmFsLn19e3tCb25ldHRvLCBHYWxsYXZvdHRpLCBHaXVs aWFuaSwgYW5kIFphbXBvbml9fX0KXGJpYmNpdGV7QkdHWjA2fXt7NDN9ezIwMDZ7fX17e0Jv bmV0dG8gZXR+YWwufX17e0JvbmV0dG8sIEdhbGxhdm90dGksIEdpdWxpYW5pLCBhbmQgWmFt cG9uaX19fQpcYmliY2l0ZXtKYTk5fXt7NDR9ezE5OTl9e3tKYXJ6eW5za2l9fXt7fX19Clxi aWJjaXRle0VQUjk5fXt7NDV9ezE5OTl9e3tKLlAuRWNrbWFubiBldH5hbC59fXt7Si5QLkVj a21hbm4sIFBpbGxldCwgYW5kIEJlbGxldH19fQpcYmliY2l0ZXtSdTA2fXt7NDZ9ezIwMDZ9 e3tSdWVsbGV9fXt7fX19ClxiaWJjaXRle0JvNzJ9e3s0N317MTk2OHt9fXt7Qm9sdHptYW5u fX17e319fQpcYmliY2l0ZXtLbDczfXt7NDh9ezE5NzN9e3tLbGVpbn19e3t9fX0KXGJpYmNp dGV7R0dMMDR9e3s0OX17MjAwNX17e0dhcnJpZG8gZXR+YWwufX17e0dhcnJpZG8sIEdvbGRz dGVpbiwgYW5kIExlYm93aXR6fX19ClxiaWJjaXRle0dhMDR9e3s1MH17MjAwNHt9fXt7R2Fs bGF2b3R0aX19e3t9fX0KXGJpYmNpdGV7Qm84NzR9e3s1MX17MTk3NH17e0JvbHR6bWFubn19 e3t9fX0KXGJpYmNpdGV7R2kwMn17ezUyfXsxOTAyfXt7R2liYnN9fXt7fX19ClxnbG9iYWwg XGNoYXJkZWYgXGZpcnN0bm90ZUBudW01MlxyZWxheCAKXGJpYmNpdGV7ZW5kbm90ZTUzfXt7 NTN9e317e319e3t9fX0KXGNpdGF0aW9ue0JvODcxYn0KXGNpdGF0aW9ue0dpMDJ9ClxjaXRh dGlvbntCbzg3MWJ9ClxiaWJjaXRle2VuZG5vdGU1NH17ezU0fXt9e3t9fXt7fX19ClxjaXRh dGlvbntCbzg4NH0KXGNpdGF0aW9ue0dhMDB9ClxnbG9iYWxcTkFUQG51bWJlcnN0cnVlClxi aWJzdHlsZXthcHNyZXZ9ClxuZXdsYWJlbHtMYXN0UGFnZX17e317MTF9fQpcbmV3bGFiZWx7 TGFzdEJpYkl0ZW19e3s1NH17MTJ9e317fXt9fQo= ---------------0606181305916--