\documentclass[11pt]{amsart}
\newtheorem{definition}{Definition}
\newtheorem{theorem}{Theorem}[section]
\newtheorem{lemma}{Lemma}[section]
\newtheorem{proposition}{Proposition}[section]
\newtheorem{corollary}{Corollary}[theorem]
\newtheorem{example}{Example}[section]
\newcommand{\nempty}{\not= \emptyset}
\newcommand{\invlim}[2]{\ensuremath{\lim\limits_{\leftarrow}\{#1,#2\}} }
\newcommand{\prodsp}[3]{\ensuremath{\prod_{#1\in #2}#3_{#1}} }
\newcommand{\distx}[4]{\ensuremath{\sum_{i=#1}^{#2}\left( \left[
\frac{|#3_i-#4_i|}{z_i} \right] \cdot \frac{1}{2^i} \right)} }
\newcommand{\B}[1]{\ensuremath{\mathbb{#1}}}
\newcommand{\C}[1]{\ensuremath{\mathcal{#1}}}
\newcommand{\chain}[1]{\{\C{#1}_i\}_{i=1}^{\infty}}
\newcommand{\hatf}[2]{\hat{#1}(#2)}
\newcommand{\grchain}[2]{\ensuremath{\C{#1}=\{
#1^{(i,j)}_1,#1^{(i,j)}_2,
\dots,#1^{(i,j)}_{#2_{(i,j)}}|(i,j)\in E_{\C{#1}}\}}}
\newcommand{\und}[1]{\ensuremath{\underline{#1}}}
\setcounter{theorem}{0}
\setcounter{definition}{0}
\setcounter{section}{0}
\begin{document}
\setlength{\unitlength}{0.01in}
\linethickness{0.01in}
\begin{center}
\begin{picture}(474,66)(0,0)
\multiput(0,66)(1,0){40}{\line(0,-1){24}}
\multiput(43,65)(1,-1){24}{\line(0,-1){40}}
\multiput(1,39)(1,-1){40}{\line(1,0){24}}
\multiput(70,2)(1,1){24}{\line(0,1){40}}
\multiput(72,0)(1,1){24}{\line(1,0){40}}
\multiput(97,66)(1,0){40}{\line(0,-1){40}}
\put(143,66){\makebox(0,0)[tl]{\footnotesize Proceedings of the Ninth Prague Topological Symposium}}
\put(143,50){\makebox(0,0)[tl]{\footnotesize Contributed papers from the symposium held in}}
\put(143,34){\makebox(0,0)[tl]{\footnotesize Prague, Czech Republic, August 19--25, 2001}}
\end{picture}
\end{center}
\vspace{0.25in}
\setcounter{page}{253}
\title[Orbits of turning points]{Orbits of turning points for maps of
finite graphs and inverse limit spaces}
\author{Brian Raines}
\address{Mathematical Institute\\
University of Oxford\\
Oxford OX1 3LB\\
United Kingdom}
\email{raines@maths.ox.ac.uk}
\subjclass[2000]{54H20, 54F15, 37E25}
\keywords{inverse limits, graph, continuum}
\thanks{Brian Raines,
{\em Orbits of turning points for maps of finite graphs and inverse limit
spaces},
Proceedings of the Ninth Prague Topological Symposium, (Prague, 2001),
pp.~253--263, Topology Atlas, Toronto, 2002; {\tt arXiv:math.GN/0204137}}
\begin{abstract}
In this paper we examine the topology of inverse limit spaces generated by
maps of finite graphs.
In particular we explore the way in which the structure of the orbits of
the turning points affects the inverse limit.
We show that if $f$ has finitely many turning points each on a finite
orbit then the inverse limit of $f$ is determined by the number of
elements in the $\omega$-limit set of each turning point.
We go on to identify the local structure of the inverse limit space at the
points that correspond to points in the $\omega$-limit set of $f$ when the
turning points of $f$ are not necessarily on a finite orbit.
This leads to a new result regarding inverse limits of maps of the
interval.
\end{abstract}
\maketitle
\section[intro]{Introduction}
Every one-dimensional continuum is an inverse limit on finite graphs, and
many, though not all, are homeomorphic to an inverse limit on a finite
graph with a single bonding map. These spaces also naturally appear in
dynamical systems. R.F. Williams showed that if a manifold diffeomorphism
$F$ has a one-dimensional hyperbolic attractor $\Lambda$ (with associated
stable manifold structure)then $F$ restricted to $\Lambda$ is
topologically conjugate with the shift homeomorphism on an inverse limit
of a piecewise monotone map $f$ of some finite graph, \cite{williams}, and
Barge and Diamond, \cite{barge&diamond1}, remark that for any map
$f:G\rightarrow G$ of a finite graph there is a homeomorphism
$F:\B{R}^3\rightarrow \B{R}^3$ with an attractor on which $F$ is conjugate
to the shift homeomorphism on $\invlim{G}{f}$. More recently, Anderson
and Putnam, \cite{anderson&putnam}, have shown that the dynamics arising
from a substitution tiling is often conjugate to the action of a shift-map
on an inverse limit of a branched d-manifold. They then demonstrate how
to use knowledge about the inverse limit space to compute the cohomology
and K-theory of a space of tilings. Extending these ideas, Barge,
Jacklitch and Vago, \cite{bargejacklitch&vago}, use inverse limits induced
by certain Markov maps on wedges of circles to analyze one-dimensional
substitution tiling spaces and one-dimensional unstable manifolds of
hyperbolic sets. Many of their results rely on showing that certain pairs
of these inverse limit spaces are not homeomorphic.
It is often quite difficult to distinguish between inverse limit spaces,
even when the dynamics of the bonding maps are very different. Many
papers have been written to this end, \cite{barge&diamond2},
\cite{barge&martin}, \cite{bruin}, \cite{kailhofer}, and \cite{raines}.
However most of the techniques have been focused on maps of the interval.
Perhaps one of the easiest ways to decide if two inverse limit spaces are
not homeomorphic is to count their endpoints. Barge and Martin have shown
that the number of endpoints of $\invlim{[0,1]}{f}$ is the same as the
number of elements in the $\omega$-limit set of the turning points of the
bonding map, $f$, when $f$ has a dense orbit and finitely many turning
points, \cite{barge&martin}.
In this paper we distinguish between these inverse limit spaces by showing
that many of the points have neighborhoods homeomorphic to the product of
a zero-dimensional set and $(0,1)$, and we show that the exceptional
points are those that always project onto the $\omega$-limit set of the
turning points. We do this for inverse limits on graphs, but of course,
the result holds for inverse limits on the interval. In the case of the
interval, our theorem is still an extension of Barge and Martin's result,
because there are many bonding maps that give rise to a three-endpoint
indecomposable continua that have more than three points in the
$\omega$-limit set of their turning points. Our theorem can be used to
easily distinguish between these inverse limit spaces.
\section[prelim]{Preliminaries}\label{prelim}
By a {\em continuum} we mean a compact, connected, metric space, and by a
{\em mapping} we mean a continuous function. We will say a mapping, $f$,
is {\em monotone} on $A$ if, and only if, $f^{-1}(x)$ is connected for all
$x\in A$. The {\em inverse limit} induced by a single bonding map, $f$, on
a continuum $M$ is defined as follows: $$\invlim{M}{f}=\{(x_0,
x_1,\dots)|x_i\in M \quad\mbox{and}\quad f(x_{i+1})=x_i\}.$$ Since $M$ is
metric and $f$ is a mapping, $\invlim{M}{f}$ is a continuum with the
metric: $$d(x,y)=\sum_{i=0}^{\infty}\frac{d_M(x_i-y_i)}{2^i},$$ where
$d_M$ is the metric on $M$ and we assume that $d_M(x,y)<1$ for all $x,y\in
M$. Define the projection maps $\pi_n:\invlim{M}{f}\rightarrow M$ by
$\pi_n(x)=x_n$, where $x=(x_1, x_2, \dots)\in \invlim{M}{f}$.
Also, define the shift homeomorphism
$h:\invlim{M}{f}\rightarrow \invlim{M}{f}$ by
$$h(x)=(f(x_0),f(x_1),f(x_2),\dots)=(f(x_0),x_0,x_1,\dots).$$
A {\em linear chaining}, or just {\em chaining}, of a continuum $M$ is a
finite sequence, $L_1,L_2,L_3,\dots ,L_n$ of open subsets of $M$ such that
$L_i$ intersects $L_j$ if and only if $|i-j|<2$. The open sets comprising
the chain are called the {\em links} of the chain. The {\em mesh} of a
chain is the largest of the diameters of its links. A continuum $M$ is
said to be {\em chainable} provided that for each positive number
$\epsilon$ there is a chaining of $M$ with mesh less than $\epsilon$, such
a chain is called an {\em $\epsilon$-chain}. It is a well-known fact that
an inverse limit of chainable continua is a chainable continuum. A {\em
closed chain} is a chain whose links are closed sets and if $i\not= j$,
then $L_i\cap L_j=$Bd$(L_i)\cap $Bd$(L_j)$ if, and only if, $|i-j|<2$. We
lose no generality in assuming that all of the chains in this paper are
taut, i.e. if $L_i\cap L_j=\emptyset$ then $\overline{L_i}\cap
\overline{L_j}=\emptyset$, \cite{nadler}. Notice that if $\C L$ is a taut
chaining of an inverse limit space and $L_i\cap L_j=\emptyset$ then it is
possible to find a positive integer, $q$, large enough so that
$\pi_q(\overline{L_i})\cap \pi_q(\overline{L_j})=\emptyset$, which implies
that $\pi_q(\C L)$ is a chain.
A {\em finite graph}, $G$, is a continuum that can be written as the union
of finitely many arcs any two of which are either disjoint or intersect at
only one of their endpoints. For any finite graph, $G$, there is a finite
set of points called {\em vertices}, $V=\{v_1,v_2,\dots,v_n\}$, and a set
of arcs, $E$, with endpoints from $V$ called {\em edges}, with the
property that if $v_k\in e_{ij}\in E$ then either $k=i$ or $k=j$, and if
two edges meet, they meet only at a single common vertex. For simplicity,
we adopt the convention that, since $e_{ij}=e_{ji}$, if we label an edge
$e_{ij}$ then $i \operatorname{diam}(T^i_j)$ and
$\delta_1 \cdot Q^i_j > \operatorname{diam}(S^i_j)$.
Let $\C{C}_1$ be a closed graph-chaining of $G$ with mesh less than
$\delta_1$ which refines $\C{T}^f$ such that, for every $i\in E'$ and
$j\le n_i$, there are $Q^i_j$ links of $\C{C}_1$ contained in $T^i_j$.
Let $\C{J}_1$ be defined similarly with respect to $g$ and $\C{S}^g$.
Let $\C{D}_1=\pi_1^{-1}(\C{C}_1)\cap \invlim{G}{f}$ and let
$\C{K}_1=\pi_1^{-1}(\C{J}_1)\cap \invlim{G}{g}$. It is easy to see that
both of $\C{D}_1$ and $\C{K}_1$ are closed graph-chainings of
$\invlim{G}{f}$ and $\invlim{G}{g}$ respectively with mesh less than
$\frac{1}{2}$.
By lemma \ref{lemma1} both of $\hat{f}(\C{C}_1)$ and $\hat{g}(\C{J}_1)$
are closed graph-chainings of $G$ which refine $\C{T}^f$ and $\C{S}^g$
respectively. Let $\delta_2$ be a positive number so that any closed
graph-chaining of $G$, $\C{H}$, with mesh less than $\delta_2$ has
$\pi_2^{-1}(\C{H})\cap \invlim{G}{f}$ and $\pi_2^{-1}(\C{H})\cap
\invlim{G}{g}$ both have mesh less than $\frac{1}{4}$.
By the hypothesis of the theorem, $\hat{f}_{p,r}(C^i_j)$ is defined if and
only if $\hat{g}_{p,r}(J^i_j)$ is defined. Notice that by the construction
of $\C{C}_1$ and $\C{J}_1$, there is a function, $\ell$, such that
$\C{C}_1$ follows pattern $\ell$ in $\C{T}^f$ and $\C{J}_1$ also follows
pattern $\ell$ in $\C{S}^g$. So for every $i\in E'$, $j\le n_i$, and
$(p,r)\in S_{\ell(i,j)}$, let $Q^{p,r}_{i,j}$ be a positive integer such
that
$Q^{p,r}_{i,j} \cdot \delta_2 >
\operatorname{diam}(\hat{f}_{p,r}(C^i_j))$ and
$Q^{p,r}_{i,j}\cdot \delta_2 > \operatorname{diam}(\hat{g}_{p,r}(J^i_j))$.
Let $\C{C}_2$
be a refinement of $\hat{f}(\C{C}_1)$ such that there are $Q^{p,r}_{i,j}$
links of $\C{C}_2$ inside each $\hat{f}_{p,r}(C^i_j)$. Let $\C{J}_2$ be a
refinement of $\hat{g}(\C{J}_2)$ defined similarly. Define $\C{D}_2$ to
be $\pi_2^{-1}(\C{C}_2)\cap \invlim{G}{f}$ and define $\C{K}_2$ to be
$\pi_2^{-1}(\C{J}_2)\cap \invlim{G}{g}$.
Notice that if $A$ is a subset of $\hat{f}_{p,r}(C^i_j)$ then
$$
\pi_2^{-1}(A)\cap \invlim{G}{f}\subseteq \pi_1^{-1}(C^i_j)\cap
\invlim{G}{f},
$$
and similarly if $A$ is a subset of $\hat{g}_{p,r}(J^i_j)$ then
$$
\pi_2^{-1}(A)\cap \invlim{G}{g}\subseteq \pi_1^{-1}(J^i_j)\cap
\invlim{G}{g}.
$$
So, since we have exactly $Q^{p,r}_{i,j}$ links of $\C{C}_2$ and $\C{J}_2$
in $\hat{f}_{p,r}(C^i_j)$ and $\hat{g}_{p,r}(J^i_j)$ respectively,
$\C{D}_2$ follows the same pattern, $h_2$, in $\C{D}_1$ that $\C{K}_2$
follows in $\C{K}_1$.
Clearly, chains of $\invlim{G}{f}$ and $\invlim{G}{g}$, $\C{D}_3$ and
$\C{K}_3$ can be constructed such that
$\operatorname{mesh}(\C{D}_3) < \frac{1}{8}$,
$\operatorname{mesh}(\C{K}_3) < \frac{1}{8}$,
and both $\C{D}_3$ and $\C{K}_3$ follow pattern $h_3$ in $\C{D}_2$ and
$\C{K}_2$ respectively.
So it is easy to see that we can build a sequence of refining chainings,
$\chain{D}$, of $\invlim{G}{f}$ such that $\C{D}_i$ follows pattern $h_i$
in $\C{D}_{i-1}$ and
$$\lim\limits_{i\rightarrow \infty} \operatorname{mesh}(\C{D}_i)=0,$$
and we can build a sequence of refining chainings,
$\chain{K}$, of $\invlim{G}{g}$ such that $\C{K}_i$ follows pattern $h_i$
in $\C{K}_{i-1}$ and
$$\lim\limits_{i\rightarrow \infty} \operatorname{mesh}(\C{K}_i)=0.$$
Thus, by Theorem \ref{smallone}, $\invlim{G}{f}$ is homeomorphic to
$\invlim{G}{g}$.
\end{proof}
This theorem provides some justification for the assumption that we make
in the next section that $f^{-1}(x)$ is completely disconnected for all
$x\in G$. It shows that for a bonding map with finitely many turning
points each on a finite orbit that we lose no generality in making this
assumption. It also has an interesting, and immediate, corollary which is
an extension of a theorem of Holte (\cite{holte1}, Theorem 3.2).
\begin{corollary}
Let each of $f$ and $g$ be Markov maps of the interval with associated
Markov partitions, $B_f=\{0=c_0M$, such that every image under
$f^{-m}$ of $B_{\delta}(x_n)$ meets the set of endpoints for $G$, $E_G$,
then clearly for every positive number, $\lambda$, we can $\lambda$-chain
$\invlim{G}{f}$ with a linear subchain that starts at $x$. Thus there is
a chainable endcontinuum in $G$ having $x$ as an endpoint, and
$B_{\epsilon}(x)$ cannot be homeomorphic to $(0,1)\times S$ where $S$ is a
zero-dimensional set.
Instead, now suppose that $\gamma$ is small enough and $n$ is large enough
so that $f^{-m}(B_{\gamma}(x_n))$ misses $E_G$ for every positive integer
$m$. Let $A=B_{\gamma}(x_n)$, and as above, enumerate the components of
the preimages of $A$, $A_1, A_2, \dots A_n$. Continuing as previously,
let $S$ be the set of sequences of positive integers, $\langle
y_i\rangle$, where $\langle y_i\rangle \in S$ if, and only if, for every
positive integer $i$, $A^{y_1,\dots y_i}_{y_{i+1}}$ is a component of
$f^{-1}(A^{y_1, \dots y_{i-1}}_{y_i})$. Let $t^{i,j}_{k,0}\in P_f$ with
$x_n\in \omega(t^{i,j}_{k,0})$. Then for infinitely many positive
integers, $m$, $t^{i,j}_{k,m}\in A$. So there are infinitely many
connected subsets of $G$, $A^{y_1\dots y_r}_{y_{r+1}}$ that contain
$t^{i,j}_{k,0}$. If there exists one of these subsets, $A^{y_1,\dots
y_r}_{y_{r+1}}$ that only meet $B_f$ at the singleton $t^{i,j}_{k,0}$
then, since these components do not contain any endpoints or vertices of
$G$, every component of the preimages of $A^{y_1, \dots y_r}_{y_{r+1}}$ is
mapped with a single fold across $A^{y_1, \dots y_r}_{y_{r+1}}$. Thus
$B_{\epsilon}(x)$ contains a subspace homeomorphic to a neighborhood of
$(0,1)$ in the $\sin(1/x)$-continuum, and it cannot be homeomorphic to
$(0,1)\times T$, where $T$ is a zero-dimensional set. If instead, for
some $A^{y_1,\dots, y_r}_{y_{r+1}}$ we have $A^{y_1,\dots
y_r}_{y_{r+1}}\cap B_f$ is a finite set then clearly we can restrict the
size of $A^{y_1, \dots, y_r}_{y_{r+1}}$ in order to make the subset meet
$B_f$ at a singleton and produce a similar subspace.
So suppose that each subset, $A^{y_1,\dots y_r}_{y_{r+1}}$ that contains a
turning point meets $B_f$ on an infinite set. Pick one of these subsets
and call it $A_1$. Let $A_2$ be a connected subset of $G$ such that
$f^{n_1}(A_2)=A_1$ and $A_2\cap P_f\nempty$. Given $A_i$ define $A_{i+1}$
to be a connected subset of $G$ with the property that
$f^{n_i}(A_{i+1})=A_i$ and $A_{i+1}$ meets $P_f$. Then since $A$ is small
enough to not meet $P_f$ at two points and since all of these subsets are
preimages of $A$, they must all meet $P_f$ at a single point. Similarly,
they must all miss $V'$ and the set of endpoints of $G$. Thus, for every
positive integer, $i$, $f|_{A_i}$ is a two-pass map, and
$f^{n_i}|_{A_{i+1}}$ is at least a two-pass map. Thus
$\invlim{A_i}{f|_{A_i}}$ is an indecomposable subcontinuum, and
$B_{\epsilon}(x)$ contains an indecomposable subcontinuum. Hence,
$B_{\epsilon}(x)$ is not homeomorphic to $(0,1)\times T$ where $T$ is a
zero-dimensional set.
\end{proof}
\begin{corollary}
Suppose that $f$ and $g$ are maps of $[0,1]$ with the properties listed
above.
Further suppose that $|\omega(P_f)|=n$ and $|\omega(P_g)|=m$.
If $n\not= m$ then $\invlim{[0,1]}{f}$ is not homeomorphic to
$\invlim{[0,1]}{g}$.
\end{corollary}
\begin{proof}
Notice that for every point in $\omega(P_f)$ there is a point in the
inverse limit that either is an endpoint or is in a neighborhood
homeomorphic to a neighborhood of $(0,1)$ in the
$\sin(1/x)$-continuum.
Also notice that any other point in the inverse limit has a neighborhood
homeomorphic to the product of $(0,1)$ with a zero-dimensional set, $S$.
These properties are preserved by homeomorphism, and so any space
homeomorphic to it must have the same number of points in the
$\omega$-limit set of its turning points.
\end{proof}
%\bibliographystyle{amsplain}
%\bibliography{24}
\providecommand{\bysame}{\leavevmode\hbox to3em{\hrulefill}\thinspace}
\providecommand{\MR}{\relax\ifhmode\unskip\space\fi MR }
% \MRhref is called by the amsart/book/proc definition of \MR.
\providecommand{\MRhref}[2]{%
\href{http://www.ams.org/mathscinet-getitem?mr=#1}{#2}
}
\providecommand{\href}[2]{#2}
\begin{thebibliography}{10}
\bibitem{anderson&putnam}
Jared~E. Anderson and Ian~F. Putnam, \emph{Topological invariants for
substitution tilings and their associated ${C}\sp *$-algebras}, Ergodic
Theory Dynam. Systems \textbf{18} (1998), no.~3, 509--537. \MR{2000a:46112}
\bibitem{barge&diamond1}
Marcy Barge and Beverly Diamond, \emph{The dynamics of continuous maps of
finite graphs through inverse limits}, Trans. Amer. Math. Soc. \textbf{344}
(1994), no.~2, 773--790. \MR{95c:58060}
\bibitem{barge&diamond2}
\bysame, \emph{Homeomorphisms of inverse limit spaces of one-dimensional maps},
Fund. Math. \textbf{146} (1995), no.~2, 171--187. \MR{96b:54048}
\bibitem{bargejacklitch&vago}
Marcy Barge, James Jacklitch, and Gioia Vago, \emph{Homeomorphisms of
one-dimensional inverse limits with applications to substitution tilings,
unstable manifolds, and tent maps}, Geometry and topology in dynamics
(Winston-Salem, NC, 1998/San Antonio, TX, 1999), Amer. Math. Soc.,
Providence, RI, 1999, pp.~1--15. \MR{2000j:37016}
\bibitem{barge&martin}
Marcy Barge and Joe Martin, \emph{Endpoints of inverse limit spaces and
dynamics}, Continua (Cincinnati, OH, 1994), Dekker, New York, 1995,
pp.~165--182. \MR{96b:54062}
\bibitem{bruin}
Henk Bruin, \emph{Inverse limit spaces of post-critically finite tent maps},
Fund. Math. \textbf{165} (2000), no.~2, 125--138. \MR{2002d:37022}
\bibitem{holte1}
Sarah~E. Holte, \emph{Generalized horseshoe maps and inverse limits}, Pacific
J. Math. \textbf{156} (1992), no.~2, 297--305. \MR{93k:58142}
\bibitem{kailhofer}
L.~Kailhofer, \emph{A classification of the inverse limit spaces of tent maps
with periodic critical points}, Preprint.
\bibitem{nadler}
Sam~B. Nadler, Jr., \emph{Continuum theory}, Marcel Dekker Inc., New York,
1992. \MR{93m:54002}
\bibitem{raines}
Brian Raines, \emph{A complete classification of inverse limit spaces generated
by tent maps with periodic critical points}, To appear in Fund. Math., 2002.
\bibitem{williams}
R.~F. Williams, \emph{One-dimensional non-wandering sets}, Topology \textbf{6}
(1967), 473--487. \MR{36 \#897}
\end{thebibliography}
\end{document}