Skip to main content

Full text of "Generic Hamiltonian Dynamics"

See other formats


(N 



r- 






> 



GENERIC HAMILTONIAN DYNAMICS 

MARIO BESSA, CELIA FERREIRA, AND JORGE ROCHA 



f^ Abstract. In this paper we contribute to the generic theory of Hamiltonians by proving 

CN that there is a C^-residual TZ in the set of C^ Hamiltonians on a closed symplectic 

H manifold M, such that, for any H G TZ, there is an open and dense set S{H) in H{M) 

^^ such that, for every e S S{H), the Hamiltonian level {H, e) is topologically mixing. 



Keywords: Hamiltonian vector field, topological transitivity, pseudo-orbits. 



C/^ 1. Introduction 

Q 

1.1. Hamiltonians and their applications. The Hamiltonian systems form a fun- 
damental subclass of all dynamical systems generated by differential equations. Their 
importance foUows from the vast range of applications throughout different branches of 
science. In fact, the laws of physics are mostly expressed in terms of differential equations, 
and a well understanded and successful subclasse of these differential equations, which 
leave invariant a symplectic structure, are the Hamiltonian equations (see [5]). 
0\ Generic properties of such continuous-time systems are thus of great importance and 

1^ interest since they give us the typical behavior in an appropriate sense that one could 

^^ expect from the class of models at hand (cf. [251 1131 E])- 

^ The main result of this paper (TheoremfTl) is the generalization of |17t[8] for Hamiltoni- 

O ans and states that "most" Hamiltonians have indecomposable energy levels in the sense 

^ that we cannot split the energy level or, in other words, there is some orbit that winds 

around the whole energy level. If we weaken the topology, and inspired on the Oxtoby 
. !^ and Ulam theoremj^ (see [23]), we expect to obtain ergodicity in the energy level. On the 

rN other hand, for stronger topologies, KAM theorem (see [27]) makes impossible to obtain 

C^ the same result (due to the persistence of invariant tori). 

There are, of course, considerable limitations to the amount of Information we can 
extract from a specific system by looking at generic cases. Nevertheless, it is of great 
utility to learn that a selected model can be slightly perturbed in order to obtain dynamics 
we understand in a reasonable way. 

Finally, our results are essentially C^-type results and therefore the Information it 
provides holds only in that topology. For instance, in mathematical physics the forces 
are C* {s > 3) objects in its essence. Let us be more precise, a Hamiltonian of class C* 



Date: February 27, 2013. 

^As far as we know there is not available yet a version for Hamiltonians for this theorem. 



2 M. BESSA, C. FERREIRA, AND J. ROCHA 

generates a Hamiltonian flow of class C*~^, and forces involve a second derivative, thus, 
the natural environment are C^ flows (or C^-Hamiltonians) . Hopefully, it is evident that 
this is not a weakness of our C^-generic results but the counterweight to its undeniable 
generality. 

1.2. The Hamiltonian framework. Let (M^'^,a;) be a symplectic manifold, where 
M = M'^'^ [d > 2) is an even-dimensional, compact, boundaryless, connected and smooth 
Riemannian manifold, endowed with a symplectic form u. Denote by C^lM, M) the set of 
C*-real-valued functions on M and call H G C*(Af , M) a C'^ -Hamiltonian, for s > 2. From 
now on, we set s = 2. Given a Hamiltonian H, we can define the Hamiltonian vector field 
XHhj 

(1) uj{XHip), u) = dpH{u), Vu e TpM, 

which generates the Hamiltonian fiow X^. Observe that H is C^ if and only if Xh is C^ 
and that, since H is smooth and M is closed, Sing^Ku) 7^ 0, where Singi^Ku) stands for 
the singularities of Xh or, in other words, the critical points of H . 

A scalar e G H{M) C M is called an energy of H. An energy hypersurface £H,e is a 
connected component of H~^{{e}), called energy level set. 

The energy level set H~^{{e}) is said to be regular if any energy hypersurface of 
H^^{{e}) is regular, i. e, does not contain any singularity. In this case, we can also say that 
the energy e is regular. Observe that a regular energy hypersurface is a X|^-invariant, 
compact and {2d — l)-dimensional manifold. Consider a Hamiltonian H G C^(M, M), 
an energy e G H(M) and a regular energy hypersurface £H,e- The triplet {H,e,SH,e) is 
called a Hamiltonian system and the pair [H, e) is called a Hamiltonian level. If {H, e) 
is regular then H~^{{e}) corresponds to the union of a finite number of closed connected 
components, that is, H~^{{e}) = ul'L^En^f.^i, for /g G N. 

Fixing a small neighborhood W of a regular energy hypersurface £H,ei there exist a 
small neighborhood U of the Hamiltonian H and e > O such that, for any H eU and for 
any e G (e — e, e + e), we have H~^{{e}) r\W = Sfj^. The energy hypersurface Sg ^ is 
called the analytic continuation oi Sh e- 



The next definition states when a Hamiltonian system is Anosov (see Definition 2.3): a 
Hamiltonian system (if , e, £^_ff,e) is Anosov if SH,e is uniformly hyperbolic for the Hamilton- 
ian flow Xjj associated to H (see ^12j for further details on Anosov Hamiltonian systems). 

1.3. Topological transitivity. The topological transitivity is a global property of a 
dynamical system. As a motivation for this notion, we may think of a real physical 
system, where a state is never measured exactly. Thus, instead of points, we should study 
(small) open subsets of the phase space and describe how they move in that space. If 
each one of these open subsets meet each other by the action of the system after some 
time, then we say that the system is topologically transitive. Equivalently, if we take a 
compact phase space, we may say that the system has a dense orbit. However, if the open 
subsets remain inseparable after some time, by the iteration of the system, then we say 



GENERIC HAMILTONIAN DYNAMICS 



that the system is topologically mixing. Obviously, a topologically inixing system is also 
a topologically transitive system. 

The concept of transitivity goes back to Birkhoff fTEl \W\ . Throughout in this paper 
transitive will always mean topologically transitive. 

There exist a lot of transitive systems, as the irrational rotations of §^, the shift niaps 
and the basic sets (see [IH]). It is also well-known that C^~^"'-Anosov conservative systems 
{a > 0) are ergodic and so transitive (see [5]). In fact, the same holds for C^-Anosov 
conservative systems because, by Poincare recurrence, its non-wandering set equals the 
whole manifold and by the Anosov closing lemma the periodic orbits are dense in the 
non-wandering set ([26j). Then, using Smale spectral decomposition ([2S]), we get only 
one piece which is transitive. Nevertheless, transitivity is not an open property. 

Question 1.1. Can the transitivity property be generic? 

Some authors have been working on this question. The first remarkable result on 
this subject is due to Bonatti and Crovisier, in [TTj. They show that, C^-generically, a 
C^-conservative diffeomorphism is transitive. Later, jointly with Arnaud, Bonatti and 
Crovisier extend this result for C^-symplectic diffeomorphisms defined on a symplectic 
manifold (see [8]). Adapting the techniques used to prove these results to the continuous- 
time case, one of the authors proved an analogous result for C^-divergence-free vector 
fields. In fact, by a result due to Abdenur et al. (see pj), the first author was able 
to show that, C^-generically, a divergence-free vector field is topologically mixing (see 
[TU]). Recently, the results in [T71 [S] get an upgrading in [2]. In the direction against 
the abundance of transitivity (ergodicity) , but with a much more exigent smoothness 
hypothesis, we recall the results of [2U] . 

Our contribution to this issue is the statement and the proof of a result that is an 



answer to Question 1.1 for Hamiltonian systems. 



Definition 1.1. A compact energy hypersurface Eu^e is topologically mixing if, for any 
open and non-empty suhsets of£H,e, say U and V , there zs r G M such that Xjj{U)nV ^ 0, 
jor any t > t. A regular Hamiltonian level {H, e) is topologically mixing if each one of 
the energy hypersurfaces of H~^{{e}) is topologically mixing. 

Accordingly with this definition, we prove the following result. 

Theorem 1. There is a residual TZ in C'^{M,M) such that, for any H &TZ, there is an 
open and dense set S{H) in H {M) such that, for every e G S{H), the Hamiltonian level 
{H, e) is topologically mixing. 

The main tool to prove the previous result is a version for Hamiltonians of the Con- 
necting Lemma for pseudo-orbits developed in [S] by Arnaud et al.. To state it, we need 



the notions of resonance relations and of pseudo-orbits, which we postpone to Section[23 

Lemma 1 (Connecting Lemma for pseudo-orbits of Hamiltonians). Take H G C'^{M,M) 
and a regular energy e G H{M), such that the eigenvalues of any closed orbit of H do 



4 M. BESSA, C. FERREIRA, AND J. ROCHA 

not satisfy non-trivial resonances. Then, for any C'^ -neighhorhoodU of H , for any energy 
hypersurface £H,e C H^^{{e}) and for any x,y G £H,e connected by an e-pseudo-orbit, for 
e > O, there exist H & U and t > O such that e = H{x) and X^-{x) = y on the analytic 
continuation £ fj ^ of £H,e- 

To prove these results, we have to resume the arguments used by Arnaud et al. [SI Ej 
and by one of the authors in [T^ and to adapt it to the Hamiltonian setting. Besides 
the perturbation techniques, the core of the proofs is the need to restrict our attention to 
the energy hypersurface, in order to perturb the Hamiltonian and keep the energy, when 
analyzing the perturbations and their supports. 

From Theorem [T| we can derive the following result concerning on the honioclinic 
class of a hyperbohc closed orbit 7 of if , which is the closure of the set of transversal 
intersections between the stable and unstable manifolds of all points p in 7 (see Section 



2.3 for more details). 

Corollary 1. There is a residual set TZ in C'^{M,'R) such that, for any H eTZ, there is 
an open and dense set S{H) in H {M) such that if e E <S{H) then any energy hypersurface 
of H~^{{e}) is a homoclinic class. 

If any energy hypersurface of H~^{{e}) is a homoclinic class, we say that H^^{{e}) is 
a homoclinic class. 

We end this section with an overview of the remaining sections of this paper. This 
paper is organized in two additional sections. In Section ^ we include some notes on 
Hamiltonian dynamics and in Section ^ we concern about the proof of Theorem [T] by 
proving the connecting lemma for pseudo orbits. In each section we also include extra 
definitions and useful auxiliary results. 

2. Hamiltonian dynamics 

2.1. More definitions. Recall that {M, u) denotes a symplectic manifold, where M is 
an even-dimensional Riemannian manifold endowed with a symplectic form u. Recall that 
a symplectic form is a skew-symmetric and non-degenerate 2-form on the tangent bundle 
TM. These properties, on the symplectic form, play an important role in the characteri- 
zation of the Hamiltonian dynamics. The non-degeneracy of the form u guarantees that 
a Hamiltonian vector field is well-defined, while the skew-symmetry of u leads to conser- 
vative properties for the Hamiltonian vector field. Once more, since u is non-degenerate, 
given H e C^(M, M) and p G M, we know that dpH = O is equivalent to Xh{p) = O, 
where dpH stands for the gradient of iJ in p G M. Therefore, the extreme values of a 
Hamiltonian H are exactly the singularities of the associated Hamiltonian vector field 
Xh- Let Per {H) denote the set of closed orbits of Xh and Sing{H) denote the set of 
singularities of Xh- 

We say that H is e-C^-close to H, for e > O fixed, if \\H — H\\c^ < e, where \\H — H\\c^ 
denotes the C^-distance between H and H. 



GENERIC HAMILTONIAN DYNAMICS 5 

Given a Hamiltonian level {H, e), let Q{H\£^ J be tlie set of non-wandering points of 
H on the energy hypersurface SH,e, that is, tlie points x G £H,e sucli tliat, for every 
neighborhood U oi x m S h, e, tliere is r > O sucli tliat X]j{U) fl ?7 7^ 0. 

Fix a Hamiltonian level (if, e). We want H~^{{e}) to decompose into a finite number 
of connected components, say H~^{{e}) = ul'L^£H,e,ii for /g G N. Let us look at the 
foUowing exaniple. 

Example 1: Write iJ : M^ ^ M sucli that 

a;^ sin ( — 1 , a; 7^ O 



H{x,y) = <^ yx 

O , x = 

It is immediate to see that, for the value of energy e = O, H^^{{e}) corresponds to an 
infinite number of connected components. This construction can be made local (torus, 
annulus). A direct consequence of the Implicit Function Theorem ensures that the absence 
of singularities is enough to ensure a finite decomposition of H~^{{e}). 

By Liouville's Theorem, the symplectic manifold {M, u) is also a volume manifold (see, 
for example, [3j). This means that the volume form u"^ = u A u induces a measure /j, on 
M, which is the Lebesgue measure associated to u^. Notice that the measure /i on M 
is preserved by the Hamiltonian flow. So, given a regular Hamiltonian level [H, e), we 
induce a volume form uj£^^ on each energy hypersurface Sn^e C H~^{{e}), where for all 

P e £H,e- 

^£H,e '■ TpSH,e X TpSH,e X TpSH,e > ^ 

{u, V, w) I )■ u'^{dpH, U, V, w) 

The volume form uJsne i^ X]:^-invariant. Hence, it induces an invariant volume measure 
fiSHe '^^ ^H,e that is finite, since any energy hypersurface is compact. Observe that, under 
these conditions, we have that /i£-^^-a.e. x G £H,e is recurrent, by the Poincare Recurrence 
Theorem. 

Now we state the definition of transitive Hamiltonian level, which is weaker than the 



definition of topologically mixing Hamiltonian level (Definition 1.1). 



Definition 2.1. A Hamiltonian vector field Xh, restricted to a energy hypersurface Sn.e, 
is transitive if, for any open and non-empty suhsets U and V of SH,e, there is t ^M. such 
that X'^{U) nV y^ ^. A regular Hamiltonian level {H, e) is transitive if the Hamiltonian 
vector field X H restricted to any energy hypersurface of H^^{{e}) is transitive. 

2.2. Transversal linear Poincare flow and hyperbolicity. Let us begin with the 
definition of the transversal linear Poincare How. After, we state some results using this 
linear fiow. 



M. BESSA, C. FERREIRA, AND J. ROCHA 



Consider a Hamiltonian vector field Xh and a regular point x in M and let e = H{x). 
Define A4 := N^ fl TxH~^{{e}), where T^H'^ {{e}) = Ker dH{x) is the tangent space to 
the energy level set. Tlius, A4 is a {dim{M) — 2) -dimensional bundle. 



$^(x) : A4 -^ ^xIj{x) 



Definition 2.2. The transversal linear Poincare flow associated to H is given by 

where Ilj^t (^) : Txt (a;) M — )► A/^* [x) denotes the canonical orthogonal projection. 

Observe tliat A4 is $^(a;)-invariant. 

It is well-known (see e.g. [3]) that, given a regular point x G SH,e, tlien ^//(a;) is a linear 
symplectomorphisni for the symplectic form U£^ ^ , that is, 

U£j^^^{u,v) = uj£j^^{^Hi^) u,^Hi^) v) for any u, v e A^. 

We recall that the set of symplectomorphisnis fornis a group under composition, denoted 
by Sp{M,u), called symplectic group. 

For any symplectomorphisni, in particular for $^(a;), we have the following result. 

Theorem 2.1. (Symplectic eigenvalue theorem, [3]j Let f G Sp{M,uj), p E M and o an 
eigenvalue of Dfp of multiplicity k. Then l/a is an eigenvalue of Dfp of multiplicity k. 
Moreover, the multiplicity of the eigenvalues +1 and —1, if they occur, is even. 

The proof of the following result can be found in [131 Section 2.3]. 

Lemma 2.2. Take a Hamiltonian H G C^(M, M) and let A be a X\j-invariant, regular 
and compact subset of M . Then A is uniformly hyperbolic for X\j if and only if the 
induced transversal linear Poincare flow $^ is uniformly hyperbolic on A. 

So, we can define a unifornily hyperbolic set as follows. 

Definition 2.3. Let H G C'^{M,M.). An X\j-invariant, compact and regular set A C M 
is uniformly hyperbolic if M\ admits a (^^^-invariant splitting M^^bM^I such that there 
is i > O satisfying 

\\^%{x)\ms\\ < - and \\^'J{X\x))\^u^^^J\ < -, for any x e A. 

We remark that the constant | can be replaced by any constant 9 G (0, 1). 

2.3. Homoclinic classes, resonance relations and pseudo-orbits. Given a hyper- 
bolic closed orbit of saddle-type 7 of a Hamiltonian H, with period tt, and p G 7. We 
define the stable and unstable manifolds of 7 by 



W^r(7) = U ^niW^'^iP))- 



0<t<TT 



GENERIC HAMILTONIAN DYNAMICS 



The homoclinic class of 7 is defined by 



^7,H = W^i^) rh iy«(7), 

where S stands for the closure of the set 5* and rh denotes the transversal intersection of 
manifolds. 

It is well-known that a non-empty homoclinic class is invariant by the flow, has a dense 
orbit, contains a dense set of closed orbits and is transitive. Moreover, the hyperbolic 
closed orbits of some index are dense in the homoclinic class. 

Consider H G C^(M, M) and recall that dim{M) = 2d. Let {ci, ...,a2d} denote the 
eigenvalues of DXh{p), if p G Sing{H), or of DXfj{q), if g G Per{H) has period vr. A 
resonance relation between {cti, ..., a2d} is an equality of the type 



2d 



n 



for some i G {l,...,2d} and ki,...,k2d natural numbers such that either ki 7^ 1, or else 
there exists j j^ i such that kj ^ 0. 

Since ^'^{q) is a symplectomorphism, the foUowing trivial resonance relations are sat- 

isfied: 

d 

<^i = 0"i]J_(o"fcCrd+A;)"N 
fc=l 

for naturals a^. A resonance relation different from these ones is called a non-trivial 
resonance relation. Robinson proved in [25] that, C^-generically, there are not non-trivial 
resonance relations. 

Theorem 2.3. p5l Theorem 1] There is a residual TZ in C'^{M,M) such that, for any 
H ElZ, any p G Sing{H) and any q G Per (H) with period n, the eigenvalues of DXh{p) 
and of DX]j{q) do not satisfy non-trivial resonance relations. 

We observe that, if we fix H in the previous residual set 7^, sometimes we say that 
Sing{H) and Per [H) do not satisfy non-trivial resonances. 
Now, we state the definition of pseudo-orbit for Hamiltonians. 

Definition 2.4. Consider a Hamiltonian system {H,e,SH,e) o-nd t > Q. A seguence 
{xi}7=o ^'^ ^H,e, with n &N, is an e-pseudo-orbit on £H,e if d{X}j{xi),Xi+i) < e, for any 
i G {O, ...,n — 1}, where d{-, ■) denotes the distance inherited by the Riemannian structure. 

The length of the pseudo-orbit is equal to n. 

Remark 2.1. For divergence-free vector fields, and so for Hamiltonian vector fields, we 
have that Q{H\£^J = £H,e- Therefore, any x,y E SH,e are connected by an e-pseudo-orbit, 
for any e > 0. 



8 M. BESSA, C. FERREIRA, AND J. ROCHA 

2.4. Lift axiom. Fix p G Per{H) and a sinall neighborhood Up of p. By the Darboux 
Theorein (see, for exaniple, [221 Theorein 1.18]), there is a smooth syinplectic change of 
coordinates ipp : Up ^ TpM, such that (pp{p) = 0. Denote by Np^s the ball centered in 
O at the normal fiber at p and with radius 6. For a given 6 > O depending on p we let 
f H '■ 'fp^{Np,s) -^ Vx^i ( )(^xi (p)) be the canonical Poincare time-one arrival associated 

toHE 

In [21], when proving the closing lemma for Hamiltonians, Pugh and Robinson show 
that the lift axioni is satisfied for Hamiltonians, and they obtain the closing from the 
lifting. In rough terms, lifting is a way of pushing the orbit along a given direction by 
a small Hamiltonian perturbation C^-close to the identity. We point out that we never 
have to push in the direction of increasing energies. 

Furthermore, we recall the key point on the using of the C^ topology of the Hamiltonian 
vector field: "...one can lift points p in prescribed directions v with results proportional 
to the support radius" ([211 pp. 266]). 

Lift Axiom for Hamiltonians. (cf. [211 §9 i^)]) Consider a Hamiltonian H G 
C^(M, M) and let U he a. C^-neighborhood of H. Then there are O < e < 1 and a 
continuous function S: M \ Sing^Ku) —j- (0, 1), both depending on H and on W, such 
that, for any p and v G NpS{p) fl ipp{H^^{H{p))), there exists H eU satisfying: 

• supp{Xfj — Xh) is contained in the flowbox T = Ute(oT) ^^(-^IblKp))' where 
B\\^\\{p) is taken in a transversal section of p and T = T{y) is such that T{p) = 1 

and X^J-^\y) G B\\^\\{Xh{p)), for any y G B\\^\\{p); 

• If several such perturbations are made in disjoint fiowboxes, then their union- 
perturbation is also realizable by a Hamiltonian. 

2.5. Perturbation flowboxes. Consider the Standard cube M^'^, tilled by smaller cubes 
by homotheties and translations. Given a symplectic chart </?:[/—)■ M^'^, for U C £H,ei 
the (/3-pre-image of any tilled cube in '{){U) is called a tiled cube of the chart {U, ip) and it 

m 

is denoted by C. Note that C = M 71, with m E N, where each Tk is called a tile ofC. 

fc=i 
Definition 2.5. Consider a Hamiltonian system {H,e,SH,e), o tiled cube of a chart C = 

m 

M Tk and a constant T > 0. We say that the pseudo-orbit {xj}"^o ^'^ £H,e, with n G N, 

fc=i 

preserves the tiling in the injective flowbox 



THiC,T)= U XJ,iC) 

t6[0,T] 



In fact, given a regular point p, we can chose any r > O less than its period, if p is periodic. 



GENERIC HAMILTONIAN DYNAMICS 




n 


"^ 




1 


1 


iEX 





J^ZL. 


1 




- 


1 




1 






: ■" 










~ 


;- 








-: 


- 






- 


; — 










- 


- 






1 




- 


1 


1 1 


II III 


1 1 


1 





^/.- 



FiGURE 1. Representation of a tiled cube of the chart {U,ip). 



a) Xo,Xn^J^H{C,T); 

b) for any i G {1, ...,n — 1}, 

• ifxi&Tk then Xjj^{xi+i) E %, for some k G {l,...,m}; 

• if Xi G Xjj{C) then Xi^i = Xl[{xi), for some j G {1, ...,T — 1}. 




FiGURE 2. Representation of a pseudo-orbit preserving the tiling. 

This definition asserts that the intersection of the pseudo-orbit {xj}"^q with the 
flowbox Fh{C,T) is an union of segments {a;^, ..., Xj+t} such that Xj G C and 



X 



j+fc = ^niUj): for every k G {1,...,T}, where i/j is a point in the same tile of Xj. 
Observe that if a pseudo-orbit preserves the tihng then we just have to take care about 
the jumps of the pseudo-orbit outside Uiefi t-ii -^ni^)- 

As Pugh and Robinson explained in pTi, §9 (a)], local perturbations on H do not change 
the energy hypersurfaces in the bottom and top of the flowboxes where the perturbations 
take place. So, we are allowed to push along the energy levels. This property motivates 
the foUowing definition of perturbation flowbox. 

Definition 2.6. Fix a Hamiltonian system {H,e,SH,e), e > O and an e-C'^-neighborhood 
U of H. A tiled cube C is an e-perturbation flowbox of length T for {H, U) if, for any 



10 



M. BESSA, C. FERREIRA, AND J. ROCHA 



pseudo-orbit {xj}"^q on Sn^e preserving the tiling in J^h{C,T), there is H ElA, such that 
H = H outside J^h{C,T — 1), and a pseudo-orbit {yj}Y=o on £^ ^, with m G N, such that: 

• y o = xo and y m = x„; 

• H{yj) = e, for any j G {O, ...,m}; 

• the intersection of the pseudo-orbit {yj}^^Q with J^h{C, T) is an union of segments 
{yi,...,yi+T} such that yi e C and yi+k = ^'^iVi), for every k G {1,...,T}. More- 

over, the segments of {|/j}^q that do not intersect M X'^{C) are segments 

te[i,T-i] 
of the initial pseudo-orbit {xi}^^Q, where the starting point belongs to Xjj{C) or 
coincides with xq and the ending point belongs to C or coincides with x„. 




FiGURE 3. Perturbation in a tiled cube. 

The set supp{C) = M XJj{C) is called the support of the perturbation fiowboxC. 
te[o,T] 

The Hayashi Connecting Lemina is a key ingredient to prove the Connecting Lemma 
for pseudo-orbits of Hamihonians (Lemma [T]) and, as stated in f2E\ , it can be adapted 
for Hamiltonians. From Definiton 2^, we can extract a shghtly stronger statement of the 
Connecting Lemma for Hamiltonians in [281 Theorem E] , which can be seen as a theorem 
of existence of perturbation flowboxes. 

Theorem 2.4. Given a Hamiltonian system {H,e,£H,e) O'^d e > O, there exists T > O 
such that if any tiled cube C on Sn^e is a fiowbox of length T then C is an e-perturbation 
fiowbox o f length T . 

From the previous definitions and theorem, the following proposition foUows inimedi- 

ately. 

Proposition 2.5. Consider a Hamiltonian system {H,e,SH,e) o^nd letlA be a C'^-neigh- 
borhood of H . For any pseudo-orbit {a;j}^^Q on SH,e preserving the tiling in a fiowbox, 
there exist H E U and t > O, such that H{xq) = e and X*-{xq) = Xn on Ef^ ^. 

In fact, flowbox after flowbox, the Connecting Lemma for pseudo-orbits of Hamiltonians 
(Lemma [I]) erases all the jumps of the pseudo-orbit. However, notice that the jumps of a 



GENERIC HAMILTONIAN DYNAMICS 



11 



pseudo-orbit have no reason to respect the tiling of soine perturbation flowbox. To deal 
with this difficulty, we introduce the concept of covering families and of avoidable closed 
orbits. 

2.6. Covering families. Given a Hamiltonian system {H,e,SH,e), we want to cover the 
orbits on Sn^e by a family of perturbation fiowboxes, with pairwise disjoint supports. Let 
W be a C^-neighborhood of H and let C denote a faniily of perturbation fiowboxes for 
{H, U), with pairwise disjoint supports, and V denote a family of non-empty open subsets 
of £H,e with pairwise disjoint supports. 

m 

Definition 2.7. The family C = \\Tk, for m gN, is a covering family o f S h, e if, for 

k=l 

any x G £H,e, there exist t > O and 1 < k < m such that X\j{x) G int{Tk)- 




FiGURE 4. Representation of a covering family of £H,e- 

In general, if SH,e contains closed orbits with small period then £H,e has not a covering 
family. In fact, this kind of closed orbits is disjoint from the perturbation flowboxes. This 
motivates the definition of covering families outside V = VSj^^Vj. The sets V j (1 < j < t) 
are, in fact, neighborhoods of these closed orbits with small period. 

The following definition is an adaption of [Hl Definition 3.2] for Hamiltonians. 

Definition 2.8. Fix a Hamiltonian system {H,e,SH,e)! e > O and an e-C'^-neighborhood 
U of H. A perturbation flowhox C for {H, U) is a covering family of Sn.e outside V 
if there are 

• t > O and e > 0; 

• an open set W j and a compact set Fj, such that F j C W j C Vj, for every j G 
{l,...,r}; 



finite family of compacts T) = \\ Di on £H,e, such that every Di is contained 



m 



i=l 



the interior of a tile of C; 

two parts Vaj and V o j of V such that the support of the tiles of C containing this 

compacts is contained in Vj, for any j G {1, ...,r}, 



12 



M. BESSA, C. FERREIRA, AND J. ROCHA 



with length greater or egual than t 



such that 

a) any segment of any e-pseudo-orbit on Sh,( 
meets a compact Fj or a compact of 'D; 

b) any segment of any e-pseudo-orbit on Sn^e starting outside Vj and ending inside 
W j meets a compact ofVaj, for any j G {1, ..., r}; 

c) any segment of any e-pseudo-orbit on SH,e starting inside W j and ending outside 
Vj meets a compact ofVoj, for any j G {1, ...,r}; 

d) for any j G {1, ...,r} and for any compact sets Da C Va,j and Do C ^oj, there 
exists a pseudo-orbit with jumps inside the tiles ofC, with starting point inside Da 
and ending point inside Do- 





FiGURE 5. Covering family of £H,e outside V. 

Roughly speaking, C is a covering family of En^e outside V if any pseudo-orbit returns 
regularly to a compact V C int{Tk), for some 1 < k < m, during the time it passes out 
of V. If the pseudo-orbit takes a long time to return to another compact set D d 'D, 
it approaches some compacts F j C Vj. For this, the pseudo-orbit must go through an 
entrance compact Da C D and then through an exit compact D^ C D. Moreover, we can 
even switch the segment of the pseudo-orbit between Da and D^ by a pseudo-orbit with 
jumps inside the tiles of C. 

2.7. Avoidable closed orbits. Consider a Hamiltonian system {H,e,SH,e) and a cfosed 
orbit 7 of iJ on Sn^e- Let U he a C^-neighborhood of H and fix T > O and p G 7. The 
next definition is adapted from [Hl Definition 3.10] for Hamiltonians. 

Definition 2.9. A closed orbit 7 is avoidable for (U, T) if, for any neighborhood Vo 
of 7 and for any t > O, there exist e > O, open neighborhoods W and V of 7, such that 
W G V G Vq, and a perturbation fiowbox C for (H, U) of length T with disjoint supports, 
such that: 

a) the support of C is contained in V; 

b) there exist two families of compacts 'Da and 'Do contained in the interior of the 
tiles of C such that 



GENERIC HAMILTONIAN DYNAMICS 



13 



• any segment of any e-pseudo- orbit on SH,e starting outside V and ending 
inside W has a point in a compact ofVa; 

• any segment of any e-pseudo-orhit on Sn^e starting inside W and ending out- 
side V has a point in a compact ofD^; 

c) for any compacts Da G Va and Do G Vo, there exist a pseudo-orhit on £H,e, with 
jumps inside the tiles of C, starting in D a and ending in Do- 

d) for any x in C, the time taking by Xjj{x) to return to supp{C) is bigger than t. 



/^.-.Ti 




^ \ v --'T / ,' / ' 






\Wk 



VoV, 







-'' 



FiGURE 6. Representation of an avoidable closed orbit 7. 

Therefore, a closed orbit 7 is avoidable for (U,T), for fixed T > O, if, for any t > O, 
there exists a family C of perturbation flowboxes for {H, U) of length T such that, given 
a pseudo-orbit with starting and ending points far from 7, but passing very close of 7, we 
can exchange the segments of the pseudo-orbit passing close of 7 by segments of another 
pseudo-orbit with jumps inside the tiles Tk {I < k < m). 

A closed orbit can be even characterized as uniformly avoidable. 

Definition 2.10. Let {H,e,SH,e) be a Hamiltonian system and U a C^ -neighborhood of 
H . The closed orbits of H on SH,e ore called uniformly avoidable if they are isolated and 
there exists a constant T > O such that any closed orbit of H on She is avoidable for 
(K,T). 

This kind of orbits is used to derive perturbation flowboxes with disjoint supports, in 
such a way that the pseudo-orbits stay away from closed orbits with small period. We 
anticipate that, if SH,e has no orbits with small period and has all the closed orbits uni- 
formly avoidable then we will be able to build a covering family of perturbation flowboxes 
for SH,e, as shown in Proposition 3.4, in SectionjSJ 

2.8. Perturbation results in the C^-topology. In this section, we state two pertur- 
bation lemmas for the Hamiltonian setting, namely the Closing Leninia and the Pasting 
Leninia. 

The first perturbation result is a version of the Closing Lemnia for Hamiltonians that we 
obtain by combining Arnaud's Closing Leninia (see [^) with Pugh and Robinson's Closing 



14 



M. BESSA, C. FERREIRA, AND J. ROCHA 



Lemma for Hamiltonians (see [21] )• It states that the orbit of a non-wandering point can 
be approximated, for a very long time, by a closed orbit of a nearby Hamiltonian. 

Lemma 2.6. (Closing Lemma for Ham,iltonians) Fix Hi G C^(M, M). Let x E M he a 
non-wandering point and e, r and r positive constants. Then, there exist H2 G C^(M, M), 
a closed orbit 7 of H2 with period vr, p G 7 and a map g : [O, r] — )■ [0,7r], close to the 
identity, such that: 

• H2 is e-C^ -close to Hi; 

• d(^XJ,^{x),X'£\p)'^ <r,0<t<T; 

• H2 = H^ on M\A where A = Uo<t<r (^r(X*,^(p))) ■ 

The next lemma is a version of the C^-Pasting Lemma (|61 Theorem 3.1]) for Hamilto- 
nians. Actually, in the Hamiltonian setting, the proof of this result is much more simple. 

Lemma 2.7. (Pasting Lemma for Hamiltonians) Fix Hi G C""(M, M), 2 < r < cxd, and 
let K be a compact subset of M and U a small neighborhood of K . Given e > O, there 
exists (5 > O such that if H2 G C^(M,M), /or 2 < s < 00, is 5 -C'^'''^''^'^ -close to Hi on U 
then there exist H^ G C*(M, M) and a closed set V such that: 

• K dV dU; 

• if 3 = if 2 on V; 

• ifs = ifi on U"; 

• H3 is e-C"'"''^''^'^ -close to ffi. 




FiGURE 7. Perturbation given by the Pasting Lemma for Hamiltonians. 



Proof. Consider {Ui, f/2} an open cover of M, such that Ui := U and U2 does not contain 
K. Then, there is a smooth partition of unity {ai,a2}, subordinate to {Ui,U2}, such 
that aj : M — )■ [0, 1] satisfies supp(ai) C Ui, for i = 1,2, and ai{x) + 0:2(3;) = 1, for any 
X e M. 

Letting V := [/| and H3 := aiH2 + (1 — ai)ifi, we have that: 

- K cV CU; 



GENERIC HAMILTONIAN DYNAMICS 15 



- H^ = H2 on V , since ai{x) = 1 and a2{x) = O, for any x & V; 

- H3 = Hi on W^, since ai{x) = O and a2{x) = 1, for any x G W^; 

~ 11-^3 — Hl\\^rnin{r,s} ^ niaX|Qfl (O;) j \\H2 — Hl\\^rnin{r,s} = 11-^2 — H l\\ (-,min{r,s} < O, 

since, by hypothesis, H2 and ifi are S-C'^'''^^^'^^-close. So, for (5 > O sufiiciently sniall, we 
are done. D 

3. Proof of Theorem [1] 

3.1. Proof of the Connecting Lemma for pseudo-orbits. This section contains the 
proof of the Connecting Lemma for pseudo-orbits of Hamiltonians. Let (M, u) denote 
a closed, symplectic 2(i-manifold, for d > 2. Take H G C^(M, M) and a regular energy 
e G H (M), such that the eigenvalues of any closed orbit of H do not satisfy non-trivial 
resonances. Then, for any C^-neighborhood U of H, for any energy hypersurface SH,e C 
H~^{{e}) and for any x,y E SH,e connected by an e-pseudo-orbit, for e > O, there exist 
H E U and t > O such that e = H{x) and Xg{x) = y, on the analytic continuation Sg^^ 

of SH,e- 




perturbation 




FiGURE 8. Perturbation given by the Connecting Lemma for pseudo-orbits. 

As explained in [Hl [17] and in [lOj, the proof of the Connecting Lemma for pseudo- 
orbits is sphtted in three main parts. The first step to prove Lemma [T] concerns on local 
perturbations. These perturbations motivate the definition of perturbation boxes whose 
support must be in the interior of small open sets, pairwise disjoint till a sufiiciently large 
number of iterates. Separately, we need to analyze the dynamics near closed orbits with 
small period because these orbits are not contained in any perturbation box. Finally, we 
must analyze the global dynamics, in order to cover any orbit with perturbation flowboxes. 

This strategy was firstly foUowed by Bonatti and Crovisier for diffeomorphisms (see 
[T7]). Later, jointly with Arnaud (see [8]), these authors proceeded with this methodology 



16 M. BESSA, C. FERREIRA, AND J. ROCHA 



to get the proof of the Connecting Lemma for pseudo-orbits of symplectomorphisins. The 
main novelties in the syinplectoinorphisms context are the need for the perturbations to 
be symplectic and also that the closed orbits can be stably elliptic. This means that the 
symplectomorphisms case cannot be reduced to the one treated in [TTj, where the closed 
orbits are assumed to be hyperbohc. That is why, in p] , the authors prove this result for 
symplectomorphisms, by doing the necessary changes. 

For the Hamiltonian case, recall that the transversal linear Poincare fiow is, is fact, a 
symplectomorphism and observe that we are assuming the absence of singularities on the 
energy hypersurfaces. Keeping in mind the strategy described in [8], the novelties in the 
proof of the Connecting Lemma for pseudo-orbits of Hamiltonian are the statement of 
adequate definitions and, since the energy hypersurfaces are invariant by the Hamilton- 
ian fiow, the need for the pseudo-orbit being completely contained in the same energy 
hypersurface. Hence, we have to ensure the creation of symplectic perturbations without 
leaving the initial energy hypersurface. Recall that the energy hypersurface is indexed to 
the Hamiltonian. Thus, it may change when we perturb the Hamiltonian. That is why, 
in the statement of Lemma [T| we want the energy of the points in the pseudo-orbit to be 
kept constant, even if we C^-perturb the Hamiltonian. However, since we are allowed to 
push along the energy levels (see [211 §9(a)]), the arguments stated in [S] can be adapted 
to the Hamiltonian case. At the end, we have a version of the Connecting Lemma for 
pseudo-orbits of Hamiltonians, where the condition on the persistence of the energy of 
the pseudo-orbit is trivially satisfied. Let us briefiy explain how to prove Lemma [ij 

Arnaud et al. proved, in [8, Proposition 4.2], that if the eigenvalues of any closed orbit 
of a symplectomorphism do not satisfy non-trivial resonance relations, then the closed 
orbits are uniformly avoidable. Therefore, since the transversal linear Poincare fiow is a 
symplectomorphism, the foUowing proposition follows directly for Hamiltonians. 

Proposition 3.1. Consider a Hamiltonian H G C^{M,M.). If, for any closed orbit p of 
H with period tt, the eigenvalues of ^'^{p) do not satisfy non-trivial resonances then the 
closed orbits of H are uniformly avoidable. 

As explained before, to prove this proposition, the authors take into account that the 
closed orbits can be hyperbohc (case analyzed in [17]) but also completely elliptic or 
elliptic. 

Observe that, by the previous proposition, Theorem 2.3| implies that the closed orbits 
of a C^-generic Hamiltonian are uniformly avoidable. 



Now, by Proposition 3.1, to prove the Connecting Lemma for pseudo-orbits of Hamil- 



tonians it is enough to show the foUowing result. 

Theorem 3.2. Consider a Hamiltonian system {H,e,£H,e) such that the closed orbits of 
H on £H,e cire uniformly avoidable. Then, for any C"^ -neighborhood U of H and for any 
x,y E £H,e, there is H eU and t > O, such that H{x) = e and X^-{x) = y, on the analytic 
continuation Sg ^ of £H,e- 



GENERIC HAMILTONIAN DYNAMICS 17 



It is obvious that Theorem 3.2 follows immediately if y G Oh{x). In fact, to prove 



LemmafTl it is enough to show Theorem 3.2 for some kind of points x, y G Sh 



Lemma 3.3. (]H1 Leinma 3.12]^ Consider a Hamiltonian system {H,e,SH,e) such that the 
closed orbits on SH,e o,fS- isolated. Toke any x,y & SH,e such that y ^ Oh{x). Then, there 
exist X and y, arbitrarily close to x and y, such that either y G Oh{x), or else x and y 
are not closed orbits. 

Recall that a uniformly avoidable closed orbit is indeed isolated. So, by the previous 
lemma, the proof of Lemma [T] is reduced to the proof of Theorem 3^, when x, y are not 



closed orbits. In fact, ii y ^ OHix) and x or y are closed orbits, we just have to apply 
Theorem |3. 2 to x and y, given by Lemma 3^ Then, a Hamiltonian perturbation of the 



identity sends x, y into x, y, and it allows us to conclude the result for any x and y in 

Recall that H satisfies the lift axioni and that any two distinct points x, y G SH,e 
are connected by an e-pseudo-orbit, for any e > 0. Therefore, by Lemma |3.3[ we can 



reduce the proof of Theorem |3. 2 and so of the Connecting Lemma for pseudo-orbits of 



Hamiltonians, to the proof of Proposition 3.4 and Proposition 3.5 bellow. 



Proposition 3.4. Take a Hamiltonian system {H, e, £H,e), such that H satisfies the lift ax- 
iom and any closed orbit of H on SH,e is uniformly avoidable. LetUo be a C"^ -neighborhood 
of H and x,y & SH,e be such that x,y ^ Per{H) and y ^ Oh{x). Then there exist a 
neighborhood IA <ZUq of H, a family of disjoint open sets V and a family of perturbation 
flowboxes C for {H, U) with disjoint supports, both V and C not containing x nor y, such 
that C is covering Eu^e outside V. 

In this case, we want to build a family of perturbation flowboxes in a neighborhood of 
closed orbits. Let us sketch the proof of this proposition, adapting the ideas of the proof 
in [Hl Proposition 3.13]. 

We want to construct finitely many disjoint perturbation flowboxes, whose union meets 
every orbit of SH,e, called topological tower of order T. Clearly, the existence of closed 
orbits with small period, even in a finite number, goes against the existence of a topological 
tower. However, if we construct a perturbation flowbox C, covering £H,e outside a finite 
family of disjoint open sets V = U^^^^V^, we can include any closed orbit with small period 
in the interior of some Vi. In this case, we have a finite family of disjoint perturbation 
flowboxes C far from closed orbits with small period. Now, it remains to show how can 
we build these disjoint perturbation flowboxes with length T. 

Remark 3.1. We state the definition of a flow, built under a ceiling function h. Consider 
a measure space S, a map i? : S —t- S, a measure fi in H and an integrable function 



18 M. BESSA, C. FERREIRA, AND J. ROCHA 

/i : S — )■ [c, +00], with C > O and J^ h{x)djl{x) = 1. The flow 

5':SxM — >SxM 



X, r h^ 



k-l X 

R\x),r + S -J2^(^'(^)n 

i=0 ^ 



fc-1 k 

where k G Z is uniguely defined by y^ h{R\x)) < r + s < 2^ h{R\x)), is called a special 

i=0 i=0 

How. In fact, the flow S^ moves the point (x,r) to {x,r + s) at velocity one, until it hits 
the graph of h. After this, the point returns to S and continues its journey. 

The Ambrose-Kakutani Theorem states that a flow having the set of critical points with 
zero Lebesgue measure is isomorphic to a special flow (see [4J/ 

Recall that any closed orbit of H on the regular energy hypersurface £H,e is uniformly 
avoidable, and so isolated. Then, H has a finite number of closed orbits with small period. 
Therefore, by Ainbrose-Kakutani's Theorem in |1], X\j is equivalent to a special flow. 
Now, foUowing |Tll Section 3.6.1], with the obvious changes, we can build a topological 
tower with very high towers in order to have enough time to perform a lot of small 
non-overlapped perturbations. 

The next proposition, jointly with Proposition 3.4, finishes the proof of LemmajT} 



Proposition 3.5. Consider a Hamiltonian system {H,e,£H,e) and a neighborhood U of 
H. Let C denote a family of periurbation flowboxes for {H, U) covering £H,e outside a 
family of open sets V. Take any x,y E £H,e outside the support of C and outside of any 
y e V. Then there exist H E U and t > O, such that H{x) = e and Xg{x) = y, on the 
analytic continuation Sfj ^ of Sn^e- 



By Proposition 2.5, if the hypothesis of the previous proposition ensure that a pseudo- 



orbit connecting x and y preserves the tiling of C, then we are done. In fact, as explained 



in Section 2.6, given that the perturbation flowbox C covers £H,e outside V, every orbit 
on SH,e spends a uniformly bounded time to return to the interior of any tile of C. It 
is straightforward to see that the same holds for any e-pseudo-orbit, with small e > 0. 
Moreover, if we choose e > O even smaller, we can modify the pseudo-orbit in such a way 
that, whenever the pseudo-orbit returns to the interior of some tile, we add at this time 
all the next jumps of the pseudo-orbit until the next return to a tile, defining, in this 
way, a new jump. The final jump respects the tile and is small, because the number of 
grouped jumps is uniformly bounded. In this way, we construct a pseudo-orbit preserving 
the tiling of C. 

3.2. Auxiliary lemmas. In this section, we state the proof of some auxiliary results for 
Hamiltonian systems defined on a 2(i-dimensional symplectic manifold, for d > 2. The 



first one (Lemma 3.6) asserts that, C -generically, the quotient between the period of two 



GENERIC HAMILTONIAN DYNAMICS 19 



distinct closed orbits of a Hamiltonian is irrational. After, in Leinina 3.7, we state that 



^5 



given a C^-generic Hamiltonian H, there exists an open and dense set in H{M) such that 
every energy taken in such a set is regular. Afterwards, we show that, given a C^-generic 
Hamiltonian H, there exists an open and dense set in H{M) such that, for every energy 



e taken in such a set, the Hamiltonian level (if, e) is transitive (Lemma 3.8) 



Lemma 3.6. There is a residual TZ in C^(M, M) such that, for any H eTZ, any distinct 

p,q E Per (H), with periods Tip and TCg, satisfy — G ]R\Q. 

ng 

Proof. Fix n E N. By Robinson's results [25], the following set 

An '■= {h g C^(M, m) : Sing{H) is hyperbolic and Pern{H) is hyperbolic or elliptic} 

is open and dense in C^(M, M). Also, define the open set 

Bn:= \h eAn-. if p, g G Pern{H) andp^g then ^ ^ {ri}r=i 

where {^j}^^ denote the positive rational numbers, with a fixed order. 

Now, this proof follows the ideas stated in the proof of [ini Lemma 2.2], but using the 
version of the Pasting Lemma for Hamiltonians, proved in Lemma 2.7[ 



Fix e > O and Hi G C^(M,]R). By density of An, there is H2 G An, e-C^-close to Hi. 



Recall that, by Proposition |3.1[ the closed orbits with period less or equal than n of H2 
are uniformly hyperbolic, and so isolated. So, Pern{H2) has a finite number of elements, 
say {pi}^i, for fixed m G N. 

Given a positive sequence {sj}™ ^, the vector field Xjj, = ^t^^j^/^j is also a Hamiltonian 
vector field, for any 1 < i < m. Actually, by (fll), Xj[, is associated to the Hamiltonian 



^q-j-iJ2- Observe that if we choose Sj arbitrarily close to O then H^ is e-C^-close to //2- 

For any 1 < i < m, consider tubular compact neighborhoods Ki oi Pi, sufiiciently small 
such that sonie open neighborhoods Wj of Ki are pairwise disjoint. The idea now is to 



apply, recursively, Lemma 2.7, in order to define Hm G C^(M, M) such that: 

• Hm converges to H2 in the C^-sense, as s j converges to 0; 

• '^Hrn,P. = (1 + ^i)'^H2,pv for 1 < i < m. 

By a good small choice of the sequence {sj}™^, we have that Hm G An and that 
^^^ ^ {rj^^i, for i ^ J. Thus, H^ G S„. 

Since Bn is open and dense in C'^{M,M.), for any n E N, the desired residual subset of 
C^{M, R) is given by 7^ := n^eN-Bn. D 

The following result is an immediate consequence of the fact that Morse functions are 
C^-open and dense among C^(M, M) ([21]). 



20 M. BESSA, C. FERREIRA, AND J. ROCHA 

Lemma 3.7. There is a C'^-open and dense subset O in C^(M, M) such that, for any 
H & O there exists an open and dense setS{H) of energies such that any energy e G S{H) 
the Hamiltonian level {H, e) is regular. 

Lemma 3.8. There is a residual set TZ in C'^[M,W) such that, for any H E 71, there is 
an open and dense set S (H) in H (M) such that, for every e e S (H), 

• H~^{{e}) is regular; 

• the closed orbits of H in H^^{{e}) do not satisfy non-trivial resonances; 

• the Hamiltonian level {H, e) is transitive. 



Proof. Let 1Zq be the resid ual set given by Theorem 2.3 and consider O and S{H), for 



H G O, as in Lemma 3.7 Observe that, if e G S{H) then H^^{{e}) = U^^^SH,e,i- In 
this case, let {Un}^ be a countable basis of open sets on M. Fix 1 < i < le and define 
U^ = Un ri SH,e,i, whenever non-empty. So, {U^}^ is a countable basis of open sets on 
SH,e,i- We say that H G Vn,m,i,e if 

Now, we define the residual set 

7^ := 7^o n O n pi {Pn,m,i,e u ir~~eT) , 

n,m 

where, given a set S, S stands for its closure and S'^ for its complementary. 

Fix H E TZ, e E S{H) and 1 < i < le- Thus, H~^{{e}) is regular and any closed orbit 
of H in SH,e,i do not satisfy non-trivial resonances. Moreover, for all integers n and m, 
we have that H G Vn,m,i,e or H E (Vn,m,i,eY- Observe that if if G Vn,m,i,e, for all integers 
n and m and any I < i < h, then {H, e) is transitive. 

So, by contradiction, assume that there are some integers n and m and I < i < le such 

all points x,y E £H,e,i 



2.1 



that H E Vn,m,i,e " ■ Choosc X E U^ and y E U^. By Remark 
are connected by an e-pseudo-orbit, for any e > 0. Moreover, since H E TZo, we can 
apply the Connecting Lemma for pseudo-orbits of Hamiltonians (Lemma [I]). So, for any 
C^-neighborhood U of H, there exists H eU nUanOn V^„,j^^ " such that e = B.{x), 

where [/* and VL are elements of the basis of the well-defined analytic continuation Er, ~ 
of £H,e,i such that x E U^ and y E U^, and there is T > O such that Xg{x) = y on 
^Hel.- Then H E "P^^^e, which is a contradiction. Hence H E Vn,m,i,e, for all integers n 
and m and for any 1 < i < le- Therefore, {H, e) is transitive, for any H E TZ and any 
e G S{H). D 

3.3. Energy hypersurfaces as homoclinic classes. In this section, we want to prove 



the foUowing coroUary of Lemma 3.8 



GENERIC HAMILTONIAN DYNAMICS 21 



Corollary 3.9. \b There is a residual set IZ in C^(M, M) such that, for any H E TZ, 
there is an open and dense set S{H) in H{M) such that if e E <S{H) then any energy 
hypersurface of H~^{{e}) is a homoclinic class. 



Proof. Let 7^ and S{H), for H E 71, he as in Lemma 3.8 Recall that if e G S{H) then 
H~^{{e}) = U^^j^£H,e,i and, fixing 1 < i < Jg, we can define a countable basis of open sets 
{U^}n on the energy hypersurface SH,e,i- 

Take a C^-neighborhood U oi H such that the analytic continuation pfj of a hyperbohc 
closed orbit p^ of H is well-defined, for any H E U. So, for any integer n, define the open 
sets 

Wn:={HEU: W^J(p^) ^^ K ^ 0}. 
We want to show that W„ is a dense subset of U, for any n E N. First, observe that 
TZu = TZriU is a. dense subset of U such that, for any H E TZu, there is an open and 
dense set S{H) C H{M) such that any e E S{H) is regular and {H, e) is transitive. So, 
fixing n E 'N, for any H E TZu and any neighborhood V of a hyperbohc closed orbit pg 
there exist j,k> O satisfying X^(y) n f/^ 7^ and X^^{V) n f/^ ^ 0, where {f/^}„ is a 
countable basis of open sets on Sg ^j. By Hayashi's Connecting Lemma of Hamiltonians 

(see [28] ), there exists a Hamiltonian H, C^-close to H, such that H E Wn- Hence, >V„ 
is dense on U, for any n E N. Therefore, 

W := fl W„ = {^ G W : lyj(p^) = ^^,,,^} 

neN 

is a residual subset of U. 

Fix iJ G 7^ n W and e G S (H). Let {f/^}n be a countable basis of open sets on the 
energy hypersurface Sg ^^j of H~^{{e}). Fix n G N and a hyperbohc closed orbit pg of 

iJ. Observe that any non-periodic x G fZ* is an accumulation point of W^^{pg). By the 
Connecting Lemma for Hamiltonians (see [28]), we construct homoclinic intersections on 
f/* and, by a small C^-perturbation, we turn it transversal. So, the set 

Zn := {H eU r\TZ : p g has a homoclinic transversal intersection on [/^} 

is open and dense on U, for any n G N. Therefore, the set 

Z := fl Z„ = {iJ G W^7^ : H^^,^ = Sg^^;^ 

neN 

is residual in U. Observe that this is valid for any small C^-neighborhood U oi H E TZ. 
So, the set 

7^l := {H E C\M, M) n 7^ : H^^^h = Sn^eA 
is residual in C^(M, M), for any I < i < le- Thus, there is a residual set T^i in C^(M, M) 
such that, for any H E TZi, there is an open and dense set S {H) such that, for e G S {H), 
any energy hypersurface of i/~^({e}) is a homoclinic class. D 



22 M. BESSA, C. FERREIRA, AND J. ROCHA 

3.4. Generic topological mixing. In this section, we conclude the proof of TheoremjT} 

Theorem 3.10. \B There is a residual IZ in C^(M, M) such that, for any H E TZ, there 
is an open and dense set S{H) in H (M) such that, for every e G S{H), the Hamiltonian 
level {H, e) is topologically niixing. 



Proof. Let T^o be the residual set given by Lemma 3.6 T^i be the residual set given by 



Lemma 3.8 and 7^2 be the residual set given by CorollaryjTJ Define 

7^:=7^o^7^l^7^2. 

Now, we follow the ideas on the proof of [H, Theorem B], making the necessary adap- 
tations to the Hamiltonian setting. 



Fix H E TZ. Since H E TZi, by Lemma p3^, there is an open and dense set S{H) such 



that, for any e E S{H), the Hamiltonian level {H, e) is transitive. So, to conclude the 
proof of Theorem [II we just have to prove that, for any e E S (H), the Hamiltonian level 
{H, e) is topologically mixing. 

Fix e E S{H) and let SH,e,i be an energy hypersurface of if ~^({e}), for 1 < z < I^. Let 
us prove that SH,e,i is topologically mixing, that is, for any open, nonempty subsets U 
and V of £H,e,i, there is r e M such that Xjj{U) fl V^ 7^ 0, for any t > r. 

Given that H E IZ2 and e E S{H), by Corollary [II £H,e,i is a homoclinic class. Since 
hyperbolic closed orbits are dense in the homoclinic class and the index is constant and 
equal to d, we can find two different hyperbolic closed orbits 71 and 72 of H, with period 
Tip and Tig, where p G 71 and q E 72, such that m(i(7i) = ind{j2) = d and 71 fl [/ 7^ and 

72 n l^ 7^ 0. Moreover, since H E TZo, we have that — E . 



^g 



Fix a; G 7i n f/, y G 72 n y and z E W'''{x) fl W" (y). Thus, there is ri > O such that 
• lim X^("^+™"'')(z) = x. 

m— >+oo 

Then, there is ti > O such that X~^''^"'"''\z) E U and, therefore, z E X^+""^(f/), for 
every m E N. Similarly, there is ^2 > O and a small e > O such that Xj^ "'^' '^(2;) G V, for 
every n eN and |s| < e. 

TT 

Since — G M\Q, observe that the set {miTp + n7r„ + s : m,n E 1j,\s\ < e} contains an 

interval of the form [T, +00), for some T > 0. This follows from the transitivity of the 
future orbits of irrational rotations of the circle. Hence, for any t > ti+t2 + T, there are 
m, rz G N and \s\ < e such that t = ti+t2+rmTp+nTCq+s. Then, Xj^ "''' ^{z) E Xjj{U)nV , 
for any t > ti + ^2 + ^^ So, SH,e,i is a topologically mixing energy hypersurface, for any 
1 < i < le- Therefore, the Hamiltonian level {H, e) is topologically mixing. D 



GENERIC HAMILTONIAN DYNAMICS 23 



ACKNOWLEDGEMENTS 

The authors were partially supported by FCT - Fundagao para a Ciencia e a Tecnologia 
through CMUP: PTDC/MAT/099493/2008. CF was supported by FCT - Fundagao para 
a Ciencia e a Tecnologia SFRH/BD/33100/2007. 



[1 
[2' 
[3' 

[4; 

[6' 

[7; 

[9 
[10 

[11 

[12 

[13: 

[14 

[15 
[16 

[17; 

[18 
[19 
[20 

[21 



References 

F. Abdenur, A. Avila and J. Bochi, Rohust transitivity and topological mixing for C^-flows. Proc. 
Amer. Math. Soc, 132, 3 (2003) 699-705. 

F. Abdenur and S. Crovisier, Transitivity and topological mixing for C^ diffeomorphisms, Preprint 
ArXiv 2011. 

R. Abrahani and J. E. Marsden, Foundations of Mechanics. The Benjamin/Cunimings Publishing 

Company. Advanced Book Program, 2"'' edition (1980). 

W. Ambrose and S. Kakutani, Structure and continuity of measurable flows. Duke Math. J., 9 

(1942) 25-42. 

D. V. Anosov, Geodesic flows on closed Riemannian manifolds of negative curvature. Proc. Steklov 

Math. Inst., 90 (1967) 1-235. 

A. Arbieto and C. Matheus, A pasting lemma and sonie applications for conservative systems. 

Erg. Th. & Dyn. Syst., 27 (2007) 1399-1417. 

M.-C. Arnaud, Le "closing lemma" en topologie C^. Mcm. Soc. Math. Fr., Nouv. Scrie, 74 (1998) 

1-120. 

M.-C. Arnaud, C. Bonatti and S. Crovisier, Dynamigue sympletigues generigues. Erg. Th. & Dyn. 

Syst. 25, 5 (2005) 1401-1436. 

L. Arnold, Random Dynamical Systems, Springer-Verlag, New York, 1998. 

M. Bessa, A generic incompressible flow is topological mixing. Coniptcs Rendus Matheniatique, 

346 (2008) 1169-1174. 

M. Bessa, The Lyapunov exponents of generic zero divergence-free three- dimensional vector fields. 

Erg. Th. & Dyn. Syst., 27, 6 (2007) 1445-1472. 

M. Bessa, C. Ferreira and J. Rocha, On the stability of the set of hyperbolic closed orbits of a 

Hamiltonian. Math. Proc. Camb. Phil. Soc, 149, 2 (2010) 373-383. 

M. Bessa and J. Lopes Dias, Generic Dynamics of 4-Dimensional C^ Hamiltonian Systems. Com- 

mun. in Math. Phys., 281 (2008) 597-619. 

M. Bessa and J. Lopes Dias, Hamiltonian elliptic dynamics on symplectic 4^-manifolds. Proc. 

Amer. Math. Soc., 137 (2009) 585-592. 

G. BirkhoflF, CoUected Mathematical Papers 1-3 (New York, 1950). 

G. BirkhofF, Dynamical Systems. Amer. Math. Soc. (Providence, RI, 1927). 

C. Bonatti and S. Crovisier, Recurrence et genericite. Invent. Math. 158, 1 (2004) 33-104. 

C. Bonatti, L. Diaz and M. Viana, Dynamics beyond uniform hyperbolicity (Springer-Verlag, New 

York, 2005, 288-289). 

R. Bowen, Eguilibrium States and the Ergodic Theory of Anosov Diffeomorphisms. Lecture Notes 

in Math (Springer-Verlag, Berlin, 1975). 

L. Markus and K. R. Meyer, Generic Hamiltonian dynamical systems are neither integrable nor 

ergodic, Memoirs of the American Mathematical Society, 144. American Mathematical Society, 

Providence, R. L, 1974. 

Y. Matsumoto. An Introduction to Morse Theory. American Mathematical Society, 2002. 



24 M. BESSA, C. FERREIRA, AND J. ROCHA 



[22] J. Moscr and E. Zchndcr, Notes on dynamical systems. Courant Lecture Notes in Mathematics, 12. 

New York University, Courant Institutc of Mathematical Sciences. Amer. Math. Soc. (Providcnce, 

RI, 2005). 
[23] J. Oxtoby and S. Ulam, Measure-preserving homeomorphisms and metrical transitivity. Ann. of 

Math., 42(2):874-920, 1941. 
[24] C. Pugh and C. Robinson, The C^ closing lemma, including Hamiltonians. Ergod. Th & Dynam. 

Sys. 3 (1983) 261-313. 
[25] C. Robinson, Generic properties of conservative systems I and II. Amer. J. Math., 92 (1970) 

562-603. 
[26] M. Shub, Global stability of dynamical systems. Springer-Verlag, New York, 1987. 
[27] J. C. Yoccoz, Travaux de Herman sur les Tores invariants, Asterisque, 206 Exp: 754, 4 (1992), 

311-344 
[28] L. Wen and Z. Xia, C^ connecting lemmas. Trans. Amer. Math. Soc, 352 (2000) 5213-5230. 

Departamento de Matematica, Universidade da Beira Interior, Rua Marques d'Avila 
e bolama, 6201-001 covilha portugal. 
E-mail address: bessa@fc.up.pt 

Departamento de Matematica, Universidade do Porto, Rua do Campo Alegre, 687, 
4169-007 Porto, Portugal 

E-mail address: celiam@fc.up.pt 

Departamento de Matematica, Universidade do Porto, Rua do Campo Alegre, 687, 
4169-007 Porto, Portugal 

E-mail address: jrocha@fc.up.pt