Hölder foliations, revisited (Q428930)

In the classical book [\textit{M. W. Hirsch}, \textit{C. C. Pugh} and \textit{M. Shub}, Invariant manifolds. Berlin-Heidelberg-New York: Springer-Verlag. (1977; Zbl 0355.58009)], normally hyperbolic foliations are proved to be structurally stable: let \(\mathcal{F}\) be a regular foliation on \(M\) preserved by a \(C^1\) diffeomorphism \(f:M\to M\), with respect to which \(\mathcal{F}\) is normally hyperbolic and plaque-expansive, then \(\mathcal{F}\) is structurally stable. That is, if \(g\) is \(C^1\)-close to \(f\), then there is a unique \(g\)-invariant foliation \(\mathcal{F}_g\) close to \(\mathcal{F}\), such that \(\mathcal{F}_g\) is also normally hyperbolic and plaque-expansive with respect to \(g\). Moreover, \((g,\mathcal{F}_g)\) and \((f,\mathcal{F})\) are canonically leaf-conjugate. The paper under review continues the previous exploration, clarifies several subtle notations and proves that the leaf conjugacies given there are Hölder for the perturbations of a class of normally hyperbolic systems. Moreover, they estimated the Hölder exponents for systems with \(C^1\) foliations. They also emphasized that the question remains open for the regularity of the conjugacies of perturbations of general normally hyperbolic systems. Two main results are: Theorem A. Let \((\mathcal{F},f)\) be normally hyperbolic such that the two distributions \(E^{\text{cu}}\) and \(E^{\text{cs}}\) are \(C^1\) (in this case \(\mathcal{F}\) is \(C^1\) and \(f\) is dynamically coherent). Then the canonical leaf conjugacy, say \(h_g\), of a \(C^1\) perturbation \(g\) of \(f\) is bi-Hölder and the holonomy maps are Hölder along all \(g\)-invariant foliations. Theorem B. Let \(\mathcal{L}\) be a lamination on a compact subset \(\Lambda\) with uniformly compact leaves, and \(f:M\to M\) be normally hyperbolic on \(\mathcal{L}\), such that the two distributions \(E^{\text{cu}}\) and \(E^{\text{cs}}\) are dynamically coherent on \(\mathcal{L}\). Then the lamination \(\mathcal{L}\) is plaque-expansive and lamination-stable. For any \(C^1\) perturbation \(g\) of \(f\), the leaf conjugacy \(h_g\) from \(\mathcal{L}_g\) to \(\mathcal{L}\) is bi-Hölder, and the leaf holonomies are also Hölder. The authors also ask whether it is possible to remove the dynamical coherence condition. In Remark 4, they construct a simple example to illustrate the difference between leaf-expansivity and plaque-expansivity.