Odd-dimensional orbifolds with all geodesics closed are covered by manifolds (Q2049953): Difference between revisions

The study of the existence of closed geodesics on Riemannian manifolds has a long and storied past beginning with the fundamental result of \textit{L. A. Lyusternik} and \textit{A. I. Fet} [Dokl. Akad. Nauk SSSR, n. Ser. 81, 17--18 (1951; Zbl 0045.20903)] that every closed Riemannian manifold possesses at least one closed geodesic and later from the results of \textit{D. Gromoll} and \textit{W. Meyer} [J. Differ. Geom. 3, 493--510 (1969; Zbl 0203.54401)] and \textit{M. Vigue-Poirrier} and \textit{D. Sullivan} [J. Differ. Geom. 11, 633--644 (1976; Zbl 0361.53058)] that every closed Riemannian manifold admits infinitely many closed geodesics as long as its rational cohomology is not generated by one element. So-called \emph{Besse manifolds}, that is, those closed Riemannian manifolds all of whose geodesics are closed have also been extensively studied [\textit{A. Besse}, Manifolds all of whose geodesics are closed. Berlin-Heidelberg-New York: Springer-Verlag (1978; Zbl 0387.53010)], and are known to have a universal covering from one of the compact rank one symmetric spaces. The study of closed geodesics has a natural extension to Riemannian orbifolds. These questions have been previously studied by the reviewer [Riemannian geometry of orbifolds. University of California, Los Angeles, USA (PhD Thesis) (1992); Pac. J. Math. 175, No. 1, 39--46 (1996; Zbl 0865.53034); Houston J. Math. 33, No. 4, 1011--1025 (2007; Zbl 1144.53057)], \textit{K. Guruprasad} and \textit{A. Haefliger} [Topology 45, No. 3, 611--641 (2006; Zbl 1099.53035)], \textit{G. C. Dragomir} [Geom. Dedicata 172, 399--411 (2014; Zbl 1301.57022)], and \textit{C. Lange} [Pac. J. Math. 294, No. 2, 453--472 (2018; Zbl 1390.53033); J. Reine Angew. Math. Journal Profile 758, 67--94 (2020; Zbl 1431.53045)], for example. For even dimensions, weighted projective spaces provide infinitely many examples of Besse orbifolds. However, for odd dimensions, the situation is much more restricted in view of the main result of the article under under review. The authors prove that any odd-dimensional Besse orbifold is of the form \((M,g)/\Gamma\) for some Riemannian manifold \(M\) homeomorphic to a sphere, a Besse metric \(g\), and some finite group \(\Gamma\) of isometries of \((M, g)\). Moreover, if \(n\ne 3\) and \(\Gamma\) contains a fixed-point-free involution, then \((M, g)\) is isometric to a round sphere and \(\Gamma\subset O(n + 1)\).