Tight Spherical Embeddings (Updated Version)

Thomas E. Cecil, Patrick J. Ryan · 2025

Read Tight Spherical Embeddings (Updated Version) by Thomas E. Cecil, Patrick J. Ryan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-25611.txt.

Introduction

Thomas E Cecil and Patrick J Ryan present an updated differential geometry paper. Their 2025 version updates a 1979 conference paper. Original exposition appears together with new notes. Those notes reference progress since the original version. The main result concerns compact isoparametric hypersurfaces. Such surfaces sit inside a sphere in Euclidean space.

Every such hypersurface is tight, they prove. Tightness means height functions have minimal critical points. Morse inequalities set that minimum number required. Since the hypersurface lies in the sphere, tautness follows. Spherical distance functions also have minimal critical points. A second result covers focal submanifolds of hypersurfaces. Those focal submanifolds must also be taut in sphere. Proofs rest on Munzner fundamental structural work. That work describes families of isoparametric hypersurfaces. Updated notes connect old results to modern progress. Readers gain both classic proof and current context. Tight embeddings remain central in global differential geometry.

Catalog record

CREATORThomas E. Cecil, Patrick J. Ryan
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-25611

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Tight Spherical Embeddings (Updated Version) Thomas E. Cecil, Patrick J. Ryan Abstract This is an updated version of the paper [ 14 ] which appeared in the proceedings of the 1979 Berlin Colloquium on Global Differential Geometry. This paper contains the original exposition together with some notes by the authors made in 2025 (as indicated in the text) that give references to descriptions of progress made in the field since the time of the original version of the paper. The main result of this paper is that every compact isoparametric hypersurface M n ⊂ S n + 1 ⊂ 𝐑 n + 2 M^{n}\subset S^{n+1}\subset{\bf R}^{n+2} is tight, i.e., every non-degenerate linear height function ℓ p \ell_{p} , p ∈ S n + 1 p\in S^{n+1} , has the minimum number of critical points on M n M^{n} required by the Morse inequalities. Since M n M^{n} lies in the sphere S n + 1 S^{n+1} , this implies that M n M^{n} is also taut in S n + 1 S^{n+1} , i.e., every non-degenerate spherical distance function has the minimum number of critical points on M n M^{n} . A second result is that the focal submanifolds of isoparametric hypersurfaces in S n + 1 S^{n+1} must also be taut. The proofs of these results are based on Münzner’s [ 23 ] – [ 24 ] fundamental work on the structure of a family of isoparametric hypersurfaces in a sphere. In his lecture at the 1979 Chern Symposium at Berkeley, N. H. Kuiper pointed out that in all known examples of tight embeddings M^{n}\subset S^{n+1} (Euclidean sphere), the manifold M^{n} was diffeomorphic to a sphere or a product of spheres. He asked if there were more examples, and if so, whether they must necessarily be algebraic. The main result of this paper is to show that every compact isoparametric hypersurface is tight. This provides an infinite collection of new examples of tight spherical hypersurfaces as can be seen from the work of Takagi and Takahashi [ 33 ] – [ 34 ] , Ozeki and Takeuchi [ 27 ] – [ 28 ] and more recently the results of Ferus, Karcher and Münzner [ 18 ] . Although the compact isoparametric hypersurfaces have not yet been classified, Münzner [ 23 ] – [ 24 ] has obtained a sufficiently explicit description of their topology and geometry to allow us to prove that they must be tight. Münzner’s proof also establishes the fact that each isoparametric hypersurface is algebraic. Thus the second part of Kuiper’s question remains open. Note 1: (added by the authors in 2025) Through the work of many researchers, the classification of isoparametric hypersurfaces in spheres has now been completed. See, for example, the survey article of Chi [ 17 ] or the book [ 15 , pp. 144–182] for descriptions of the classification. Our second result is that the focal submanifolds of isoparametric hypersurfaces must also be tight. This path of exploration leads to some interesting questions concerning tightness and the structure of the focal set which we list at the end of the paper. Our results also provide many new examples of taut submanifolds (both compact and non-compact) of Euclidean space. These are obtained as the images of isoparametric hypersurfaces and their focal sets under stereographic projection. 1. Preliminaries. 1a. Tight and taut immersions. Let f:M\rightarrow E^{m} be an immersion of a smooth manifold into Euclidean space. Let S^{m-1} denote the unit sphere in E^{m} . For \xi\in S^{m-1} , the linear height function \ell_{\xi}:M\rightarrow{\bf R} is defined by \ell_{\xi}(x)=\langle f(x),\xi\rangle , where \langle\ ,\ \rangle is the Euclidean inner product. For p\in E^{m} , the Euclidean distance function L_{p}:M\rightarrow{\bf R} is defined by L_{p}(x)=|f(x)-p|^{2} . One shows using Sard’s theorem that almost all of the functions of each type are non-degenerate on M . The immersion f is called tight , respectively, taut , if every non-degenerate \ell_{\xi} , respectively L_{p} , has the minimum number of critical points required by the Morse inequalities. Tightness was first discussed in the form of minimal total absolute curvature by Chern and Lashof [ 16 ] . The formulation in terms of critical point theory is due to Kuiper [ 19 ] . For a two-dimensional manifold M , tightness, respectively, tautness, is equivalent to the two-piece property of Banchoff with respect to Euclidean hyperplanes [ 1 ] , respectively, hyperspheres [ 2 ] . Since a hyperplane can be approximated arbitrarily closely by a hypersphere of large radius, one sees that tautness implies tightness for two-dimensional manifolds. By a similar argument, one shows that this is true for arbitrary manifolds [ 7 ] . Moreover, a taut immersion must be an embedding [ 7 ] , whereas a tight immersion need not be. An immersion f:M\rightarrow E^{m} is called spherical if the image of f lies in a Euclidean hypersphere S^{m-1}\subseteq E^{m} . If f is tight and spherical, it is easy to show that f is taut [ 2 ] . Moreover, using the fact that tautness is a conformal property [ 7 ] , one then can obtain a taut embedding of M into E^{m-1} by composing f with stereographic projection into E^{m-1} . Thus, tautness is equivalent to the combination of tight and spherical in the sense that if M is a compact manifold which is not a sphere, then there exists a substantial non-spherical taut embedding of M into E^{m-1} if and only if there exists a substantial tight spherical embedding of M into E^{m} . 1b. Distance functions and focal points in the sphere. It is advantageous at times to consider distance functions in the sphere rather than linear height functions. Let f:M\rightarrow S^{m} be an immersion. For p\in S^{m} , x\in M , the spherical distance function d_{p} has the form d_{p}(x)=\cos^{-1}\langle p,f(x)\rangle. Clearly d_{p} and the linear height function \ell_{p} have the same critical points. As in the case of Euclidean distance functions [ 20 , p. 32–38] , there is a natural formula for the index of a critical point of d_{p} in terms of the focal points of M in S^{m} which we will now develop. Let NM denote the normal bundle of M in S^{m} with projection map \pi onto M . The exponential map E:NM\rightarrow S^{m} is defined by letting E(\xi) be the point reached by traversing a distance |\xi| along the geodesic in S^{m} with initial tangent vector \xi . A point p\in S^{m} is called a focal point of (M,x) of multiplicity \nu>0 if p=E(\xi) , where x=\pi\xi and the Jacobian of E has nullity \nu at \xi . Assume now that |\xi|=1 and let x=\pi\xi . A straightforward computation shows that p=E(t\xi) is a focal point of multiplicity \nu of (M,x) if and only if \cot(t) is an eigenvalue of the shape operator A_{\xi} of multiplicity \nu . Note that each principal curvature \lambda of A_{\xi} gives rise to two distinct focal points of (M,x) of the form p=\cos t\ x+\sin t\ \xi for t=\cot^{-1}\lambda and t=\cot^{-1}\lambda-\pi . (For further discussion, see [ 11 ] .) The relationship between focal points and spherical distance functions is embodied in the next result

Read the complete text