Spherical Bernstein's problem explained

The spherical Bernstein's problem is a possible generalization of the original Bernstein's problem in the field of global differential geometry, first proposed by Shiing-Shen Chern in 1969, and then later in 1970, during his plenary address at the International Congress of Mathematicians in Nice.

The problem

Are the equators in

Sn+1

the only smooth embedded minimal hypersurfaces which are topological

n

-dimensional spheres?

Additionally, the spherical Bernstein's problem, while itself a generalization of the original Bernstein's problem, can, too, be generalized further by replacing the ambient space

Sn+1

by a simply-connected, compact symmetric space. Some results in this direction are due to Wu-Chung Hsiang and Wu-Yi Hsiang work.

Alternative formulations

Below are two alternative ways to express the problem:

The second formulation

Let the (n − 1) sphere be embedded as a minimal hypersurface in

Sn

(1). Is it necessarily an equator?

By the AlmgrenCalabi theorem, it's true when n = 3 (or n = 2 for the 1st formulation).

Wu-Chung Hsiang proved it for n ∈  (or n ∈, respectively)

In 1987, Per Tomter proved it for all even n (or all odd n, respectively).

Thus, it only remains unknown for all odd n ≥ 9 (or all even n ≥ 8, respectively)

The third formulation

Is it true that an embedded, minimal hypersphere inside the Euclidean

n

-sphere isnecessarily an equator?

Geometrically, the problem is analogous to the following problem:

Is the local topology at an isolated singular point of a minimal hypersurface necessarily different from that of a disc?

For example, the affirmative answer for spherical Bernstein problem when n = 3 is equivalent to the fact that the local topology at an isolated singular point of any minimal hypersurface in an arbitrary Riemannian 4-manifold must be different from that of a disc.

Further reading