Hopf sign conjecture

OPENMajorConjectureProposed c. 1931 · Canonical special case

Canonical statement

If M2mM^{2m} is a closed connected Riemannian manifold with sectional curvature K0K\le0, then
(1)mχ(M)0. (-1)^m\chi(M)\ge0.
View source LaTeX
If \(M^{2m}\) is a closed connected Riemannian manifold with sectional curvature \(K\le0\), then \[ (-1)^m\chi(M)\ge0. \]

The Hopf sign conjecture asserts that nonpositive curvature controls the sign of the Euler characteristic: a closed connected Riemannian manifold M2mM^{2m} with sectional curvature K0K\le0 should satisfy (1)mχ(M)0(-1)^m\chi(M)\ge0. The expectation goes back to Hopf's lectures around 1931 on curvature and vector fields [Hopf1931CurvatureEuler], though the exact first written formulation is historically uncertain.

In dimension 22 the statement follows at once from the Gauss–Bonnet theorem, and in dimension 44 it follows from the Chern–Gauss–Bonnet formula, whose integrand is pointwise nonnegative under K0K\le0 there; this pointwise argument does not extend to dimension six and higher. Beyond low dimensions the conjecture is known for various locally symmetric and other high-symmetry classes, and the Euler characteristic of nonpositively curved manifolds has been studied more broadly [Davis1995EulerNonpositive]. Recent work examines the borderline case of nonpositively curved 44-manifolds with vanishing Euler characteristic [ConnellRuanWang2026Euler].

The conjecture is open in every even dimension at least 66; a resolution would require a mechanism relating sectional curvature to the Euler characteristic that avoids the sign analysis of the Gauss–Bonnet integrand.

The boxed statement is the canonical open formulation — not a stronger variant or a related research program. The status reflects the catalog's last review; do your own literature search before investing serious effort.