A discrete view of Gromov's filling area conjecture

A compact metric surface $M$ isometrically fills a closed metric curve $C$ if $\partial M=C$ and $d_M(x,y)=d_C(x,y)$ for every $x,y\in C=\partial M$; that is, $M$ does not introduce any ``shortcuts’’ between points on its boundary. Gromov’s filling area conjecture in differential geometry from 1983 asserts that among all compact, orientable Riemannian surfaces which isometrically fill the Riemannian circle, the one with the smallest surface area is the hemisphere. Gromov demonstrated that this is indeed the case if $M$ is homeomorphic to the disk. While Gromov’s conjecture has since been verified in some other cases, the full conjecture remains unresolved. In this paper, we consider a discrete analogue of Gromov’s problem, which is likely natural to those who study graph embeddings on arbitrary surfaces. Using standard graph-theoretic tools, such as Menger’s theorem, we obtain reasonable asymptotic bounds on this discrete variant. We then demonstrate how these discrete bounds can be translated to the continuous setting, showing that any isometric filling of the Riemannian circle of length $2π$ has surface area at least $1.36π$ (the hemisphere has surface area $2π$). This appears to be the first quantitative lower-bound on Gromov’s problem that applies to arbitrary isometric fillings.

💡 Research Summary

The paper revisits Gromov’s filling‑area conjecture, which asserts that among all compact orientable Riemannian surfaces that isometrically fill a circle of length 2π, the hemisphere has the smallest area (2π). While the conjecture is proved for disks and genus‑one surfaces, the general case remains open. The authors propose a discrete analogue that is natural for researchers studying graph embeddings on surfaces, and they use combinatorial tools to obtain quantitative lower bounds that apply to any filling, regardless of topology.

First, the authors formalize the notion of an “isometric filling” for a graph. Given a cycle Cₙ on n vertices, an abstract triangulation K (a 2‑dimensional simplicial complex where each edge belongs to one or two triangles) is an isometric filling of Cₙ if its boundary ∂K equals Cₙ and the graph distance d_K between any two boundary vertices coincides with the distance d_{Cₙ} along the cycle. They also introduce a relaxed version: a δ‑Lipschitz filling, where d_K(x,y) ≥ δ·d_{Cₙ}(x,y) for all boundary vertices, with δ∈(0,1].

The central combinatorial result (Theorem 2.1) states that any δ‑Lipschitz filling K of Cₙ must contain at least
|V(K)| ≥ (δ³/8)(n−1)² + (n−1)/2
vertices. For the exact isometric case (δ = 1) this yields |V(K)| ≥ n²/8, which is tight up to the constant 1/8. The proof hinges on two lemmas. Lemma 2.2 is a Sperner‑type statement: if a set S of interior vertices separates the two arcs of Cₙ obtained by deleting non‑adjacent boundary vertices x and y, then S must contain all interior vertices of some x‑y path in the 1‑skeleton. Proposition 2.3 gives a lower bound on the sum of distances between k distinct pairs of vertices taken from opposite arcs of the cycle; it shows that Σ d_{Cₙ}(ℓ_i,r_i) ≥ k(k+2)/2. Combining these with Menger’s theorem (which guarantees many vertex‑disjoint paths) yields the vertex‑count bound.

Euler’s formula then translates the vertex bound into a bound on the number of triangles: Corollary 2.4 shows |T(K)| ≥ (δ³/4)(n−1)² + 1 − 2χ, where χ is the Euler characteristic of the closed surface obtained by capping off the boundary.

To connect the discrete model with the original continuous problem, the authors construct “balanced triangulations” of any piecewise‑linear (PL) metric surface (Lemma 3.1). For a sequence ε→0 they produce triangulations K(ε) in which every edge has length ε ± o(ε), every boundary edge has length ε ± o(ε), and all but O(1/ε) triangles are equilateral of side ε. The construction uses Dirichlet’s approximation theorem to partition each original edge into nearly equal sub‑segments and then fills each PL triangle with a near‑regular hexagonal lattice of equilateral triangles, carefully handling the boundary to avoid introducing new vertices.

Applying the discrete lower bound to these balanced triangulations yields Theorem 3.3: if a PL surface M is a δ‑Lipschitz filling of a circle of circumference ℓ, then
Area(M) ≥ √3 · δ³ · ℓ² / 16.
For the exact isometric case (δ = 1) and ℓ = 2π this gives Area(M) ≥ √3·(2π)²/16 ≈ 1.36π, a concrete quantitative lower bound that holds for any isometric filling, irrespective of genus or orientability. Corollary 3.4 extends the same bound to arbitrary compact Riemannian surfaces by a standard smoothing argument.

Thus the paper provides the first general quantitative lower bound for Gromov’s filling‑area problem, derived entirely from combinatorial graph theory. The constants are not optimal; improving the 1/8 factor in Theorem 2.1 or reducing the loss in the δ‑Lipschitz approximation could bring the bound closer to the conjectured 2π. The authors suggest that further refinement of the discrete analysis, as well as extensions to higher dimensions or non‑Euclidean metrics, are promising directions for future work.