What you're seeing
A flat head of seeds, like the face of a sunflower. They are not placed by any plan. In Emergent mode you are watching them form one at a time: a new bud (a primordium) buds off near the center, then drifts slowly outward as still-newer buds appear behind it, pushing the whole head to grow from the middle. Each new bud lands wherever the most room is — crowded into the biggest gap the recent buds leave open. Out of nothing but that local jostling, two things appear that nobody put there: the seeds arrange into interlocking spirals winding both ways, and the angle between one bud and the next settles on a single value near 137.5°. Count the spiral arms and you get consecutive Fibonacci numbers — 34 one way and 55 the other, or 55 and 89 in a bigger head.
The rule
The whole local rule is: each new primordium forms in the most open space — repelled by its
neighbors. Concretely, when a bud is about to appear on the small ring at the center, it goes to the
angle that is farthest from the buds that formed just before it: the spot that minimizes how hard those
neighbors push on it (a sum of 1/distance² repulsions). Older buds have already drifted outward, so
it is the fresh front of recent buds that does the steering. That is all. The simulation never knows about the
golden ratio and is never told to use 137.5° — that angle is what the rule converges to, not what it is
given. (In Explore mode we do the opposite: we impose a divergence angle δ and place
bud number n directly at angle n·δ and radius c·√n, the geometry Helmut
Vogel used in 1979 to draw a sunflower. That mode lets you dial the angle and see why the golden one wins — but
it does not explain where the angle comes from. The Emergent mode does.)
Why it matters
The selected angle is the golden angle ≈137.5°, which is exactly
360°/φ² where φ = (1+√5)/2 ≈ 1.618 is the golden ratio. Why this angle and not
another? Because φ is, in a precise sense, the "most irrational" number — the one
least well approximated by any simple fraction. If successive buds turned by a rational fraction of a
full circle — say 90° = ¼ turn — then every fourth bud would line up in the same direction and the head would
collapse into a few radial spokes with big empty wedges between them. The golden angle is the angle that
never closes up into spokes at any scale, so the buds stay spread out and the packing stays dense and
even whether the head has ten seeds or ten thousand. That even, gap-free packing is exactly what a plant "wants"
from a sun-collecting leaf or a seed-bearing head — and it is why sunflower seeds, pinecone scales, and the
leaves spiralling up a stem so often show Fibonacci spiral counts.
Crucially, plants do not compute φ. Stéphane Douady and Yves Couder showed in
1992 that the angle emerges from the physical repulsion of new buds growing into available space — they
even reproduced it with no biology at all, by letting drops of magnetized ferrofluid fall one at a time onto a
dish and repel each other; the drops arranged themselves at the golden angle. So this entry is honest about its
flavor of emergence: like the Atlas's L-systems, phyllotaxis is a
generative, geometric kind of emergence — a form unfolding under a simple growth rule — rather
than the real-time dynamical jostling of, say, the flocking boids. But the headline holds: a single local rule,
repeated, produces a precise global number that the rule never names.
In the wild
Spiral phyllotaxis is one of the most widespread patterns in botany. You can find Fibonacci parastichies (the visible spiral arms) in:
- Sunflower and daisy seed heads — the classic case, often 34/55, 55/89, even 89/144 in large heads.
- Pinecones and pineapples — count the scales spiralling up and down: usually 8 and 13, or 13 and 21.
- Cacti, romanesco, artichokes, and the spiral of leaves up a stem, which spaces successive leaves about 137.5° apart so each shades the ones below as little as possible.
It is worth being accurate: real phyllotaxis is statistical, not perfect. Not every sunflower hits 137.5° exactly, growing heads have defects and transitions, and a minority of plants show non-Fibonacci (e.g. Lucas-number) spirals or other arrangements entirely. The Fibonacci dominance is a strong statistical tendency produced by a robust growth mechanism — not an iron law obeyed by every specimen.
Try this
- In Explore mode, sit at the default 137.51° and admire the dense, even pack. Then nudge the divergence slider by a tenth of a degree and watch the spirals jump between Fibonacci families — the arm counts in the readout flip from one consecutive pair to another.
- Set the angle to 137.0° or 90° and watch the packing fail: spokes and big empty gaps open up. The golden angle's near neighbors all pack worse — that's the whole point of it being "most irrational."
- Switch to Emergent mode and just watch. No angle is imposed; buds simply grow into the biggest gap. The measured divergence angle in the readout climbs and settles near the golden angle — about 138° in this discrete model — entirely on its own — the rule discovers the golden angle without being told it.
References
- Vogel, H. (1979). "A better way to construct the sunflower head." Mathematical Biosciences 44(3–4),
179–189. (The
n·137.5°,c·√nplacement used in Explore mode.) - Douady, S. & Couder, Y. (1992). "Phyllotaxis as a Physical Self-Organized Growth Process." Physical Review Letters 68(13), 2098–2101. (The repulsion mechanism behind Emergent mode, and the ferrofluid-drop experiment.)
- Jean, R. V. (1994). Phyllotaxis: A Systemic Study in Plant Morphogenesis. Cambridge University Press.
- Adler, I., Barabé, D. & Jean, R. V. (1997). "A History of the Study of Phyllotaxis." Annals of Botany 80(3), 231–244.