The explanation
The golden angle is the "most irrational" turn, the hardest to approximate by fractions, so seeds never line up in rows and space is used best; the Fibonacci spiral counts come from the same fact.
01One seed, one angle
New seeds keep appearing at the centre of a sunflower head and push older ones outwards. In 1979 the mathematician Vogel wrote this as a simple model: seed sits at
where is the angle each seed is turned from the one before. The radius gives every seed roughly the same area. The whole pattern depends on the single number .
02Why fractions fail
Turn by of a circle each time and every seeds you are back facing the same way: the seeds line up in straight rows, and the space between them is wasted.
An angle that is merely close to a fraction, such as 137°, close to of a turn, gives 21 slightly curved spokes, with gaps that widen further out.
So the best angle should stay as far as possible from every simple fraction.
03The most irrational number
Written as a continued fraction, the golden ratio has a 1 at every level:
Large numbers in a continued fraction make a number easy to approximate by fractions; all ones make it the hardest of all irrational numbers to approximate.
The golden angle is the smaller part of a full turn split in the golden ratio:
Turn by it and the seeds never line up in spokes. The "minimum spacing" readout compares the two most crowded seeds at the current angle with the golden angle: at 137.6° only about 63% is left.
04Why Fibonacci numbers
The spirals your eye picks out join each seed to its nearest neighbours. At the golden angle, the nearest neighbours of seed are seeds and (on the outer ring of 1000 seeds), so you see 55 and 89 spirals.
The reason is that the best fractional approximations of the golden ratio are ratios of consecutive Fibonacci numbers: . Further out there are more seeds and the visible pair moves up: 13 and 21, 21 and 34 nearer the centre, 55 and 89 at the edge. Real sunflowers often show 34 and 55 or 55 and 89 spirals.
Questions
Do sunflowers really compute the golden angle?
They do not need to. A widely accepted explanation is that each new bud forms where there is the most room, furthest from the existing ones, and this local rule drives the angle towards the golden angle by itself. Real plants deviate a little, so spiral counts are occasionally not Fibonacci numbers.
Why is the radius √n?
A disc of radius $r$ has area proportional to $r^2$. With $r=\sqrt n$, the first $n$ seeds fill an area proportional to $n$, so every seed gets about the same space.
Where else does this pattern show up?
Pine cone scales, the hexagons on a pineapple, cauliflower and romanesco, succulents and the leaf arrangement on many stems all show consecutive Fibonacci numbers.
The syllabus side
Related curves
Sources: Phyllotaxis — Wikipedia · Golden angle — Wikipedia