Title: Phyllotaxis
1Phyllotaxis
- A mathematical theory?
- Sjors van der Stelt
2Fibonacci sequence
- F_11F_21F_n2 F_n1 F_n.
- Gives 1,1,2,3,5,8,13,21,34,55,89,144,233, .
- F_n1/F_n converges to f 0.61803 which is a
root of x²x-1
0.This is the golden section. The corresponding
angle - f360º 137.5 º is called the golden angle.
- Other elementary property gcd(f_n1,f_n)1
3Phyllotactic Patterns in nature
4(No Transcript)
5(No Transcript)
6(No Transcript)
7(No Transcript)
8(No Transcript)
9(No Transcript)
10(No Transcript)
11(No Transcript)
12(No Transcript)
13Wrong picture taken from De Andere Linkerkant
Links en rechts in de evolutie from Tijs
Goldschmidt
14Nobody understands?
- From Tijs Goldschmidt
- Door Goulds essay over Thompson nam ik ook
voor het eerst kennis van Fibonacci-reeksen. Een
Fibonacci-reeks vorm je door steeds de twee
vorige getallen in de reeks bij elkaar op te
tellen en daarmee het volgende getal te vormen.
Dus 1, 1, 2, 3, 5, 8, 13, 21. () Zon
dennenkegel bijvoorbeeld kan, als je de
afzonderlijke schubben volgt, zijn opgebouwd uit
13 links- en 21-rechtsdraaiende spiralen. Dus
niet 16 en 24, of 9 en 19, maar uitgerekend 13 en
21, allebei getallen uit de Fibonacci-reeks.
Waarom volgen zoveel structuren deze curve die
wiskundigen een logaritmische spiraal noemen?
Thompsons antwoord op de vraag was destijds
geweest dat deze abstracte vormen de best
mogelijke oplossingen zijn voor algemene
problemen.
15(Serious) literature on phyllotaxis
- Kepler and Leonardo were among the first to
recognize the phenomenon - Extensively studied during nineteenth en
twentieth century articles were mostly published
in German or French during the nineteenth
century, later English. However, English and
French people rarely speak German!!(to be
continued)
16Preliminaries
- The seeds or petals that form the phyllotactic
patterns are called primordia. - There are n spirals, called parastichies,
spiralling to the one direction (right, par
exemple) and m parastichies going to the left.
(Of course, there are others, but these are the
best discernable. They are also called contact
parastichies.) - If we number the primordia from youngest to
oldest, then two neighbours on a n-parastichy
differ n. - The angle between two primordium and the next
oldest primordium is called the divergence
angle - Wrong model fixed divergence angle radial
growth of vk - Assume divergence angle d is a rational number
(as a fraction of the number of revolutions), for
example d 0.125 - Bravais Bravais (1830s) proved that if a
phyllotactic structure has d f then the
parastichy numbers are equal to Fibonacci numbers - Tait (1872), Coxeter (70s) (pictures taken from
Michael Naylor 02)
17Numbering of primordia
18We then get
19Other examples
20What if d f?
21And what about d v2?
22What if we choose d p?
23So d p does not give an equal distribution of
the petals on the disk, but f and v2 do. Why?
- The continued fraction expansion of p converges
relatively rapidly - Now f and v2 have continued fraction expansions
that converge slowly (according to a criterion of
Klein)
24Hofmeisters rules (1870s)
- (Expended version Snow Snow (1950s), Couder
Douady (1996)) - The apex has an axis of symmetry.
- Primordia form at edge of the apex and, due to
the shoots growth, they move away radially from
the center with radial velocity only depending on
their distance to the edge of the apex. - New primordia are formed periodically the period
is called the plastochrone. - The incipient primordium forms in the largest
available space left by the previous ones.
25Van Iterson (1907)
- Mathematische und mikroskopische-anatomische
Studien uber Blattstellungen - Es besteht natürlich die Möglichkeit, daß die
Schraubenlinie durch die Punkte a_0 und b_0 alle
Punkte des Systems umfaßt aber im allgemeinen
wird dieses nicht eintreffen, und es wird
anzunehmen sein, daß Punkte außerhalb derselben
liegen. Wählen wir den Punkt g_0, der von allen
diesen Punkten am dichtsten an der betrachteten
Schraubenlinie liegt, und schrauben wir ... (p9)
(Etcetera etc.!) - Van Itersons (nor Schoutes) theory is not
mentioned as often as it should be in the
abundant literature on phyllotaxis. This might be
due to the fact that he wrote his thesis in
German.
26General principle
- There are n spirals, called parastichies,
spiralling to the one direction (right, par
exemple) and m parastichies going to the left.
(Of course, there are others, but these are the
best discernable. They are also called contact
parastichies.) - If we number the primordia from youngest to
oldest, then two neighbours on a n-parastichy
differ n. - If a primordium k on an n-parastichy gets farther
from the apex, it looses contact with the its
younger neighbour k-n a bifurcation happens
27Numbering of primordia
28Some calculations from Van Iterson
29Some calculations from Van Iterson
30Van Iterson diagram
31Douady Couder (1996) experimental results
- Douady Couder settled out an experiment in
which magnetized oil droplets were periodically
dropped on a disk. The experiments showed
phyllotactic patterns to a high degree. - Numerical results published in the same sequence
of articles showed the same. - A dynamical model proposed by the authors gave
rise to a bifurcation diagram which is exactly
the Van Iterson diagram
32Dynamical systems approach
33Dynamical systems approach
34Dynamical systems approach
35Dynamical systems approach
36Dynamical systems approach
37Dynamical systems approach
38Dynamical systems approach
39Dynamical systems approach
40Dynamical systems approach
41Dynamical systems approach
42Dynamical systems approach
43Further theories
- Turing (52) morphogenesis
- Cellular automata
- Lindenmayer systems
44Main References
- Atela P, Gole C and Hotton, S. A Dynamical System
for Plant Pattern Formation. Nonlinear Science
2002 - Douady S and Couder Y, Phyllotaxis as a Self
Organising Process Part I,II and III. Journal of
Theoretical Biology 1996 - Van Iterson, G., Mathematische und
mikroskopische-anatomische Studien uber
Blattstellungen. Jena Verlag 1907