A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Turing instability occurs when a spatially uniform steady state that is stable without diffusion becomes unstable when diffusion is added — the counterintuitive result that diffusion (which normally smooths out variations) can create spatial patterns. In a reaction-diffusion system with an activator and an inhibitor, if the inhibitor diffuses much faster than the activator, the uniform state becomes unstable to spatial perturbations of specific wavelengths, and stable patterns (spots, stripes, labyrinths) emerge spontaneously. This mechanism explains pattern formation in chemistry (Belousov-Zhabotinsky reaction), biology (animal coat markings, morphogenesis), and physics (convection cells).
In 1952, Alan Turing — already famous for his work on computation and codebreaking — published a paper titled "The Chemical Basis of Morphogenesis" that proposed a radical idea: the patterns on animal skins, the arrangement of leaves on plants, and the segmentation of embryos could all arise from simple chemical reactions coupled with diffusion. The mechanism he identified — now called Turing instability — is one of the most beautiful and counterintuitive results in mathematical biology and nonlinear dynamics.
The setup is a reaction-diffusion system: two or more chemical species that react with each other and diffuse through space. Consider two chemicals, an activator (A) that promotes its own production and that of an inhibitor (B), and the inhibitor that suppresses both. In a well-mixed solution (no spatial variation), the system has a stable equilibrium — A and B reach a balance. Now allow them to diffuse. Intuition says diffusion should make things smoother — it should stabilize the uniform state. Turing showed the opposite: if the inhibitor diffuses much faster than the activator, the uniform state can become unstable to spatial perturbations.
The mechanism is local activation, long-range inhibition. Imagine a small random fluctuation creates a spot with slightly more activator. The activator amplifies itself locally (autocatalysis), but the inhibitor it produces diffuses away rapidly, creating a halo of inhibition that suppresses the activator at a distance. The result: the activator peaks grow at regularly spaced intervals, separated by inhibitor-dominated valleys. The spacing is set by the balance between the reaction time scales and the diffusion length scales. Too close together, and neighboring peaks' inhibition halos overlap and suppress them. Too far apart, and new peaks can nucleate in the gaps. The selected wavelength is a prediction of the theory, and it matches experimental observations.
The patterns that emerge depend on geometry, dimensionality, and the specific nonlinearities. In one spatial dimension, the Turing instability produces periodic stripes. In two dimensions, the same parameters can produce stripes, spots, or labyrinthine patterns depending on the nonlinear interactions between different spatial modes. On domains of different shapes, the available modes change: a wide domain supports 2D patterns (spots), while a narrow domain constrains the system to 1D patterns (stripes). This explains why animal coat markings transition from spots on the body to stripes on the tail, and why spotted animals can have striped tails but not vice versa. The Turing mechanism has been confirmed experimentally in chemical systems (the CIMA reaction) and is increasingly supported as a mechanism for biological patterning, from fish skin pigmentation to digit spacing in vertebrate limbs.