The Logistic Map
One line of arithmetic, one knob. Turn it up and a steady number splits into two, then four, then breaks into chaos you can never predict. No randomness anywhere. Find the edge.
Nearest landmark: The Storm. A negative λ means orbits pull together (order); a positive λ means they fly apart (chaos).
Determined and unpredictable at the same time. At r = 3.7000 there is no randomness in this equation, not one coin flip, and yet you could never guess the next value. Two orbits a billionth apart are tearing away from each other as fast as the arithmetic allows. The Lyapunov exponent is positive (0.356), and a positive exponent is the definition of chaos: any error in the starting value, however tiny, doubles and doubles until it swamps everything. Same rule, same parabola as the calm settings. This is what deterministic chaos looks like.
Tip: hit ▶ sweep and watch the live orbit settle, split, split again, and shatter as the dial climbs. Or click straight onto the fig tree above to land anywhere.
The reframe
What gets me about this is that there is no randomness anywhere in it. One line of arithmetic, no coin flips, no noise, and below r = 3.5699 it is the tamest thing in the world: it settles to a number, or ticks politely between a few. Then you nudge the dial a thousandth past the edge and it becomes genuinely unpredictable. Not because we are missing information. Because any error in the starting value, even in the fifteenth decimal place, doubles every few steps until it is the size of the answer. That is the positive Lyapunov exponent, and it is exactly the number the "twin orbits" readout is measuring.
The route in is the strangest part. The splittings, 2, 4, 8, 16, come faster and faster, and the ratio of one window's width to the next converges to a single number, Feigenbaum's constant 4.6692016. The shock is that the same number shows up in a dripping faucet, a fibrillating heart, a column of convecting fluid, and an electronic oscillator, systems that share nothing with populations. The road to chaos has universal constants, the way circles have π.
And chaos is not the opposite of order. Drag into the chaotic band and find the period-3 island: clean, perfect order sitting in the middle of the storm. Zoom into the fig tree near it and you would find a smaller copy of the whole tree, windows of order inside chaos inside order, all the way down. Determinism never promised predictability. This little equation is where mathematics learned the difference.
The history
Pierre-François Verhulst wrote the logistic equation down around 1838 to model how a population grows when resources run out. For over a century it was a quiet curve in an ecology textbook. Then in 1976 the biologist Robert May published a short paper in Nature, "Simple mathematical models with very complicated dynamics," showing that this innocent population toy goes wildly chaotic for large r, a warning to every ecologist that a simple model can behave unpredictably and a founding document of chaos theory. Around the same time, working at Los Alamos with a pocket HP calculator, Mitchell Feigenbaum noticed the period-doublings converged at a fixed rate, 4.6692016, and that the same constant governed a whole family of unrelated maps, the discovery that chaos has universal numbers. Tien-Yien Li and James Yorke gave the field its name in 1975 with the paper "Period Three Implies Chaos," building on Oleksandr Sharkovskii's 1964 ordering of the periods, and Edward Lorenz had already, in 1963, found the sensitive dependence on initial conditions that the world would come to call the butterfly effect. A line of arithmetic from a nineteenth-century ecology book that turned out to be one of the cleanest doors into chaos we have.