"All right," Malcolm said. "Let's go back to the beginning." He paused, staring at the ceiling. "Physics has had great success at describing certain kinds of behaviour: planets in orbit, spacecraft going to the moon, pendulums and springs and rolling balls, that sort of thing. The regular movement of objects. These are described by what are called linear equations, and mathematicians can solve those equations easily. We've been doing it for hundreds of years."

"Okay," Gennaro said.

"But there is another kind of behaviour, which physics handles badly. For example, anything to do with turbulence. Water coming out of a spout. Air moving over an airplane wing. Weather. Blood flowing through the heart. Turbulent events are described by nonlinear equations. They're hard to solve - in fact, they're usually impossible to solve. So physics has never understood this whole class of events. Until about ten years ago. The new theory that describes them is called chaos theory."

"Chaos theory originally grew out of attempts to make computer models of weather in the 1960s. Weather is a big complicated system, namely the earth's atmosphere as it interacts with the land and the sun. The behaviour of this big complicated system always defied understanding. So naturally we couldn't predict weather. But what the early researchers learned from computer models was that, even if you could understand it, you still couldn't predict it. Weather prediction is absolutely impossible. The reason is that the behaviour of the system is sensitively dependent on initial conditions."

"You lost me," Gennaro said.

"Use a cannon to fire a shell of a certain weight, at a certain speed, and a certain angle of inclination - and if I then fire a second shell with almost the same weight, speed, and angle - what will happen?"

"The two shells will land at almost the same spot."

"Right," Malcolm said. "That's linear dynamics."

"Okay."

"But if I have a weather system that I start up with a certain temperature and a certain wind speed and a certain humidity - and if I then repeat it with almost the same temperature, wind, and humidity - the second system will not behave almost the same. It'll wander off and rapidly will become very different from the first. Thunderstorms instead of sunshine. That's nonlinear dynamics. They are sensitive to initial conditions: tiny differences become amplified."

"I think I see," Gennaro said.

"The shorthand is the 'butterfly effect.' A butterfly flaps its wings in Peking, and weather in New York is different."

"So chaos is all just random and unpredictable?" Gennaro said. "Is that it?"

"No," Malcolm said. "We actually find hidden regularities within the complex variety of a system's behaviour. That's why chaos has now become a very broad theory that's used to study everything from the stock market, to rioting crowds, to brain waves during epilepsy. Any sort of complex system where there is confusion and unpredictability. We can find an underlying order. Okay?"

"Okay," Gennaro said. "But what is this underlying order?"

"It's essentially characterized by the movement of the system within phase space," Malcolm said.

"Jesus," Gennaro said. "All I want to know is why you think Hammond's island can't work."

"I understand," Malcolm said. "I'll get there. Chaos theory says two things. First, that complex systems like weather have an underlying order. Second, the reverse of that - that simple systems can produce complex behaviour. For example, pool balls. You hit a pool ball, and it starts to carom off the sides of the table. In theory, that's a fairly simple system, almost a Newtonian system. Since you can know the force imparted to the ball, and the mass of the ball, and you can calculate the angles at which it will strike the walls, you can predict the future behaviour of the ball. In theory, you could predict the behaviour of the ball far into the future, as it keeps bouncing from side to side. You could predict where it will end up three hours from now, in theory."

"Okay." Gennaro nodded.

"But in fact," Malcolm said, "it turns out you can't predict more than a few seconds into the future. Because almost immediately very small effects - imperfections in the surface of the ball, tiny indentations in the wood of the table - start to make a difference. And it doesn't take long before they overpower your careful calculations. So it turns out that this simple system of a pool ball on a table has unpredictable behaviour."

"Okay."

"And Hammond's project," Malcolm said, "is another apparently simple system - animals within a zoo environment - that will eventually show unpredictable behaviour."

"You know this because of . . ."

"Theory," Malcolm said.

"But hadn't you better see the island, to see what he's actually done?"

"No. That is quite unnecessary. The details don't matter. Theory tells me that the island will quickly proceed to behave in unpredictable fashion."

"And you're confident of your theory."

"Oh, yes," Malcolm said. "Totally confident." He sat back in the chair. "There is a problem with that island. It is an accident waiting to happen."