Chapter 83 15.1 Tom Ray's Electric Evolution Machine

As soon as Tom Ray put the programmed gizmo into the computer, it multiplied rapidly until several hundred copies filled the available storage space.Ray's gadget was sort of an experimental computer virus, and since it couldn't reproduce once it left his computer, it wasn't dangerous.He just wanted to see what would happen if viruses had to compete with each other in a confined space. Ray's world is so ingeniously designed that, among the thousands of clones of the virus's ancestor, about 10 percent have undergone slight mutations as they replicated themselves.Originally that guy was an "80" - so named because it encoded 80 bytes long.Some 80s mutate a little at random to become 79s or 81s.Some variants of these new viruses soon took over Ray's virtual world.They in turn mutate into more species.Virus 80 is nearly driven to the brink of extinction by this rapidly growing army of new "species."It survived, though, and some time after the 79, 51, and 45 peaked in numbers, the 80 resurfaced.

In just a few hours, Tom Ray's electric evolution machine has evolved a "pot of culture fluid", and nearly a hundred computer viruses are fighting to survive in this isolated world.After spending months writing code, Ray conceived artificial evolution on his first attempt. When Ray was a soft-spoken, shy Harvard undergrad, he collected ant colonies in Costa Rica for the famous Ant-Man Wilson.Wilson's Cambridge lab needed live leafcutter ant colonies, and Ray was hired to find and capture field colonies in good condition in the dense tropical jungles of Central America for shipment to Harvard.He found himself particularly good at this job.His trick is to dig through the jungle soil with surgeon-like dexterity, removing the core of the ant colony.What needs to be removed is the queen's complete inner chamber, including the queen herself, her caretakers, and a miniature ant garden stocked with enough food to ensure the colony doesn't starve during transport.Young, newborn ant colonies are ideal.The core of this ant colony fits perfectly in a teacup.Another trick is to find tiny ant nests hidden under forest vegetation.In just a few years, this palm-sized colony can fill a large room.

While collecting ants in the tropical rainforest, Lei also discovered an unknown species of butterfly that would follow the marching route of army ants.The ferocious habit of army ants devouring all animals in their path can drive a swarm of flying insects into a panic.A species of bird has grown into the habit of following this horde of predators, feasting happily on the scurrying insects that scatter through the air.And behind the flying birds following the army of army ants, butterflies followed one after another.Butterflies follow to feast on the antbird's dung—a source of much-needed nitrogen for egg laying.Ants, "antbirds", "antbirds, butterflies" and maybe who knows what else followed behind, forming a motley army, sweeping across like a group of connected gypsies This jungle.

Ray was overwhelmed by such an exquisite complex combination.This is simply a nomadic society!In the face of all things in the universe, most attempts to understand ecological relationships appear ridiculous.In the vast universe, how did these three populations (one kind of ants, three kinds of butterflies, and more than a dozen kinds of birds) form such a strange interdependent relationship?Why is this so? When Ray finished his Ph.D., he felt that ecological science was stagnant because it could not give a satisfactory answer to the above important questions.Ecology lacks a good theory that summarizes the wealth of observational data accumulated in each wilderness.It is plagued by a mass of partial knowledge: without an overarching theory, ecology is little more than a library full of charming fairy tales.The life cycle of a barnacle colony, the seasonal morphological changes of a buttercup field, or the behavior of a bobcat family are well known, but what principles, if any, govern the variation in all three?Biology needs a science of complexity to answer the unsolved mysteries of form, history, and development—all very interesting questions—and backed up with field data.

Like many biologists, Ray believes that the hope of biology lies in shifting the focus of its research from biological time (the thousand-year lifespan of forests) to evolutionary time (the million-year lifespan of tree species).Evolution at least has a theory.Yet an obsession with detail also tends to plague evolutionary research. "I'm frustrated," Ray told me, "because I don't want to study the products of evolution—climbing vines, ants, butterflies, whatever. I want to study evolution itself." Tom Ray's dream is to build an electric evolution machine.Using a black box that "contains" evolution, he was able to shed light on the historical laws of ecology—how rainforests descended from earlier forests, how ecosystems emerged from the same primordial forces that produced species .If he could develop an evolution machine, he would have a test bed on which to conduct real ecological experiments.He can pick a colony and experiment over and over with different combinations, such as generating a pond without algae, a forest without termites, a meadow without gophers, or, to avoid generalizations, a jungle with gophers and grass with algae.He could start by creating a virus and see where it all takes him.

Ray had been bird-watching, insect-collecting, and flower-growing—no computer nerd at all—and he was convinced that such a machine could be built.He remembers a decade ago when he was learning the Japanese game of Go from an MIT computer whiz who used biological metaphors to explain the rules of the game.Ray states, "He said to me, 'You know what? It's possible to write a computer program that replicates itself.' At that moment, what I was dreaming of was what I'm doing now. I asked him what to do. , he said, 'Oh, it's a piece of cake, it's not worth mentioning.' But I don't remember what he said, or if he really understood. When I think of that conversation, I put the novel aside and hold the Here comes the computer manual."

Lei's electric evolution machine plan is to start with simple replicas, give them a comfortable habitat, and a lot of energy and space to be filled.The closest thing to these guys are self-replicating fragments of RNA.This daunting task appears feasible.He intends to prepare a culture solution for computer viruses. It was 1989, and news magazines were flooded with cover stories about computer viruses that were worse than the plague, the worst evil technology could achieve.But Ray glimpsed the birth of a new science in the simple code of a computer virus: experimental evolution and ecology.

To protect the outside world (and to keep his own computer from crashing), Ray used a virtual computer to run his experiments.A virtual computer is an intelligent software that simulates a specific computer in the subconscious depths of a real computer.By confining the self-replicating little guys to this shadow computer, Ray isolated them from the outside world, allowing him to mess around with important functions like the computer's memory without endangering the mainframe. "After reading computer manuals for a year, I sat down to write code. Two months later, the little thing was running. Within the first two minutes of running the program, I had obtained creatures that could evolve."

Ray planted a gizmo he had written—80 bytes of program code—on the world he called Earth—and put it in the memory of his virtual computer.This little guy first found a blank memory space of 80 bytes, and then occupied the site with a copy of himself, thus realizing self-replication.In a few minutes, the memory is full of 80 copies. Ray added two important features that turned this Xerox copier-like copying machine into an evolutionary machine: his program occasionally messed up a few bits of code in the copying, and he gave priority to the executioners among these "creatures" .In short, he introduced mutation and death.

Computer scientists have told him that if he randomly changes the computer code (all his creatures are actually code), the changed program may not work properly, or even crash the computer.They decided that the chances of getting a working program by randomly introducing bugs into the code were so low that his scheme was a waste of time.Ray also knew that the perfection needed to keep a computer running was too weak—a bug would kill a process.However, since his creation program runs in his shadow computer, once it mutates into a severely "deformed" thing, his executioner program - which he named "The Harvester" - will kill it, And the rest of his "earth" continued to function. "Earth" actually finds the vulnerable programs that cannot be copied and drags them out of the virtual computer.

The Harvester, however, will pass by the very few valid variants, that is, those that happen to form a genuine replacement program.These legitimate variants are capable of replicating and producing other variants.If you run "Earth" for a billion computer cycles like Ray did, out of those billion chances, there will be a surprising amount of randomly generated stuff.To make the system more dynamic, Ray also age-stamped the little things he built so that the older ones would die. "The Harvester kills the oldest guys as well as the toughest ones," Ray said with a laugh. In Earth's first run, random mutation, death, and natural selection all played a role.Within minutes, Ray witnessed the birth of an ecosystem—a system of those new creatures that competed for the computer's cycles.Competition rewards smaller guys because they require fewer cycles, while ruthless Darwinian evolution weeds out ravenous consumers, sickly species, and old guys.Species 79 (one byte less than 80) are lucky.It worked like a charm and quickly crossed 80. Ray also found something very strange—a variant that was only 45 bytes long.Its code efficiency is extremely high and outnumbers all other variants. “I was blown away by how quickly the system optimized itself,” recalls Ray. "Survivors in the system have shorter and shorter genes, and I can map that rate." On further examination of 45's code, Ray was surprised to find that it was a parasite.It only contains the code needed to survive.In order to reproduce, it "borrowed" 80's breeding code to reproduce itself.As long as there are enough 80 hosts around, 45 will thrive.But if 45 is too much in the limited range, there won't be enough 80 to provide a source of reproduction.As 80 goes down, so does 45.The couple dances a co-evolved tango, advancing and retreating like a fox and a rabbit in a northern woods. “It seems to be a universal property of life that all successful systems attract parasites,” Ray reminded me.Parasites are so common in nature that hosts quickly co-evolve immunity against them.The parasite then evolved strategies to trick that immunity.The hosts then co-evolve defenses against them as a result.These actions are not actually alternating, but two forces that are constantly interacting. Ray learned to conduct ecological experiments in "Earth" with parasites.He put 79 into his "culture fluid" because he thought 79 might be immune to parasite 45.indeed so.But as 79 thrived, a second parasite evolved to prey on them.This one is 51 bytes long.When Ray sequenced its genes, he found that the reason 45 became 51 was caused by a "genetic event": "Seven instructions of unknown provenance replaced one somewhere in the middle of 45." order," turning an incapacitated parasite into a potent new species.But that's not all - a new species that is immune to 51 has evolved.And the process is still going on. In the long-running "culture solution," Ray discovered hyperparasites that host other parasites: "Hyperparasites are like neighbors who steal electricity from your wires. They use your electricity to , you pay the electricity bill, and you're kept in the dark." On "Earth," organisms like 45 found themselves without having to "carry" a lot of code to reproduce themselves because there was enough of it around them.Ray quipped, "It's like we use amino acids from other animals [when we eat them]." On further inspection, Ray found that hyper-hyperparasites thrived, and the parasitism escalated to the third level.He discovered the "social cheater" - a creature that uses the code of two cooperating hyperparasites (the "cooperative" hyperparasites also steal from each other!).Social cheaters need a fairly well-developed ecosystem.As for hyper-ultra-ultra-parasites, I haven't seen them yet, but maybe there are.In his world, this game of getting something for nothing may never end.