[M.D. Holloway]

Other scientists disagree. Organisms are dealt a finite number of genes and so must choose from a limited menu of evolutionary options, narrowing the directions the organisms can go in a particular environment. “The evolutionary routes are many, but the destinations are limited,” says Simon Conway Morris, a paleontologist at the University of Cambridge in England. As a result, disparate organisms often end up independently developing the same sorts of structures to solve a particular problem. Take eyes. Although the details of how eyes work vary between species, the basic structures are similar.

But since it’s impossible to turn back time (no matter how easy Superman makes it look) and replay all of evolution again, scientists have devised other ways of investigating the issue.

Richard Lenski, an evolutionary biologist at Michigan State University in East Lansing, is among the scientists hitting the rewind button on evolution. Meter-high letters taped to the windows of his lab spell out the lab’s motto: EVOLVE. In the center of the “O,” the face of Charles Darwin peers out toward the football stadium.

Inside the lab, a dozen glass flasks containing what looks like clear liquid swirl in a temperature-controlled incubator. Although the naked eye can’t see them, millions of E. coli bacteria grow in the flasks, doing what the window exhorts. Lenski started the cultures in 1988, intending to follow the course of natural selection for several hundred generations. Now, 20 years later, the cultures are still growing and have produced more than 45,000 generations of bacteria each.

Lenski inoculated each of the 12 flasks with bacteria from the same ancestor, so they all started out with the same genetic deck of cards. Only one gene differed among the bacteria at the beginning — six flasks contain a marker gene that makes the bacteria red when grown on certain media while bacteria in the other six grow white. The marker gene doesn’t affect the strains’ fitness — the term biologists use to describe the capacity of an organism to compete against others to pass on its genes — but it does help researchers distinguish between two lineages of the bacteria during competition experiments.

Each flask contains media with only the minimum requirements for survival — some glucose (a sugar that bacteria use for food) and a few other nutrients. The bacteria replicate, or divide, six or seven times daily, creating a new generation with each round. Each division shuffles the cards and produces genetic changes and mutations, some of which may help or hurt the bacteria’s ability to compete for glucose and win the evolutionary poker game. The next day, a dilution is done, with 10 percent of the culture within a flask transferred to a new flask, and a new hand is played. (Every 500 generations or so, the remaining 90 percent is frozen and stored for later experiments.) The dilution acts as a population bottleneck, randomly selecting a subset of the bacteria (and so a subset of accumulated genetic changes) to continue the experiment.

These 12 flasks “represent the stripped-down bare essentials of evolution,” says Zachary Blount, a graduate student in Lenski’s lab. The environment never changes. No new genes enter the system from migrating microorganisms. And the scientists take no action to affect the course of evolution within the flasks. Only the intrinsic, core processes of evolution influence the outcome, Blount says.

Lenski and his colleagues have watched the game play out, occasionally analyzing DNA to peer over the players’ shoulders and find out what cards they hold. On the surface, the populations in the 12 flasks seem to have traveled similar paths — all have grown larger and become more efficient at using glucose than their ancestors. And many of the strains have accumulated mutations in the same genes. Notably, though, no one strain has developed exactly the same genetic changes as another.

Randomness is an important part of the evolutionary equation, as the experiment illustrates. During the first 2,000 generations, all of the 12 populations rapidly increased in size and fitness. But then cell size changes and reproductive and glucose efficiency gains began to slow down, hitting the evolutionary equivalent of a dieter’s plateau.