[M.D. Holloway]

Morris and Gould that one could ever hope to see,” Lenski says. “Conway Morris ‘wins’ based on the number of changes that fit his pattern, but Gould might prevail if weighted by the profundity of change. Both perspectives are important contributions, and they are not mutually exclusive.”

Now the researchers are watching to see if citrate-eating bacteria will evolve in other flasks, and if citrate eaters will eventually reject glucose and feast only on citrate. Such a transformation would probably herald the birth of a new species. “It would be amazing,” says Blount. “It would be like teenagers who no longer like to eat pizza — they prefer broccoli.”

When the game changes

While Lenski’s experiment takes place in a constant environment, natural evolution must cope with a messier reality. In Steven Finkel’s lab at the University of Southern California in Los Angeles, a long-term experiment is showing how evolution plays out in a closed and ever-changing environment, more like the real world. This allows Finkel to focus on how evolution and environment are interwoven.

Finkel didn’t start out to test evolution in changing environments. The experiment was prompted by a graduate student’s casual remark that E. coli will live a long time. Finkel asked, “How long?”

“A long time,” the student responded.

“So we set up some experiments to see how long they would live, and they just would never die,” Finkel says. The immortal bacterial cultures are teaching scientists a few things about how organisms change their environments and adapt to changes wrought by outside forces.

Some of the cultures in Finkel’s laboratory have been growing for more than five years. The bacteria started out with the same genetic background, or so the researchers like to tell themselves. In reality, each flask started with a single bacterial colony, containing perhaps 50 billion individuals. Given that DNA replication systems aren’t perfect, one in every 10,000 cells probably carries a typo in at least one letter of its instruction manual. Such DNA typos are known as point mutations.

Finkel and his colleagues placed the bacteria in a rich broth full of sugars and many other nutrients and then just let them grow. After a short initial lag, the bacteria began growing like gangbusters, a phenomenon known to microbiologists as “log phase” because the bacteria increase their numbers logarithmically. Once the nutrients start to run out, the bacteria stop growing so quickly and settle into a senescent state. After a few days, millions of bacteria die, spilling their guts into the surrounding media and providing food for survivors.