Changes in ecological settings are thought to drive the evolution of one species into many — a process called adaptive radiation. Darwin's finches are a classic example, evolving into species with different beaks adapted to food availability in their island environments. However, few have considered the converse: that evolution might affect ecology. In an experimental demonstration of this idea, Luke Harmon of the University of Idaho in Moscow and his colleagues show that fish speciation can have profound effects on freshwater ecosystems.

Harmon credits his co-author Dolph Schluter of the University of British Columbia in Vancouver, Canada, with the idea of testing the reverse process. For the work, Harmon chose freshwater threespine sticklebacks (Gasterosteus aculeatus), which evolved from marine sticklebacks fairly recently — in the 10,000 years since the glaciers retreated from British Columbia. In some lakes, the sticklebacks evolved further into two separate subspecies: a bottom-dwelling fish that feeds on invertebrates in the mud, and an open-water fish that skims near the surface eating zooplankton. For Harmon, this represented the very first step of adaptive radiation — the divergence of one species into two. But what happens to a lake ecosystem as a result of that speciation?

Harmon's group created artificial ponds in 1,136-litre tanks and added fresh pond muck to deliver invertebrates and zooplankton. They then introduced either sticklebacks from single-species lakes, which Harmon calls “generalists”, or the more specialized sticklebacks from lakes with two subspecies.

As the 10-week experiment progressed, some tanks bloomed with algae, whereas others showed limited growth. The researchers measured changes in plant and animal populations in the tanks, but the results, Harmon says, were “pretty chaotic. We could tell that there was an effect, but we couldn't explain it.”

One part of Harmon wanted to be done with it. “I could've written a paper at that point that said, 'Evolution matters, but in unpredictable ways',” he says. The other part of him believed that important answers lay in his tanks' muddy waters, so he sought advice from Blake Matthews at the Swiss Federal Institute of Aquatic Science and Technology in Kastanienbaum, an expert in aquatic ecosystems. He pointed out dozens of physical measurements that Harmon's team had overlooked, such as the amount of dissolved organic carbon (DOC) and light transmission.

When the researchers ran the experiment a second time and included these measurements, they found that DOC was a key factor. DOC is essentially organic detritus and gives ponds their tea-like colour. In tanks with the one generalist fish species, most DOC molecules were small enough that light could penetrate to the bottom of the tank, and algae bloomed. In tanks with the two specialist subspecies, DOC molecules were larger, blocking out light. The authors suggest that differences in feeding between the generalists and the the two specialists is what alters DOC size (see page 1167).

Harmon says he didn't appreciate how much “organisms can alter the physical structure of their environment through behaviours such as feeding”. And he's not the only one. Lake water chemistry, he adds, “is one of those huge areas of ecology that most evolutionary biologists don't think about”.