BY MARK SHWARTZ

A team of California scientists made headlines four years ago when it reported finding one of the largest insect colonies in the world -- a 600-mile-long subterranean network of Argentine ants stretching from Northern California to the Mexican border. According to the researchers, this "supercolony" is made up of billions of closely related workers -- all direct descendants of a small group of Argentine ants that were accidentally introduced into California more than a century ago.

Graduate student Nicole Heller, left, and biological sciences Professor Deborah Gordon discussed the Jasper Ridge project in Gordon's office. The project analyzed the DNA from ants in the biological preserve and found obvious genetic differences among some of them, suggesting that the ants came from different colonies. Photo: L.A. Cicero

But new studies by Stanford University scientists are raising serious doubts about the existence of a single supercolony running through the Golden State. The Stanford team questions the notion that Los Angeles ants are descended from the same founding population as San Francisco ants, which live 400 miles away. A more likely explanation, they say, is that California has been infested by numerous colonies of genetically distinct Argentine ants during the last 100 years.

Stanford scientists found ants with genetic differences at Jasper Ridge.

Challenging the supercolony paradigm is more than an academic exercise, says Stanford biologist Deborah M. Gordon. Argentine ants have had a major impact in many parts of the world, she says, and understanding how they reproduce and colonize is essential if scientists hope to develop realistic strategies that will keep their populations in check.

"Our data show that it's not the case that the whole California coastline is one genetically homogenous supercolony," says Gordon, a professor of biological sciences. "We find a lot of genetic diversity here, which indicates that there were probably many introductions in the past."

An authority on ant behavior, Gordon has spent more than 20 years studying native and invasive species, including the Argentine ant, or Linepithema humile, which has displaced many of California's indigenous ant species since it was first introduced in the state around 1900.

Common scents
For years, Gordon and other scientists have tried to figure out why these tiny invaders are such remarkably successful colonizers, not only in the United States but also in the Mediterranean and other parts of the world.

In May 2000, biologist Neil D. Tsutsui and his colleagues from the University of California-San Diego (UCSD) came up with an apparent solution (for California, at least). Their results, published in the May 2000 Proceedings of the National Academy of Sciences (PNAS), were based on genetic and behavioral experiments on worker ants in California and Argentina.

In field studies conducted in Argentina, the Tsutsui team observed that ants taken from different nests often fought when placed side by side in an experimental vial. But in California, the scientists found that ants rarely engaged in combat, even if they were collected from nests hundreds of miles apart.

Why would workers in Argentina exhibit intercolony aggression but their California cousins behave so downright friendly? The Tsutsui team turned to genetics for an answer. Previous studies of other ant species have shown that insects from the same colony don't fight among themselves, because each one carries a common scent identifying it as a nestmate.

Could it be that laid-back California ants are all members of the same family, while aggressive workers in Argentina engage in territorial battles because they belong to different colonies that are genetically distinct?

To find out, the UCSD scientists compared the DNA of workers in California and Argentina by focusing their attention on microsatellites -- short sequences of DNA that occur in unique patterns that are passed down from generation to generation. Unlike genes, microsatellites are nonfunctional and carry no genetic information that would affect the ant's development or behavior. But because microsatellite patterns are inherited, scientists use them as genetic markers to determine if individuals from different populations are related.

Laboratory analysis by the UCSD team revealed that ants in Argentina had more than twice as much microsatellite variation in their DNA than ants from California. As a result of their greater genetic diversity, ants from different nests in Argentina do not recognize each other as family and therefore display territorial aggression, the researchers concluded.

But in California, all of the ants must be closely related because they're so genetically alike, according to the UCSD scientists. That would explain why California workers hardly ever fight with one another -- even if they're from opposite ends of the state.

"We found that if ants are not genetically similar, they are typically aggressive toward each other," said Tsutsui, who is now on the faculty at the University of California-Irvine. He explained that the loss of genetic diversity in the California population probably began a century ago, when the state was invaded by a small population of Argentine ants whose offspring have continued to work cooperatively ever since.

"California's Argentine ants are very genetically similar, probably because they share a common ancestry," Tsutsui said. "There had to have been at least 10 mated queens in the founding population a century ago, but it could have been much larger. It would be as if all of the people in the United States were descended from the Pilgrims who came here in 1620."

Instead of fighting, Tsutsui explained, each new generation of California ants works as a team -- gathering food, wiping out competitors and building new nests year after year. The result: an enormous supercolony that currently extends from San Diego to Ukiah.

These findings "suggest a possible control strategy for the Argentine ant," Tsutsui and his co-authors wrote in PNAS. Their proposal: Consider introducing genetically unrelated colonies of Argentine ants into California to compete with the supercolony. By aggressively defending their territory, these new colonies "should decrease the density of Argentine ants, allowing native ant species to compete more effectively, thereby facilitating the recovery of invaded ecosystems," the scientists offered.

Genetic barriers
But introducing more unwanted ants into California is not a solution, say Gordon and her colleagues, arguing that there are major flaws in the supercolony theory.

"There's a perception out there in the public eye that Argentine ants cover this wide, uninterrupted swath from San Diego to San Francisco," said postdoctoral fellow Krista K. Ingram. "But we've found that there are breaks in their distribution and barriers they do not cross."

In a study published in the journal Ecology last November, Ingram and Gordon analyzed the DNA of 768 ants collected from 48 nests in and around Stanford's Jasper Ridge Biological Preserve, a 1,200-acre research site located in the foothills above the main campus. The preserve is bounded by Sand Hill Road and the Stanford Linear Accelerator Center (SLAC) -- two imposing barriers, from an ant's point of view. The SLAC facility is 1.9 miles long, 30 feet tall and 45 feet deep -- a particularly formidable obstacle for an insect only an eighth of an inch long.

Using microsatellite data, Ingram and Gordon found obvious genetic differences between nests located inside the preserve and those on the opposite side of SLAC and Sand Hill Road, less than 300 feet away. These results contradict the UCSD study, which found genetic similarity up and down the California coast.

"How can they be genetically homogenous across wide, wide scales, when they're not even genetically homogenous across hundreds of meters?" asked Nicole E. Heller, a graduate student in the Gordon lab who has conducted field research in Argentina and California.

"We suspect that the reason we got different results is that the UCSD lab relied on DNA microsatellite markers that look virtually the same in Argentine ants throughout California," Gordon noted. "In our study, we added different markers that are more variable, which made it easier to detect genetic variation among different colonies."

The genetic work at Jasper Ridge shows that there are definite barriers to gene flow, Heller added. "We've also found that the distribution of Argentine ants on Jasper Ridge is patchy," she noted. "There are places where you just don't find any, so it's not like they're blanketing the entire state."

While agreeing that there are "big gaps" in the California population, Tsutsui maintained that the Jasper Ridge study did nothing to undermine the notion of a statewide supercolony.

"The big supercolony in California extends for almost 1,000 kilometers [600 miles]," he said. "Although it is not continuous throughout this length, it is technically a single big colony, because workers can be freely exchanged among different nests without triggering any aggression. That is, throughout this 1,000 kilometers, virtually all the nests that we looked at did not display territorial behavior toward each other."

But, asked Gordon, does it really matter if an ant in San Diego doesn't fight with an ant in San Francisco, if they will never actually encounter each other in nature? "The UCSD lab argues that the lack of aggression among California ants can be explained by the lack of microsatellite variation," she said. "But our Jasper Ridge study revealed plenty of variation, so how can we say that genetic differences have anything to do with aggression?"

European ‘union'
The Stanford study at Jasper Ridge reinforced an April 2002 PNAS study by European scientists, who reported unearthing an even bigger Argentine ant supercolony stretching from Portugal to Italy along the Mediterranean coast -- a distance of about 3,500 miles. Unlike the UCSD researchers, the European team found a great deal of microsatellite diversity in the Mediterranean supercolony but also observed very little aggression among nests. How did they explain this?

"The argument by the European group ran like this," Gordon said. "The reason ants in the Mediterranean don't fight is that, although they are genetically diverse, they have lost the diversity in the genes that are involved in recognition and thus don't recognize each other as different. This runs counter to the argument made by the UCSD lab -- that California ants don't fight because they're related and therefore don't recognize each other as different."

The European and UCSD groups did agree on one point: The lack of aggression within the supercolony is the main reason Argentine ants have been more successful in the Mediterranean and California than in their native Argentina, where frequent territorial battles have kept the population in check.

But Gordon and her colleagues have found evidence contradicting that neat explanation. Heller, for example, has documented instances of both cooperation and combat among colonies in California. And in Argentina -- where fighting prevails, according to the UCSD group -- Heller observed widespread cooperation among densely populated colonies. Her findings raised questions about the hypothesis that aggressive ants have managed to keep Argentina's native population in check.

"So it's not simply a matter of native-versus-introduced or genetic diversity," Ingram said. "There are also ecological factors contributing to the ants' success."

Gordon agreed, pointing to an experiment by North Carolina State University in which researchers were able to turn nonaggressive nestmates into combatants simply by changing their diet.

"It seems really clear that aggression in Argentine ants, as in most ants, is related to how they smell, and that seems to be related to food or environmental influences," Gordon noted. "Again, there isn't any evidence that there's any relationship between microsatellite diversity and aggression."

Science and media
Gordon and her coworkers maintain that too many researchers have accepted the supercolony theory without carefully scrutinizing the data. The Stanford team also is troubled by the recent spate of news reports in major American and European media presenting the supercolony theory as unquestioned fact.

"The image of a homogenous, continuous supercolony overshadows the important questions about invasive Argentine ants: specifically, where they're actually invading, where they can't invade and what happens after they invade," Ingram explained.

"This story has really captured the imagination of the public, and it's somewhat frustrating," Heller said. "But it's such a neat story, people sometimes don't want to hear conflicting evidence."

Added Gordon: "I think real ants are much more interesting than the stories we make up about ants. We'd have better stories to tell if we started with the actual data."

The Gordon Lab's Argentine ant research has been supported with grants from the U.S. Department of Agriculture, the National Science Foundation and the Stanford Center for Evolutionary Studies.