Ants Practice Nepotism, Study Finds

The highly social and complex world of ants is not void of selfish acts. Worker ants of the species Formica fusca apparently can distinguish who their closest relatives are and kill their more distant relations.

"That workers capitalize on this ability simply means that the workers use the information they have to enhance their genetic contribution to future generations," said Liselotte Sundström, an entomologist at the University of Helsinki in Finland.

Science has long held that ants who share the same genetic make up have a vested interest in favoring their closest kin at the expense of the social cohesion of the entire ant colony. But evidence for ant nepotism has been difficult to find.

Sundström and colleague Minttumaaria Hannonen took advantage of advances in genetic analysis to prove the theory correct. They report the results of their study in the February 27 issue of Nature.

Kenneth Ross, an entomologist at the University of Georgia in Athens, said that researchers have looked for nepotism in social insects for the last 10 to 15 years and not found it. Given that fact, Ross said he was surprised at the success of Sundström and Hannonen's study, which he described as nicely done.

"The theory tells us it should be there," he said. "But the implication from data from the last 10 to 15 years tells us it shouldn't be there. In light of these negative results, I am somewhat surprised."

Ant Study

Unlike bee colonies, ant colonies are often ruled by more than one queen. Among some species, such as fire ants, a single ant colony can have as many as 500 queens. On the surface, the ants in such colonies seem to work cooperatively for the interest of the entire group.

Scientists have often wondered if evidence for nepotism is present in such a complex social environment. But each time they have looked for it their results come up negative. Sundström said that perhaps their studies were hindered by logistical constraints.

"For instance, there has not been any way to very accurately distinguish the brood produced by different queens, especially in the field where the actual number of queens is not known," she said.

In addition, Sundström said scientists have not had a way to accurately determine how closely individuals within a colony are related to each other. As a result, they could not predict which ants should favor others if nepotism were indeed taking place.

But advances in genetic analysis enabled Sundström and Hannonen to test for nepotism. The pair were able to distinguish one ant's DNA from another's and use certain genetic markers to determine what queen was the mother of various ant offspring.

They conducted their experiment on laboratory colonies with two mother queens. The colonies were set up in the early spring, just before the queens lay their brood. At the time the colonies were established, they consisted of only the queens and adult worker ants.

"When the queens started to lay eggs in our laboratory colonies, we collected a sample of about 50 eggs and left about the same amount in the colony. The eggs left in the colony were allowed to mature into adults," said Sundström.

The researchers then screened all the queens and eggs and a sample of mature individuals from each colony. Based on their genetic analysis, the researchers determined the proportion of eggs laid by each queen and also the proportion of the adult brood that belonged to each queen. The researchers then calculated how the proportions had changed.

"We predicted that if the workers on average are more closely related to one queen than the other they could gain genetic benefits by favoring the offspring laid by this queen," said Sundström. "Conversely, if the workers are equally related to both queens they would have no incentive to favor the offspring of one queen over the other."

If the researcher's theory is correct, in colonies where there is a high proportion of worker ants more closely related to one queen than the other, the brood belonging to the queen with all the close working relatives will get a higher proportion of surviving offspring.

The results from Sundström and Hannonen's experiment were in line with this prediction, implying that "the workers manipulated brood composition to fit their interests," said Sundström.

Ross says there is little reason to question the methodology of Sundström and Hannonen, but their result does raise the question of why nepotism was found in this species of ant when it has not been found in so many others.

"You can't argue that the earlier studies were systematically flawed, so why is this one different?" he said.

Nepotistic Behavior

While the researchers do not yet have direct evidence that workers remove eggs or larvae to achieve their nepotistic goals, they say that such action is the most plausible explanation for the observed nepotistic behavior among Formica fusca ants.

"The rationale behind this argument builds on kin selection theory," said Sundström. "In social insects such as ants and bees, workers are sterile females, which help their mother, or some other female relative to reproduce."

Since the workers share half their genes with their mothers, they also share half their genes in common with their siblings. Since siblings have more genes in common than do cousins, if a worker wants to see its genes passed on to the next generation, it's to the ant's advantage to favor its close relatives, assuming it can tell which ants are its closest relatives.

"Accurate kin discrimination abilities are a prerequisite for nepotism. Because we found evidence for nepotism, we can conclude that workers have the necessary ability to discriminate between closer and more distant relatives," said Sundström

That nepotism is now known to exist at least among one species of ant in one situation, Ross says it puts ant nepotism research back 15 years, posing such questions as, "How common is it? How universal is it? What sort of conditions should we expect to see it in?" he said.

Sundström finds such questions to be the most rewarding aspect of her research. "Every time one resolves one question, there will be ten new ones emerging."