Friendly groups of marmots could help settle an enduring debate in evolution – whether the characteristics of a group can be more important to an individual’s chances of survival than the characteristics of that individual. An analysis of their behaviour is the first evidence in wild animals for this evolutionary idea, known as multilevel selection
“What we found is that the group traits are under just as strong selection as the individual traits, if not slightly stronger,” says then at the University of California, Los Angeles. “So the group matters just as much, if not just a little bit more.”
Evolutionary fitness is often portrayed as being about an individual’s strength or ability to fight. But biologists from Charles Darwin onwards have realised that being part of a successful group can be crucial for survival, and that there could be natural selection at the group level as well as at the individual level.
Advertisement
At the other end of the spectrum, evolutionary biologist George Williams suggested in the 1960s that selection occurs at the level of genes, now known as the selfish gene concept.
Others think that because the microorganisms living in and on an individual can play a big role in their success, there is selection for individuals plus their microbiome – together creating what they call the hologenome.
These ideas have been the subject of fierce debate. Some have dismissed the notion of group selection entirely or have claimed it is so weak that it is inconsequential. But many biologists now think that natural selection can act simultaneously at several different levels, known as multilevel selection.
“We think about these things all the time as humans, whether or not we’re calling it multilevel selection,” says Philson. For instance, people are more likely to succeed as part of a successful team or company, he says.
There is already a lot of evidence for multilevel selection from evolutionary experiments done in labs, says Philson. But in this research, animals are typically kept in controlled conditions with no predators. “In order to really understand what is under selection, studying them in the wild is so valuable,” he says.
To do this, Philson and his colleagues took advantage of a long-running study of yellow-bellied marmots (Marmota flaviventer)Â that began in 1962.
These ground squirrels live in groups with 10 to 20 members. The researchers analysed 20 years of data on the interactions between 700 individuals, who were part of 172 groups over this time.
They looked at individual characteristics such as how many other members of a group each individual interacted with. Philson and his team also examined group characteristics, such as how many connections between individuals there were in each group. They then worked out how these traits affected an individual’s chances of surviving – that is, the strength of selection for each trait.
For marmots, being part of a group is important because other group members help spot predators. Better connected groups are less likely to break apart, says Philson, so being part of one boosts an individual’s chance of survival.
But interacting with lots of other group members seems to be costly and reduces survival, the researchers found. So there is selection for more social groups, but against more social individuals.
The analysis also takes account of the fact that individuals are affected by the group they are in, and that there is feedback between the group and individual levels, says Philson. “One of the reasons this is the first study that looks at this in the wild is because you need a lot of data.”
“It is very exciting to see these results in a wild system,” says at Swarthmore College in Pennsylvania. “This preprint demonstrates, in a very elegant way, that position in and structure of social networks are important traits with fitness consequences that cascade across levels of biological organisation.”
“I think their claim that this is one of the first good studies of multilevel selection in nature is correct,” says at the Georgia Institute of Technology. In some fields of biology, such as microbiology, the idea of multilevel selection is now widely accepted, he says. In others, such as animal behaviour, it remains controversial.
Some of those who are sceptical about multilevel selection have no doubt that there is selection of plants and animals at the individual level, says Ratcliff. “Yet all multicellular organisms consist of groups of cells, with natural selection acting on heritable group-level traits. This is, in fact, group selection.”
Philson says that what his study can’t show is whether any of the selective pressures on marmots leads to evolutionary change. This would require demonstrating genetic changes, he says, which is extremely difficult with traits related to social behaviour.
bioRxiv
