Ant evolutionary conundrum

How can natural selection favour the evolution of non-reproductive worker ants?
23 December 2015

Interview with 

Martin Nowak, Harvard University


Social insects, like ants, seem to fly in the face of evolution, because workers have to sacrifice their own opportunity to reproduce for the overall good of the colony. Ant social evolutionThis is particularly hard to reconcile with situations where the queen has mated with multiple males, meaning that the workers, were they to breed, would have the potential to introduce greater diversity into the population. But now a new mathematical analysis, by Harvard's Martin Nowak, shows that, so long as the reproductive rate of the colony remains highly efficient, it's still a viable evolutionary proposition, even if the queen has mated with multiple males. Chris Smith hears how..

Martin - Socially, insects are specific for most biological adaptation because they're hugely successful. So only 2 per cent of insect species are social but they make up 50 per cent of insect biomass. So here, we have a phenomenon called the super organism, which is seen as a distinct form biological organisation. The big question is, why does natural selection build individuals that do not reproduce themselves but help an other individual namely the queen reproduce? It is a conundrum that was already realised by Charles Darwin.

Chris - How do we think this came about then? What's the origin of this behaviour and how do we think natural selection allows it?

Martin - Natural selection can allow this behaviour if the overall efficiency of this form of reproduction is better than the solitary one. So what we have to say is that solitary reproduction is straightforward exponential power of growth. So if you want to change that lifestyle, you have to beat the power of exponential growth. That can happen if basically, the queen gets a huge benefit from the offspring that stay around. That benefit has to be coming in terms of a much greater rate at which she can produce fertile offspring and also defend those offspring.

Chris - How have you proved that those are the constraints and those are the requirements for this to happen? How did you do that?

Martin - So, that's a simple calculation in terms of the mathematics of evolutionary dynamics. So you set up the different genetic variants and you write down equations of population genetics to ask what is the most successful, the more successful way of reproduction. In that paper here, we actually analysed a second problem and this is assuming that insect colonies are already there who should lay the male eggs. In the ants for example, the queen makes male and female eggs. The workers that are not fertilised, they could only lay male eggs. In the colony where there's one fertilised females namely the queen and all other females are not fertilised, it is very clear that all the female offspring have to come from the queen. But there's a conflict over who lays the male eggs because they could come either from the queen or from the worker.

Chris - In the average colony though the workers are going to carry any genetic information relevant to being male which is identical to that genetic information for maleness carried by the queen. So you wouldn't actually know and there wouldn't really be any advantage, would there or perhaps there would to the workers having male offspring or the queen laying male offspring?

Martin - The genetic material is not exactly identical because the queen could have made it with one male or could have made it with several males. It was previously believed before we did the analysis for our paper that it is very important for the queen to mate with only this one male and this would then allow for the evolution of sterile workers. If the queen has mated with several males, workers would still prefer to lay their own eggs.

Chris - What is the actual status quo though? Does the queen routinely mate with multiple males because bees certainly do don't they?

Martin - It depends on the species. Some species queens lay mate is one male and other species queen mate with several males. So it does depend on the species. What our analysis actually shows is that for the evolution of non-reproductive workers, whether or not, the queen has mated with one or several males is actually not the primary question. So we find that alleles that lead to worker sterility can be favoured by natural selection in both cases. In cases where the queen has mated once and in cases that the queen has mated several times. The more crucial relationship is how does the colony efficiency increase with the fraction of sterile workers and that's a quantity that was largely neglected in the previous literature.

Chris - So putting all of this together, how does this paper move us forward and what are the ultimate conclusions here? What have you found that changes our thinking?

Martin - I think the way how the paper moves us forward is by developing a precise mathematic theory for social evolution. A theory that really asks the question, how does natural selection act on genes that operate in colonies and in groups of social insects? I think this is what really leads us forward.


Add a comment