Fungus-farming ants (the “attines”) are one of the most widely recognized groups of ants, particularly in the form of leaf-cutter ants. They even feature prominently at the beginning of the best classic Disney movie, Ant Lion King:


But a lesser-known fact is that fungus-farming ants, as a group, represent a wide range of mutualistic specialization. The attines are separated into two basic categories: the “lower attines” and the “higher attines”. Lower attines facultatively farm fungus, which is to say that the fungus they farm can freely live independently of their ant hosts. In contrast, the fungus farmed by the higher attines are obligately associated with their hosts, meaning they cannot live freely independently of their ant farmers. So, two key questions about the evolution of ant agriculture are: How did fungus farming originate in the ants, and how did the jump from lower attine farming to higher attine farming occur?

Enter Dr. Michael Branstetter and friends. In a study published last week, these researchers report their findings from an evolutionary investigation of the geographical origins of ant farming. Using new, robust genomic techniques coupled with species locality information, Branstetter and colleagues found that ant agriculture originated in the tropical rainforests of South America, which is what pretty much everyone expected. But, quite unexpectedly, they also found that higher attine agriculture – the obligate mutualism with the farmed fungus – originated in seasonally dry habitat.


Acromyrmex ants with their friendly fungus. Photo: Alex Wild

Why does this matter? Well, the longstanding consensus among myrmecologists has been that higher attine farming evolved in tropical rainforests, given the abundance of farmed fungal species and their relatives in such habitats. However, it would seem that the evolution of domesticated fungal species, which is what the obligate mutualistic fungal species represent, must involve some level of isolation from free-living members of the same species. It seems unlikely that this apparently necessary isolation could occur in wet, tropical rainforests with abundant fungal populations, while it seems much more likely that in dry habitats, such isolation would occur and therefore drive a necessity to rely on ant hosts for survival in dry habitat. So, the evolutionary geographic findings of Branstetter and colleagues both challenge a prevailing view of ant agriculture evolution and provide a reasonable mechanism that bolsters their claim.

This study is therefore a very nice investigation of ant agriculture evolution that provides a robust hypothesis of the habitat where it happened.