Species of ants that practice a complex form of fungi agriculture developed their knack for farming about 50 million years ago and have employed several different, successful strategies to culture their crops in the intervening millennia, according to a study published in PNAS today.

Smithsonian Institution entomologist Ted Schultz told The Scientist that humans - who developed an agriculture that is still fraught with difficulties only about 10,000 years ago - could learn from the nearly 50 million year history of ant agriculture. "There has been a co-evolution [between ants and the fungi they farm] that's very much like the kind of co-evolution that has taken place between humans and our cultivars," he said. Some fungi-farming ant species have even weathered crop diseases by developing their own forms of antibiotic control. "It's possible that we could learn something from these incredibly ancient, stable states."

Schultz and his colleague Sean Brady constructed the most complete phylogeny for 65 ant species belonging to tribe Attini, a group of ants found throughout South and Central America that sow, grow, fertilize, weed, harvest and eat gardens of fungi.

"Without this phylogeny, you could only speculate about the stepwise accretion of behavioral and ecological characters that led to some complicated evolutionary endpoint," said Schultz, who specializes in the evolution of attine ants. "With the tree and with well supported nodes that tell you what those steps were, you can move beyond speculation."

Schultz and Brady used standard PCR techniques to sequence four nuclear, protein-coding genes, placing ants known to practice five distinct systems of farming their fungal crops on an evolutionary tree based on the molecular data. The phylogeny concurs with speculation that fungi farming arose in ants about 50 million years ago in a warmer, more humid South America. These original ant agriculturists - whose descendants and farming methods survive today - grew a diverse array of fungal species by collecting fresh detritus from leaf-littered forest floors to fertilize their crops.

Over the next 30 million years or so, four other farming strategies arose as the ants and the fungi they cultivated evolved. Schultz and Brady identified potential transitional species between these different systems of fungus farming. The most recently derived form of ant agriculture that the researchers placed on the tree is practiced by the leaf-cutter ants of tropical Central and South America. Instead of scrounging from the forest floor, leaf-cutters tailor-make their fertilizer by cutting leaves and leaf fragments, often from living plants.

Leaf-cutters, said Schultz, have complex social and agricultural systems, with several different sizes of worker ants to perform the various tasks necessary to grow symbiotic, "domesticated" fungi that do not occur outside of ant gardens. Schultz and Brady found, much to their surprise, that this system evolved fairly recently by insect standards. "If I had to pick the biggest surprise," Schultz said, "it would be that the leaf-cutter ants could be less than 10 million years old."

Schultz said that leaf cutters are the dominant herbivores throughout the New World tropics, with some colonies containing the biomass of a large mammalian plant-eater. "They've taken fungus growing and turned it, arguably, into one of the greatest ecological success story in all animals," he said.

According to Schultz, this phylogenetic reconstruction is only the start of understanding the diversity and evolution of fungi-farming ants. More than 230 species of the ants have been described, and Schultz said that there may be more agricultural systems than the four derived forms that he and Brady addressed among them. Schultz also said that he and Brady's tree does not answer the question of how or what ant moved from not farming fungi to an agricultural lifestyle. "This tree is essentially mute about that point," he said.