1. The Leaves Are Soil for a Fungus, and the Fungus Is the Food
Leafcutter ants (the best-known groups are called Atta and Acromyrmex) cut fresh leaves and carry them to an underground nest. Inside, the leaves become the base, or substrate, for a fungus that lives with the ants[1]. The fungus grows threads called hyphae. Some threads swell into little round tips called gongylidia, and these are what the ants eat[1].
Fungus and ants need each other. This kind of partnership is called a symbiosis. The ants feed and protect the fungus, and the fungus feeds the ants. Because the ants supply the plant material and the starter fungus, scientists call it a kind of farming.
2. How the Ants Turn a Leaf Into Garden Soil
Where does the fertilizer come from? A research team studying Acromyrmex echinatior found that the ants use a fluid from their own waste. The ants spread tiny droplets of fecal fluid on new leaf material. Inside those droplets were enzymes (natural chemical tools) that break down the walls of plant cells. Seven of the 33 proteins the team identified in the droplets were enzymes of this kind[1].
The surprising part is where the enzymes come from. They are made by the fungus, mostly in the swollen gongylidia. When an ant eats the gongylidia, the enzymes pass through its body without being digested, and they are still working when they come out[1]. In this way, a tool made by the fungus is eaten, carried through the ant, and used on the next leaf. The researchers showed that the enzymes appeared in the droplets only when the ants had been able to eat from a fungus garden[1].
Fungi that attack living plants use the same kind of enzyme to get past plant cell walls, and the authors point out the likeness[1].
3. How Ants Learn to Stop Carrying a Leaf That Hurts the Fungus
Here is the puzzle. A leaf that is harmful to the fungus is not necessarily harmful to the ants. How could the ants know which leaves to avoid?
In one experiment, researchers soaked natural leaves in a fungicide called cycloheximide, which hurts the fungus but which the ants cannot detect. They used small lab colonies of Acromyrmex lundi. At first the ants carried the treated leaves into the nest as usual. About ten hours after the leaves went into the garden, the ants began to reject that kind of leaf. The refusal was specific to the plant species that had been treated, and it spread even to ants that had not carried those leaves themselves but had touched the garden. The ants kept avoiding the plant for at least nine weeks. When the researchers forced the ants to handle untreated leaves of that plant, they accepted them again after about three weeks[2]. In other words, the "no" comes late, it targets one plant, and it can be unlearned.
A second team found that the colony's waste dump carries a clue. Waste from a garden fed with treated privet leaves was enough to teach untouched ants to avoid privet, and about one-third of the marked foragers visited the dump on their own. Even ants shut in the dump for three hours learned to avoid the plant that had been used[3]. The authors suggest that clues at the dump let foragers predict which plants are unsuitable without ever having seen the damage in the garden. What exactly the clue is (a smell, a sign of sick fungus, or both) is not settled by their summary.
Both studies used small colonies of Acromyrmex in the lab. We cannot assume the same thing happens in the same way in Atta colonies or in the wild.
4. Why Each Nest Grows Just One Kind of Fungus
Farmers who plant a single crop everywhere risk losing it all to one disease. Do leafcutter ants take that risk? A team studied nests of two Atta species, Atta texana in Texas and Atta cephalotes in Panama, using DNA fingerprinting. They resampled the nests over a five-year period. Except for tiny mutations, every garden in a nest held the same fungus, and nothing was replaced during the five years[4]. The paper describes Atta nests as enormous (up to 5,000 gardens, with lifespans of 10 to 20 years), so keeping one fungus going for years in such a nest is a real feat[4].
The researchers also looked at the young queens that leave to start new colonies. A queen carries a starter ball of fungus, which the paper counts as at least several thousand fungus cells[4]. So a new garden starts from a piece of the old one.
Why would a colony stay with one fungus? The authors explain that one fungus removes competition between rival fungi, a competition that could favor traits harmful to the ants[4]. This uniformity is also a risk, as with any single crop, but I did not read a source that measured how the ants cope with disease, so this article leaves that out.
5. How Old Is Ant Farming, and When Did Leafcutters Appear?
Ant farming is much older than human farming. A study that used fossils and DNA of many fungus-growing ants estimated that the first ant agriculture began about 50 million years ago in the tropics of the Americas. Leafcutter ants appeared much later, about 8 to 12 million years ago, and they grow a single fungus species[5]. Dates like these come from computer models and are estimates, so other studies may give somewhat different numbers.
Colony size is also impressive. Wikipedia describes leafcutter societies as among the largest and most complex animal societies on Earth other than humans, and mentions a nest with a few million ants[6].
6. How to Look Closer at Leafcutter Ants Yourself
Leafcutter ants live in the Americas, mainly in warm regions, so most people will meet them through videos, books, zoos, or insect museums. If you watch a video, follow one leaf from the plant to the nest and ask where it goes next. You can then draw the loop: leaf, chewed and mixed with droplets, fungus, gongylidia, ant, and back to the next leaf. Try explaining the loop to someone in your own words.
To learn more, read the abstracts of sources [2] and [3]. Ask what a ten-hour delay might mean inside a nest, and what the researchers still did not know when they finished.