Atta mexicana - Mexican Giant Leafcutter Ant

Atta mexicana (Smith, 1858) is one of the great leafcutter ants of North America and Mesoamerica. Its range covers much of Mexico and reaches the southernmost tip of the United States, with populations in the Sonoran Desert of Arizona: it is the only Atta that lives in that state. It inhabits everything from arid regions and thorn scrub to tropical dry forests, riparian areas and mountain environments, and shows considerable ecological plasticity.

It is also one of the most spectacular ants you can keep. It does not eat the leaves it cuts: it uses them to grow a fungus that feeds the whole colony. Its nests in Arizona cover hundreds of square metres, its colonies probably number millions of workers and they can live for more than twenty years. In this care sheet we always distinguish between what is documented by science, what other keepers report and what we have learnt keeping it at Antomas.

Biology

The leafcutter that reached the desert

Leafcutter ants of the genus Atta are, above all, ants of the American tropics. A. mexicana is one of those that stray furthest from them: it is found across much of Mexico, reaches Guatemala, El Salvador and Honduras to the south and, to the north, southern Arizona. There, in Organ Pipe Cactus National Monument and its surroundings, right by the border, lies its northern limit, in the heart of the Sonoran Desert. On the Gulf of Mexico side it also extends northwards, through Tamaulipas, Nuevo León and Coahuila, although, according to surveys by amateurs, it does not come close to the Texas border.

What climate does it prefer? A 2023 study of the Atta of Mexico modelled its climatic niche and found that the most important variable is isothermality: the ratio between how much the temperature changes within a day and how much it changes over the year. It is followed by the rainfall of the wettest month and that of the warmest quarter. The most favourable conditions for the species are in the pine and oak forests of the Sierra Madre Occidental and in the dry forests of the Bajío. This fits with what is seen in Arizona: an ant of climates with summer rains, which shelters from heat and cold underground. The same study warns that, under the highest-emissions scenario, it could lose more than half of its climatically suitable area by the end of the century.

Close cousin of Atta texana: the Archeatta clade

It was described by the British entomologist Frederick Smith in 1858 as Oecodoma mexicana, from queens and males with a locality as terse as “Mexico”, which is where its name comes from. The reference specimens are kept at the Natural History Museum in London. It then spent almost a century without a name of its own: Roger moved it to Atta in 1863, but he himself, Mayr, Forel and others considered it a mere synonym of Atta fervens, and Emery downgraded it to a subspecies of A. insularis, the Cuban leafcutter. Gonçalves placed it in the subgenus Archeatta in 1942, and Borgmeier redescribed it in 1959 as a valid species, which is what it is today.

Three species of Atta live in Mexico: A. mexicana and A. texana, of the Archeatta group, and A. cephalotes, of the group Atta in the strict sense. A. texana, the leafcutter of Texas and Louisiana, is its closest relative, and the two are the only Atta in the United States. This is how to tell them apart:

• A. mexicana: the lobes at the back of the head and the first segment of the gaster without erect hairs; the notch between the lobes shallow, and the pronotal spines relatively short.
• A. texana: with erect hairs, a matt integument, a deeper notch and long pronotal spines. As a curiosity, its head gives off a minty smell when crushed.

And in captivity they also differ in character: in our experience, A. mexicana has a single queen per colony, whereas A. texana queens can live together (see ‘The queen who travels with her garden in her mouth’).

Confusion with A. cephalotes is less likely: it belongs to another subgenus and to humid tropical forest, and their climatic niches barely overlap in a small area of Mexico.

Cities beneath the desert

The Arizona figures come from three decades of censuses at Organ Pipe Cactus National Monument, and they are dizzying. Large colonies can occupy a central area of about 20 × 20 m, can come to exploit more than 2 hectares of land and use lateral tunnels of up to about 130 m to reach their foraging areas. An established colony can bring roughly 200 kg of dry plant matter a year into its farming system. Nests are usually 75–200 m apart, sometimes only 30–40 m, and colonies almost never move: the size of the craters and refuse piles even makes it possible to estimate the age of a nest.

Unlike Acromyrmex octospinosus, Atta mexicana develops clearly multi-chambered nests. A mature society maintains numerous fungus gardens spread across different underground chambers, linked by an extensive network of tunnels. Nobody has yet published the complete excavation of an A. mexicana nest, but it has been measured in other species of the genus: one A. laevigata nest had more than 7,000 chambers and reached a depth of up to 8 m.

This architecture keeps temperature and humidity much more stable than at the surface. Even the Sonoran Desert populations keep their gardens in deep, sheltered chambers. And a nest like this needs to breathe: fungus and ants consume oxygen and give off CO₂, heat and water vapour. In the giant nests of A. vollenweideri, a South American leafcutter, it was shown that the driving force of ventilation is the wind: stale air leaves through the central, higher tunnels of the mound, and fresh air enters through the peripheral ones.

In captivity, this organisation is reproduced much better with several connected fungus containers than by trying to keep a large colony in a single gigantic container. As the population grows, new modules are added, and it is the ants themselves that decide how much fungus they keep in each one (see ‘Recommended set-up’).

Leafcutters on the night shift

In the desert, heat rules. In 1979 Alex Mintzer followed a colony in northern Sonora that harvested over an area of about 8,000 m², with surface trails and underground tunnels radiating from the nest in all directions. The workers foraged by day in winter and spring, and switched to foraging mainly at night in summer. At Organ Pipe, surveys in December and January found them foraging by day, in the middle of winter. Those tunnels of more than a hundred metres also let them reach the plants without crossing the scorching ground.

What do they cut in the desert? In Arizona, the use of palo verde (Parkinsonia, formerly Cercidium), ironwood or palo fierro (Olneya tesota), mesquite (Prosopis), catclaw (Acacia) and creosote bush or gobernadora (Larrea tridentata) has been documented. To these are added numerous annual plants that are only available in certain seasons. The diet of a wild colony varies enormously with locality and season.

Don’t be surprised if, in summer, your colony works more at night: it is in its nature.

A garden renewed every month

Atta mexicana does not consume leaves like a conventional herbivore. It uses plant material to grow a symbiotic fungus, Leucoagaricus gongylophorus, which is the functional centre of the colony.

The workers cut leaves, flowers and other tissues, carry them to the nest and subject them to progressive processing: the material is cleaned, cut into smaller and smaller fragments, chewed and conditioned before being incorporated into the garden. The smallest workers then continue the work directly inside the fungus.

The fungus develops specialised nutritive structures, swollen hyphae rich in nutrients that specialists call gongylidia, which are especially important for feeding the larvae. In practice, the fungus is the food of the whole colony: in Arizona it is considered the only food of larvae and adults. The workers also make use of the plant juices released when the leaves are cut and processed.

And the garden is not eternal. The plant substrate of each portion of fungus is exhausted in about a month; the workers then remove it and carry it out to the refuse piles. That is why a leafcutter colony never stops working: the garden is constantly being renewed.

Nor should the garden be seen as a mere mass of fungus. It is a living ecosystem, where the cultivated fungus lives alongside bacteria and other microorganisms whose composition varies even between different parts of the same garden. A healthy colony maintains that ecosystem by cleaning, removing ageing areas, selecting plant material and constantly carrying out waste.

Taking the rubbish outside

Here A. mexicana breaks with the norm. Most Atta keep their waste in underground refuse chambers: in A. capiguara, for example, about 4–6 m from the gardens. A. mexicana, like A. colombica, does the opposite: it takes its rubbish out of the nest and piles it up on the surface. In Arizona, large colonies dump their spent fungus in sizeable outdoor piles.

This is no minor detail. Leafcutter waste is dangerous: it is full of spent fungus and of microorganisms harmful to the colony. In A. colombica it has been seen that the workers of the waste dump form a separate work caste, among the most “expendable” in the colony, with hardly any contact with the garden. The lesson for keeping them is clear: a separate, external and dry waste dump that you can clean without touching the fungus (see ‘Waste dump’).

Antibiotics from a gland, not from bacteria

Every garden has its pests, and the leafcutters’ garden has a highly specialised one: Escovopsis, a parasitic fungus that lives only in the gardens of fungus-growing ants and feeds on the cultivated fungus. To defend themselves, many fungus-growing ants carry a whitish layer of bacteria on their cuticle, Pseudonocardia, which produce antibiotics.

But Atta plays a different hand. A 2009 study of nine species from five genera of fungus-growers compared their defences.

• Acromyrmex and Trachymyrmex keep abundant Pseudonocardia on the cuticle: it is the white powdery coating you can see, for example, on Acromyrmex octospinosus.
• Atta has lost that visible layer, or reduced it drastically. Instead, when a pathogen appears, Escovopsis included, the workers step up their grooming with the secretion of the metapleural gland, a gland on the thorax that produces antimicrobial substances.

The authors link this more elaborate chemical defence, and the lower dependence on bacteria, to the enormous size of Atta colonies. Later, one of its weapons was identified: the phenylacetic acid of the metapleural gland, with which Atta holds Escovopsis in check, whereas Acromyrmex relies on its bacteria. On top of that comes “weeding”: the workers locate diseased or ageing pieces of fungus, pull them out and carry them to the waste dump.

For keepers: in A. mexicana there is no need to look for the white coating of Acromyrmex. A clean, shiny worker is normal.

Leaf tasters

Atta mexicana is polyphagous: it exploits a wide variety of plants. But it does not cut just any plant it comes across. Leaf toughness, water content, the presence of hairs or trichomes and chemical composition all influence the choice. Two studies carried out with this species have measured it.

The 2020 study. In the laboratory, the workers preferred the leaves of rose (Rosa × alba) and of two Mexican trees, Trema micrantha and Styrax glabrescens, and rejected those of coffee, guava and mandarin (Citrus reticulata). The rejected leaves contained compounds, some with an antifungal or deterrent effect: coffee, many alkaloids and phenols such as chlorogenic acid; guava, shikimic and ellagic acids; and mandarin, scopoletin. In the case of mandarin, however, the authors attribute the rejection more to the physical characteristics of the leaf than to its chemistry.

The 2026 study. Published in the journal Botanical Sciences with colonies from Veracruz, it compared three plants:

• Palicourea padifolia, known as flor de mayo, cafecillo or flor de cera depending on the region;
• Platanus mexicana, the Mexican sycamore (álamo blanco, álamo mexicano or haya);
• Eriobotrya japonica, the loquat.

The ants showed a very marked preference for P. padifolia. Its leaves were softer, lacked trichomes and contained a lot of water. The loquat and the Mexican sycamore had tougher, hairier leaves, with more phenols, flavonoids and tannins, and were harvested much less intensively. In addition, the extract of P. padifolia favoured the growth of the fungus, whereas those of the other two species caused moderate inhibition. None proved toxic to the ants.

In the case of the loquat, the workers even removed the trichomes from the leaves before cutting them, a very delicate manipulation that shows how carefully they assess and prepare the material before farming it.

The fungus gives the warning: delayed rejection

Workers may enthusiastically cut a plant that, nevertheless, harms the fungus. How do they find out they have made a mistake? Through the fungus itself. As early as 1996 it was proposed that the fungus “controls” the behaviour of leafcutters, and later experiments have confirmed it in a very literal way.

In a 2008 study with another leafcutter, Acromyrmex lundi, colonies were offered leaves treated with a fungicide harmless to the ants. At first they cut them and carried them to the garden quite normally. About 10 hours after incorporating them, they began to reject them, and the rejection lasted at least 9 weeks. It was also specific: it only affected the plant species that had been treated, and over time it could be reversed. In Atta colombica, that “memory” lasted up to 18 weeks. The ants learn to associate the plant with the damage their garden suffers, and they do not forget it for a long time.

This phenomenon has two very practical consequences for keepers:

• Being cut does not mean it is safe. A leaf treated with fungicides can damage the garden hours before the ants react. Their ability to select does not replace our own control (see ‘Plant safety’).
• If they suddenly stop cutting a plant they used to accept, remove it and don’t insist. Offer a different one. That plant, or that batch, may have disagreed with the fungus, and the colony may take weeks to accept it again.

From minims to soldiers: one size, one trade

The division of labour is one of the most striking features of Atta mexicana. In Arizona, workers range from about 2 mm to about 12 mm, and each size range specialises in particular tasks:

• The minims, extremely small, work mainly inside the garden: they handle the fungus, clean it and tend the eggs and larvae.
• The media workers do much of the cutting and carrying of plant material.
• The large majors or soldiers deal mainly with defence and with tasks that require great mandibular strength.

And there is one more trade, the least glamorous: that of waste worker. In the leafcutters studied, the workers that work in the waste dump keep away from the garden, so as not to carry the microbes from the refuse into it. When you watch your colony, take a look: the ants going in and out of the waste dump are not usually the same ones carrying leaves.

One milligram to go around the world

Communication between them all relies on trail pheromones, contact signals, colony odour and other chemical stimuli. When a few scouts discover a suitable plant, the route can quickly go from a handful of workers to a continuous stream of laden workers. Plant selection is also part of that information system: the workers that process the material before taking it into the garden are a further filter before the plant comes into contact with the fungus.

The trail pheromones of these ants are astonishingly potent. The first ant trail pheromone to be identified chemically was that of its closest relative, Atta texana, in 1971: methyl 4-methylpyrrole-2-carboxylate. The workers detected trails of just 0.08 picograms per centimetre. With that figure, the authors calculated that 0.33 mg would in theory be enough to lay a detectable trail around the Earth; hence the popular version that “one milligram would go around the world three times”. Note: it is a theoretical calculation and a figure for A. texana. The pheromone of A. mexicana has not yet been studied, although, given how closely related the species are, it is reasonable to think it works in a similar way.

As well as smelling, leafcutters “listen”. By rubbing two body segments together they produce vibrations, stridulation, which travel through the ground and through the leaves. A 2017 study with Atta sexdens showed that workers not only perceive these vibrations, but can tell which direction they come from.

The queen who travels with her garden in her mouth

In Arizona, mature colonies raise winged queens and males every spring, and release them in summer with surprising punctuality: in the darkness before dawn, on the morning after a monsoon storm of more than 1 cm of rain. After heavy rain, of about 5 cm, there can be flights on at least two nights in a row; rain or a breeze in the early hours, on the other hand, can hold them back. It is a dangerous moment: bats, nighthawks and the Sonoran Desert toad (Incilius alvarius) feast on the alates, and badgers and spotted skunks later attack the queens and young colonies.

Founding is claustral and independent. Before the nuptial flight, the future queen stores a small portion of the fungus from her home colony in her infrabuccal cavity, a small pouch that ants have beneath the mouth. After mating, she sheds her wings, digs a vertical shaft, deposits that small portion in her chamber and starts tending the new garden herself. She raises the first generation of workers without going out to collect leaves, and the first workers appear after 1–2 months. Queens do not usually travel far: dispersal is probably less than 5 km.

The survival of the founding colony depends directly on keeping that small initial fragment of fungus. A queen without fungus is not in the same situation as a healthy founding colony, even if she has started laying eggs.

Founding is, by far, the hardest part of an Atta’s life. In a 2023 study with Atta queens collected in the field after the nuptial flight and taken to the laboratory, 98.3 % died in one year and 100 % in another; the authors concluded that it is much more effective to start with young nests that are already established. That is why at Antomas we work with colonies with queen, workers and their own garden already up and running.

A single queen. In Antomas’ experience, A. mexicana is monogynous: each colony has a single queen, who founds it alone. Queens do not tolerate one another, not even when newly mated: they show considerable aggression and, if put together, they kill each other. That is why they must always be kept apart. Curiously, its closest relative behaves the other way round: in our experience, A. texana is polygynous, and its queens can live together without aggression.

One night, several fathers

You often read that Atta queens mate only once. That is only half true: they mate only during the nuptial flight and never again, but on that occasion they usually do so with several males. This is called polyandry.

In Atta colombica, a genetic study found that 33 out of 36 colonies had two or more fathers, with an average of 2.6 fathers per colony. And across the fungus-growing ants as a whole, the shift from single mating to obligate multiple mating happened only once, in the lineage formed by Atta and Acromyrmex, the leafcutter ants. One of the advantages is quantity: in A. colombica, queens that mate with more males store more sperm. There are no data of its own for A. mexicana, but everything suggests it follows the rule of the genus.

The sperm the queen stores that night has to last her whole life, years or decades. That is why there is no point at all in exposing her to unnecessary risks, such as excessive heat (see ‘Temperature’).

Millions of workers and twenty years of life

Mature colonies reach enormous sizes, probably of millions of individuals. And what is most interesting is that their lives have been followed for three decades. In the Organ Pipe censuses, A. mexicana colonies normally lived 10–20 years; several passed 20, but none lasted the 30 years the monitoring went on. Of the colonies counted in 1985, only 4 or 5 were still alive in 2005–2006, and of the 26 recorded in 2005, nine were still active ten years later. The total number of colonies also rose and fell over the years: 36 in 1985–1986, 45 in 1995–1996, 25 in 2005–2006 and 83 in 2015–2016.

In captivity, growth is what most surprises anyone starting out with leafcutters. An Atta mexicana colony can go from a few hundred individuals to an enormous society, and the fungus grows with it. That is why the set-up must be designed from the start with expansion in mind (see ‘Growth and expansions’).

A colony of Atta leafcutter ants growing at Antomas (Antomas channel).

From farm pest to delicacy: arrieras and chicatanas

In Mexico, Atta are known as hormigas arrieras, “muleteer ants”, for their endless caravans of leaves, and not always fondly. Being polyphagous, A. mexicana can cut many cultivated plants, and the genus Atta is recognised as a major agricultural pest, with a significant economic impact. In the amaranth fields of southern Puebla, for example, a study with 61 growers found that the arriera ant is the main pest: more than half of them use chemical control, and 11 native plants were identified that farmers use as natural repellents or insecticides.

But the same ant is also a delicacy. Chicatanas are the winged queens of the arrieras, which are caught during the nuptial flight, with the first rains of the season. In Oaxaca, Chiapas, Veracruz, Guerrero and Puebla they are toasted on the comal and ground into salsas, moles or tamales. It is a pre-Hispanic tradition, already recorded by Friar Bernardino de Sahagún in the 16th century, and the name comes from Nahuatl: tzicatl, “ant”, and tanatli, “palm-leaf bag”, because of the queen’s swollen gaster.

And they are a very nutritious food: in an analysis carried out in Huehuetla (Hidalgo), toasted A. mexicana chicatanas contained 43.7 % protein, slightly more than those of A. cephalotes. The name “chicatana” is also applied to other Atta depending on the region. They fetch high prices and, according to the Mexican press, some communities already limit collection, because every queen caught is an ant nest that will never be born.

Size and morphology

Atta mexicana shows one of the most spectacular polymorphisms among North American ants: workers of about 2 mm and soldiers of about 12 mm live together in the same colony.

• Minims. Very small, delicate bodies that allow them to work inside the narrow spaces of the garden.
• Media workers. Mandibles adapted to cutting and carrying vegetation.
• Majors. They develop an enormous head and extraordinarily powerful mandibular muscles. Here we really are talking about true soldiers. In Antomas’ experience, the largest majors can make cuts that go through the skin and draw blood. They should not be handled directly, even though the species lacks the dangerous sting of a Myrmecia or a Pogonomyrmex.
• Queen. Enormous, with a very bulky gaster. According to Mintzer, who measured her in Arizona, she is the largest known ant in North America.
• Colour. It ranges from reddish brown to dark red or maroon tones.
• Surface. Without the whitish bacterial coating of Acromyrmex: the workers look clean (see ‘Antibiotics from a gland, not from bacteria’).
• Features that distinguish it from A. texana. No erect hairs on the back of the head or on the first segment of the gaster, and relatively short pronotal spines (see ‘Close cousin of Atta texana’).
• Pupae. Naked, without a cocoon.

Approximate measurements

• Minims: about 2 mm
• Media workers: all the intermediate sizes
• Majors (soldiers): up to about 12 mm
• Queen: 20–22 mm and around half a gram (about 500 mg)
• Male: no published measurements that we have been able to verify

These are measurements from the Arizona populations. The higher figures circulating in some shops are not backed by published measurements. The first workers of a young colony are small, and the large majors appear once the colony has grown.

Care in captivity

Difficulty: 4/5 – Advanced

At Antomas we consider it an advanced-level species, but it is not a species just for experts. An established colony does not require complicated interventions every day. What it asks for is consistency with three things:

• The fungus. The colony depends on it completely: if the garden dies, the colony dies. It needs high humidity, without puddles, and a stable temperature, never above 30 °C continuously, and it does not tolerate treated plants.
• The leaves. You have to give it fresh, varied, pesticide-free leaves regularly, all year round, including in winter.
• The space. Its growth potential is enormous, so you have to keep expanding the set-up with new modules as it grows.

Added to this are the majors, strong and with mandibles that cut skin, and an abundant production of waste. A mini-colony can be simple; the real challenge comes when the fungus and the population start to multiply. If you have access to untreated leaves and space to grow, it is perfectly manageable.

And it is worth it like few others. Watching a highway of cut leaves entering the garden, the minims working inside the fungus and the majors on patrol is a glimpse into one of the most complex societies in the animal kingdom: a civilisation of insects that grows its own food.

Recommended parameters

• Fungus temperature: 23–26 °C, as stable as possible.
• Never above 30 °C continuously: in Antomas’ experience, the queen can become infertile (see ‘Temperature’).
• Foraging area: at room temperature.
• Humidity: high in the fungus chamber, but without free water or puddles; the foraging area drier, and the waste dump dry.
• Ventilation: constant, without drying out the outside of the garden.
• Water: the humidity tray of the growing chamber. A drinker in the foraging area is optional: they get plenty of water from the fungus itself and from the humidity in the set-up.
• Light: a normal day-night cycle. Avoid direct sunlight on the fungus chamber, because it heats it up.
• Diapause: not needed (see ‘Diapause’).

Temperature

For the garden we recommend approximately 23–26 °C, kept stable and avoiding strong heating. This is the range commonly used to keep leafcutters in captivity, and it fits with what we know about the species: natural fungus chambers remain much more sheltered than the surface, and even in the Sonoran Desert conditions inside the nest are relatively constant.

Be very careful with heat. At Antomas we apply a very important precaution: the fungus chambers, and especially the chamber where the queen is, must not be above 30 °C continuously. In our experience with Atta, queens can become infertile if they spend time above that temperature, and the fungus itself also deteriorates.

We know of no experiment that has measured that threshold in A. mexicana: 30 °C is a handling limit in captivity, based on experience. And it makes perfect sense: the queen mates only during the nuptial flight, albeit with several males, and the sperm she stores that day has to last her whole life (see ‘One night, several fathers’). There is no advantage whatsoever in exposing her to unnecessarily high temperatures: the aim is to keep her well below that limit.

If you need extra heat in winter, a gentle, thermostat-controlled source on one side of the chamber is better than one that heats the whole fungus evenly. And in summer, make sure the room temperature does not soar.

Humidity, ventilation and drainage

This is one of the most important sections of the care sheet. The fungus needs high ambient humidity, but it must not stay wet. Letting it dry out is a serious mistake: a dehydrated garden shrinks and may not recover.

A large mass of fungus contains an enormous amount of water. Between the ants’ respiration, the metabolic activity of the garden and the evaporation from the culture itself, the container can quickly reach very high humidity. If the walls or the lid are slightly colder, that humidity starts to condense. The droplets fall back onto the garden and, if the system cannot drain the water away, they end up accumulating at the bottom until they form puddles. This can become a serious problem.

In Antomas’ experience, free water accumulating around the fungus must be drained. We should not leave the lower part of the garden submerged for days, waiting for the water to evaporate on its own. When liquid builds up, we recommend:

• Draining or removing the water.
• Increasing ventilation slightly.
• Checking whether there is too much moisture input or an excessive temperature difference between the inside and the walls.

The system therefore needs some mechanism to remove accidentally accumulated water. In the Anthouse Fungarium, the fungus rests on a raised grid, above a water tray: this way it receives humidity without ever being in contact with the water (see ‘Recommended set-up’).

Ventilation is just as important. Fungus and ants consume oxygen and release CO₂, heat and water vapour. The natural nests of leafcutters have an architecture that favours gas exchange: in A. vollenweideri it has been shown that the wind renews the air in the nest, which leaves through the central tunnels and enters through the peripheral ones. In captivity we must reproduce that function:

high humidity + constant ventilation + no waterlogging

We are not looking for an airtight container. Nor do we want so much ventilation that the outside of the garden keeps drying out. The right balance is easy to recognise: moist, active fungus, without dry crusts, but without droplets constantly falling or puddles at the bottom.

Two more tips. Never spray water directly onto the fungus: humidity should come from the atmosphere of the chamber, not from wetting the garden. And don’t obsess over a number: some keepers observe that A. mexicana, which comes from seasonal, dry climates, tolerates swings in humidity better than rainforest Atta. What counts is how the fungus looks.

Recommended set-up

For this species we recommend the Anthouse Fungarium Kit, a modular system designed specifically for leafcutter ants, Atta and Acromyrmex, with modules measuring 19.5 × 19.5 × 26 cm:

Fungarium Kit for leafcutter ants

It fits with everything we have seen in this care sheet:

• Growing chamber with passive humidity. The fungus rests on a raised grid, above a water tray. This keeps the atmosphere humid and stable without the garden ever touching the water, and any excess water stays below the fungus.
• Reducing covers. Three transparent, interchangeable covers adjust the volume of the chamber to the actual size of the fungus in the early stages, and are removed one by one as the garden grows.
• Built-in hygrometer and thermometer, to monitor the two parameters that matter most.
• Foraging and maintenance box, connected to the chamber by wide tubes, through which the workers pass laden with leaves.
• Fully closing rotating doors, optional, to isolate each module when cleaning, expanding or moving.

An initial set-up should separate at least:

fungus chamber → foraging area → waste dump

With a small colony, the foraging box can also serve as the waste dump; when the colony grows, add another module for them to use as a refuse tip (see ‘Waste dump’). And as the fungus grows, add new growing containers. In Atta mexicana this modularity is especially important, because it is a naturally multi-chambered species (see ‘Cities beneath the desert’). We do not recommend trying to concentrate the whole garden in a single box indefinitely.

In the foraging area, place the fresh leaves and, if you like, a drinker or a test tube with water and cotton wool, which is optional. Since they climb smooth surfaces, protect it with a strip of mineral oil along the top of the walls and keep it well closed (see ‘Escapes and barriers’).

Arrival and installation in the Fungarium

Our colonies always arrive with their fungus, in a container with the queen and her workers. We often also include a test tube with reinforcement fungus, in case the main garden arrives a little the worse for the journey.

Get everything ready before opening anything: the tray with water and the grid in place, the cover suited to the size of the fungus, the foraging box connected, with a few tender leaves and the strip of mineral oil, and the lids checked. Then follow this order:

• First, the reinforcement tube. If there is one, remove all its ants and set it aside, closed, at the same temperature as the colony.
• The main fungus, in one block. Move the garden from the queen’s container to the growing chamber in one go, with the workers on it, without crumbling it or breaking it up unnecessarily. If it has to be moved in several parts, they should be large, complete blocks.
• The reinforcement, only if needed. If the main garden has arrived in poor shape, place the fungus from the tube next to it, with no ants left on it at all, and your colony’s workers will incorporate it. If it arrives in good condition, there is no need to use it.
• The colony’s stragglers, by tipping. The workers left in the queen’s container are gently tipped into the foraging area; they will make their own way back to the fungus. Never tip the ants from the reinforcement tube in there.
• Calm. Leave them in peace for a day or two, with tender leaves within reach, and let them reorganise the garden at their own pace. It is normal for them to put more effort into rearranging the fungus than into cutting at first.

For expansions there is no need to move anything: just connect the new module, and it is the ants themselves that decide how much fungus they take to each one.

Founding

At Antomas we work with established colonies, but if you ever look after a founding queen, these are the keys. A newly mated queen should be kept with her small original portion of fungus in a quiet chamber. She does not need leaves during the first claustral phase: adding plant material unnecessarily to a tiny culture only increases the risk of contamination. What really matters is keeping the humidity, temperature and ventilation of that first chamber stable.

The initial fungus can be extremely small, and this should not be seen as a problem as long as it stays alive and keeps growing. In the wild, the first workers take one or two months to appear. If after that time there are no workers and the fungus is not growing, it is a bad sign. Never put several queens together: in our experience, A. mexicana is monogynous and the queens kill each other. Each queen in her own set-up.

Handling the fungus

A healthy garden should be handled as little as possible. We do not recommend taking it apart periodically to look inside: it can be observed perfectly well through the walls of the chamber. If it has to be moved, we move it in complete blocks, avoiding breaking it up unnecessarily. The workers themselves are able to recognise deteriorated areas, remove them and take them to the waste dump.

Nor would we routinely mix fungus from different colonies. A healthy reinforcement can be useful in specific circumstances, such as the reinforcement fungus we sometimes send with the colony, and always without its ants (see ‘Arrival and installation in the Fungarium’). But each garden contains much more than Leucoagaricus: it includes an associated microbial community that can vary even between different parts of the same culture.

Bear in mind too that the garden grows upwards. One keeper reports that, with a colony in full expansion, the fungus multiplied several times over in a few weeks, reached the lid and ended up toppling over. If you see the fungus already filling almost the whole chamber, it is time to remove a cover or add a new module.

Waste dump

The waste dump must be physically separate from the fungus chambers. The colony generates plant remains, fragments of spent culture, corpses and other waste. In the wild, A. mexicana takes all that rubbish out of the nest and piles it up on the surface, unlike most Atta (see ‘Taking the rubbish outside’). That is why an external waste dump, in a separate and rather dry module, is true to its biology.

• Let them choose. In captivity it is best to let the ants choose the container they will use as a refuse tip. Once it is established, we would not keep changing its location.
• Easy to clean. The container should be easy to disconnect and clean without opening the fungus chambers.
• Grey or brown fungus in the waste dump: normal. Each portion of fungus is exhausted in about a month, so seeing old, greyish or brownish fungus come out regularly is part of the normal functioning of the garden, not a disease.

Feeding

Leafcutters do not eat leaves: they feed their garden with them. That is why, in captivity, the diet must be broad, varied and always free of treatments. Antomas’ experience with Atta mexicana shows very good acceptance of different plants that are easy to find in Spain.

The more fresh leaves, the more fungus and the bigger the colony. This is the Antomas philosophy: if you do daily maintenance and remove what they don’t use, it is better to offer plenty, because the colony understands that it has resources and grows faster. Offer fresh leaves regularly, ideally daily for a large colony, and remove wilted or dry leaves they have not cut before they go mouldy. If you won’t be able to clean often, it is better to match the amount to what they cut.

Plants we use at Antomas:

• Rose (Rosa spp.). Leaves and petals. One of the most useful resources, provided it comes from a plant that is completely free of treatments. It is also backed by science: rose was one of the leaves preferred by A. mexicana in the laboratory.
• Hibiscus (Hibiscus spp.). They accept the leaves and, depending on the colony, the flowers too. Ornamental hibiscus must come from plants we know have not been treated.
• Orange tree and other citrus (Citrus spp., especially Citrus sinensis). In our experience young orange leaves work well, being much more tender than mature ones, and they eat orange as a fruit quite readily. As a nuance, in the 2020 study the workers rejected mandarin leaves, because of their physical characteristics rather than their chemistry. That is why we prefer young, tender leaves to mature ones.
• Almond (Prunus dulcis). Both leaves and flowers can be used in season, and the colony usually works it well.
• Mulberry (Morus spp.). It can be offered from time to time. Tender leaves are usually easier to handle.
• Bramble (Rubus spp.). It can also go into the rotation. Other keepers name it among their favourites.

None of these plants should make up the whole diet on its own. Variety allows the colony to keep selecting the material that works best for it at any given time.

In winter. When the almond and the mulberry lose their leaves, rely on evergreens such as bramble, citrus or hibiscus, and on these resources, which we use a lot at Antomas at that time of year:

• Dried rose petals, which are sold ready to use. Check that they are fit for consumption and free of treatments.
• Supermarket fruit: apple, banana and orange, which they eat quite readily.
• Vegetables: lettuce, cauliflower and broccoli, which they accept well.
• Oat flakes, especially when there is a lot of humidity in the set-up: they are a dry resource that adds no water.

Other plants used by European keepers: raspberry, elder, lime (linden) and privet. We have no experience of our own with them in this species: if you try them, start with small amounts.

With caution: there are plants recommended by some shops or widely used in the hobby for which there are no data on their effect on the fungus, such as box, Virginia creeper or oak, which is rich in tannins, compounds that the 2026 study associated with the least chosen leaves. Ivy contains substances with known antifungal activity: better to avoid it.

Plants to avoid:

• Coffee and guava, which A. mexicana rejected in the laboratory.
• Any treated plant: cut flowers from florists, newly bought ornamental or nursery plants, and plants from gardens, parks or fields that are sprayed.
• Plants you cannot identify or whose history you do not know.

Fruit and vegetables, as a supplement. Besides leaves and flowers, at Antomas we offer other resources from time to time. Among fruits, they accept apple (Malus domestica), banana (Musa spp.) and orange (Citrus sinensis), and among vegetables, lettuce (Lactuca sativa), cauliflower (Brassica oleracea var. botrytis) and broccoli (Brassica oleracea var. italica). They should be seen as a supplement, not as a permanent substitute for leaves and flowers. Because of their high water content and, in fruit, their sugars, it is best to give them in small amounts and remove the leftovers before they ferment or grow microorganisms.

No insects. Unlike almost all the ants we sell, leafcutters need no animal protein: their food is the fungus. The sugars the workers use come mainly from the juices of the leaves and from fruit.

Nectar, only now and then. It is not necessary. If you wish, you can offer a little nectar or sugary liquid very occasionally, as a supplement, roughly every 15 days. Always in a drinker, never on cotton wool, which goes mouldy and spoils the liquid straight away, and remove whatever is left over. Don’t turn it into a routine: leaves are the basis.

Water, optional. You can put a drinker or a test tube with water and cotton wool in the foraging area, but it is not necessary: in our experience, they get plenty of water from the fungus itself and from the humidity that evaporates in the set-up, as well as from the fresh leaves. What is a serious mistake is letting the fungus dehydrate (see ‘Humidity, ventilation and drainage’).

Plants documented in the wild and in the laboratory. In the Sonoran Desert it uses palo verde, ironwood, mesquite, catclaw and creosote bush, as well as seasonal annual plants (see ‘Leafcutters on the night shift’). In the laboratory it preferred rose, Trema micrantha, Styrax glabrescens and flor de mayo (Palicourea padifolia), whose extracts favoured the growth of the fungus, and chose the Mexican sycamore and the loquat much less (see ‘Leaf tasters’). There is no need to obtain these plants to keep the colony properly: they help to understand how diverse its natural diet is.

Plant safety

Every plant intended for the colony must be completely free of pesticides, acaricides, herbicides and fungicides. We would never use cut flowers from florists or leaves from newly bought ornamental plants.

• Washing a leaf does not guarantee it is safe. Many plant protection products are systemic and remain inside the tissue.
• The ants are not a quality control. They may reject certain resources, but their ability to select does not replace our own control. A plant they are cutting should not automatically be considered harmless to the fungus either: a fungicide can damage the garden hours before they react (see ‘The fungus gives the warning: delayed rejection’).
• New things, little by little. That is why it is always worth introducing new foods in small amounts first, and watching the fungus over the following days.
• Check the leaves. Garden leaves can bring small lodgers, such as red spider mites or aphids: shake them and take a look before giving them.

Diapause

Atta mexicana does not need a cold diapause in captivity. Even the northern populations, which endure relatively cool winters, keep their fungus in deep chambers where the temperature is greatly buffered, and in Arizona they go on foraging by day in the middle of winter.

We would not deliberately cool the Fungarium during the European winter. We would keep roughly the same growing conditions all year round, allowing small natural variations. The colony may change its rhythm of foraging or growth, but it does not need a cold period like that of a temperate-climate species.

Growth and expansions

An Atta mexicana colony can go from a few hundred individuals to an enormous society, and its modular system should be designed from the outset with expansion in mind.

• More fungus, more modules. When the garden fills the chamber, remove a cover and, later, connect a new growing chamber. The ants share out the fungus on their own.
• Wide connections. A connection that seems sufficient for a mini-colony can become too small when tens of thousands of workers are trying to carry leaves at the same time.
• More foraging and more waste dump. As the colony grows, so do the amount of leaves it needs and the waste it produces.
• Regular checks. All lids, tubes and joints should be checked regularly.

Behaviour in captivity

Leaf highways. The most striking thing is to see how a new plant turns, within a few hours, into a continuous stream of workers laden with pieces of leaf, often larger than themselves, on their way to the garden (see ‘One milligram to go around the world’).

Night shifts. In summer, don’t be surprised to see more activity at night than by day: in the desert, they forage mainly at night at that time of year.

Each to her own job. The minims inside the fungus, the media workers on the trails, the majors patrolling and defending, and the waste workers in the refuse tip, kept apart from the rest (see ‘From minims to soldiers: one size, one trade’).

Watch without fear. At Antomas we believe ants are there to be enjoyed, observed and learnt from, not kept hidden away. It is enough to observe them gently, intervening and stressing them as little as possible. The garden can be seen through the walls of the chamber, without needing to open it.

Signs that all is well: moist, spongy, growing fungus; workers cutting and carrying leaves; and old fungus regularly coming out to the waste dump.

Warning signs:

• The fungus shrinks, dries out or forms crusts: not enough humidity.
• Droplets constantly falling or puddles at the bottom: too much water or not enough ventilation (see ‘Humidity, ventilation and drainage’).
• They suddenly stop cutting a plant they used to accept: remove it (see ‘The fungus gives the warning: delayed rejection’).
• The fungus stops growing or darkens over large areas, and the colony abandons it: check the last plants you gave them, the temperature and the humidity.
• Workers crowd outside the fungus, with no room inside: time to expand.

Escapes and barriers

A large colony has thousands of scouts, the majors are very strong and the minims get through tiny gaps. What is more, in our experience, they climb smooth surfaces. At Antomas we recommend:

• A strip of mineral oil along the top of the walls of the foraging area, thin and continuous. In our experience, it works quite well as a barrier. Renew it when it collects dust or debris.
• Well-fitting lids on all modules, kept closed while you are not working.
• Checking tubes and joints regularly, especially after each expansion.
• Vents with fine mesh, which lets air through but not the minims.
• Nothing that can act as a bridge: no leaves or branches resting near the barrier.
• The rotating doors of the Fungarium let you isolate a module while you clean or expand it.

Handling and safety

The large majors deserve a specific warning. Their mandibles are powered by enormous muscles inside the head and can exert considerable force. In Antomas’ experience, the large soldiers can make cuts that go through the skin and draw blood. They should not be picked up with your fingers or handled directly: a major that gets a good grip on the skin with its mandibles can cause a much worse wound than you would expect from an ant without a sting.

To work with large colonies, use forceps and a set-up that lets you feed, clean and expand the installation without putting your hands among the workers. Do not put queens or different colonies together, and do not routinely mix their fungi. It is an exotic species and, in its natural range, an agricultural pest: under no circumstances should it be released.

What not to do

• Spraying water onto the fungus, or leaving puddles in the growing chamber for days.
• Letting the fungus dry out.
• Giving leaves from florists, nurseries, newly bought ornamental plants or any treated plant.
• Basing the diet on a single plant.
• Turning nectar into a routine: something every 15 days, at most, is enough.
• Insisting on a plant they have stopped cutting.
• Keeping the queen or the fungus above 30 °C continuously: the queen can become infertile. Nor in direct sunlight.
• Opening, taking apart or breaking up the garden to see inside.
• Mixing the ants from the reinforcement tube with those of the colony: they must be removed before using that fungus.
• Putting queens together, as they kill each other, or putting different colonies together.
• Routinely mixing fungi from different colonies.
• Neglecting the oil barrier or leaving gaps: they climb smooth surfaces.
• Putting the waste dump inside the fungus chamber, or constantly moving it.
• Mistaking the spent fungus coming out to the waste dump for a disease.
• Giving it a cold diapause.
• Running out of space: not expanding in time.
• Picking up the majors with your fingers.

Scientific sources

This care sheet is based on the scientific literature on Atta mexicana and leafcutter ants, in particular the field censuses by A. C. Mintzer at Organ Pipe Cactus National Monument (Arizona), published in Park Science in 2018 (open access), and on the experience of Antomas and other keepers. Main references:

1. Infante-Rodríguez, D. A., Velázquez-Narváez, A. C., Aguilar-Méndez, M. J., Fuentes-Jacques, L. J., Monribot-Villanueva, J. L., Guerrero-Analco, J. A. & Valenzuela-González, J. (2026). Selection of leaves of Palicourea padifolia, Platanus mexicana and Eriobotrya japonica by Atta mexicana (Hymenoptera: Formicidae) and its influence on the growth of the mutualistic fungus Leucoagaricus gongylophorus. Botanical Sciences 104: 81–97. doi:10.17129/botsci.3671
2. Infante-Rodríguez, D. A. et al. (2020). Phytochemical characteristics of leaves determine foraging rate of the leaf-cutting ant Atta mexicana (Smith) (Hymenoptera: Formicidae). Chemoecology 30: 147–159. doi:10.1007/s00049-020-00306-4
3. Mintzer, A. (1979). Foraging activity of the Mexican leafcutting ant Atta mexicana (F. Smith), in a Sonoran Desert habitat (Hymenoptera, Formicidae). Insectes Sociaux 26: 364–372. doi:10.1007/BF02223555
4. Gómez-Díaz, J. A., Baena, M. L., González-Zamora, A. & Delfín-Alfonso, C. A. (2023). Potential present and future distributions of the genus Atta of Mexico. PLoS ONE 18: e0292072. doi:10.1371/journal.pone.0292072
5. Fernández-Marín, H., Zimmerman, J. K., Nash, D. R., Boomsma, J. J. & Wcislo, W. T. (2009). Reduced biological control and enhanced chemical pest management in the evolution of fungus farming in ants. Proceedings of the Royal Society B 276: 2263–2269. doi:10.1098/rspb.2009.0184
6. Heine, D. et al. (2018). Chemical warfare between leafcutter ant symbionts and a co-evolved pathogen. Nature Communications 9: 2208. doi:10.1038/s41467-018-04520-1
7. Currie, C. R., Mueller, U. G. & Malloch, D. (1999). The agricultural pathology of ant fungus gardens. PNAS 96: 7998–8002. doi:10.1073/pnas.96.14.7998
8. Herz, H., Hölldobler, B. & Roces, F. (2008). Delayed rejection in a leaf-cutting ant after foraging on plants unsuitable for the symbiotic fungus. Behavioral Ecology 19: 575–582. doi:10.1093/beheco/arn016
9. Saverschek, N., Herz, H., Wagner, M. & Roces, F. (2010). Avoiding plants unsuitable for the symbiotic fungus: learning and long-term memory in leaf-cutting ants. Animal Behaviour 79: 689–698. doi:10.1016/j.anbehav.2009.12.021
10. Tumlinson, J. H., Silverstein, R. M., Moser, J. C., Brownlee, R. G. & Ruth, J. M. (1971). Identification of the trail pheromone of a leaf-cutting ant, Atta texana. Nature 234: 348–349. doi:10.1038/234348b0
11. Fjerdingstad, E. J., Boomsma, J. J. & Thorén, P. (1998). Multiple paternity in the leafcutter ant Atta colombica — a microsatellite DNA study. Heredity 80: 118–126. doi:10.1038/sj.hdy.6882470
12. Villesen, P., Murakami, T., Schultz, T. R. & Boomsma, J. J. (2002). Identifying the transition between single and multiple mating of queens in fungus-growing ants. Proceedings of the Royal Society B 269: 1541–1548. doi:10.1098/rspb.2002.2044
13. Santos, T. T. M., Santana, M. V., Mascarin, G. M., Arthurs, S. P. & Fernandes, P. M. (2023). Collecting young nests of Atta spp. is more efficient to establish laboratory colonies than collecting founder queens after nuptial flights. International Journal of Tropical Insect Science 43: 1837–1841. doi:10.1007/s42690-023-01060-0
14. Kleineidam, C., Ernst, R. & Roces, F. (2001). Wind-induced ventilation of the giant nests of the leaf-cutting ant Atta vollenweideri. Naturwissenschaften 88: 301–305. doi:10.1007/s001140100235