Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
SAF Archives  search ask the scientists in the classroom cool science
scientists from previous shows
cool careers in science
ask the scientists

Photo of Ulrich Mueller Champion Chompers

In this story, Ulrich Mueller of the Department of Zoology at the University of Maryland, keeps a close watch on the forest floor to study the activities of leaf-cutter ants. Following the Frontiers special Expedition Panama: Champion Chompers, Ulrich answered viewers' questions in an Ask the Scientists panel. Discover more about the role of ants in the complex of ecology of the rain forest. Here are viewers' questions and his answers:

QWhat role do the ants play in the downfall or the uprise of their habitat, the complex rainforest? In other words, what are the up and down sides of having this insect around?

Leafcutter ants have a major impact on the ecology of neotropical forests. According to some estimates, leafcutter ants may cut up to 20% of all foliage per year in a given rainforest. That estimate is certainly at the high end, but even only a 5% damage (which is more realistic) would greatly surpass the impact of any other single herbivore. For the average rainforest tree or shrub, therefore, leafcutter ants are a major threat; consequently most rainforest plants have evolved defenses to minimize damage by leafcutter ants.

The main defense for most rainforest plants is leaf chemistry: Plants incorporate compounds in their leaves that are toxic to potential herbivores. With respect to leafcutters, such chemical defenses include both insecticides (the ants habitually drink some of the plant juices during cutting) or fungicides (which would kill or reduce growth rates of the symbiotic fungus). The ants are very sensitive to leaf fungicides and prudently avoid cutting vegetation high in fungicides. Throughout millions of years of coevolution with leafcutter ants, a high proportion of rainforest plants have developed such chemical defenses. This is best illustrated by the fact that virtually all plants that humans have introduced to South America (garden vegetables, crops such as citrus fruits, apples, etc.) are strongly preferred by leafcutter ants over native plants. The reason for this is that the introduced plants have not undergone a coevolutionary history with leafcutter ants (because they are of European, African, or Asian origin; leafcutter ants only occur in the New World) and thus have not been able to evolve adequate defenses against leafcutters. (Gardening in the tropics is often an impossible task because of leafcutter ants.)

Rainforest plants, in contrast, have developed an arsenal of defenses and therefore can coexist with leafcutter ants. In fact, leafcutter ants occur at surprisingly low levels in primary rainforest; it is only in disturbed areas (gaps created in the forest by falling trees; forest cleared by humans and allowed to regrow) that leafcutter ants thrive. The density of leafcutter ant colonies in such "secondary" forest can be ten times the one of "primary" (undisturbed) forest. Can you think of a reason why leafcutter ants would do better in secondary forest? (Hint 1: Growth rates of plants in secondary forest are higher than plants in primary forest. Hint 2: Any plants has only limited resources, which it can allocate either to growth, maintenance of structures, or chemical defenses).

One more thought: Many pharmaceuticals are derived from the chemical defenses that plants evolved in response to herbivores or pathogens. Plant herbivores such as leafcutter ants have greatly accelerated the evolution of such chemical defenses; they have thus contributed indirectly to the evolution of the chemical diversity that is now tapped by pharmaceutical industry. How many potential drugs do you think still remain undiscovered in rainforest plants?

QYou and Alan talked about the ants "farming" the fungus gardens for millions of years. Why do you think they have survived so long? Have you any knowledge of any adaptations they've made over this time?

One of the greatest living myrmecologits and evolutionary biologists, Edward O. Wilson, has called the attine ant-fungus symbiosis one of the most significant "breakthroughs" in all of evolutionary history. He has placed the transition from the ancestral predatory lifestyle to the vegetarian ("fungitarian") lifestyle of fungus-growing ants at par with such transitions as the move from aquatic to terrestrial life, the development of photosynthesis in plants, or the evolution of flight in insects and birds. While the relative importance of any of these transitions can be debated, it is clear that the evolution of the attine symbiosis must have had as drastic an impact on the ants as the transition of humans from a hunting/nomadic to an agricultural lifestyle.

The main difference between human and ant agriculture is that humans developed true agriculture about 10,000 years ago (actually agriculture developed independently in at least three places around the globe at that time), while the attine ancestors came to rely on fungiculture about 50- 60 million years ago, that is, at a time when the last dinosaurs had just become extinct. Ant agriculture therefore is 5000 times older than human agriculture.

Why have fungus-growing ants have been able to survive for such a long time as farmers (will humans still be farming 50 million years from now?)? I can offer a two-part answer to explain this long-term persistence, but would like to add immediately that, despite 100 years of research on the attine ant-fungus symbiosis, we still have only a partial understanding of the ecological success and impact of these ants:

(a) The switch from a predatory to a vegetarian lifestyle opened up a novel resource to fungus-growing ants, a resource that is incompletely utilized by other herbivores (like herbivorous mammals or insects). This would explain the ecological success of fungus-growing ants over other (predatory) ants (i.e., fungus-growing ants no longer have to compete with predatory ants).

(b) The success of the symbiosis between ants and fungus is based on a unique constellation of complementary skills: (i) the ability of the fungi to convert plant matter into sugars and proteins (which the ants consume) while detoxifying the plant matter; and (ii) the ability of the ants to prepare the plant matter in such a way so as to maximize fungal growth. The fungi cultured by fungus-growing ants (fungi in the family Lepiotaceae) are specialist litter decomposers. The choice by the ants of these fungi therefore is not arbitrary; the ants have selected fungal specialists that are good at subsisting on plant matter. However, most plant matter is chemically protected (plants incorporate "fungicides" into leaves or the leaf-cuticle to prevent fungal invasion) and is thus not readily available to any fungus. The ants have figured out which plant substrate is "best" for their fungi (low in fungicides; the ants are very sensitive to leaf fungicides and avoid foraging on leaves high in fungicides). The ants also appear to have developed complicated preparations of a plant "manure" for their fungus (e.g., by licking off fungicides present on leaf-cuticles; "aging" leaf pieces inside the nest until toxic compounds have degraded). Now, apart from fungicides, there are also insecticidal compounds present in many plants (e.g., insecticides that the plant incorporates to deter attack by beetle or butterfly larvae), but these are generally not harmful to the fungus, or may even be degraded biochemically by the fungus. So, in a sense, the ants are utilizing the fungus to detoxify and enhance a food source (leaves) that is abundant in the forest. Lastly, the ants have evolved "herbicidal" glands producing antifungal secretions that are harmful to alien fungi but not to the symbiotic fungi (note: the ants have evolved the use of "pesticides" millions of years before humans). So the fungus receives protection by the ants and a never-ending supply of substrate brought in by the ant foragers. The ecological and evolutionary success of the attine ant-fungus symbiosis probably rests on all these mutualistic exchanges between ant and fungus.

I have two more questions for thought: Projecting fifty million years from now into the future, would you predict that ant agriculturists or human agriculturists, or both, will still be present on this planet?

Would you think that, during the past fifty million years of evolution, the ants have figured out one or the other agricultural practice that we could learn from? After all, they had figured out pesticides millions of years ago, while humans began using pesticides last century.

For more technical information on fungus-growing ants, I recommend reading Chapter 17 in "The Ants" by B Hoelldobler and E.O. Wilson (1990, Harvard University Press).

QHow can such a small organism "lift" up an object that is way over its own weight (mass) and size?

This question is a bit outside my area of expertise. But I will try.

Many biological variables are not linearily related in a 1 : 1 relationship, but exponentially. This is best illustrated by the relationship between body mass and body surface. As body mass increases by one unit, body surface increases by less than one unit. Specifically, any increase in size causes a cubic increase in mass (volume is a cubic function), but only a square increase in body surface (area is a square function). This is why babies are prone to become hypothermic in the cold (they loose body heat quickly because of their large body surface RELATIVE to their body mass, which generates and stores heat), while big people quickly overheat (they can't loose heat fast enough because they have insufficient body surface RELATIVE to their body mass).

Similarly, body size and body strength are exponentially related to each other. Any increase in body size by one unit entails an increase in body strength by less than one unit. There is probably a physiological and mechanistic basis for this, which may be something like (I am hypothesizing here), that muscle strength increases less and less with larger and larger size. Hence, a human can lift masses of about his/her own weight, while an ant can lift masses of many times its own weight. So here is a question: Projecting to animals larger than humans (say dinosaurs), if large dinosaurs could have lifted weights, would they have been able to lift weights greater than their own weight, about their own weight, or smaller than their own weight?

QThe show didn't really explain how you get the estimate that the ants harvest 20 percent of the forest canopy each year. What do you have to do to get that figure? How far off could it be, and what's the part of the calculation that you're least sure is right?

To estimate the percent foliage cut by leafcutter ants per year, one would need to know (i) the total leaf area in a given forest; and (ii) the area cut per unit of time by leafcutters in the same area of forest. I don't know exactly how (i) is estimated, but I would guess it probably involves careful measurements of total leaf area on a subsample of branches subsampled from a representative set of trees and shrubs, then extrapolating across the entire forest (across all branches on all tress and all shrubs) To estimate (ii), researchers have collected representative samples of leave-fragments carried by leafcutter ants (e.g., as they are about to enter their nest) and derived the total leaf area cut (summed over all the different fragments carried by different ants) by exact measurements of fragment area and extrapolation across all the foraging ants active in all nests in a given forest.

For some leafcutter ant species, researchers have observed nests throughout the year, so they have a fair understanding of the annual activity profile (they know, for example, that leafcutters are more active during the early rainy season, a time when many plants develop new leaves; new leaves are not as tough and more nutritious than older leaves). Weighing across the yearly activity profile allows an estimate of the leaf area cut within an entire year. This in turn, allows direct calculation of the percent area cut in a given forest, assuming that one has accurately estimated (i).

For more reading on ants, check out the following web site:

QWhy did you choose to study ants?

First, because I have an inordinate fondness for social insects (I don't know why; some childhood trauma maybe). Second, because ants are fascinating (don't you think so?). Third, because ants provide superb model systems to answer questions that cannot be addressed with other model systems (e.g., the evolution of agriculture in ants; social organization). Fourth, because ants are social (like humans) and they live out the same balance between social cooperation and conflict that dominates human interactions (a close look at many ant societies reveals that not everything is harmonious). Fifth, because, in terms of number of individuals and biomass, ants represent the single, ecologically most dominant group of animals. Sixth, because they are there.

Why ants? I really don't know. I currently study ants, but it probably could have been anything. My experience has been that, no matter what phenomenon or organism you are looking at, the closer you look, the more fascinating the things you see. That thrill of discovery justifies every day of my work.

QOn the show we saw ants transporting leaves, acting as bodyguards, garbage-disposers and traffic controllers. Are there other jobs the ants do? And how do the ants know what job they're supposed to do?

First a correction: It is true that there are specialist leafcutters/transporters (the larger ants), specialist gardeners (tiny ants that fit into the crevices of the spongy fungus garden), and specialist guards (the really big ones, the ones that can pierce your skin, if you let them). But there are no specialist traffic controllers; at least, I haven't seen them and, as far as I know, nobody else has.

The various tasks get sorted out largely by size. Smaller ants have different behavioral tendencies than larger ants. From studies on ants and other social insects (e.g., honeybees), we know that some portions of these behavioral tendencies are learned (they change as a result of experience), some portions are dependent on age (e.g., older individuals do proportionally more foraging, younger ants do more nest duties), and some portions are more hardwired (not learned) and dependent on the size of the ant. Size (and overall morphology) itself is controlled by the amount of food received by a developing larva (the more food, the larger the adult; adults cannot grow further).

How does each ant knows what it is "supposed" to do? That, actually, is a very complicated process that is intensively studied by many researchers around the world. It is clear that social insects don't have a central command post that issues orders to each ant (so ant activities are fundamentally different from human activities that are regulated by central decision makers, like a government, or a board of supervisors). Rather, each ant acquires partial information on particular needs of its colony (e.g., by patrolling part of the colony), then making decisions based on this partial information. The information that any single ant has is incomplete, so some ants do make "mistakes" (as seen from the level of the entire colony). But there are so many ants that, overall, the mistakes don't matter much; most ants do make "correct" decisions and perform what is best for their colony. It appears that the amount of information that each single ant gathers, and the size-and age-dependent decision rules that each ant uses to act on this information, are finely tuned such that, overall, few mistakes are made and the colony operates as a functional whole.

If you are interested in these issues (colony organization, division of labor, task allocation), I recommend reading "The Wisdom of the Hive" by Thomas D. Seeley. These are areas where humans may be able to learn from the ways of the ants.

QOn Expedition Panama we saw how you count the ants and it looked real hard! Those guys move fast! What is the top speed of an ant? How do you actually count them? How can you be sure you didn't count the same one twice?

The person counting was actually counting the leaf fragments carried along the trail. The number of leaf fragments carried is actually much less than the number of ants traveling (not every ants carries a leaf). The top speed of an ant (without a leaf) is (I would guess) probably around 1-3 meters per minute; I know that some researchers have measured this precisely, but I can't find the figures at this moment.

QWhat is the average lifespan of Leafcutter ants? Do they sting or bite people (like fire ants). How many ants are in a colony?

In the lab, leafcutter workers can live over a year. In the field, worker lifespan is probably much shorter. Leafcutter queens, of course, live longer than workers. The record for a queen of a lab colony is SEVENTEEN years. In the field, queens may live up to 4-8 years, depending on the species; but most queens (nests) perish during their first year. A nest dies once the queen dies (that is, nests cannot requeen in leafcutters). Males live a day or two after they leave the nest; they try to find a mate, may or may not be successful, then die. Mature leafcutter colonies may have 2-10 million workers, depending on the species. All of these workers are sisters (daughters of the same queen). (Just think about remembering birthdays for 2 million siblings.)

QWhat is the largest fungus garden known? How much of the fungus is nutritional? How much of the fungus do the ants throw away every year?

Single fungus-gardens never grow much larger than the size (and approximate shape) of about a football. The ants never excavate chambers larger than a football, so a garden can't exceed that size. A single mature colony has about 1000 (thousand) chambers, each filled completely with garden.

The fungus of leafcutter ants produces special structures (aggregations of terminal swellings of hyphal tips) that the ants consume, or harvest to feed to their brood. These structures, called gongylidia, are rich in proteins and sugars. I have tasted the fungus and it is, indeed, quite sweet. Of the entire garden biomass, gongylidia biomass comprises only a small fraction; I would guess, less than 5-10%. Most of the fungal biomass is "rooted" in the manure (taking up nutrients).

The ants add tiny leaf fragments (somewhat mushed-up so as to form a manure) to the top of the garden, then plant tiny inocula of the fungus on top of these fragments. As the fungus matures, new fragments are added on top, then more, and so on. Meanwhile, old garden (fungus that exhausted the manure) is removed by workers at the bottom of the garden, then dumped. So leaf fragments gradually move downward through the garden. The entire cycle from initial incorporation into the garden to final dumping takes about six weeks. Once exhausted, all garden is dumped by the ants. Some species have their dumps above ground (as you saw in the movie), some species construct huge (many, many times the size of their gardens) underground chambers that serve as dumps.

QCan you please tell us more about the special 'senses' or abilities that the ants use to mark and follow a trail?

The ants are extremely sensitive to their trail peromones. Even minute amounts of trail pheromone can guide foraging workers. One researcher quantified this sensitivity and estimated that the amount of trail peromone carried by a single worker in her permonone gland (which is not a particularly large gland) is enough to lay a trail at the equator around the circumference of the entire earth.

QHow far will leaf cutter ants go to find food? If food becomes scarce, would they move the colony?

Most trails are less than 100 meters long. But I have read reports of trails of about 600 meters long. Trail length varies between species, nest density (the more nests per area, the shorter the trails), and habitat (average distance to suitable food source). There have been a number of reports on colony movements; why the ants relocated in each of these cases is not know for certain. Flooding and devastation of fungal gardens by pathogens are probably the most likely causes. Relocating a nest is dangerous to the queen (she is exposed for several days while in transit; researchers have observed this), so, if there is a local food-shortage, the ants probably will opt to carry food for longer distances, rather than moving the nest closer to a food source.

QYou never see a sleeping ant. It seems like they are always working. Do the ants ever sleep?

Outside the nest, ants are moving (working) pretty much all the time; they venture outside to get work done, and, since it is more dangerous to be outside than inside, they are generally very efficient at what they have to do outside. In the nest, ants spend alot of time resting (not moving). Whether this counts as sleep, I would not be able to say. You would have to ask an ant-psychologist.


Scientific American Frontiers
Fall 1990 to Spring 2000
Sponsored by GTE Corporation,
now a part of Verizon Communications Inc.