Amazon Ants by Adrienne Sussman
The Rainforest’s Micro-Engineers Copyright BBC Natural History Unit
^ Another angle of a M. schumanni worker ant attacking a foreign tree by injecting formic acid into a leaf.
I
magine walking through the Amazon rainforest. Trees surround you, blocking out the sky, plants and insects swarm on every surface, and unseen animals rustle through the canopy above. Suddenly, you step into a clearing. The area is bare except for the evenly spaced trunks of a single species of tree, Duroia hirsuta. No other vegetation is in sight, and there are few signs of animal life. Above the stunted branches, you can see the sky. As you walk into one of these “devil’s gardens”, you might feel you have entered a supernatural place. Local legends attribute the groves to an evil forest spirit. Ecologists, in search of a natural explanation, hypothesized that the phenomenon was related to the tropical ant Myrmelachista schumanni. Recent work by Megan Frederickson, a graduate student in the lab of Stanford Biology Professor Deborah Gordon, has finally demonstrated that ants are indeed the cause. Her re-
The Amazon Rainforest
www.gsfc.nasa.gov/goddardnews/20020712/lba.html
V
24 Stanford Scientific
search on the construction of devil’s gardens reveals ants’ importance in the ecology of the rainforest.
Into the Devil’s Garden Frederickson began her study of devil’s gardens four years ago at the Madre Selva Biological Station in western Peru. She found striking patches of forest, some consisting of more than 500 D. hirsuta trees. Noting the swarms of ants in the groves, Frederickson formulated an initial hypothesis: the ants somehow kill all plants other than D. hirsuta. To test her conjecture, she planted two Spanish cedar saplings in ten devil’s gardens she had picked as study sites. One of these saplings was treated with a sticky insect barrier to protect it from ants. She also planted two saplings outside of each devil’s garden as a control group. Again, one of these trees was protected from insects, while the other was not. The results of the experiment were unmistakable: the sapling without protection from ants and inside the devil’s garden turned brown and died within days, while the treated sapling and both control trees survived. How did the ants kill the sapling? Frederickson observed that the ants swarmed up the untreated cedar’s trunk, grasped the leaves with their mandibles, and injected a substance into the leaf. The leaves subsequently turned brown and fell off, and the body of the tree soon died as well. She suspected that the substance was formic acid—other species of ants in M.
Photographed by Megan Frederickson
^ A M. schumanni worker ant injecting formic acid into a foreign tree.
^ A Devil’s Garden, a mysterious clearing amidst the lush rainforest.
schumanni’s family use formic acid for hunting or protection. A chemical analysis of the ants’ poison glands confirmed this hypothesis. Another test verified the lethal effects of formic acid, which spreads through the plant’s circulatory system and poisons it. Next, Frederickson addressed the question of how the ants distinguished foreign plants from D. hirsuta. As the ants nest inside the stalks of D. hirsuta, she predicted that they might make their decision based on the presence of these hollow stems, called domatia. Frederickson’s next experiment was to create artificial domatia in a foreign tree species by placing foil-wrapped test tubes partially filled with cotton inside the stem. If the ants were looking for good nesting sites, these artificial domatia would be attractive. For this experiment, she planted two foreign saplings at each test site: one untreated, and one with artificial domatia. She also planted two D. hirsuta at each site: again, one sapling for control and one sapling with an artificially implanted nesting space.
Interestingly, the experiment indicated that the presence of domatia had no effect on the ants’ behavior: at all sites, both of the foreign trees were killed while both of the D. hirsuta were spared. These results suggest that the ants use chemical markers or some other means to identify which trees to attack and which trees to cultivate.
The ant colonies are ‘as old as many of the oldest trees in the rainforest.’
How Much Control Do Ants Have Over the Devil’s Garden? M. schumanni’s attempts at engineering its environment are remarkably successful. During the study, Frederickson measured the growth rate of 26 different devil’s gardens and used this factor to estimate each colony’s age. The findings were impressive: one of the colonies seems to be over 800 years old. According to Frederickson, this suggests that the ant colonies are “as old as many of the oldest trees in the rainforest.” “The system,” she says, “is remarkably stable.” The findings indicate that ants have an astonishing degree of control over their environment. Interestingly, ants are actually one of the most dominant animals in the tropical rainforest, making up an estimated 10 percent of the total rainforest biomass. Vertebrate species, meanwhile, make up only 3 percent of the total. Ecologically, ants are a sustainable force in shaping the structure of rainforests.
Layout designed by Fah Sathirapongsasuti
Volume IV 25
Amazon Ants
Adrienne Sussman is a junior majoring in Biological Sciences with a minor in Spanish Language. She is interested in neurobiology, coffee, and travel.
26 Stanford Scientific
^ The red wood ant drags a sawfly larva back to its nest; it has killed the larva using formic acid.
http://www.wcrl.ars.usda.gov/cec/photo/d13.htm
Frederickson’s results have received attention in the scientific community not only because she solved the mystery of the devil’s gardens, but also because she found the first example of ants using formic acid to farm plants on such a large scale. Like many tropical organisms, D. hirsuta and M. schumanni seem to have carved out a very specific ecological relationship that is mutually beneficial. “This case is really amazing because the ant indirectly helps the plant by killing off the competition,” explains Prof. Gordon. Yet the mutualism is not as straightforward as it first appears. Many other rainforest plants would make good nests for the ants. Anecdotal evidence suggests that M. ^ A common brown ant mag- schumanni ants in southern Peru and nified x150 using a scanning Ecuador may have different host prefelectron microscope erences, and create devil’s gardens of different types of trees. If this is true, then the ants’ adaptive preference for a single tree species warrants closer research. Fredrickson’s next study may resolve this mystery—and others. She plans to investigate how ants chemically recognize D. hirsuta, how they kill foreign plants in their gardens, and whether there are limits to the size of the plants they can exterminate. Meanwhile, other members of her research team are analyzing insect diversity in devil’s gardens to test whether the insect population of the gardens differs from that of the rest of the rainforest. If there are significant differences, the group may conduct further experiments to determine whether the ants are attacking invading insects as well as foreign plants. Either way, relationships such as the one between M. schumanni and D. hirsuta provide a good opportunity to study the evolution of symbiotic relationships in nature. As Frederickson put it, “we do need to understand how they work if we want to understand all kinds of things about tropical rainforests: why they are so biodiverse, their role in all kinds of things like carbon storage and other global ecosystem processes, and if we want to make a case for people not to cut them down.” S
^ Again, red wood ants attacks two Cetonia cuprea beetles, using formic acid.
http://www.srs.fs.usda.gov/4501/townants.htm
Further Uncovering the Role of Rainforest Ants
http://www.ucrl.ars.usda.gov/cec/photo/formica2.htm
The age of the devil’s gardens also indicates a social system unusual for ants; such a large and durable colony must have multiple queens at any one time, each of whom is replaced as she dies. Since only one other species of ant in the world is known to have such long-lasting colonies, the longevity of devil’s gardens is an attractive subject for future research. Frederickson also wants to research the extent of the ants’ control over a devil’s garden: specifically, how large the gardens can grow. While the largest devil’s garden in Frederickson’s study was more than 1300 square meters, there are rumors of gardens as large as football fields within the forest. Such an enormous ant colony might provide an interesting case study for population biologists interested in probing the limits of the colony’s expansion. Frederickson anticipates that factors such as disease, parasitism, and aggression with other ant species may eventually stop colony growth, or that after a certain point, the ants may simply allocate more resources towards reproduction than towards development.
^ A town ant bites a human, in the same manner as M. schumanni attack tree leaves.