The Nature Institute

The Nature Institute 
20 May Hill Road, Ghent, New York 12075  Tel: (518) 672 0116

Home | Our Education Programs | Our Publications | Content Areas | Writings Ordered by Author | Resources and Links | Contact Us | Search

In Context #20 (Fall 2008, pp. 3-5); copyright 2008 by The Nature Institute

Ants, Acacias and Herbivores
Craig Holdrege

acacia tree showing fruit and thorn
Acacia trees in Africa often provide a home and food for different specialized ants. In the case of the whistling thorn tree (Acacia drepanolobium), four different species of ants form colonies in the tree crowns. Usually one ant species inhabits any given individual tree to the exclusion of other ants. The most common of these ant species, Crematogaster mimosae (C. mimosae), raises its brood and houses its workers in the cavities formed at the swollen bases of modified thorns. The ants feed on the rich nectar that is produced by glands at the bottom of leaves. When large herbivores - such as giraffes, antelopes, or elephants - browse on the trees, the ants aggressively swarm out and bite. Whether the ants are actually “protecting” the tree, which is what we so easily imagine, is not so clear. Field observations show, in the case of giraffes, that to some extent the ants may be warding off young giraffes, but they have little effect on the browsing behavior of adult giraffes. At least we can say that these specialized ants are an integral part of the whistling thorn, a kind of active, animal extension of the plant. But what about the large herbivores that feed on whistling thorns? Do they just take from the trees, entering into a one-sided, exploitative relation with it?

Evidently it’s not that simple, as ecologist Todd Palmer, who studies plant and animal interactions in East Africa, discovered. He was in Kenya in an area where plots of acacia forests had been fenced in ten years before to exclude large herbivores. He noticed that the trees in the enclosures seemed less healthy than the ones outside that were fed on by giraffes and other large herbivores. What was going on? Weren’t the enclosures there to protect the trees? Does browsing somehow help the trees stay healthy? These questions led Palmer and his colleagues from different universities in research centers in Kenya, Canada, and the U.S. to carry out an intensive study comparing the ecology of whistling thorn trees in enclosures and those in open areas.

They confirmed Palmer’s initial impression that in the enclosed areas the whistling thorns were not thriving: the trees often grew more slowly and more of them died than in the unprotected areas. Surprisingly, the trees produced less nectar and fewer swollen thorns, an unexpected change that was associated with a shift in the ant species. While in unprotected whistling thorn stands, C. mimosae is the most prevalent of the four specialized ant species, after ten years of large herbivore exclusion in the fenced areas, its colonies inhabited fewer trees and the size of individual colonies shrank. Of all the ant specialists, C. mimosae is most reliant on nectar for food and the swollen thorns for housing. The colonies of this species that remained on the unbrowsed trees changed their feeding behavior - they began feeding on the sap provided by sap-sucking scale insects. Remarkable ecological plasticity, expressed through both the trees and the ants.

The predominant ant species on herbivore-excluded whistling thorns became C. sjostedti. In contrast to the other species of acacia-specializing ants, it feeds mainly on small invertebrates and does not nest in the swollen thorns but rather in stem cavities that have been made by the larvae of long-horned beetles. While other specialist ants prevent beetle infestation, C. sjostedti in some unknown way seems to attract or invite the beetles: when the ants were experimentally removed from whistling thorn trees, attack by long-horned beetles also decreased significantly. It is not clear why this should be. In any case, since trees inhabited by C. sjostedti were the ones that grew most slowly and showed highest death rates and these are the trees infested with these stem boring insects, the researchers believe that the insects may be a significant factor in reducing the vitality of the trees.

giraffe feeding on acacia tree
So what is the role of large herbivores that normally feed on whistling thorns? Evidently, their browsing stimulates directly or indirectly the development of swollen thorns and the production of leaf nectar, which provide the ideal environment for C. mimosae as the primary ant specialist colonizing whistling thorn. When this stimulation or interaction between mammalian browser and tree is lacking, everything changes. C. mimosae recedes and a new constellation of organisms and interactions develops. The ant species C. sjostedti increases, as does infestation by the stem-boring long-horned beetles, and the trees grow more slowly and are more likely to die.

All this provides a wonderfully concrete picture of how things in life are interconnected. The lives of tree, ant, beetle, and browsing mammal are all tightly interwoven. When one feature changes - in this case the removal of the large leaf-eating mammals - the dynamics of the previously-existing relations change and a new constellation of relations develops.

What’s additionally interesting is that it is not a matter of a clear-cut chain of causal connections. You can’t simply say that lack of browsing causes the trees to form less sap and make fewer swollen thorns, which in turn causes the one species of ants to leave, which in turn causes another species of ants to inhabit the tree, which in turn causes the long-horned beetle larvae to infest the tree, which infestation, finally, causes trees to grow slowly or die.

Nonetheless, our overriding habit of thought is to think in terms of causal chains, and that is what scientists are often after. We focus on the links between individual species and individual occurrences. Or perhaps we should say more correctly: we isolate such links out of the totality of the phenomena given and focus on them. An ecological phenomenon such as this one does not allow us to isolate any simple, one-directional relations. Any given aspect - the ants or the trees - shows itself to be related to an array of others. In fact, if we find something “simple,” we should be wary, for the simplicity is more likely to be an artifact of our narrow view than an expression of an actual cause-effect relation.

I remember learning in a college biology class that in science one chooses the simplest explanation for any given problem or phenomena. If we take ecological phenomena seriously, they teach us to look especially critically at simple explanations, and I think we could go as far as to say: expect at every turn greater complexity than you at first imagine; then you will be more likely to touch what’s going on. It wouldn’t be a bad idea for college ecology courses (actually, all biology courses - but that would probably be too revolutionary) to teach the maxim: if you find a simple explanation, don’t trust it, but search for complexity.

So we can step back from “causal chain” thinking and also move beyond the idea of a network (conceived as built “bottom up” from single causal links). As a step toward a different way of thinking about ecological constellations and how they change, it is helpful to ponder how a single organism responds to an alteration induced from outside. As Kurt Goldstein discovered in his careful observations of human behavior and bodily functions, “with any change in one locality in the organism simultaneous changes occur in other localities.” (Goldstein 1995, p. 173). In other words, the organism as a whole responds and shifts its functions and activities. Similarly, we can begin to look at ecological changes as organism-like shifts. We can discover the living quality of holistic response and flexibility that we find embodied in individual organisms.

The difficulty is that we have no bounded spatial “body” in ecology as we do in organismic biology. In organismic biology the bounded body gives us a fairly clear point of reference. We soon learn, however, that the boundaries of an organism are in fact interfaces and that an organism extends as far into the environment as it has relations with light, air, soil, or other organisms. In ecology we don’t have a body as a point of reference. Our focus is on the relations themselves and we begin to get a sense of the weaving of life in these relations. We can begin to paint a vivid and dynamic picture of these relations. There remain many questions in this picture, something that in fact contributes to its vitality. In contrast to a causal network, such a picture - as a kind of suggestive painting - can speak.


Goldstein, Kurt (1995). The Organism. New York: Zone Books.

Holdrege, Craig (2005). The Giraffe’s Long Neck: From Evolutionary Fable to Whole Organism, especially pp. 65-9. Ghent, NY: The Nature Institute.

Palmer, T.M. et al. (2008). “Breakdown of an Ant-Plant Mutualism Follows the Loss of Large Herbivores from an African Savanna,” Science vol. 319, pp. 191-5.

- Back to top
About Us | Become a Friend | Bookstore | Contact Us | Search | Calendar of Events | Our Education Programs | Our Publications | Content Areas | Writings Ordered by Author | Resources and Links | Home