Figure 1. Two views of a small fungus comb

Macrotermes colonies host a remarkable symbiotic relationship with a basidiomycete fungus, Termitomyces. The termites cultivate the fungi in a fungus garden, an assemblage of structures built from chewed up grass and wood, and inoculated with fungal spores. Each year, these fungi produce a crop of large mushrooms (pictured at left), known locally as omajowa, which are highly prized as a delicacy.
Unlike the fungi cultivated by leaf-cutter ants, which the ant colony uses as food, the Termitomyces culture in a Macrotermes nest aids in the breakdown of cellulose and lignin into a more nutritious compost which serves as the termites actual food. The fungus garden is, therefore, a kind of extracorporeal digestive system, to which termites have 'outsourced' cellulose digestion.
The fungi also play a significant role in the emergence of social homeostasis in Macrotermes colonies. Indeed, in a remarkable way, it is not the termites that cultivate the fungi, but the fungi that are cultivating the termites.
To find out how, read on!
J S Turner. 2002. A superorganism's fuzzy boundary. Natural History 111 (6). July-August 2002: 62-67.

Read more about:

What the fungi do

The fungi are part of an extracorporeal digestive system that converts undigested woody material in plants into higher quality oligosaccharides and more easily digestible complex sugars. There is some nitrogen fixation that also takes place.

The fungi are grown in structures called fungus combs (Figure 1). Combs are made from macerated woody material, gathered by foraging workers, that is chewed up and swallowed. When the foragers return to the nest, they evacuate this material very quickly as pseudofeces, passing it on to nest workers which take this material and mold it into the fungus comb.

Somewhere along the way, perhaps in the digestive tracts of foragers or nest workers, this woody slurry is inoculated with a variety of fungal spores. Once deposited in the comb, the Termitomyces spores germinate and begin spreading hyphae through the comb. As these grow, they delignifiy and digest cellulose, converting it to simpler sugars and nitrogen. The termites then consume this enriched fodder for food. The structure of the combs is dynamic. Fresh material is continually added to the top, and digested material is consumed from the bottom. Food "flows through" the comb, just as silage flows through a silo.

A colony amasses a large number of fungus combs, gathered into a series of galleries atop the nest called a fungus garden. The collection of fungus combs in the photograph to the right (Figure 2) represents a small sample of a single colony's fungus garden.

Part of a fungus garden of an excavated mound (Figure 3. the nest has been removed). Each fungus comb is placed in a semi-enclosed space called a gallery. The total mass of fungus combs typically exceeds the colony's entire mass of termites by about eight fold - roughly 40 kg of fungus comb per colony.

Each year, the fungi produce mushrooms which eventually emerge from the base of the mound (fFigure 4). Prior to mushroom emergence, however, each fungus comb begins to sprout a number of stems that will penetrate through the hard soil of the mound, coalescing in a mushroom that literally erupts throught the mound surface.

Several sprouting fungus combs are evident in the fungus garden (Figure 3). A close-up of a sprouting fungus comb is shown in Figure 4. Below that (Figure 5) is a mushroom shortly after its eruption from the mound. The fracturing of the mound surface is clearly visible.

What triggers emergence of the mushrooms is unknown, but it typically follows the onset of heavy rains in late January - early February. Top

Mushrooms and moundlets

The mushrooms produced by Termitomyces may play a major role in mound morphogenesis and social homeostasis.

The connection lies in a common structural feature of Macrotermes mounds, small accessory mounds, or moundlets, that appear around the base (Figure 6).

The origin of these moundlets has always been mysterious. Recently, my student, Wendy Park, and Grace Shihepo, of the National Museum of Namibia (pictured in Figure 7), have discovered that the moundlets originate as repair work built over rotting mushrooms.

Shortly after emergence, the mushroom is invaded by numerous beetles and flies that lay eggs. Within a day or so, the mushroom has begin to rot.

As the mushroom rots, the rotting stem leaves a large tunnel for egress by the termites (Figure 8a). As with any other breach of the mound, the termites quickly begin to build new mound surface over the mushroom, eventually covering it completely (Figure 8). The covering continues to grow, until a moundlet is formed. Top

Moundlets, "boils" and mound growth

The mushroom-moundlet connection opens a door to a more general mechanism of mound growth and a better understanding of how the mound serves as an organ of social homeostasis.

Homeostasis in any form requires negative feedback of some sort - a response that opposes any perturbation of the system. To regulate the nest atmosphere, the mound's architecture must be part of a negative feedback control system. Yet, the termites' principal mechanism for mound building is stigmergy, which is, at root a positive feedback system - any perturbation, such as a deposition of a daub of soil, exaggerates the perturbation (more daubs of soil). Left to itself, stigmergy cannot produce the elegantly designed system of tunnels found within a typical termite mound. Stigmergy would drive the system to a solidly filled-in structure. We have observed this for termites engaged in repair activities within pipes.

What keeps this from happening is an ongoing disturbance to mound architecture that keeps the positive-feedback mechanism of stigmergy from ever playing out to completion.

This ongoing disturbance can arise in a number of ways. The colony's concentrated metabolism, for example, continually pushes CO2-laden air upwards, which induces a continual level of rebuilding there. As a consequence, there is ongoing remodeling of the mound which results in, among other things, the prominent spire that caps the mound.

More significant, though, might be the annual disturbance caused by erupting mushrooms.

A prominent feature of many termite mounds is structures we have named 'boils': areas of fulminating growth that erupt along the flanks of the mound's conical base (Figure 9). Boils cover a zone of intense rebuilding activity that has many curious features, including a large space just below the boil, reminiscent of a blister, that caps a wide tunnel that extends directly to the colony (Figure 10).

In light of the moundlet-mushroom connection, it seems clear that boils are actually moundlets that are growing along the flanks of the mound.

Thus, the fungi appear to be major agents of ongoing disturbance to mound architecture, which has, among other things, the activation of the ongoing remodeling that results in social homeostasis of the nest atmosphere. Top

Who's cultivating whom?

If fungi play an important role in shaping mound architecture, one must ask why they do this?

Fungi are normally enemies of termites, because they are normally parasites on the rich trove of food termites concentrate in their nests. Most termites have evolved defence mechanisms against these parasites, including various kinds of chemical fungistats.

A slow-growing fungus like Termitomyces poses little threat, however. Because it absorbs its digestate slowly, Termitomyces leaves ample 'leavings' that provide termites a rich food. Indeed, this is one reason why Macrotermes is such a prodigious 'ecosystem engineer' - its fungus-enriched food enables it to mobilize energy at faster rates than most termites.

However, Termitomyces' tepid growth also makes it no match for more aggressive fungal competitors, like the common cellulose-digesting fungus Xylaria. Termitomyces seems to use the Macrotermes colony as a sort of safe haven. Although fungus combs are laden with spores of more than two dozen different fungi, including Xylaria, it is only the Termitomyces spores that germinate and grow. When a comb is removed from the nest, these other spores germinate and grow, quickly overwhelming the Termitomyces culture. Something about being in the nest gives Termitomyces an edge over its more aggressive competitors.

Just what this edge is unknown. Certainly fungistatic agents play a role, as well as active 'weeding' of fungus combs by termites. This is not the whole cause, though - a fungus comb removed from the nest with a retinue of termites delays the takeover by Xylaria, but Xylaria wins in the end.

Some have suggested that Termitomyces' competitive edge is provided by some attribute of the nest environment, specifically, the high CO2 concentration there. Termitomyces is supposedly more tolerant of high CO2 concentrations and acidic environments than fungi like Xylaria. These are precisely the conditions that prevail within the nest. Furthermore, these conditions are reliably provided by the homeostatic tendencies of the termites.

Thus, the remarkable social physiology of Macrotermes colonies may actually arise from fungi cultivating termites, rather than the more conventional view that it is the termites doing the cultivating. By co-opting termites' homeostatic tendencies to provide a favorable environment for themselves, the Termitomyces' principal fungal competitors are effectively fended off.

Id quot circumiret, circumveniat! Top
(What goes around, comes around!)

Figure 4. A fungus comb, showing incipient mushroom stalks.

Figure 7. A collection of moundlets. (l) Grace Shihepo, (r) Wendy Park

Figure 5. A freshly erupted omajowa.

Figure 3. Part of the fungus garden in a Macrotermes nest

Figure 6. Two large moundlets and Little G (official project dog).

Figure 10. Mushroom emergence tunnel and moundlet formation. Mushrooms leave behind a tunnel which elicits intense building activity, forming a "boil." Inset shows a tunnel of a mushroom stalk.

Figure 8. Stages in moundlet formation.

Figure 9. A Macrotermes mound with 'boils' along its flank.

Figure 2. A sample of fungus combs from a single colony