Adventures among Ants. Mark W. Moffett
their mandibles are long, with pitchfork-like teeth only at their far ends, and they can open 180 degrees or more. In many cases, the jaws come equipped with trigger hairs. While the ants can be slow, their “bear-trap” jaws are not: the fastest muscular-driven action for any animal is achieved by the jaws of one group of these ants, Odontomachus.10 These speed-biters nab insects and also ply their mandibles as defensive tools, striking them against the ground when harassed; the resulting recoil sends them flying head over heels to safety. In Surinam, I’ve seen schoolchildren, betting over candy, make a game of encouraging the Odontomachus ants’ bouncing behavior while trying to avoid their searing stings.
Long jaws are great for catching prey but impractical at mealtime. Asian Myrmoteras, another group of creeping trapjaw ants that nest in any dark corner of the leaf litter, chew their prey from afar using the spiked tips of absurdly thin mandible blades that they can open an extraordinary 280 degrees. After chewing, they walk forward to place their mouths on the victim and feed at the oozing wound, then circle back to chew some more—the most awkward and labor-intensive approach to dining I have witnessed in all my travels.11
Acanthognathus have a partial solution to this logistical problem. While they use their long jaws to seize skittish springtails, they avoid the arduous dining experience of Myrmoteras by having a face like a Swiss army knife, with an entire arsenal of utensils at their disposal. To eat, they open their jaws wide, revealing a pair of what look like normal mandibles but are actually curved teeth, sprouting near the base of the longer bear-trap blades. The workers masticate their springtail meals to a pulp with these minijaws. As the small jaws are of a piece with the rest of the mandible, chewing with them sets the bear-trap blades waving to such a degree that feeding ants often knock over their neighbors.
Marauder ants have no elaborate built-in tools with which to seize springtails. Instead they must rely on commonplace, workaday mandibles (which have several small teeth along their forward margins, as do those of most ants). Furthermore, the marauder’s massive, frenetic societies are at the opposite extreme from those of the slow, stealthy Myrmoteras and Acanthognathus. The tempo of an ant species tends to relate to its colony size.12 Workers in small societies tend to be slow and cautious—a sensible way to approach elusive prey like the springtail on a low-energy budget. (Is the per capita energy quota of a small colony indeed likely to be smaller than that of a large one? Picturing a colony as a superorganism, a physiologist might predict that this would be the case. Since larger creatures are relatively efficient, burning fewer calories per unit of weight—or when measured microscopically, per cell—this gives them energy to spare.13 We can extrapolate that the same would be true for superorganisms, resulting, for example, in decreased labor demands for each individual in a large nest.14 If so, life must be precarious for Acanthognathus colonies—which, in my experience, are very rare, with no more than eighteen workers nesting in the rotted-out core of a single twig on the rainforest floor.)
How does the marauder ant, with its numbers and seeming chaos, nab the wily springtail? Lots of the ants seem to be doing the same thing at once, with sloppy overlap in their activities. But the effectiveness of large societies often has to do with redundancy rather than precision: although an individual ant may not be reliable, the density and overlapping actions of multiple ants ensure success for the raid. As each point on the ground is probed exhaustively, every critter, no matter how small, is rooted out.15 Once flushed, a springtail leaps about as one ant after another frightens it. Sooner or later, one of the minor workers will snare the springtail and make the kill. The raid, in its entirety, becomes the colony’s bear trap.
The effectiveness of this form of predation lies in exhausting the victim. Lions and wild dogs accomplish much the same thing. Although a solitary cheetah may have the edge on them in terms of speed, working as a pack the group predators can kill a gazelle that easily outruns them, wearing it down by chasing it sequentially, like relay runners, or by driving the animal toward an individual lying in wait. Marauder attacks aren’t as subtle or as calculated, but given the ants’ massive numbers, they may not need to be.
THOSE VORACIOUS OMNIVORES
The marauder ants’ predatory skills are only part of the picture. “The voraciousness of these ants is very great,” wrote a Vietnamese phytopathologist named Pham-tu-Thien in 1924. “We are dealing with a species whose greediness has fully developed its capacity for work.” Pham recorded marauder ants consuming insects, seeds, and fruit.16 What they take varies widely according to availability—they nibble on such oddities as leaves, flowers, bird droppings, and fungi when few other resources are available. But even when foods are bounteous, marauder ants tend to be wide-ranging gourmands.
Swarm-raiding army ants, often said to have among the Earth’s broadest diets, don’t compete with marauders in this regard. In particular, army ants are poor vegetarians, while marauders collect equal amounts of plant and animal material. Vegetable matter contains cellulose that many carnivores find indigestible. The only army ant approaching the marauder’s omnivorous diet is south Asia’s Dorylus orientalis, which, like the marauder, is considered an occasional agricultural pest—though it eats tubers such as potatoes, rather than the rice and other grains fancied by the marauder.17
The marauder ant species—Pheidologeton diversus—shows a proficiency at seed harvesting equal to that of many of its seed-harvesting relations in the group to which Pheidologeton belongs, the Myrmicinae, and I imagine the ancestor of Pheidologeton was like many of these relatives in eating seeds while scrounging for dead insects and perchance killing the occasional live one.18 On my Indian palm plantation, instead of taking their seeds straight to the nest as they did prey, the workers established caches along trails, carrying grain down holes or under leaves, where it was stored or milled to an edible flour by medias and majors. The ants also harvested an herb called goatweed by dropping its seeds to the ground, where workers of all sizes congregated to chop them up for immediate consumption.
Marauders are even more organized when they harvest grasses, one of their pastimes in the Singapore Botanic Gardens. When a raid passes a fruiting grass plant, only the minor workers and small medias can climb the slim stalk. The first minors gnaw the attached seeds ineffectually, but productivity skyrockets when a media arrives. The ants now set up a little assembly line, in which the media extracts one seed after another and then appears to hand it to a minor to haul away. What is really happening, however, is that the minor, who is too weak to pull a seed free from the stalk on her own, snatches the seed from the media before the larger ant can depart with it. The media dutifully plucks another seed, which another minor grabs. With minor workers so numerous, a media seldom has an opportunity to exit with her find.
Windfall fruit and vertebrate carcasses draw much larger crowds that defend and often consume them where they lie. Tens of thousands of workers will dismantle a mango or a dead bird. When I spilled a bag of canary food next to a trail, the ants arrived by the thousands to carry off 300 grams of seeds in eight hours, ten minutes. Under ordinary circumstances, the workers never seemed to become finicky or grow tired of a food, but this overfed colony refused over the next several days to touch any more of the seed.
An assembly line of marauder ants on a grass stalk in Singapore. A media worker extracts grass seeds, which the minor workers carry away.
The one food source that marauders forgo is another kind of bonanza, the populous nests of social insects. Tackling well-fortified bees, wasps, termites, and other ant species requires a convergence of forces to break through the foe’s weak points—a military tactic that marauders lack, though army ants display it in abundance.19 Indeed, almost all army ants gang-raid social insects routinely; many species especially relish the eggs, larvae, and pupae seized from colonies of their ant relations.
Marauder ants don’t just steer clear of social insect nests; they actively avoid making meals of them. When marauders kill another ant species in a skirmish, they cover the bodies with soil and abandon them. Despite this odd and unexplained aversion to cannibalism, the marauders evolved mass foraging in part for the same reason army ants did, as an